html_url
stringlengths 48
51
| title
stringlengths 5
268
| comments
stringlengths 70
51.8k
| body
stringlengths 0
29.8k
| comment_length
int64 16
1.52k
| text
stringlengths 164
54.1k
| embeddings
sequence |
|---|---|---|---|---|---|---|
https://github.com/huggingface/datasets/issues/1981 | wmt datasets fail to load | I'll do a patch release for this issue early tomorrow.
And yes we absolutly need tests for the wmt datasets: The missing tests for wmt are an artifact from the early development of the lib but now we have tools to generate automatically the dummy data used for tests :) | on master:
```
python -c 'from datasets import load_dataset; load_dataset("wmt14", "de-en")'
Downloading and preparing dataset wmt14/de-en (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /home/stas/.cache/huggingface/datasets/wmt14/de-en/1.0.0/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e...
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 578, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 634, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt14/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e/wmt_utils.py", line 760, in _split_generators
extraction_map = dict(downloaded_files, **manual_files)
```
it worked fine recently. same problem if I try wmt16.
git bisect points to this commit from Feb 25 as the culprit https://github.com/huggingface/datasets/commit/792f1d9bb1c5361908f73e2ef7f0181b2be409fa
@albertvillanova | 50 | wmt datasets fail to load
on master:
```
python -c 'from datasets import load_dataset; load_dataset("wmt14", "de-en")'
Downloading and preparing dataset wmt14/de-en (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /home/stas/.cache/huggingface/datasets/wmt14/de-en/1.0.0/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e...
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 578, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 634, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt14/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e/wmt_utils.py", line 760, in _split_generators
extraction_map = dict(downloaded_files, **manual_files)
```
it worked fine recently. same problem if I try wmt16.
git bisect points to this commit from Feb 25 as the culprit https://github.com/huggingface/datasets/commit/792f1d9bb1c5361908f73e2ef7f0181b2be409fa
@albertvillanova
I'll do a patch release for this issue early tomorrow.
And yes we absolutly need tests for the wmt datasets: The missing tests for wmt are an artifact from the early development of the lib but now we have tools to generate automatically the dummy data used for tests :) | [
-0.3290721774,
-0.0584941357,
-0.0104082599,
0.5395735502,
0.3104315698,
0.0041043535,
0.20977965,
0.0872704089,
0.3129349947,
0.1078864411,
-0.0224996582,
-0.1256613433,
-0.2969317734,
0.1887944937,
0.1114165783,
0.1848196089,
-0.1364243478,
0.0058543496,
-0.8216048479,
0.0158688203,
-0.1838494837,
0.1984839439,
-0.167342037,
-0.0969724581,
-0.5068029165,
0.2375525087,
-0.0300085619,
0.2536429167,
-0.2113710195,
-0.4816322923,
0.5141909122,
0.0309754163,
0.0518512577,
0.5576845407,
-0.0001141044,
0.151525557,
0.2935244441,
-0.0102459118,
-0.2466028631,
-0.2974120677,
-0.5779015422,
-0.3827318549,
0.0219601616,
0.2245495617,
-0.1884618104,
0.0820964426,
-0.240066871,
-0.3640689552,
0.2820507288,
0.2943234742,
0.2214949876,
0.415157944,
0.2789460719,
-0.309194088,
0.0510979444,
0.2907715142,
-0.1379879266,
-0.0280021429,
0.0135745909,
-0.1352174878,
0.0232768059,
0.0316455141,
0.0733230338,
0.3297719657,
0.3457939029,
-0.1587233692,
0.2857410014,
-0.0081028454,
0.2792252898,
0.0577791333,
0.196494475,
-0.0757790208,
-0.2360398173,
-0.0594308451,
-0.1306953132,
-0.4276535511,
0.2592033744,
0.2658744454,
0.0303440131,
0.2660059631,
-0.3589413464,
0.0739578009,
0.2212082744,
0.0317870677,
-0.3881678283,
0.377141118,
-0.311494112,
0.0018199123,
0.2749302089,
-0.1902685463,
0.1660660654,
-0.1114595681,
-0.0187489055,
0.0837207213,
-0.5072566271,
0.2616918683,
0.0255594216,
0.2756205201,
-0.080057241,
0.158236891,
-0.2520164549,
0.0149280876,
0.2089378983,
-0.0155651644,
0.1787930727,
0.1386242807,
-0.1099446937,
-0.0626591966,
0.2749222219,
0.0550981387,
-0.1853418052,
0.1351206601,
-0.1703791618,
-0.4433654249,
-0.0891755894,
-0.1894305944,
0.1525506973,
-0.0720275789,
-0.259921968,
0.0635320395,
0.1355188936,
-0.3142133057,
0.0204191059,
0.2855195999,
0.0484090075,
0.1702363491,
0.1179608256,
0.0036532003,
-0.0042029768,
-0.0357705802,
-0.356447041,
-0.3906193972,
-0.2991642356,
0.0611202568,
0.4078394175,
-0.218572557,
0.228302747,
0.0004106164,
-0.1310046017,
-0.1273209155,
-0.2269294858,
-0.0077419803,
0.218893528,
0.4824033678,
0.0710929483,
0.2990788817,
0.244366318,
0.1595918089,
-0.0081310496,
-0.0074382517,
-0.099539116,
-0.2956004739,
0.129521057,
0.2581729889,
-0.0815231055,
0.1570642591,
-0.2294502705,
-0.1136632562,
0.1050200462,
0.3215994835,
-0.1928714961,
-0.2417547405,
-0.1637252569,
-0.066638723,
0.5321370959,
0.5977163911,
0.0095657185,
-0.1942434609,
-0.3947505057,
-0.0471857563,
0.2344344556,
0.3322233558,
-0.0298237316,
0.1007022113,
-0.2609951198,
-0.2470604479,
-0.0820464492,
-0.0594501421,
-0.183284685,
0.1371449828,
-0.1313238442,
0.2751172185,
0.0324520469,
-0.1263801455,
0.0145311803,
-0.2265356332,
0.3428908288,
0.0891364068,
-0.1580776423,
0.0836277902,
-0.0794163719,
-0.3164909482,
0.2474279106,
0.1886272281,
0.1936812997,
-0.1896170974,
0.2157250792,
0.0242884457,
0.0684090033,
0.0570892282,
0.1292504668,
0.0385189466,
0.0848583654,
-0.012778338,
-0.0889182165,
0.0534031913,
-0.447301656,
0.2964881361,
0.269192636,
0.0437410511,
0.1173467636,
0.1208165959,
-0.1622186601,
0.13357687,
-0.3434742987,
-0.1887685806,
0.0876791999,
0.1349028051,
-0.0035157129,
-0.1450344622,
-0.0388086475,
-0.0239786468,
-0.2259171605,
0.1487138569,
0.2445332706,
0.0667201504,
-0.1433454007,
-0.1252084225,
0.2802848816,
0.2743450105,
0.1635606736,
-0.1697694361,
-0.1630631387,
0.0908573046,
0.1399755478,
0.3951956034,
-0.0871864706,
0.1039212644,
0.1791914403,
-0.4736975431,
0.1699436903,
0.1361609995,
-0.1402691752,
-0.2992375493,
0.120428741,
0.0424620919,
-0.1103157252,
0.2675850391,
0.1175594181,
0.0357344747,
0.1648181081,
0.070990026,
0.1453139335,
-0.3879216909,
0.2801464796,
-0.0923160538,
0.3428555727,
0.2670230269,
0.0239536017,
-0.0425493941,
0.3805397749,
0.1634019911,
0.2022345662,
0.0150982998,
-0.3856003881,
-0.0856612027,
-0.0661869422,
0.0501855165,
0.3967727721,
0.0735270008,
0.0935864449,
0.1786359251,
0.114971295,
-0.3025265336,
0.4467021823,
-0.0749784783,
-0.0767069533,
0.3043871224,
-0.0235594586,
-0.0722774118,
-0.2620557547,
0.5669638515,
0.0582102686,
0.0062173288,
-0.1975154579,
-0.2630079687,
-0.4342723489,
0.235784024,
-0.350056529,
-0.5691423416,
-0.205263555,
-0.2057436854,
-0.2758672237,
0.1181361526,
-0.0988535583,
0.2478711605,
0.0687461495,
0.2188890874,
-0.0490783267,
0.3465975523,
-0.0977027416,
-0.2259618938,
-0.1064975113,
0.0391055197,
0.531206429,
-0.2995471656,
0.3275354505,
-0.1957136095,
0.0639886782,
-0.2002594918,
-0.2282969803,
0.0872323737,
-0.0478701815,
0.3465406597,
0.2117685527,
-0.0232824236,
0.364433527,
-0.1089352071,
0.3259529173,
-0.1572976112,
-0.0153348558,
0.0034700036,
-0.0445439406,
-0.142670542,
-0.2228536606,
-0.3446230292,
-0.4408051372,
-0.38834548,
-0.0106526744,
-0.0261883438,
-0.041893553,
0.0917574763,
-0.1342886686,
-0.0532306172,
-0.0178117566,
-0.0258909091,
-0.3486142755,
-0.2984478474,
0.2000669688,
-0.3020643294,
-0.2819241881,
0.2392929345,
0.1354366541,
0.4544391036,
0.0401276685,
-0.3937389553,
0.580820024,
-0.1391109526,
-0.1142628118,
0.1804574281,
0.3090637922,
-0.0905659199,
0.1654050052,
-0.0495855138,
-0.1657475233,
-0.264600873,
-0.1806015968,
-0.2320662737,
0.1591045111,
-0.1456700712,
0.0627860576,
0.0104182586,
0.8047190309,
0.0654782131,
-0.2943533063,
0.1827456653,
0.1267128289,
0.22185269,
-0.0006961301,
-0.5895357728,
0.2010112554,
-0.0986171663,
0.2215834707,
0.3626621366,
0.0432907455,
0.1737743169,
-0.1139222234,
0.1939598769,
-0.1887793094,
-0.3513717055,
-0.1281633973,
0.2005108893,
0.384968102,
0.0900530443,
0.1219890267,
0.0273914561,
-0.0954924524,
-0.0095352717,
0.3547010124,
0.1222474873,
0.0885327458,
-0.1050639749,
-0.0560129024,
-0.2369753569,
0.0535893925,
-0.1932053417,
0.4575623572,
-0.0324506685,
-0.1230477318,
-0.0423157625,
-0.018290678,
0.5026881099,
0.196738705,
0.392770201,
0.1022482887,
-0.4248225689,
-0.2392999828,
-0.1275756508,
0.0902224034,
-0.0735769495,
0.1285207719,
0.1846205294,
-0.2020289898,
-0.5072083473,
0.2025788128,
0.110791266,
-0.1888124496,
-0.1198265851,
-0.1453298032,
0.0249493048,
-0.2902212739,
-0.0777402893,
-0.2248148024,
0.1797914803,
-0.4049386382,
-0.023972854,
-0.1545184255,
-0.0788380727,
0.1511588097,
0.0657216012,
0.0934650078,
0.0442998596,
0.4923816025,
0.3122971356,
0.3200490177,
0.5650812387,
0.615952909,
-0.1010635719,
0.0344185531,
0.0835056677,
0.0163797066,
0.052282501,
0.2831603289,
-0.0781764388,
0.1111001968,
0.0438617133,
0.1107479557,
-0.5960832238,
0.1240012124,
0.2019079924,
0.0680201054,
-0.357242465,
-0.0997589529,
0.1946074665,
0.0268734321,
0.2006981224,
0.1760098636,
0.1702614427,
-0.3972955346,
-0.4666161835,
0.0168192871,
0.494731009,
-0.0735877305,
0.2222425938,
-0.1423189342,
-0.1103766337,
0.0566480383,
0.2778766751,
-0.1172653213,
-0.1914952397,
0.1790857166,
0.0319306925,
-0.1585558057,
0.0860942155,
0.1152570099,
-0.074036397,
0.2761102319,
0.0473108888,
0.5473350286,
0.1399029195,
0.0827613249,
-0.1333880872,
-0.2570033967,
-0.2534720302,
0.1469008625,
-0.040101774,
0.3355660141,
-0.2314385772,
-0.1019250154,
0.3119747043,
-0.0665351152,
-0.2452949733,
0.006580241,
-0.6495986581,
0.1033414826,
-0.0357615538,
-0.1031016409,
0.3692238927,
-0.2313648611,
0.1811163127,
0.3199765384,
-0.1902631074,
0.101099968,
-0.3137950599,
0.0383314565,
-0.2302719355,
0.2646709085,
0.1008015499,
-0.0336168185,
-0.0603650734,
-0.1086391211,
-0.040744178,
-0.0539291054,
0.0338524282,
0.0507716499,
-0.0552918538,
-0.049335558,
-0.3194902837,
0.0118200779,
0.01315099,
-0.1511109769,
0.1543609649,
0.0290808994,
-0.0549282245,
0.3291850686,
-0.0549550839,
-0.1067547947,
-0.0585500076,
0.4145898521,
-0.2623614073,
0.0229861513,
0.3038800061,
0.4151065946,
-0.0175478719,
-0.2813272774,
-0.0685153008,
0.0908067673,
-0.2448109835,
0.27016747,
0.1164213791,
-0.0899816453,
0.3801681995,
0.561144352,
0.3422603309,
-0.1498251259,
0.0122462288,
-0.2868362665,
-0.1911177933,
0.06081523,
-0.1435581297,
0.1625946313,
0.0277143493,
0.4404928684,
-0.174705416,
0.3926444352,
-0.3059358299,
0.2749379277,
-0.0947568417,
-0.0000579925,
0.2123729736,
-0.1585533768,
0.1591585428,
-0.092134729,
0.0412249677,
-0.0340268984,
-0.5342018604,
-0.2137607932,
-0.1178592443,
0.094538033,
0.0980823785,
-0.2711147368,
0.1690744013,
-0.1080356762,
0.0807956457,
-0.1697111577,
0.3470470011,
0.0660693496,
-0.1222603619,
0.0229018033,
0.0703008622,
0.0658249557,
-0.1689614058,
0.1390581727,
0.0714532286,
-0.0677271932,
-0.004992824,
-0.0410777628,
0.0441779867,
0.0208380446,
0.1150303781,
0.0102870464,
-0.0216588713,
0.1182219386,
0.5093998313,
-0.3056087196,
-0.2525116503,
0.2638111115,
0.2856666148,
0.1686179787,
0.0513615236,
0.0614273101,
-0.1690220535,
0.1712699234,
-0.295384407,
-0.0795395225,
-0.0377572887,
0.1637817621,
0.0857126564,
0.0499727502,
0.3570747375,
-0.2151892185,
-0.2337025553,
0.3549884558,
0.3251186609,
-0.1116922349,
0.1748536229,
0.4612627625,
0.0847785696,
0.2086673677,
0.1075343117,
-0.1653206944,
0.10477487,
0.3765082955,
-0.2396265268,
0.6803712249,
0.0646546483,
0.0063732327,
0.2407553494,
-0.4143283069,
0.1107710898,
0.2690974474,
-0.1568294764,
0.0244205967,
-0.2630803585,
0.1377332509,
-0.3402011395,
-0.2731532454,
-0.1855367422,
0.1984927356,
-0.0655511543,
-0.1254134029,
-0.2556758821,
-0.1166858748,
-0.0780744553,
0.0273623541,
-0.0752226561,
-0.1022675335,
0.2595927715,
0.1732608676,
-0.1564743519,
-0.4046474099,
-0.3694781065,
0.0161442123,
-0.0983204991,
-0.1228213087,
0.2560083866,
0.4048366845,
-0.2123370916,
0.1384947449,
0.4051621854,
0.4715678096,
0.3013880551,
-0.019286247,
-0.0860341489,
0.0871977657,
-0.0851399675,
-0.2775003314,
0.4197261631,
-0.256807059,
-0.1573418081,
0.0049491078,
0.1699633002,
-0.1862419993,
-0.36442554,
0.1278568655,
-0.312604636,
-0.2835619748,
0.1925960928,
-0.0886968076,
-0.0576488152,
-0.046404805,
0.3801314831,
-0.0463572331,
0.227106601,
0.6677178144,
-0.0275352802,
0.1280926317,
-0.2423587441,
0.1004885435,
0.0852620155,
0.7199052572,
0.2930562198,
-0.0817145929,
-0.2865350842,
-0.5099943876,
-0.7815914154,
0.0482604392,
0.1542910635,
0.4362874031,
0.0736527219,
0.0439882576,
0.1192258596,
-0.022530023,
0.1537166834,
0.1220185757,
0.2925760448,
-0.2151335776,
-0.3548650146,
0.1796961129,
-0.0950151533,
-0.1906720847,
-0.0053810701,
-0.3220207691,
0.1466880143,
-0.2788922787,
0.0247031078,
-0.2417233735,
0.423589766,
0.0029747784,
0.2875815034,
0.2188121825,
0.0880904496,
0.3063713312,
0.0107985735,
-0.0630578622,
-0.4197328091,
-0.311278224,
0.0145377479,
0.0860738903,
-0.1074681431,
0.3646842241,
-0.0537011325,
-0.2685218155,
-0.4432089329,
0.1050859094,
0.1143686622,
0.2023361176,
-0.1769938469,
0.2222105861,
-0.2147685289,
-0.1733142138,
0.3443752229,
0.4282422364,
-0.0292371623,
0.1878371388,
-0.2106765211,
-0.3873320818,
0.2546340823,
-0.2887779176,
-0.1629849672,
-0.4629842937,
0.1772361249,
-0.0954382569,
-0.1698629558,
-0.4217800498,
0.1528435349,
0.1712167561,
-0.1016814485,
-0.2147470117,
0.372612536,
-0.046054516,
-0.0684142113,
-0.0185814574,
0.1694672108,
0.1803844571,
-0.3343187273,
-0.1139620915,
-0.0598332062
] |
https://github.com/huggingface/datasets/issues/1981 | wmt datasets fail to load | still facing the same issue or similar:
from datasets import load_dataset
wtm14_test = load_dataset('wmt14',"de-en",cache_dir='./datasets')
~.cache\huggingface\modules\datasets_modules\datasets\wmt14\43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e\wmt_utils.py in _split_generators(self, dl_manager)
758 # Extract manually downloaded files.
759 manual_files = dl_manager.extract(manual_paths_dict)
--> 760 extraction_map = dict(downloaded_files, **manual_files)
761
762 for language in self.config.language_pair:
TypeError: type object argument after ** must be a mapping, not list | on master:
```
python -c 'from datasets import load_dataset; load_dataset("wmt14", "de-en")'
Downloading and preparing dataset wmt14/de-en (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /home/stas/.cache/huggingface/datasets/wmt14/de-en/1.0.0/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e...
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 578, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 634, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt14/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e/wmt_utils.py", line 760, in _split_generators
extraction_map = dict(downloaded_files, **manual_files)
```
it worked fine recently. same problem if I try wmt16.
git bisect points to this commit from Feb 25 as the culprit https://github.com/huggingface/datasets/commit/792f1d9bb1c5361908f73e2ef7f0181b2be409fa
@albertvillanova | 52 | wmt datasets fail to load
on master:
```
python -c 'from datasets import load_dataset; load_dataset("wmt14", "de-en")'
Downloading and preparing dataset wmt14/de-en (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /home/stas/.cache/huggingface/datasets/wmt14/de-en/1.0.0/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e...
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 578, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 634, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt14/43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e/wmt_utils.py", line 760, in _split_generators
extraction_map = dict(downloaded_files, **manual_files)
```
it worked fine recently. same problem if I try wmt16.
git bisect points to this commit from Feb 25 as the culprit https://github.com/huggingface/datasets/commit/792f1d9bb1c5361908f73e2ef7f0181b2be409fa
@albertvillanova
still facing the same issue or similar:
from datasets import load_dataset
wtm14_test = load_dataset('wmt14',"de-en",cache_dir='./datasets')
~.cache\huggingface\modules\datasets_modules\datasets\wmt14\43e717d978d2261502b0194999583acb874ba73b0f4aed0ada2889d1bb00f36e\wmt_utils.py in _split_generators(self, dl_manager)
758 # Extract manually downloaded files.
759 manual_files = dl_manager.extract(manual_paths_dict)
--> 760 extraction_map = dict(downloaded_files, **manual_files)
761
762 for language in self.config.language_pair:
TypeError: type object argument after ** must be a mapping, not list | [
-0.3290721774,
-0.0584941357,
-0.0104082599,
0.5395735502,
0.3104315698,
0.0041043535,
0.20977965,
0.0872704089,
0.3129349947,
0.1078864411,
-0.0224996582,
-0.1256613433,
-0.2969317734,
0.1887944937,
0.1114165783,
0.1848196089,
-0.1364243478,
0.0058543496,
-0.8216048479,
0.0158688203,
-0.1838494837,
0.1984839439,
-0.167342037,
-0.0969724581,
-0.5068029165,
0.2375525087,
-0.0300085619,
0.2536429167,
-0.2113710195,
-0.4816322923,
0.5141909122,
0.0309754163,
0.0518512577,
0.5576845407,
-0.0001141044,
0.151525557,
0.2935244441,
-0.0102459118,
-0.2466028631,
-0.2974120677,
-0.5779015422,
-0.3827318549,
0.0219601616,
0.2245495617,
-0.1884618104,
0.0820964426,
-0.240066871,
-0.3640689552,
0.2820507288,
0.2943234742,
0.2214949876,
0.415157944,
0.2789460719,
-0.309194088,
0.0510979444,
0.2907715142,
-0.1379879266,
-0.0280021429,
0.0135745909,
-0.1352174878,
0.0232768059,
0.0316455141,
0.0733230338,
0.3297719657,
0.3457939029,
-0.1587233692,
0.2857410014,
-0.0081028454,
0.2792252898,
0.0577791333,
0.196494475,
-0.0757790208,
-0.2360398173,
-0.0594308451,
-0.1306953132,
-0.4276535511,
0.2592033744,
0.2658744454,
0.0303440131,
0.2660059631,
-0.3589413464,
0.0739578009,
0.2212082744,
0.0317870677,
-0.3881678283,
0.377141118,
-0.311494112,
0.0018199123,
0.2749302089,
-0.1902685463,
0.1660660654,
-0.1114595681,
-0.0187489055,
0.0837207213,
-0.5072566271,
0.2616918683,
0.0255594216,
0.2756205201,
-0.080057241,
0.158236891,
-0.2520164549,
0.0149280876,
0.2089378983,
-0.0155651644,
0.1787930727,
0.1386242807,
-0.1099446937,
-0.0626591966,
0.2749222219,
0.0550981387,
-0.1853418052,
0.1351206601,
-0.1703791618,
-0.4433654249,
-0.0891755894,
-0.1894305944,
0.1525506973,
-0.0720275789,
-0.259921968,
0.0635320395,
0.1355188936,
-0.3142133057,
0.0204191059,
0.2855195999,
0.0484090075,
0.1702363491,
0.1179608256,
0.0036532003,
-0.0042029768,
-0.0357705802,
-0.356447041,
-0.3906193972,
-0.2991642356,
0.0611202568,
0.4078394175,
-0.218572557,
0.228302747,
0.0004106164,
-0.1310046017,
-0.1273209155,
-0.2269294858,
-0.0077419803,
0.218893528,
0.4824033678,
0.0710929483,
0.2990788817,
0.244366318,
0.1595918089,
-0.0081310496,
-0.0074382517,
-0.099539116,
-0.2956004739,
0.129521057,
0.2581729889,
-0.0815231055,
0.1570642591,
-0.2294502705,
-0.1136632562,
0.1050200462,
0.3215994835,
-0.1928714961,
-0.2417547405,
-0.1637252569,
-0.066638723,
0.5321370959,
0.5977163911,
0.0095657185,
-0.1942434609,
-0.3947505057,
-0.0471857563,
0.2344344556,
0.3322233558,
-0.0298237316,
0.1007022113,
-0.2609951198,
-0.2470604479,
-0.0820464492,
-0.0594501421,
-0.183284685,
0.1371449828,
-0.1313238442,
0.2751172185,
0.0324520469,
-0.1263801455,
0.0145311803,
-0.2265356332,
0.3428908288,
0.0891364068,
-0.1580776423,
0.0836277902,
-0.0794163719,
-0.3164909482,
0.2474279106,
0.1886272281,
0.1936812997,
-0.1896170974,
0.2157250792,
0.0242884457,
0.0684090033,
0.0570892282,
0.1292504668,
0.0385189466,
0.0848583654,
-0.012778338,
-0.0889182165,
0.0534031913,
-0.447301656,
0.2964881361,
0.269192636,
0.0437410511,
0.1173467636,
0.1208165959,
-0.1622186601,
0.13357687,
-0.3434742987,
-0.1887685806,
0.0876791999,
0.1349028051,
-0.0035157129,
-0.1450344622,
-0.0388086475,
-0.0239786468,
-0.2259171605,
0.1487138569,
0.2445332706,
0.0667201504,
-0.1433454007,
-0.1252084225,
0.2802848816,
0.2743450105,
0.1635606736,
-0.1697694361,
-0.1630631387,
0.0908573046,
0.1399755478,
0.3951956034,
-0.0871864706,
0.1039212644,
0.1791914403,
-0.4736975431,
0.1699436903,
0.1361609995,
-0.1402691752,
-0.2992375493,
0.120428741,
0.0424620919,
-0.1103157252,
0.2675850391,
0.1175594181,
0.0357344747,
0.1648181081,
0.070990026,
0.1453139335,
-0.3879216909,
0.2801464796,
-0.0923160538,
0.3428555727,
0.2670230269,
0.0239536017,
-0.0425493941,
0.3805397749,
0.1634019911,
0.2022345662,
0.0150982998,
-0.3856003881,
-0.0856612027,
-0.0661869422,
0.0501855165,
0.3967727721,
0.0735270008,
0.0935864449,
0.1786359251,
0.114971295,
-0.3025265336,
0.4467021823,
-0.0749784783,
-0.0767069533,
0.3043871224,
-0.0235594586,
-0.0722774118,
-0.2620557547,
0.5669638515,
0.0582102686,
0.0062173288,
-0.1975154579,
-0.2630079687,
-0.4342723489,
0.235784024,
-0.350056529,
-0.5691423416,
-0.205263555,
-0.2057436854,
-0.2758672237,
0.1181361526,
-0.0988535583,
0.2478711605,
0.0687461495,
0.2188890874,
-0.0490783267,
0.3465975523,
-0.0977027416,
-0.2259618938,
-0.1064975113,
0.0391055197,
0.531206429,
-0.2995471656,
0.3275354505,
-0.1957136095,
0.0639886782,
-0.2002594918,
-0.2282969803,
0.0872323737,
-0.0478701815,
0.3465406597,
0.2117685527,
-0.0232824236,
0.364433527,
-0.1089352071,
0.3259529173,
-0.1572976112,
-0.0153348558,
0.0034700036,
-0.0445439406,
-0.142670542,
-0.2228536606,
-0.3446230292,
-0.4408051372,
-0.38834548,
-0.0106526744,
-0.0261883438,
-0.041893553,
0.0917574763,
-0.1342886686,
-0.0532306172,
-0.0178117566,
-0.0258909091,
-0.3486142755,
-0.2984478474,
0.2000669688,
-0.3020643294,
-0.2819241881,
0.2392929345,
0.1354366541,
0.4544391036,
0.0401276685,
-0.3937389553,
0.580820024,
-0.1391109526,
-0.1142628118,
0.1804574281,
0.3090637922,
-0.0905659199,
0.1654050052,
-0.0495855138,
-0.1657475233,
-0.264600873,
-0.1806015968,
-0.2320662737,
0.1591045111,
-0.1456700712,
0.0627860576,
0.0104182586,
0.8047190309,
0.0654782131,
-0.2943533063,
0.1827456653,
0.1267128289,
0.22185269,
-0.0006961301,
-0.5895357728,
0.2010112554,
-0.0986171663,
0.2215834707,
0.3626621366,
0.0432907455,
0.1737743169,
-0.1139222234,
0.1939598769,
-0.1887793094,
-0.3513717055,
-0.1281633973,
0.2005108893,
0.384968102,
0.0900530443,
0.1219890267,
0.0273914561,
-0.0954924524,
-0.0095352717,
0.3547010124,
0.1222474873,
0.0885327458,
-0.1050639749,
-0.0560129024,
-0.2369753569,
0.0535893925,
-0.1932053417,
0.4575623572,
-0.0324506685,
-0.1230477318,
-0.0423157625,
-0.018290678,
0.5026881099,
0.196738705,
0.392770201,
0.1022482887,
-0.4248225689,
-0.2392999828,
-0.1275756508,
0.0902224034,
-0.0735769495,
0.1285207719,
0.1846205294,
-0.2020289898,
-0.5072083473,
0.2025788128,
0.110791266,
-0.1888124496,
-0.1198265851,
-0.1453298032,
0.0249493048,
-0.2902212739,
-0.0777402893,
-0.2248148024,
0.1797914803,
-0.4049386382,
-0.023972854,
-0.1545184255,
-0.0788380727,
0.1511588097,
0.0657216012,
0.0934650078,
0.0442998596,
0.4923816025,
0.3122971356,
0.3200490177,
0.5650812387,
0.615952909,
-0.1010635719,
0.0344185531,
0.0835056677,
0.0163797066,
0.052282501,
0.2831603289,
-0.0781764388,
0.1111001968,
0.0438617133,
0.1107479557,
-0.5960832238,
0.1240012124,
0.2019079924,
0.0680201054,
-0.357242465,
-0.0997589529,
0.1946074665,
0.0268734321,
0.2006981224,
0.1760098636,
0.1702614427,
-0.3972955346,
-0.4666161835,
0.0168192871,
0.494731009,
-0.0735877305,
0.2222425938,
-0.1423189342,
-0.1103766337,
0.0566480383,
0.2778766751,
-0.1172653213,
-0.1914952397,
0.1790857166,
0.0319306925,
-0.1585558057,
0.0860942155,
0.1152570099,
-0.074036397,
0.2761102319,
0.0473108888,
0.5473350286,
0.1399029195,
0.0827613249,
-0.1333880872,
-0.2570033967,
-0.2534720302,
0.1469008625,
-0.040101774,
0.3355660141,
-0.2314385772,
-0.1019250154,
0.3119747043,
-0.0665351152,
-0.2452949733,
0.006580241,
-0.6495986581,
0.1033414826,
-0.0357615538,
-0.1031016409,
0.3692238927,
-0.2313648611,
0.1811163127,
0.3199765384,
-0.1902631074,
0.101099968,
-0.3137950599,
0.0383314565,
-0.2302719355,
0.2646709085,
0.1008015499,
-0.0336168185,
-0.0603650734,
-0.1086391211,
-0.040744178,
-0.0539291054,
0.0338524282,
0.0507716499,
-0.0552918538,
-0.049335558,
-0.3194902837,
0.0118200779,
0.01315099,
-0.1511109769,
0.1543609649,
0.0290808994,
-0.0549282245,
0.3291850686,
-0.0549550839,
-0.1067547947,
-0.0585500076,
0.4145898521,
-0.2623614073,
0.0229861513,
0.3038800061,
0.4151065946,
-0.0175478719,
-0.2813272774,
-0.0685153008,
0.0908067673,
-0.2448109835,
0.27016747,
0.1164213791,
-0.0899816453,
0.3801681995,
0.561144352,
0.3422603309,
-0.1498251259,
0.0122462288,
-0.2868362665,
-0.1911177933,
0.06081523,
-0.1435581297,
0.1625946313,
0.0277143493,
0.4404928684,
-0.174705416,
0.3926444352,
-0.3059358299,
0.2749379277,
-0.0947568417,
-0.0000579925,
0.2123729736,
-0.1585533768,
0.1591585428,
-0.092134729,
0.0412249677,
-0.0340268984,
-0.5342018604,
-0.2137607932,
-0.1178592443,
0.094538033,
0.0980823785,
-0.2711147368,
0.1690744013,
-0.1080356762,
0.0807956457,
-0.1697111577,
0.3470470011,
0.0660693496,
-0.1222603619,
0.0229018033,
0.0703008622,
0.0658249557,
-0.1689614058,
0.1390581727,
0.0714532286,
-0.0677271932,
-0.004992824,
-0.0410777628,
0.0441779867,
0.0208380446,
0.1150303781,
0.0102870464,
-0.0216588713,
0.1182219386,
0.5093998313,
-0.3056087196,
-0.2525116503,
0.2638111115,
0.2856666148,
0.1686179787,
0.0513615236,
0.0614273101,
-0.1690220535,
0.1712699234,
-0.295384407,
-0.0795395225,
-0.0377572887,
0.1637817621,
0.0857126564,
0.0499727502,
0.3570747375,
-0.2151892185,
-0.2337025553,
0.3549884558,
0.3251186609,
-0.1116922349,
0.1748536229,
0.4612627625,
0.0847785696,
0.2086673677,
0.1075343117,
-0.1653206944,
0.10477487,
0.3765082955,
-0.2396265268,
0.6803712249,
0.0646546483,
0.0063732327,
0.2407553494,
-0.4143283069,
0.1107710898,
0.2690974474,
-0.1568294764,
0.0244205967,
-0.2630803585,
0.1377332509,
-0.3402011395,
-0.2731532454,
-0.1855367422,
0.1984927356,
-0.0655511543,
-0.1254134029,
-0.2556758821,
-0.1166858748,
-0.0780744553,
0.0273623541,
-0.0752226561,
-0.1022675335,
0.2595927715,
0.1732608676,
-0.1564743519,
-0.4046474099,
-0.3694781065,
0.0161442123,
-0.0983204991,
-0.1228213087,
0.2560083866,
0.4048366845,
-0.2123370916,
0.1384947449,
0.4051621854,
0.4715678096,
0.3013880551,
-0.019286247,
-0.0860341489,
0.0871977657,
-0.0851399675,
-0.2775003314,
0.4197261631,
-0.256807059,
-0.1573418081,
0.0049491078,
0.1699633002,
-0.1862419993,
-0.36442554,
0.1278568655,
-0.312604636,
-0.2835619748,
0.1925960928,
-0.0886968076,
-0.0576488152,
-0.046404805,
0.3801314831,
-0.0463572331,
0.227106601,
0.6677178144,
-0.0275352802,
0.1280926317,
-0.2423587441,
0.1004885435,
0.0852620155,
0.7199052572,
0.2930562198,
-0.0817145929,
-0.2865350842,
-0.5099943876,
-0.7815914154,
0.0482604392,
0.1542910635,
0.4362874031,
0.0736527219,
0.0439882576,
0.1192258596,
-0.022530023,
0.1537166834,
0.1220185757,
0.2925760448,
-0.2151335776,
-0.3548650146,
0.1796961129,
-0.0950151533,
-0.1906720847,
-0.0053810701,
-0.3220207691,
0.1466880143,
-0.2788922787,
0.0247031078,
-0.2417233735,
0.423589766,
0.0029747784,
0.2875815034,
0.2188121825,
0.0880904496,
0.3063713312,
0.0107985735,
-0.0630578622,
-0.4197328091,
-0.311278224,
0.0145377479,
0.0860738903,
-0.1074681431,
0.3646842241,
-0.0537011325,
-0.2685218155,
-0.4432089329,
0.1050859094,
0.1143686622,
0.2023361176,
-0.1769938469,
0.2222105861,
-0.2147685289,
-0.1733142138,
0.3443752229,
0.4282422364,
-0.0292371623,
0.1878371388,
-0.2106765211,
-0.3873320818,
0.2546340823,
-0.2887779176,
-0.1629849672,
-0.4629842937,
0.1772361249,
-0.0954382569,
-0.1698629558,
-0.4217800498,
0.1528435349,
0.1712167561,
-0.1016814485,
-0.2147470117,
0.372612536,
-0.046054516,
-0.0684142113,
-0.0185814574,
0.1694672108,
0.1803844571,
-0.3343187273,
-0.1139620915,
-0.0598332062
] |
https://github.com/huggingface/datasets/issues/1977 | ModuleNotFoundError: No module named 'apache_beam' for wikipedia datasets | I sometimes also get this error with other languages of the same dataset:
File "/dara/libs/anaconda3/envs/code/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 99, in pyarrow.lib.check_status
OSError: Memory mapping file failed: Cannot allocate memory
@lhoestq
| Hi
I am trying to run run_mlm.py code [1] of huggingface with following "wikipedia"/ "20200501.aa" dataset:
`python run_mlm.py --model_name_or_path bert-base-multilingual-cased --dataset_name wikipedia --dataset_config_name 20200501.aa --do_train --do_eval --output_dir /tmp/test-mlm --max_seq_length 256
`
I am getting this error, but as per documentation, huggingface dataset provide processed version of this dataset and users can load it without requiring setup extra settings for apache-beam. could you help me please to load this dataset?
Do you think I can run run_ml.py with this dataset? or anyway I could subsample and train the model? I greatly appreciate providing the processed version of all languages for this dataset, which allow the user to use them without setting up apache-beam,. thanks
I really appreciate your help.
@lhoestq
thanks.
[1] https://github.com/huggingface/transformers/blob/master/examples/language-modeling/run_mlm.py
error I get:
```
>>> import datasets
>>> datasets.load_dataset("wikipedia", "20200501.aa")
Downloading and preparing dataset wikipedia/20200501.aa (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /dara/temp/cache_home_2/datasets/wikipedia/20200501.aa/1.0.0/4021357e28509391eab2f8300d9b689e7e8f3a877ebb3d354b01577d497ebc63...
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 1099, in _download_and_prepare
import apache_beam as beam
ModuleNotFoundError: No module named 'apache_beam'
``` | 55 | ModuleNotFoundError: No module named 'apache_beam' for wikipedia datasets
Hi
I am trying to run run_mlm.py code [1] of huggingface with following "wikipedia"/ "20200501.aa" dataset:
`python run_mlm.py --model_name_or_path bert-base-multilingual-cased --dataset_name wikipedia --dataset_config_name 20200501.aa --do_train --do_eval --output_dir /tmp/test-mlm --max_seq_length 256
`
I am getting this error, but as per documentation, huggingface dataset provide processed version of this dataset and users can load it without requiring setup extra settings for apache-beam. could you help me please to load this dataset?
Do you think I can run run_ml.py with this dataset? or anyway I could subsample and train the model? I greatly appreciate providing the processed version of all languages for this dataset, which allow the user to use them without setting up apache-beam,. thanks
I really appreciate your help.
@lhoestq
thanks.
[1] https://github.com/huggingface/transformers/blob/master/examples/language-modeling/run_mlm.py
error I get:
```
>>> import datasets
>>> datasets.load_dataset("wikipedia", "20200501.aa")
Downloading and preparing dataset wikipedia/20200501.aa (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /dara/temp/cache_home_2/datasets/wikipedia/20200501.aa/1.0.0/4021357e28509391eab2f8300d9b689e7e8f3a877ebb3d354b01577d497ebc63...
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 1099, in _download_and_prepare
import apache_beam as beam
ModuleNotFoundError: No module named 'apache_beam'
```
I sometimes also get this error with other languages of the same dataset:
File "/dara/libs/anaconda3/envs/code/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 99, in pyarrow.lib.check_status
OSError: Memory mapping file failed: Cannot allocate memory
@lhoestq
| [
-0.2131412327,
-0.3383001685,
0.0015385412,
0.3398340642,
0.2197448909,
0.2354418933,
0.2507146895,
0.2680621147,
0.1814853102,
-0.0144521594,
-0.035932079,
0.036490161,
-0.1710567474,
0.1019869745,
0.1627782583,
-0.4244244099,
0.1850163043,
-0.1324691772,
-0.2871961296,
-0.0623329878,
-0.222919032,
0.3467074633,
-0.2787267268,
0.0804544091,
-0.0931129903,
-0.1117044464,
-0.0233222581,
0.1777576804,
-0.0540726483,
-0.1476041675,
0.1096981764,
-0.2670556307,
0.2810364366,
0.3396104872,
-0.0001204881,
0.0703728944,
0.3882136643,
-0.2105152607,
-0.3313602209,
-0.350505501,
0.1729956865,
-0.1294178665,
0.1381457448,
-0.2263072133,
-0.2553460002,
-0.1438834667,
0.152582109,
-0.2568331957,
0.628241837,
-0.0624547489,
0.1526159048,
0.3100945652,
0.3425412774,
-0.0296259429,
0.1350965798,
0.2360669225,
-0.0238423534,
0.3365864158,
-0.1699713171,
-0.1384140104,
0.1368694007,
0.5407431722,
-0.2211784124,
-0.012528046,
0.4623479545,
-0.2002916485,
0.0987882465,
-0.5092288852,
0.1167535484,
0.2758405805,
0.5031312704,
-0.2679270506,
-0.1555137336,
-0.0499548167,
0.033807531,
-0.0353642888,
0.0232852623,
0.2298913002,
-0.1179499626,
0.0734569132,
0.2295570374,
-0.3673402965,
-0.1967659295,
0.2305392921,
0.2532693148,
0.2152471989,
-0.066951111,
0.2225006223,
0.1748568416,
-0.1025936604,
-0.4621141851,
-0.0402749516,
0.1561261714,
0.4422610998,
-0.2312312573,
0.002823256,
-0.1472999901,
0.2895693779,
0.2478104383,
-0.1703073382,
-0.4538458288,
0.1153054088,
0.271007061,
0.0321764573,
0.014241457,
0.014833428,
0.1804624349,
-0.1404559761,
0.2022215128,
0.1521082222,
0.0284993723,
0.0471677855,
-0.083432287,
-0.0926124826,
-0.6663612723,
-0.0144302165,
0.2571782172,
-0.1104114652,
-0.1701253504,
0.0090114847,
-0.5486818552,
-0.2558386922,
-0.1178340316,
0.4914839566,
-0.0876234695,
-0.0432360955,
0.3066840172,
0.2725156546,
-0.3196043372,
-0.332946986,
-0.0259877071,
0.3970681429,
-0.4042426348,
0.2580560744,
0.1624426097,
-0.1852097064,
0.3300157785,
-0.0969915837,
0.0612079203,
-0.0126663968,
0.0938327163,
0.0160336271,
-0.2645788193,
0.1376618296,
0.2601520121,
0.2935314775,
0.334785521,
-0.1072973311,
-0.1303372979,
-0.0130059663,
-0.2017821074,
-0.078586638,
0.0849884003,
0.0104479874,
-0.1748301089,
0.1516971886,
-0.1632312834,
0.2782361507,
-0.0621967502,
0.0589513779,
-0.0485333018,
0.2518561482,
-0.0731162354,
-0.1545419991,
0.4467759728,
0.7152787447,
-0.0924547315,
-0.1832815409,
-0.1248662323,
0.0888793916,
-0.1330817491,
0.0378333107,
-0.0201482065,
0.2038289756,
-0.1944640875,
0.0074124634,
0.2688694596,
-0.4080123901,
0.0281281024,
0.0515704267,
0.0758030787,
0.0689503178,
0.1730379462,
-0.1063590944,
-0.2876464725,
-0.0430137254,
-0.0504937246,
0.2006361634,
0.1387091726,
-0.0972392038,
-0.1221534014,
-0.2608412206,
0.093887493,
0.233961612,
0.1046051234,
-0.0214356221,
0.0598471463,
0.5199068785,
0.225365907,
-0.1468616426,
0.1068722606,
0.365106076,
-0.1032073274,
0.2124214768,
0.080100894,
-0.2059350908,
-0.1896694601,
0.0508288294,
-0.2155386358,
0.5097806454,
-0.1102504358,
-0.0496578589,
-0.2062757164,
-0.062301565,
-0.1766990572,
-0.5256923437,
0.0174632929,
-0.1082245409,
0.1392479837,
0.3834106922,
-0.102400884,
0.2250565439,
0.0055665374,
0.1990667582,
-0.8539184332,
0.1584815681,
-0.1625045389,
0.0418304503,
0.1188351363,
0.2998980284,
0.14185597,
-0.2450332642,
0.1450188756,
0.2261149883,
0.0557806231,
0.0124183632,
0.0420553945,
-0.1097350344,
0.1478804201,
-0.384585917,
0.2552295923,
0.0223893858,
0.0874388218,
0.059449017,
0.0067109661,
0.2668874562,
0.1542758793,
0.3922152817,
0.1603033096,
0.1677395254,
-0.0652835891,
0.0154547542,
-0.2106283009,
-0.3001225591,
0.3233875036,
0.1925310791,
0.3315280676,
-0.2222318649,
-0.1197224781,
-0.3948609531,
0.2577235699,
-0.0251157917,
0.2270758152,
-0.056015864,
-0.6535446644,
0.2904532254,
0.1373189092,
-0.2333567142,
0.1929878145,
0.2338313907,
-0.1762010604,
0.2047180384,
0.1758314371,
0.0648407266,
0.2654978037,
0.0882536918,
0.4785857201,
-0.0819544047,
-0.2294494063,
-0.111442402,
-0.2178027928,
-0.179396376,
-0.2691537142,
0.1448301226,
-0.352633357,
0.1320126653,
-0.2323409319,
-0.5582600832,
-0.3051196337,
-0.0217381138,
-0.5030120611,
-0.0974114388,
-0.3454747796,
0.1530037671,
0.1994353235,
0.4281540513,
0.2881689072,
-0.0364839137,
0.0833427459,
-0.1567294002,
-0.1880507767,
-0.150294587,
-0.2787017524,
-0.0067597143,
0.3752516806,
0.139342919,
0.1550536752,
-0.2120731324,
-0.0861392021,
-0.038353771,
-0.4270919561,
0.224488616,
-0.2542715669,
0.2628420293,
0.1885617673,
0.4130574763,
-0.1757035255,
-0.0502507575,
0.3604339063,
0.0225336924,
-0.0979226455,
-0.0710497722,
0.0758107007,
0.0790888444,
0.0070313737,
-0.2876691222,
-0.274710238,
-0.3001061976,
0.1849600077,
0.1132890284,
-0.0709091425,
0.378770858,
0.2008129507,
0.1534816474,
-0.0516652726,
0.0577601269,
-0.1449172795,
0.0399069339,
0.3708284497,
-0.2393714637,
-0.3559530377,
0.1864735782,
-0.1147245616,
0.219479546,
-0.03918989,
-0.2681618333,
-0.1775779426,
0.3354677856,
-0.1826423705,
-0.0108238757,
0.3228729367,
0.2339253724,
-0.1341774315,
0.1562603414,
-0.1174858436,
-0.0410773121,
-0.0801705122,
-0.4929056764,
0.3673176467,
0.2488037497,
0.6314402223,
-0.0803146511,
0.8895261288,
0.2763774395,
0.2054400742,
0.1781021655,
-0.0261578355,
0.3253339529,
-0.0767206997,
-0.2912578583,
-0.1696089804,
0.0166210383,
0.138505891,
0.3068870604,
-0.0287206396,
0.0690140873,
-0.5301668048,
-0.2567856014,
-0.1413564384,
-0.234993875,
0.0289407242,
-0.1157509387,
0.40203771,
0.0677074641,
0.2030598521,
0.0631274655,
-0.0334787928,
0.3315421343,
0.5468325019,
0.0733551905,
0.1779474318,
-0.0101673827,
-0.1991822869,
-0.5310753584,
0.4021536112,
-0.1498973668,
0.0007157736,
-0.0902604684,
-0.1064203903,
0.1710890085,
0.0208134316,
0.7178605795,
-0.0888773203,
-0.0936145857,
0.1358971,
0.0015432462,
-0.730561316,
0.1050548106,
0.0151005313,
0.1899869889,
0.157246232,
0.2613233626,
-0.4709849656,
-0.1298936605,
0.4364515543,
0.1856459826,
-0.1291808784,
-0.15219374,
-0.5246489048,
-0.1736599803,
-0.4372414947,
-0.190143615,
-0.0122953579,
0.1366356164,
0.1450327635,
0.3733566701,
0.0267636161,
-0.1719457656,
0.2619965374,
0.2708521187,
0.3428228199,
-0.0041510202,
-0.1550676227,
0.1976629049,
0.3420521319,
-0.0263309516,
0.1929227859,
-0.0081904298,
-0.4109746516,
-0.048152484,
0.0178304985,
0.2324724197,
0.1311832964,
-0.1344867498,
0.1006975621,
0.1931824833,
-0.0921951085,
-0.1482955068,
-0.0953786969,
0.2387635857,
0.2665475607,
-0.3958497047,
-0.416685462,
0.6590349078,
0.0648798496,
0.0568406135,
0.1441073567,
0.2705354691,
-0.3297988772,
0.3083681166,
0.1000935435,
1.0329891443,
-0.0303209908,
0.1867603809,
0.2142025977,
0.1614116132,
0.8284101486,
-0.4468647838,
0.1807176918,
-0.320856452,
0.1913250387,
-0.0089975335,
-0.0728423446,
0.4384718239,
0.0767247379,
-0.1231355369,
0.3307363987,
0.220532462,
0.083788,
-0.1420508623,
0.458617717,
-0.0194511339,
-0.2219653726,
-0.4577874541,
0.0059213527,
-0.1063209847,
0.6052085161,
-0.2764024138,
-0.188620016,
-0.2304376811,
-0.2535759509,
-0.1999852061,
-0.0153510198,
-0.353202343,
0.2437591255,
-0.1092240736,
-0.2311708033,
0.3377097845,
0.2006299049,
0.2454079539,
0.0233235229,
-0.3331475854,
0.1841921508,
-0.3711701334,
-0.4309201539,
-0.2007158399,
0.0908435807,
0.3212770224,
-0.1738762259,
-0.1705290824,
0.0283659436,
-0.074469544,
0.0023994073,
-0.1516769826,
-0.1976523101,
0.2312404513,
0.0417473689,
-0.17325598,
0.2482138723,
0.1364870369,
0.1078019664,
0.0466101132,
-0.1768839955,
-0.1747899652,
-0.0595202483,
0.1811486781,
-0.1508110315,
-0.0135540422,
0.4774490595,
0.1515923291,
-0.0545031205,
0.5850340128,
0.4320333898,
-0.1466628611,
-0.1480634362,
0.1150297076,
0.0594773665,
-0.3519893587,
-0.0794995278,
0.3436743617,
0.2288254201,
-0.1090678573,
0.0155703705,
0.066021204,
-0.220502004,
0.1175456047,
-0.4755688906,
-0.1650406569,
0.4451122284,
-0.2226312757,
-0.1648504138,
0.2192070037,
0.2640366256,
0.2220731378,
-0.091098085,
-0.1593146324,
-0.1072033122,
0.1454122216,
0.0105219651,
0.5708832145,
0.0572467484,
0.1781496108,
0.1490952373,
0.0441930108,
0.0211144686,
-0.2832331061,
-0.0636304021,
-0.0720101669,
0.2172733843,
-0.187689364,
-0.1258063614,
0.0418039672,
-0.220944345,
-0.0505201295,
-0.1861304045,
0.022619769,
-0.0529785231,
0.0284199435,
0.0480479673,
0.0441346541,
0.2348096073,
-0.4368966222,
0.0437425748,
-0.068809472,
0.1023069993,
0.2306393683,
-0.0978874862,
0.3817754686,
0.0426771939,
-0.3169850707,
0.0096644368,
0.0188737586,
-0.0928130969,
0.2114265561,
0.0173060782,
-0.1756636947,
-0.0178816281,
0.0450216383,
0.2304663956,
-0.1178765222,
0.0271786526,
0.2184093893,
0.0492903218,
-0.4587967992,
0.3201699257,
0.4787879288,
-0.2735530436,
-0.0876100138,
0.2480542958,
0.0477450043,
0.3263323307,
0.0658335835,
0.1816515326,
0.2887360454,
0.1045156121,
0.3736542463,
0.0241868384,
0.1239091903,
0.1329518259,
0.2444556504,
0.1555861235,
0.21123676,
0.4075266421,
-0.2197605968,
0.2934308946,
-0.2903069258,
0.0693988651,
-0.0584796928,
-0.1634106934,
0.1687479317,
0.3123418689,
0.0832755789,
0.1723279059,
-0.2653711438,
0.4888259172,
-0.0398294441,
-0.2026304901,
-0.1932530701,
0.261872679,
-0.0716299564,
-0.1267793328,
-0.1107864827,
-0.2217426896,
-0.1446864307,
-0.034208376,
0.1442164481,
-0.6080905795,
0.1658253521,
0.1780466586,
-0.1325175464,
-0.1697185785,
-0.1060792953,
0.2342262566,
0.1096987948,
-0.0739860013,
-0.0205646101,
0.0220542252,
0.0508320592,
-0.0317691937,
0.0707946047,
0.2845671773,
0.0694994107,
-0.3261434734,
-0.0886075944,
0.0281690173,
0.0773351192,
0.1540692151,
-0.03257237,
0.148021549,
0.2344671935,
0.1302426606,
0.0422346294,
-0.0877720714,
-0.1766920686,
0.1153910011,
0.0173511319,
-0.3497012556,
-0.1444830894,
-0.5136170387,
0.1191737205,
-0.3950392902,
0.1520346254,
-0.6309111714,
0.3177125752,
0.24209328,
0.1043870747,
-0.23572734,
-0.2944584489,
-0.0183673538,
-0.2088823617,
0.4825664759,
0.2201555818,
0.1175151765,
-0.2192917615,
-0.3498179913,
-0.5625338554,
-0.0253224224,
-0.1389908642,
-0.0824831948,
-0.2705151737,
-0.0041029491,
0.2746978402,
0.2762577534,
-0.0568131953,
-0.0429307446,
-0.0837090164,
0.0761819482,
-0.1995840967,
-0.2112316936,
-0.3830578029,
0.1804931611,
-0.1026849225,
-0.0495565906,
-0.0805136114,
-0.0298244115,
-0.1119439602,
-0.25309816,
-0.1765735298,
0.1710341871,
-0.1491252184,
0.3884035349,
-0.0165547971,
0.3359569311,
0.0793646723,
-0.1874969602,
-0.1123817861,
0.0434570722,
-0.0540189818,
0.3798468709,
0.0634099245,
0.2255604714,
-0.2784629166,
-0.3539012969,
-0.3619381785,
0.5960180759,
0.1159182787,
-0.0575563014,
-0.3417953849,
0.1050304323,
-0.0232631341,
0.1765859723,
-0.0566634163,
0.2489114404,
-0.1592448056,
0.0920141786,
-0.4346442819,
-0.0472104549,
0.5104723573,
-0.8678759933,
-0.2605214715,
0.003188137,
0.0732157901,
-0.0388209112,
-0.2040876597,
-0.6355009079,
0.1892228723,
0.301679641,
0.1750741452,
-0.0031790584,
0.2277703881,
-0.1221884489,
-0.0525796637,
-0.0560737178,
-0.3247558773,
-0.0901464671,
-0.1945208311,
0.0034506842,
-0.2350437045
] |
https://github.com/huggingface/datasets/issues/1977 | ModuleNotFoundError: No module named 'apache_beam' for wikipedia datasets | Hi ! Thanks for reporting
Some wikipedia configurations do require the user to have `apache_beam` in order to parse the wikimedia data.
On the other hand regarding your second issue
```
OSError: Memory mapping file failed: Cannot allocate memory
```
I've never experienced this, can you open a new issue for this specific error and provide more details please ?
For example what script did you use to get this, what language did you use, what's your environment details (os, python version, pyarrow version).. | Hi
I am trying to run run_mlm.py code [1] of huggingface with following "wikipedia"/ "20200501.aa" dataset:
`python run_mlm.py --model_name_or_path bert-base-multilingual-cased --dataset_name wikipedia --dataset_config_name 20200501.aa --do_train --do_eval --output_dir /tmp/test-mlm --max_seq_length 256
`
I am getting this error, but as per documentation, huggingface dataset provide processed version of this dataset and users can load it without requiring setup extra settings for apache-beam. could you help me please to load this dataset?
Do you think I can run run_ml.py with this dataset? or anyway I could subsample and train the model? I greatly appreciate providing the processed version of all languages for this dataset, which allow the user to use them without setting up apache-beam,. thanks
I really appreciate your help.
@lhoestq
thanks.
[1] https://github.com/huggingface/transformers/blob/master/examples/language-modeling/run_mlm.py
error I get:
```
>>> import datasets
>>> datasets.load_dataset("wikipedia", "20200501.aa")
Downloading and preparing dataset wikipedia/20200501.aa (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /dara/temp/cache_home_2/datasets/wikipedia/20200501.aa/1.0.0/4021357e28509391eab2f8300d9b689e7e8f3a877ebb3d354b01577d497ebc63...
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 1099, in _download_and_prepare
import apache_beam as beam
ModuleNotFoundError: No module named 'apache_beam'
``` | 84 | ModuleNotFoundError: No module named 'apache_beam' for wikipedia datasets
Hi
I am trying to run run_mlm.py code [1] of huggingface with following "wikipedia"/ "20200501.aa" dataset:
`python run_mlm.py --model_name_or_path bert-base-multilingual-cased --dataset_name wikipedia --dataset_config_name 20200501.aa --do_train --do_eval --output_dir /tmp/test-mlm --max_seq_length 256
`
I am getting this error, but as per documentation, huggingface dataset provide processed version of this dataset and users can load it without requiring setup extra settings for apache-beam. could you help me please to load this dataset?
Do you think I can run run_ml.py with this dataset? or anyway I could subsample and train the model? I greatly appreciate providing the processed version of all languages for this dataset, which allow the user to use them without setting up apache-beam,. thanks
I really appreciate your help.
@lhoestq
thanks.
[1] https://github.com/huggingface/transformers/blob/master/examples/language-modeling/run_mlm.py
error I get:
```
>>> import datasets
>>> datasets.load_dataset("wikipedia", "20200501.aa")
Downloading and preparing dataset wikipedia/20200501.aa (download: Unknown size, generated: Unknown size, post-processed: Unknown size, total: Unknown size) to /dara/temp/cache_home_2/datasets/wikipedia/20200501.aa/1.0.0/4021357e28509391eab2f8300d9b689e7e8f3a877ebb3d354b01577d497ebc63...
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/dara/temp/libs/anaconda3/envs/codes/lib/python3.7/site-packages/datasets-1.3.0-py3.7.egg/datasets/builder.py", line 1099, in _download_and_prepare
import apache_beam as beam
ModuleNotFoundError: No module named 'apache_beam'
```
Hi ! Thanks for reporting
Some wikipedia configurations do require the user to have `apache_beam` in order to parse the wikimedia data.
On the other hand regarding your second issue
```
OSError: Memory mapping file failed: Cannot allocate memory
```
I've never experienced this, can you open a new issue for this specific error and provide more details please ?
For example what script did you use to get this, what language did you use, what's your environment details (os, python version, pyarrow version).. | [
-0.2131412327,
-0.3383001685,
0.0015385412,
0.3398340642,
0.2197448909,
0.2354418933,
0.2507146895,
0.2680621147,
0.1814853102,
-0.0144521594,
-0.035932079,
0.036490161,
-0.1710567474,
0.1019869745,
0.1627782583,
-0.4244244099,
0.1850163043,
-0.1324691772,
-0.2871961296,
-0.0623329878,
-0.222919032,
0.3467074633,
-0.2787267268,
0.0804544091,
-0.0931129903,
-0.1117044464,
-0.0233222581,
0.1777576804,
-0.0540726483,
-0.1476041675,
0.1096981764,
-0.2670556307,
0.2810364366,
0.3396104872,
-0.0001204881,
0.0703728944,
0.3882136643,
-0.2105152607,
-0.3313602209,
-0.350505501,
0.1729956865,
-0.1294178665,
0.1381457448,
-0.2263072133,
-0.2553460002,
-0.1438834667,
0.152582109,
-0.2568331957,
0.628241837,
-0.0624547489,
0.1526159048,
0.3100945652,
0.3425412774,
-0.0296259429,
0.1350965798,
0.2360669225,
-0.0238423534,
0.3365864158,
-0.1699713171,
-0.1384140104,
0.1368694007,
0.5407431722,
-0.2211784124,
-0.012528046,
0.4623479545,
-0.2002916485,
0.0987882465,
-0.5092288852,
0.1167535484,
0.2758405805,
0.5031312704,
-0.2679270506,
-0.1555137336,
-0.0499548167,
0.033807531,
-0.0353642888,
0.0232852623,
0.2298913002,
-0.1179499626,
0.0734569132,
0.2295570374,
-0.3673402965,
-0.1967659295,
0.2305392921,
0.2532693148,
0.2152471989,
-0.066951111,
0.2225006223,
0.1748568416,
-0.1025936604,
-0.4621141851,
-0.0402749516,
0.1561261714,
0.4422610998,
-0.2312312573,
0.002823256,
-0.1472999901,
0.2895693779,
0.2478104383,
-0.1703073382,
-0.4538458288,
0.1153054088,
0.271007061,
0.0321764573,
0.014241457,
0.014833428,
0.1804624349,
-0.1404559761,
0.2022215128,
0.1521082222,
0.0284993723,
0.0471677855,
-0.083432287,
-0.0926124826,
-0.6663612723,
-0.0144302165,
0.2571782172,
-0.1104114652,
-0.1701253504,
0.0090114847,
-0.5486818552,
-0.2558386922,
-0.1178340316,
0.4914839566,
-0.0876234695,
-0.0432360955,
0.3066840172,
0.2725156546,
-0.3196043372,
-0.332946986,
-0.0259877071,
0.3970681429,
-0.4042426348,
0.2580560744,
0.1624426097,
-0.1852097064,
0.3300157785,
-0.0969915837,
0.0612079203,
-0.0126663968,
0.0938327163,
0.0160336271,
-0.2645788193,
0.1376618296,
0.2601520121,
0.2935314775,
0.334785521,
-0.1072973311,
-0.1303372979,
-0.0130059663,
-0.2017821074,
-0.078586638,
0.0849884003,
0.0104479874,
-0.1748301089,
0.1516971886,
-0.1632312834,
0.2782361507,
-0.0621967502,
0.0589513779,
-0.0485333018,
0.2518561482,
-0.0731162354,
-0.1545419991,
0.4467759728,
0.7152787447,
-0.0924547315,
-0.1832815409,
-0.1248662323,
0.0888793916,
-0.1330817491,
0.0378333107,
-0.0201482065,
0.2038289756,
-0.1944640875,
0.0074124634,
0.2688694596,
-0.4080123901,
0.0281281024,
0.0515704267,
0.0758030787,
0.0689503178,
0.1730379462,
-0.1063590944,
-0.2876464725,
-0.0430137254,
-0.0504937246,
0.2006361634,
0.1387091726,
-0.0972392038,
-0.1221534014,
-0.2608412206,
0.093887493,
0.233961612,
0.1046051234,
-0.0214356221,
0.0598471463,
0.5199068785,
0.225365907,
-0.1468616426,
0.1068722606,
0.365106076,
-0.1032073274,
0.2124214768,
0.080100894,
-0.2059350908,
-0.1896694601,
0.0508288294,
-0.2155386358,
0.5097806454,
-0.1102504358,
-0.0496578589,
-0.2062757164,
-0.062301565,
-0.1766990572,
-0.5256923437,
0.0174632929,
-0.1082245409,
0.1392479837,
0.3834106922,
-0.102400884,
0.2250565439,
0.0055665374,
0.1990667582,
-0.8539184332,
0.1584815681,
-0.1625045389,
0.0418304503,
0.1188351363,
0.2998980284,
0.14185597,
-0.2450332642,
0.1450188756,
0.2261149883,
0.0557806231,
0.0124183632,
0.0420553945,
-0.1097350344,
0.1478804201,
-0.384585917,
0.2552295923,
0.0223893858,
0.0874388218,
0.059449017,
0.0067109661,
0.2668874562,
0.1542758793,
0.3922152817,
0.1603033096,
0.1677395254,
-0.0652835891,
0.0154547542,
-0.2106283009,
-0.3001225591,
0.3233875036,
0.1925310791,
0.3315280676,
-0.2222318649,
-0.1197224781,
-0.3948609531,
0.2577235699,
-0.0251157917,
0.2270758152,
-0.056015864,
-0.6535446644,
0.2904532254,
0.1373189092,
-0.2333567142,
0.1929878145,
0.2338313907,
-0.1762010604,
0.2047180384,
0.1758314371,
0.0648407266,
0.2654978037,
0.0882536918,
0.4785857201,
-0.0819544047,
-0.2294494063,
-0.111442402,
-0.2178027928,
-0.179396376,
-0.2691537142,
0.1448301226,
-0.352633357,
0.1320126653,
-0.2323409319,
-0.5582600832,
-0.3051196337,
-0.0217381138,
-0.5030120611,
-0.0974114388,
-0.3454747796,
0.1530037671,
0.1994353235,
0.4281540513,
0.2881689072,
-0.0364839137,
0.0833427459,
-0.1567294002,
-0.1880507767,
-0.150294587,
-0.2787017524,
-0.0067597143,
0.3752516806,
0.139342919,
0.1550536752,
-0.2120731324,
-0.0861392021,
-0.038353771,
-0.4270919561,
0.224488616,
-0.2542715669,
0.2628420293,
0.1885617673,
0.4130574763,
-0.1757035255,
-0.0502507575,
0.3604339063,
0.0225336924,
-0.0979226455,
-0.0710497722,
0.0758107007,
0.0790888444,
0.0070313737,
-0.2876691222,
-0.274710238,
-0.3001061976,
0.1849600077,
0.1132890284,
-0.0709091425,
0.378770858,
0.2008129507,
0.1534816474,
-0.0516652726,
0.0577601269,
-0.1449172795,
0.0399069339,
0.3708284497,
-0.2393714637,
-0.3559530377,
0.1864735782,
-0.1147245616,
0.219479546,
-0.03918989,
-0.2681618333,
-0.1775779426,
0.3354677856,
-0.1826423705,
-0.0108238757,
0.3228729367,
0.2339253724,
-0.1341774315,
0.1562603414,
-0.1174858436,
-0.0410773121,
-0.0801705122,
-0.4929056764,
0.3673176467,
0.2488037497,
0.6314402223,
-0.0803146511,
0.8895261288,
0.2763774395,
0.2054400742,
0.1781021655,
-0.0261578355,
0.3253339529,
-0.0767206997,
-0.2912578583,
-0.1696089804,
0.0166210383,
0.138505891,
0.3068870604,
-0.0287206396,
0.0690140873,
-0.5301668048,
-0.2567856014,
-0.1413564384,
-0.234993875,
0.0289407242,
-0.1157509387,
0.40203771,
0.0677074641,
0.2030598521,
0.0631274655,
-0.0334787928,
0.3315421343,
0.5468325019,
0.0733551905,
0.1779474318,
-0.0101673827,
-0.1991822869,
-0.5310753584,
0.4021536112,
-0.1498973668,
0.0007157736,
-0.0902604684,
-0.1064203903,
0.1710890085,
0.0208134316,
0.7178605795,
-0.0888773203,
-0.0936145857,
0.1358971,
0.0015432462,
-0.730561316,
0.1050548106,
0.0151005313,
0.1899869889,
0.157246232,
0.2613233626,
-0.4709849656,
-0.1298936605,
0.4364515543,
0.1856459826,
-0.1291808784,
-0.15219374,
-0.5246489048,
-0.1736599803,
-0.4372414947,
-0.190143615,
-0.0122953579,
0.1366356164,
0.1450327635,
0.3733566701,
0.0267636161,
-0.1719457656,
0.2619965374,
0.2708521187,
0.3428228199,
-0.0041510202,
-0.1550676227,
0.1976629049,
0.3420521319,
-0.0263309516,
0.1929227859,
-0.0081904298,
-0.4109746516,
-0.048152484,
0.0178304985,
0.2324724197,
0.1311832964,
-0.1344867498,
0.1006975621,
0.1931824833,
-0.0921951085,
-0.1482955068,
-0.0953786969,
0.2387635857,
0.2665475607,
-0.3958497047,
-0.416685462,
0.6590349078,
0.0648798496,
0.0568406135,
0.1441073567,
0.2705354691,
-0.3297988772,
0.3083681166,
0.1000935435,
1.0329891443,
-0.0303209908,
0.1867603809,
0.2142025977,
0.1614116132,
0.8284101486,
-0.4468647838,
0.1807176918,
-0.320856452,
0.1913250387,
-0.0089975335,
-0.0728423446,
0.4384718239,
0.0767247379,
-0.1231355369,
0.3307363987,
0.220532462,
0.083788,
-0.1420508623,
0.458617717,
-0.0194511339,
-0.2219653726,
-0.4577874541,
0.0059213527,
-0.1063209847,
0.6052085161,
-0.2764024138,
-0.188620016,
-0.2304376811,
-0.2535759509,
-0.1999852061,
-0.0153510198,
-0.353202343,
0.2437591255,
-0.1092240736,
-0.2311708033,
0.3377097845,
0.2006299049,
0.2454079539,
0.0233235229,
-0.3331475854,
0.1841921508,
-0.3711701334,
-0.4309201539,
-0.2007158399,
0.0908435807,
0.3212770224,
-0.1738762259,
-0.1705290824,
0.0283659436,
-0.074469544,
0.0023994073,
-0.1516769826,
-0.1976523101,
0.2312404513,
0.0417473689,
-0.17325598,
0.2482138723,
0.1364870369,
0.1078019664,
0.0466101132,
-0.1768839955,
-0.1747899652,
-0.0595202483,
0.1811486781,
-0.1508110315,
-0.0135540422,
0.4774490595,
0.1515923291,
-0.0545031205,
0.5850340128,
0.4320333898,
-0.1466628611,
-0.1480634362,
0.1150297076,
0.0594773665,
-0.3519893587,
-0.0794995278,
0.3436743617,
0.2288254201,
-0.1090678573,
0.0155703705,
0.066021204,
-0.220502004,
0.1175456047,
-0.4755688906,
-0.1650406569,
0.4451122284,
-0.2226312757,
-0.1648504138,
0.2192070037,
0.2640366256,
0.2220731378,
-0.091098085,
-0.1593146324,
-0.1072033122,
0.1454122216,
0.0105219651,
0.5708832145,
0.0572467484,
0.1781496108,
0.1490952373,
0.0441930108,
0.0211144686,
-0.2832331061,
-0.0636304021,
-0.0720101669,
0.2172733843,
-0.187689364,
-0.1258063614,
0.0418039672,
-0.220944345,
-0.0505201295,
-0.1861304045,
0.022619769,
-0.0529785231,
0.0284199435,
0.0480479673,
0.0441346541,
0.2348096073,
-0.4368966222,
0.0437425748,
-0.068809472,
0.1023069993,
0.2306393683,
-0.0978874862,
0.3817754686,
0.0426771939,
-0.3169850707,
0.0096644368,
0.0188737586,
-0.0928130969,
0.2114265561,
0.0173060782,
-0.1756636947,
-0.0178816281,
0.0450216383,
0.2304663956,
-0.1178765222,
0.0271786526,
0.2184093893,
0.0492903218,
-0.4587967992,
0.3201699257,
0.4787879288,
-0.2735530436,
-0.0876100138,
0.2480542958,
0.0477450043,
0.3263323307,
0.0658335835,
0.1816515326,
0.2887360454,
0.1045156121,
0.3736542463,
0.0241868384,
0.1239091903,
0.1329518259,
0.2444556504,
0.1555861235,
0.21123676,
0.4075266421,
-0.2197605968,
0.2934308946,
-0.2903069258,
0.0693988651,
-0.0584796928,
-0.1634106934,
0.1687479317,
0.3123418689,
0.0832755789,
0.1723279059,
-0.2653711438,
0.4888259172,
-0.0398294441,
-0.2026304901,
-0.1932530701,
0.261872679,
-0.0716299564,
-0.1267793328,
-0.1107864827,
-0.2217426896,
-0.1446864307,
-0.034208376,
0.1442164481,
-0.6080905795,
0.1658253521,
0.1780466586,
-0.1325175464,
-0.1697185785,
-0.1060792953,
0.2342262566,
0.1096987948,
-0.0739860013,
-0.0205646101,
0.0220542252,
0.0508320592,
-0.0317691937,
0.0707946047,
0.2845671773,
0.0694994107,
-0.3261434734,
-0.0886075944,
0.0281690173,
0.0773351192,
0.1540692151,
-0.03257237,
0.148021549,
0.2344671935,
0.1302426606,
0.0422346294,
-0.0877720714,
-0.1766920686,
0.1153910011,
0.0173511319,
-0.3497012556,
-0.1444830894,
-0.5136170387,
0.1191737205,
-0.3950392902,
0.1520346254,
-0.6309111714,
0.3177125752,
0.24209328,
0.1043870747,
-0.23572734,
-0.2944584489,
-0.0183673538,
-0.2088823617,
0.4825664759,
0.2201555818,
0.1175151765,
-0.2192917615,
-0.3498179913,
-0.5625338554,
-0.0253224224,
-0.1389908642,
-0.0824831948,
-0.2705151737,
-0.0041029491,
0.2746978402,
0.2762577534,
-0.0568131953,
-0.0429307446,
-0.0837090164,
0.0761819482,
-0.1995840967,
-0.2112316936,
-0.3830578029,
0.1804931611,
-0.1026849225,
-0.0495565906,
-0.0805136114,
-0.0298244115,
-0.1119439602,
-0.25309816,
-0.1765735298,
0.1710341871,
-0.1491252184,
0.3884035349,
-0.0165547971,
0.3359569311,
0.0793646723,
-0.1874969602,
-0.1123817861,
0.0434570722,
-0.0540189818,
0.3798468709,
0.0634099245,
0.2255604714,
-0.2784629166,
-0.3539012969,
-0.3619381785,
0.5960180759,
0.1159182787,
-0.0575563014,
-0.3417953849,
0.1050304323,
-0.0232631341,
0.1765859723,
-0.0566634163,
0.2489114404,
-0.1592448056,
0.0920141786,
-0.4346442819,
-0.0472104549,
0.5104723573,
-0.8678759933,
-0.2605214715,
0.003188137,
0.0732157901,
-0.0388209112,
-0.2040876597,
-0.6355009079,
0.1892228723,
0.301679641,
0.1750741452,
-0.0031790584,
0.2277703881,
-0.1221884489,
-0.0525796637,
-0.0560737178,
-0.3247558773,
-0.0901464671,
-0.1945208311,
0.0034506842,
-0.2350437045
] |
https://github.com/huggingface/datasets/issues/1973 | Question: what gets stored in the datasets cache and why is it so huge? | Echo'ing this observation: I have a few datasets in the neighborhood of 2GB CSVs uncompressed, and when I use something like `Dataset.save_to_disk()` it's ~18GB on disk.
If this is unexpected behavior, would be happy to help run debugging as needed. | I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you! | 40 | Question: what gets stored in the datasets cache and why is it so huge?
I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you!
Echo'ing this observation: I have a few datasets in the neighborhood of 2GB CSVs uncompressed, and when I use something like `Dataset.save_to_disk()` it's ~18GB on disk.
If this is unexpected behavior, would be happy to help run debugging as needed. | [
-0.0360605493,
-0.0662574247,
-0.1086276025,
0.5382814407,
0.1768517494,
0.3046649992,
-0.0687095374,
0.2563448548,
-0.1498930156,
-0.1115517691,
-0.0044144988,
-0.238436386,
-0.1599155068,
-0.2295419574,
0.1550872624,
0.2438127697,
0.1417648196,
-0.0338773914,
-0.088298209,
-0.0986521989,
-0.1149731055,
0.0371303335,
0.0000263602,
-0.1759964526,
-0.5011149049,
-0.1951545328,
-0.1269188374,
-0.0226084702,
-0.1835439503,
-0.1439902931,
0.2084014118,
-0.090036273,
0.3271475136,
0.4473099411,
-0.0001115048,
-0.3621961772,
0.3227519393,
-0.0084093474,
-0.3896472752,
0.3312594593,
-0.3773739338,
-0.3923428357,
-0.1046533585,
-0.1323722452,
0.3790540695,
-0.0305663235,
-0.0775614902,
-0.4155649245,
-0.1001972705,
0.236139223,
0.2581386268,
-0.2053124905,
-0.3442169726,
0.1373019516,
0.1090574414,
0.0737134367,
-0.1573265195,
-0.2792851627,
0.2627801895,
0.2356610596,
-0.0979842544,
0.1034085304,
0.0833160356,
-0.0201199669,
0.2829734385,
-0.0790503994,
-0.3153595626,
-0.2074171454,
0.6619532704,
0.2052865624,
0.868317008,
-0.3482805789,
-0.080823347,
-0.2358754277,
-0.1497368515,
-0.2302530706,
0.1697390527,
0.4599977732,
0.050563924,
0.0040057208,
-0.5909301639,
-0.4143618047,
-0.1268030554,
-0.1330241412,
-0.2453346252,
-0.2898694277,
-0.3402656913,
0.037111111,
0.1638762653,
-0.0343986079,
0.3052970767,
-0.3209096193,
-0.0971698016,
0.2020125687,
-0.350128144,
-0.2808209062,
-0.1091755331,
0.4334006608,
0.2432491332,
0.0263104849,
-0.2546924055,
-0.1597528756,
-0.1192892715,
-0.0493231677,
0.2371747494,
0.4153014421,
-0.1042508408,
0.0482305288,
-0.0493972152,
-0.3486228585,
-0.1449478269,
-0.1352478713,
0.1227401942,
0.0719256401,
0.4723912179,
-0.4745313525,
-0.1642239392,
-0.3412799835,
0.178727299,
0.2651433051,
0.0549174249,
0.0278724432,
0.0875426382,
0.1612012535,
-0.113969028,
0.2380317152,
-0.2866079807,
-0.1712260097,
0.0806445852,
-0.1550653577,
-0.2483569831,
-0.1152963638,
-0.0712154433,
0.2489612103,
0.3250065446,
-0.2040179968,
0.2558794022,
0.0902936459,
-0.058172483,
-0.122166045,
0.1817219704,
-0.3663789034,
0.4594607055,
0.3354998827,
-0.115059182,
0.4209493101,
-0.1799027026,
0.1078725085,
-0.2113254666,
0.2094639689,
-0.4173653126,
-0.4349771738,
-0.0622865558,
0.0899464265,
-0.2316666096,
0.007756386,
-0.3615059853,
0.0903246552,
0.4243741632,
0.1884948462,
0.2217033207,
-0.1052606925,
-0.3592416048,
-0.2606437504,
-0.1278335154,
0.1828977913,
-0.5496001244,
0.2316556573,
-0.0171850696,
0.2038090527,
0.0265681744,
0.3862607479,
-0.0676845759,
0.0514890589,
-0.0738904178,
-0.1387888044,
-0.0273052529,
0.0410685837,
-0.6250156164,
0.1765685678,
0.3490018249,
-0.0522841215,
-0.0821779817,
0.3963994384,
0.0915792063,
-0.0462789834,
-0.113797009,
0.2983601987,
-0.2702447474,
-0.142247498,
-0.3710265458,
-0.1000406072,
0.0486654527,
-0.0800514072,
0.0584716126,
-0.2000576556,
-0.0648279339,
0.2171606421,
-0.0068970509,
0.1431342065,
0.2213948071,
0.3800556064,
-0.0400363468,
0.2462721169,
0.2545775473,
0.0045568421,
-0.3890990615,
0.2957886755,
0.3068703711,
-0.6919739842,
0.1563737541,
-0.1010133773,
0.1169685423,
0.1049529836,
-0.0330853835,
-0.1894970536,
-0.0202210732,
0.0810447931,
0.0512781143,
-0.1019685045,
-0.0221285522,
0.4722426534,
0.0037873462,
-0.172062099,
-0.3683247268,
-0.2058073133,
0.0847272947,
0.0366326943,
-0.1773811132,
0.0202002786,
0.0359574258,
-0.1158991754,
-0.1611274183,
0.1403149068,
0.1411768049,
-0.1652726531,
0.2010639012,
0.5026516318,
-0.009897341,
0.113393344,
0.3529169858,
-0.0542696081,
0.1295191199,
-0.1824493855,
-0.2455494553,
0.0227385499,
-0.1056363508,
0.0891261026,
-0.0018705204,
-0.1094688922,
0.0670590326,
0.0806368142,
0.2521678805,
0.0496882871,
0.1556519568,
-0.1102391779,
0.3004677892,
0.3928954601,
-0.0488351807,
0.2108540237,
0.4110555947,
-0.2141381949,
0.0022680499,
0.1635853797,
-0.287011534,
-0.4692260027,
0.09584748,
0.361961931,
0.440828681,
0.0798804909,
0.3315308094,
-0.1054328308,
0.2634229064,
-0.1867279112,
0.083248131,
0.0455253609,
0.0130669326,
-0.1717015952,
0.1634771079,
-0.0028155949,
-0.0295012146,
0.2800489664,
-0.0815470591,
0.2360969335,
-0.0313673466,
0.4049851,
-0.2644426823,
-0.0564953089,
-0.0029136017,
0.3724590242,
-0.2667138278,
-0.126384452,
-0.1069569811,
-0.078147158,
0.232027024,
-0.2643205523,
-0.253031075,
0.6069621444,
0.0467412323,
0.1223387644,
0.1749621779,
-0.0454662144,
-0.3047429025,
0.1147945449,
0.1704498082,
-0.3243662715,
0.2530547976,
0.2005160004,
0.1197219342,
-0.2608418763,
-0.18692635,
-0.1394842565,
0.1243321672,
0.2161898315,
-0.3505870104,
0.0301975552,
-0.2612143457,
0.1456108689,
-0.3643142283,
-0.1169226691,
-0.0494122505,
0.1684371382,
-0.0416088216,
-0.0869178325,
0.1603960097,
-0.0047822781,
-0.24828659,
0.0057885647,
0.2127773464,
-0.1623180211,
-0.0298233069,
0.1968927085,
-0.1227709204,
0.0286220759,
-0.1135850921,
-0.0954577401,
-0.3736052513,
-0.7964874506,
0.3856806457,
0.1086681783,
-0.2260925472,
0.1569164544,
0.1852318048,
-0.028862739,
0.1390070319,
-0.7766139507,
0.2781105638,
-0.2700858712,
0.185999617,
0.0507348478,
-0.04825937,
0.2186819911,
-0.2456575185,
0.0444293171,
-0.1558119506,
-0.3363452554,
-0.2957925797,
0.1520011127,
0.5615341067,
-0.2364706397,
-0.034599334,
0.3850359917,
0.5469371676,
0.4004919529,
-0.0670355558,
-0.0679813325,
0.4184623659,
0.5475921035,
-0.311563164,
-0.0791824311,
0.233995989,
-0.1497087479,
-0.0996914357,
0.3620905876,
0.3107396364,
-0.1633524001,
0.2869631648,
0.0401580632,
0.1326719373,
0.0942766368,
0.2241771072,
-0.3098754287,
0.1005248874,
0.1926444322,
-0.1880111396,
-0.051875066,
-0.3989801407,
-0.1218361408,
-0.0243606977,
0.061001163,
0.264626801,
-0.197054565,
-0.0600720644,
-0.0649955422,
0.1443314254,
0.002743233,
0.0161043368,
-0.0121990219,
0.2155072987,
0.2796203494,
0.2686990499,
0.4034876227,
-0.2655992806,
-0.0943299979,
-0.1338837147,
-0.1440499723,
-0.0571441203,
-0.0475821532,
0.0855803043,
0.2596668303,
0.109712556,
0.1156006604,
0.3220531642,
-0.0970591009,
-0.0447797626,
0.4560160935,
-0.3914614618,
-0.2944032848,
0.0297592282,
-0.1912536919,
0.2687642574,
-0.0891848803,
-0.0929319188,
-0.2434484363,
-0.0358284488,
-0.0312524736,
-0.1641992778,
0.134291485,
0.1278089583,
-0.1605035663,
0.1407567114,
0.2580204308,
0.1434350014,
0.0538883246,
0.2126131207,
0.1216578409,
0.2759926021,
-0.0779655278,
0.1932588071,
0.0723068789,
-0.0612021945,
0.1887558401,
0.0845036283,
-0.2157592773,
-0.1994143724,
-0.260589689,
-0.0752065033,
-0.1938836724,
0.2318018973,
0.0760554895,
0.1843229681,
-0.2688039243,
-0.5139296055,
0.4054386318,
0.2316490412,
-0.059795998,
0.1844305098,
-0.3162144125,
-0.0129250009,
-0.0800292045,
0.0073938817,
0.6194279194,
-0.2254929692,
0.3763306439,
-0.1509216428,
-0.0443960205,
0.3291867077,
-0.497579217,
-0.0549178198,
-0.040544115,
-0.2693859935,
-0.0064737163,
-0.0443165526,
-0.116427362,
0.0063691512,
-0.0639233515,
0.3358240128,
-0.1385195851,
0.0161020346,
0.1028255969,
0.0630029142,
0.0982588232,
-0.1343481243,
0.0600646213,
0.2710026503,
-0.1289545149,
-0.1042522639,
-0.0486170426,
0.1003939658,
0.0181987211,
0.2117583305,
-0.3484104574,
-0.1806468666,
0.0193876848,
0.3263838887,
0.0004045344,
-0.193110168,
-0.024747476,
0.1287295073,
0.2211126834,
0.2069507241,
-0.1576768458,
0.2406378537,
-0.0560302623,
0.1770827174,
0.2221261263,
-0.1428225338,
0.1603878736,
-0.0400771797,
0.0599727929,
-0.0360327736,
0.0636526793,
-0.1874228418,
-0.0689611882,
0.2709846497,
0.1704345047,
-0.053402096,
-0.2605213821,
0.046616964,
0.1419522762,
0.0665946081,
0.1456883848,
0.3092529178,
-0.0304542761,
0.535805881,
-0.2437468618,
-0.444132477,
-0.0491460674,
0.0895826072,
0.4706987739,
-0.2969498038,
0.2461408228,
-0.0117873922,
-0.1994108409,
-0.0052338392,
0.3188086748,
0.067256555,
0.0313019529,
0.103018865,
-0.2232809216,
0.2156287879,
-0.1635623425,
-0.4461264312,
0.1223969609,
-0.3968325555,
-0.2381709665,
-0.1169587448,
-0.4305740595,
-0.0959670693,
-0.2070608288,
0.2224826068,
0.0288558081,
-0.4427158237,
0.1704186797,
0.4168641269,
-0.2408591211,
0.1030588448,
0.1197369769,
0.2150498182,
0.0983551219,
-0.0639103055,
0.0655844659,
-0.1661857367,
-0.1151801348,
-0.0573547967,
-0.1430148482,
-0.2526409924,
0.040196158,
0.119419232,
0.1940989196,
-0.0846005827,
-0.1608937234,
-0.3413434029,
0.0277586803,
-0.2151892185,
0.0458554253,
0.3281028271,
0.2502225339,
0.0525251552,
-0.0190481655,
-0.2545651793,
0.0606616884,
0.2935043871,
0.0707680061,
0.3296493888,
0.0834346712,
0.1396348029,
0.1571823955,
-0.2573202848,
-0.4219651222,
0.18723315,
-0.1020593792,
-0.0509156696,
0.2260446101,
-0.2251012921,
0.249586165,
0.4410790205,
-0.1233841777,
0.1734839678,
-0.080980137,
-0.1859038174,
0.0557178259,
0.1488508582,
0.268877387,
-0.1151914746,
0.0335802771,
0.4176068008,
-0.2711030245,
0.0664891377,
0.1544736028,
-0.0757161677,
0.0487861335,
0.1051443219,
0.4411543012,
-0.2353552133,
0.1328931004,
0.1203706563,
0.2203483582,
-0.0414521061,
0.391582787,
0.4607021213,
0.066889219,
0.5477505922,
0.1909266561,
0.273395896,
0.4535448253,
-0.1988290846,
-0.0916389972,
-0.344214201,
-0.1206760108,
0.3691905737,
-0.1119835526,
0.1148327589,
-0.0498375893,
0.2011880875,
0.1710633337,
-0.4117726982,
-0.1741051674,
-0.0463554747,
-0.3184819221,
-0.0562827922,
0.4661231041,
0.009787336,
0.1980103701,
0.3031122684,
-0.0922477469,
-0.3561851978,
0.4774301052,
-0.0334121846,
-0.2023449391,
-0.2770078182,
0.2927496731,
0.1348911375,
-0.1748968065,
-0.1015884802,
0.2509456873,
-0.2589649558,
-0.0646930188,
-0.0242525823,
0.0933385491,
0.1487420499,
0.1655648351,
0.0571862906,
0.3473055363,
0.1391281486,
-0.0707048997,
-0.1802371591,
0.232895419,
-0.1752927899,
-0.1302568614,
-0.069315806,
0.1011552289,
0.0940847546,
0.3056694865,
0.0186946876,
0.2601127028,
-0.2928521931,
-0.1882197559,
0.0067084804,
-0.0192580372,
-0.2305588126,
0.2583488822,
-0.3910730481,
-0.1737106144,
0.6758269072,
-0.2580589056,
0.1359238923,
-0.1140784621,
0.0960623026,
-0.1910626292,
0.4656357169,
0.4458967447,
-0.0579665713,
-0.2991361618,
0.0074319839,
-0.2848969102,
0.1120931283,
-0.2606486976,
0.4734936059,
-0.1006662846,
0.5078718066,
-0.1443120241,
-0.0784373805,
-0.1094810292,
-0.0335024334,
-0.0562615693,
-0.0785240754,
-0.2845356762,
-0.0505346581,
0.1140258461,
0.1673987508,
0.0565687269,
-0.1680807918,
0.2948612273,
-0.2146661282,
-0.0311390348,
-0.0888765454,
0.0323958248,
0.077620998,
0.3440175951,
0.080825299,
0.0452026203,
0.0137081221,
0.1630225778,
0.1854195595,
-0.3548763394,
0.0036778264,
0.1407098025,
0.0632231981,
0.0539237484,
-0.0235885158,
-0.5196057558,
0.1973179132,
0.1435050964,
0.0603237674,
-0.1435730159,
-0.0266825203,
0.0854973122,
-0.3765102029,
-0.2722698748,
0.3478080928,
0.4459003806,
0.1441459656,
-0.1681193411,
-0.050699912,
-0.2086984962,
-0.0880707875,
0.2826551199,
0.0581456125,
-0.3323767185,
-0.0007812418,
0.1203753874,
0.5062115788,
-0.1627058387,
-0.335636735,
0.1049754173,
0.4289160967,
-0.1710664928,
-0.2342907935,
0.2778822184,
-0.262394011,
0.1653284431,
-0.1241285354,
0.617325902,
0.0001727473,
-0.0431102887,
0.2192084491,
-0.1861626506
] |
https://github.com/huggingface/datasets/issues/1973 | Question: what gets stored in the datasets cache and why is it so huge? | Thanks @ioana-blue for pointing out this problem (and thanks also @justin-yan). You are right that current implementation of the datasets caching files take too much memory. We are definitely changing this and optimizing the defaults, so that the file sizes are considerably reduced. I will come back to you as soon as this is fixed. | I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you! | 55 | Question: what gets stored in the datasets cache and why is it so huge?
I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you!
Thanks @ioana-blue for pointing out this problem (and thanks also @justin-yan). You are right that current implementation of the datasets caching files take too much memory. We are definitely changing this and optimizing the defaults, so that the file sizes are considerably reduced. I will come back to you as soon as this is fixed. | [
-0.0833510756,
0.0121251792,
-0.1291755885,
0.5186681151,
0.1493167579,
0.2426842004,
-0.0757895112,
0.3019647598,
-0.0877330527,
-0.0359356254,
-0.0295592919,
-0.2534341216,
-0.0936841071,
-0.2008036673,
0.0723076984,
0.1365636289,
0.0901815444,
-0.0653723329,
-0.0765933543,
-0.0797736645,
-0.1062293276,
-0.0433258712,
-0.0135117024,
-0.1718013585,
-0.5766088367,
-0.1823076904,
-0.1253971159,
-0.0051542595,
-0.2007427961,
-0.1783479452,
0.1808401048,
-0.029162541,
0.3237578869,
0.3823103905,
-0.0001080761,
-0.3183886707,
0.35694924,
0.0258178413,
-0.3599814177,
0.3442704678,
-0.5000376701,
-0.2974618077,
-0.1312580258,
-0.136044234,
0.3404592872,
-0.0457880348,
-0.0374123007,
-0.3942117095,
-0.0077164546,
0.2449935377,
0.2759653926,
-0.2257372439,
-0.3055833578,
0.092758067,
0.2078105807,
0.0236135349,
-0.1338834763,
-0.2668146491,
0.2703862786,
0.1654698849,
-0.1730730534,
0.1816095412,
0.0901527628,
-0.0411951914,
0.2849495113,
-0.0867874548,
-0.2421435714,
-0.203481257,
0.6005624533,
0.219785586,
0.933111608,
-0.3052319586,
-0.1002646089,
-0.1678257436,
-0.1438966393,
-0.151291132,
0.2134750187,
0.4073272049,
0.0708831102,
-0.0055509489,
-0.5161409974,
-0.4197382927,
-0.1441659033,
-0.1212615222,
-0.1735285521,
-0.2512041032,
-0.2740027308,
0.0589072257,
0.1617287099,
-0.0618542619,
0.4042037427,
-0.2412799895,
-0.1609717757,
0.1731910408,
-0.3817915022,
-0.2755791545,
-0.2229239196,
0.4864889085,
0.2402953058,
0.0038444493,
-0.260057807,
-0.1092918962,
-0.1039831266,
-0.0202955343,
0.2225334346,
0.3948449492,
-0.1059478372,
0.1032148376,
-0.0339699797,
-0.3564935029,
-0.2345283329,
-0.1782753319,
0.1071285158,
0.0232656971,
0.4347732663,
-0.4517470598,
-0.1963356733,
-0.3539863825,
0.2049479038,
0.1451403052,
-0.020994693,
-0.0245748851,
0.1459485888,
0.1945822686,
-0.1634187102,
0.231179595,
-0.2916729748,
-0.2418981493,
0.014578104,
-0.2142632753,
-0.2492372394,
-0.0623591654,
-0.1481209993,
0.3357027173,
0.3038077354,
-0.1790135652,
0.2806242108,
0.0677446723,
-0.0248506106,
-0.142868489,
0.2385396659,
-0.357353121,
0.4010711908,
0.3282656074,
-0.1311095059,
0.2729565501,
-0.2001670599,
0.1455437392,
-0.1907132566,
0.1966865957,
-0.5129280686,
-0.4545749426,
-0.0051718717,
0.1176293492,
-0.1054760143,
0.0378942899,
-0.3192511201,
0.0472879559,
0.4130819738,
0.1691476405,
0.1751926839,
-0.1207906902,
-0.4018246531,
-0.2324274927,
-0.1263591349,
0.1870447993,
-0.389220804,
0.2134616673,
0.0099464841,
0.2114602625,
0.0577604361,
0.3836232722,
-0.1483821869,
0.077675283,
-0.0870076418,
-0.1209374592,
0.0001560673,
-0.0795796588,
-0.6367112994,
0.121822767,
0.3812291026,
-0.1184330732,
-0.0189935192,
0.4304907918,
0.0283242464,
-0.0962787122,
-0.1710996628,
0.2358876169,
-0.2705536485,
-0.1395442486,
-0.2917769551,
-0.1208136082,
-0.0083785625,
-0.0380461216,
0.0865780488,
-0.1643363982,
-0.0146147497,
0.3307112753,
-0.0058597215,
0.1374898553,
0.0867401138,
0.3432821929,
-0.0301415287,
0.2613254189,
0.3053286672,
0.0460470617,
-0.4341362715,
0.27020818,
0.1865424812,
-0.6862900257,
0.1906164289,
-0.1346431822,
0.1436361521,
0.0420823172,
-0.0126282759,
-0.1946732104,
0.0207597688,
0.0416757762,
0.0679129884,
-0.2097644508,
-0.0188925713,
0.5739930868,
-0.0988554507,
-0.1863342822,
-0.3965404034,
-0.2163491249,
0.0431044921,
0.0391308963,
-0.0731343925,
-0.0137702376,
0.0558141246,
-0.0892110243,
-0.1688679755,
0.1361526102,
0.2321188003,
-0.1308438182,
0.2174191326,
0.5695485473,
-0.0355294459,
0.0977271497,
0.3712708056,
-0.1007413119,
0.0756175965,
-0.1984051764,
-0.2419495732,
-0.0179811046,
-0.1506021917,
0.1063491851,
0.027817674,
-0.1154013574,
0.0016547367,
0.0606405362,
0.2964201868,
0.0728689805,
0.1380549967,
-0.1058863774,
0.3245922625,
0.438267827,
-0.0780996829,
0.17650038,
0.460590452,
-0.150939092,
-0.041019585,
0.1564590931,
-0.2423933744,
-0.4789296985,
0.1506141871,
0.3311381936,
0.3907627463,
0.1092211157,
0.285066247,
-0.1033122987,
0.2868416309,
-0.1608241349,
0.081222415,
0.0667130202,
0.021948766,
-0.2884733379,
0.0932123661,
-0.0410420001,
-0.0126919299,
0.2164715827,
-0.0556182526,
0.2983462214,
-0.0109923407,
0.3491845429,
-0.2550541759,
-0.1621958762,
-0.0070123281,
0.3860312402,
-0.2329474092,
-0.1407173574,
-0.1002915427,
-0.0584417991,
0.266043067,
-0.1486242712,
-0.2300080359,
0.5778571367,
0.0820519626,
0.1423580647,
0.0880351067,
-0.0799467415,
-0.3060519099,
0.1110750586,
0.249334991,
-0.3045138717,
0.1999037713,
0.1163128689,
0.0619491674,
-0.2768292129,
-0.13323991,
-0.1086860448,
0.1835203469,
0.1987148374,
-0.3027623296,
0.0109497905,
-0.3178110123,
0.1380321532,
-0.3626976609,
-0.1848438382,
-0.1165110618,
0.148491323,
-0.0446066409,
-0.0062333196,
0.2158530354,
-0.0736985356,
-0.1628846526,
-0.0196523517,
0.1723953038,
-0.1615503579,
0.0438600704,
0.3077878952,
-0.1309025139,
0.0314121693,
-0.2782612741,
-0.0471067019,
-0.3229492307,
-0.8187429905,
0.2975213528,
0.1009912714,
-0.2260666937,
0.1321329176,
0.2460590601,
0.0601856783,
0.1901033372,
-0.7528631091,
0.1599372476,
-0.2401579767,
0.1551104933,
-0.0018201759,
-0.0307550468,
0.2011797279,
-0.1798162907,
0.0073360279,
-0.1245793179,
-0.1967882663,
-0.2765601873,
0.1456561238,
0.4915037453,
-0.1925028861,
-0.0633853078,
0.3370989263,
0.5671011806,
0.5190480947,
0.0446700677,
-0.070525445,
0.3782271743,
0.4487403333,
-0.3152042031,
-0.0880660862,
0.2686748505,
-0.1435835212,
-0.0421173871,
0.4030550122,
0.3427498341,
-0.1661661267,
0.2124954462,
-0.0696433038,
0.1701491773,
0.0360478014,
0.199965328,
-0.2818180919,
0.1630784422,
0.1915143728,
-0.2135231793,
-0.0805289522,
-0.4090752006,
-0.0097259134,
-0.0490241461,
0.060440857,
0.3139722347,
-0.2041912675,
-0.0216280092,
-0.035479486,
0.0954540595,
0.0373731963,
0.0340190828,
-0.0738481656,
0.2846225798,
0.2186447382,
0.3076268137,
0.3758860528,
-0.1845825464,
-0.1280827671,
-0.1948769391,
-0.1542894393,
0.0497721806,
0.0045884699,
0.0770996362,
0.3125365973,
0.1059293598,
0.1206215918,
0.3195888698,
-0.1694951355,
-0.0632759929,
0.4400776327,
-0.350592643,
-0.3430588245,
0.0311943293,
-0.1552644372,
0.2621400654,
-0.1166708916,
-0.0711958483,
-0.2251217067,
0.0282578766,
-0.03998347,
-0.1176182404,
0.0885977298,
0.1425407827,
-0.1457241774,
0.1115200445,
0.2330705971,
0.1177609786,
0.1649422795,
0.1629832387,
0.1325432062,
0.2992494404,
-0.10561084,
0.1794900745,
0.0435790643,
-0.0517016575,
0.1908408552,
0.0886838883,
-0.2106290013,
-0.1995036751,
-0.2176278979,
-0.1043854952,
-0.1774015427,
0.2469909042,
0.0499367192,
0.1593242437,
-0.2842360437,
-0.589230895,
0.4678242207,
0.2875882983,
-0.0503032431,
0.2816500664,
-0.3000888824,
-0.026212994,
-0.1170724779,
0.0719798207,
0.640000701,
-0.2170850933,
0.3547426164,
-0.1607197523,
-0.0057513714,
0.3252634704,
-0.4948315024,
-0.0203318857,
-0.0317214727,
-0.2943031192,
0.0027603358,
-0.0793970153,
-0.1042398587,
0.0160079636,
-0.0375993215,
0.3806454539,
-0.0770796835,
0.0171461124,
0.102303192,
0.2111572921,
0.0683917925,
-0.1219477504,
0.1412031204,
0.282170862,
-0.1376040429,
-0.1486559063,
-0.0250104964,
0.1032503843,
0.0334873796,
0.2228503078,
-0.2927026153,
-0.1754135042,
-0.028216891,
0.3199634552,
-0.0239457153,
-0.1103062555,
0.0224181712,
0.1508464068,
0.1966027617,
0.220222488,
-0.1755136549,
0.2917009592,
-0.0590802245,
0.1441817433,
0.2262176275,
-0.1373252273,
0.1802844405,
-0.0922552496,
-0.0155957043,
-0.0061954837,
0.0756169111,
-0.2075469196,
-0.1815808266,
0.2590363622,
0.1753004044,
-0.0318704769,
-0.180973947,
0.0111448895,
0.0968551189,
0.0720031559,
0.1782183498,
0.3341042399,
-0.0147101991,
0.545096457,
-0.1917359531,
-0.4085083008,
-0.039024625,
0.0376458652,
0.4952645302,
-0.2649905682,
0.296145767,
0.0108358487,
-0.2105102539,
-0.030077856,
0.2401514649,
0.1065252423,
-0.0192219317,
0.0464053228,
-0.1987548172,
0.2384645194,
-0.1002334133,
-0.3699205518,
0.0672203675,
-0.3693828583,
-0.1530644596,
-0.115230836,
-0.4739056826,
-0.0556580462,
-0.2485064715,
0.2175405025,
-0.003382653,
-0.4405683875,
0.1212479249,
0.4964891374,
-0.2847179174,
0.0173556283,
0.0766952336,
0.2561145127,
0.0128195323,
-0.1073097438,
0.0661792904,
-0.1396750212,
-0.0889872164,
-0.0818247795,
-0.1462652832,
-0.2698169053,
0.0640408099,
0.1060322672,
0.1788945198,
-0.09701702,
-0.1527721137,
-0.3858830333,
-0.004049927,
-0.2405212522,
0.0508169457,
0.2874115109,
0.2743218541,
0.0970946699,
0.028157087,
-0.3459439278,
0.1409606934,
0.3457015157,
0.0663541108,
0.2746602297,
0.0574158095,
0.1722718477,
0.1525083184,
-0.26756078,
-0.317443639,
0.191609785,
-0.1096863896,
-0.110775575,
0.1850314289,
-0.2056341171,
0.2195076942,
0.4261026978,
-0.1534729302,
0.1396081448,
-0.0626814067,
-0.1825829446,
0.1420995295,
0.1769848913,
0.2107807398,
-0.1552996039,
0.0270278379,
0.3825098276,
-0.2631723881,
0.1017863303,
0.1827223301,
-0.0995819494,
0.1150417477,
0.091678381,
0.3488767743,
-0.3227352202,
0.0893636644,
-0.0018473435,
0.2343329042,
-0.147089839,
0.4342315495,
0.5094912648,
0.0556070805,
0.5676189065,
0.2129419595,
0.3438633978,
0.4268434048,
-0.1354212165,
-0.07074745,
-0.2690300345,
-0.2201734185,
0.4318251014,
-0.1616874635,
0.060006313,
-0.0915106535,
0.2003554404,
0.2525607347,
-0.3557160795,
-0.1063613147,
-0.0910751969,
-0.2840982676,
-0.0544142202,
0.4328335822,
0.0195151865,
0.1626626253,
0.2804258168,
-0.0502701886,
-0.3196656406,
0.4784550965,
0.0176237412,
-0.2315437645,
-0.2045262307,
0.2657099664,
0.2216133475,
-0.096250236,
-0.1503150016,
0.2840473652,
-0.2878264189,
-0.0521471016,
-0.0020573067,
0.1353498399,
0.1079748422,
0.1173433959,
0.0900893584,
0.2902497947,
0.1260775924,
-0.1177471504,
-0.1577920914,
0.2456449717,
-0.1479881257,
-0.1631242633,
-0.0584238023,
0.1051508635,
0.0860580504,
0.1809345037,
-0.0109627657,
0.2041436136,
-0.2903203666,
-0.1878152341,
0.0826534629,
-0.0291629285,
-0.1981455982,
0.30620116,
-0.3989688754,
-0.1746676564,
0.6400594711,
-0.2770072818,
0.0793247819,
-0.1532567143,
0.1183214039,
-0.2559438944,
0.5393369198,
0.4638427496,
0.0643788427,
-0.2885503471,
-0.066926837,
-0.2184085548,
0.0996606648,
-0.3270795643,
0.4932588339,
-0.0297249854,
0.4698669612,
-0.2054979354,
-0.1049921662,
-0.0908683389,
-0.1021398157,
-0.0105252154,
-0.1008915007,
-0.3023271859,
-0.0487515032,
0.1350873411,
0.1893384159,
0.0312538892,
-0.1479460895,
0.3177108765,
-0.2440877557,
0.0139771253,
-0.1394676268,
-0.0066867396,
0.1513637304,
0.3554558754,
0.0827657431,
0.0488250628,
-0.0478209965,
0.1564614922,
0.1493868232,
-0.3969500363,
-0.05113611,
0.08510077,
-0.0666463226,
0.0280114822,
-0.0466874465,
-0.5293644071,
0.1965583563,
0.0999140814,
0.1286892295,
-0.2723076344,
-0.0669395626,
0.061558798,
-0.385614872,
-0.1979496777,
0.3814427853,
0.3347819746,
0.1350271553,
-0.1180506051,
-0.1002389565,
-0.2879346907,
-0.0645390004,
0.2868935168,
0.008070562,
-0.3462713361,
0.065700978,
0.1039325446,
0.447666049,
-0.1230790764,
-0.395904839,
0.1256315857,
0.4089437723,
-0.0978846401,
-0.1693585217,
0.2278121114,
-0.1347614229,
0.2264449745,
-0.0989770591,
0.577196002,
-0.0549579971,
-0.0072086304,
0.2037743926,
-0.1564336121
] |
https://github.com/huggingface/datasets/issues/1973 | Question: what gets stored in the datasets cache and why is it so huge? | Thank you! Also I noticed that the files don't seem to be cleaned after the jobs finish. Last night I had only 3 jobs running, but the cache was still at 180GB. | I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you! | 32 | Question: what gets stored in the datasets cache and why is it so huge?
I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you!
Thank you! Also I noticed that the files don't seem to be cleaned after the jobs finish. Last night I had only 3 jobs running, but the cache was still at 180GB. | [
-0.1613563597,
0.0733159035,
-0.1333290786,
0.5806110501,
0.0784171075,
0.2742606103,
-0.0560065918,
0.2457105666,
-0.1244979948,
-0.0830653161,
-0.0408941209,
-0.2797771692,
-0.1030301303,
-0.1785739809,
0.0967970416,
0.2181341052,
0.0879468545,
-0.1207298934,
-0.1814734936,
-0.1622636169,
-0.1023327485,
-0.0179167613,
0.0224895477,
-0.118005164,
-0.652338922,
-0.2212354243,
-0.1131068021,
0.055173181,
-0.140467748,
-0.0933793485,
0.143976748,
-0.0460860208,
0.3950193822,
0.5104982853,
-0.0001139643,
-0.2925595641,
0.3823720217,
0.0545069426,
-0.4045551717,
0.3270024359,
-0.4570841193,
-0.3279613853,
-0.1085928082,
-0.1003948674,
0.4394214749,
-0.0204239823,
0.0014370624,
-0.3428090215,
-0.0186628215,
0.2640450895,
0.2358755171,
-0.3048946261,
-0.3635191321,
0.1135003418,
0.1681372672,
0.1231588274,
-0.0441246852,
-0.2613187432,
0.2455045879,
0.1298613995,
-0.1326355934,
0.1569849849,
0.1074742004,
-0.0296069067,
0.1850612015,
-0.0502867773,
-0.2164558172,
-0.2998595834,
0.7091447115,
0.2274084985,
0.8417989016,
-0.2384046018,
-0.0828320086,
-0.2749613225,
-0.1479181647,
-0.1580326259,
0.2207848877,
0.5174493194,
0.1128784716,
-0.0317041501,
-0.511755228,
-0.3802756965,
-0.0655806214,
-0.1840290427,
-0.1793048084,
-0.2046903819,
-0.3094809651,
0.0825137347,
0.1481610537,
-0.0059114965,
0.3916046321,
-0.2668060064,
-0.1414570063,
0.2316581607,
-0.3380454779,
-0.2481550127,
-0.1725747734,
0.5365598202,
0.1817600876,
-0.0329653844,
-0.3444202542,
-0.1124419346,
-0.2114591151,
-0.000093203,
0.2603340149,
0.3116801381,
-0.0861701518,
0.1187397093,
0.0085731484,
-0.3854295611,
-0.321998924,
-0.1497398913,
0.0372535661,
0.0221334957,
0.3994229734,
-0.4164709449,
-0.1734294146,
-0.361415863,
0.1985379308,
0.1956637651,
-0.0042335466,
-0.0722751543,
0.1121164262,
0.2183164805,
-0.1871266961,
0.2036714852,
-0.2611677349,
-0.2313589007,
0.0284115747,
-0.1532806009,
-0.2285508513,
-0.0909059122,
-0.1327640414,
0.2995836139,
0.2765156329,
-0.139654845,
0.3309638202,
0.036887683,
-0.0648177192,
-0.1563340127,
0.2549434006,
-0.3582815528,
0.3951000869,
0.4093770683,
-0.112428695,
0.3327395618,
-0.2217196971,
0.2115444094,
-0.1827076972,
0.2137181908,
-0.5258376598,
-0.4418341517,
-0.0159433838,
0.0983831957,
-0.1572526544,
0.0768251941,
-0.3453712761,
-0.0189488474,
0.4571041763,
0.2275613099,
0.2256823629,
-0.1646017134,
-0.3279087543,
-0.1894805878,
-0.2278371304,
0.1464493871,
-0.3910381794,
0.2511364222,
0.0179405473,
0.1454430819,
0.0711840689,
0.2979475856,
-0.1191541851,
0.1315396726,
-0.0978694186,
-0.1348294318,
0.0279833749,
-0.04060838,
-0.5370237827,
0.1421535015,
0.4007992744,
-0.1622822583,
-0.0745798871,
0.3459576666,
0.0290506333,
-0.0679927692,
-0.1514481604,
0.2860540152,
-0.2108335942,
-0.1279174984,
-0.2654141784,
-0.088537991,
-0.0383826084,
-0.018470753,
0.0998504013,
-0.2232895344,
-0.0368052125,
0.1881960779,
0.0287928395,
0.1677855402,
0.0976625234,
0.3305998147,
-0.1397837698,
0.4280254841,
0.2879229784,
0.0306460336,
-0.43177706,
0.234340176,
0.1349745393,
-0.6966283917,
0.1623964608,
-0.1269115955,
0.2554291487,
0.0729258433,
-0.0731002092,
-0.2337893099,
-0.0279713906,
0.1845547557,
0.0301330015,
-0.1857675314,
-0.0711148009,
0.4955204725,
-0.1140732765,
-0.1607789695,
-0.4297209978,
-0.1848451495,
0.1431549639,
-0.002323579,
-0.1459565759,
0.0411188267,
0.0786861405,
-0.1117612571,
-0.1637636125,
0.1110548675,
0.247766301,
-0.1484709382,
0.2682817876,
0.567351222,
-0.0548565984,
0.1951504648,
0.3002672493,
-0.0716318637,
0.0636249781,
-0.1734066904,
-0.2402173579,
-0.0508689508,
-0.2077073157,
0.1525837481,
0.001532793,
-0.1117343456,
0.0542667806,
0.0190218836,
0.2636958957,
0.1293394268,
0.1632872373,
-0.0756035447,
0.2168033272,
0.444778651,
-0.1519057006,
0.1747809649,
0.4706741571,
-0.1771166325,
-0.0516840927,
0.1113192886,
-0.2850252986,
-0.4193943739,
0.1050442383,
0.4016013443,
0.4452230036,
0.0496108681,
0.3604389131,
-0.0756826997,
0.277546525,
-0.1837285161,
0.0208874419,
0.074898988,
-0.0487932861,
-0.1522929817,
0.0795726627,
-0.0078955758,
-0.0303095207,
0.1681206226,
-0.073871471,
0.2904920578,
-0.0363347791,
0.3572682738,
-0.2580593526,
-0.1316238642,
-0.0732911378,
0.3482124805,
-0.2150787413,
-0.136512652,
-0.1094917953,
-0.0709861517,
0.2380697578,
-0.1624612808,
-0.2139178067,
0.5364332795,
-0.0142116696,
0.150057584,
0.0735522509,
-0.0893294066,
-0.3087904453,
0.0690321401,
0.2572315335,
-0.2098607123,
0.2047889233,
0.0922112539,
0.0664837211,
-0.2757596076,
-0.123782225,
-0.1766271293,
0.1951683164,
0.2061213404,
-0.2882379591,
0.0006669164,
-0.2626597285,
0.0905252844,
-0.3399085999,
-0.1441927701,
-0.0432294495,
0.0609962493,
-0.069648169,
0.0072283819,
0.2078794092,
-0.0054569282,
-0.2065437436,
-0.0260301232,
0.1936480701,
-0.1558229923,
-0.0068773627,
0.2063535899,
-0.2218586355,
0.0161886513,
-0.2453487962,
-0.0151542444,
-0.3820661604,
-0.9283385277,
0.2592500448,
0.036852207,
-0.1697472036,
0.2085365355,
0.1965107024,
0.0232021641,
0.1615055799,
-0.7610756159,
0.2020996809,
-0.2372548431,
0.1262794435,
0.0077729113,
0.0311024506,
0.2027195692,
-0.1744493842,
0.0423538014,
-0.1398318708,
-0.2366735339,
-0.3054128885,
0.1870911121,
0.523781538,
-0.2106371075,
-0.0840957984,
0.1811248958,
0.5365936756,
0.4866471887,
-0.0381648168,
-0.1205074787,
0.4225434959,
0.4848380983,
-0.291248858,
-0.0783842579,
0.291151464,
-0.086643286,
-0.0904414356,
0.4184950888,
0.3539144397,
-0.0811956376,
0.2511946559,
-0.0332704857,
0.0404648483,
0.0854170769,
0.2601002455,
-0.3480748236,
0.1371673048,
0.1285500675,
-0.166455254,
-0.0566878729,
-0.3869126141,
-0.0410663821,
-0.0087419003,
0.0842422619,
0.2667951286,
-0.1963016987,
-0.0937535018,
0.0104852952,
0.1013095155,
-0.0088181682,
-0.0358107202,
-0.0618789829,
0.2916465104,
0.2779177129,
0.3221156597,
0.3650981486,
-0.3677513301,
-0.0626017526,
-0.1375655532,
-0.1767583489,
0.0598023608,
-0.0522747375,
0.0341919065,
0.2674117684,
0.1784341037,
0.1854331046,
0.3122752011,
-0.1331876814,
0.0565021522,
0.4340999126,
-0.3935497403,
-0.314712584,
0.0180360675,
-0.1934536397,
0.2668859363,
-0.09675318,
-0.1134521291,
-0.2543801665,
-0.0134461224,
-0.0458249226,
-0.1563432962,
0.1244402528,
0.1552248895,
-0.1609626412,
0.0811807737,
0.3141676486,
0.1119782329,
0.1702612191,
0.2174155116,
0.1782796532,
0.3218364418,
-0.1787978709,
0.2074739933,
0.09105739,
-0.068598561,
0.278559804,
0.0779271051,
-0.1736048162,
-0.2642306089,
-0.2368405312,
-0.0679098666,
-0.1778145134,
0.2747411132,
0.0785838291,
0.2560665309,
-0.2771655321,
-0.5995447636,
0.3857510388,
0.2862666249,
-0.1092853099,
0.2671397924,
-0.4895543456,
-0.010604281,
-0.1321151853,
-0.0013050921,
0.6466197968,
-0.2087162435,
0.3260280788,
-0.1999160796,
-0.0667584836,
0.3643881679,
-0.522583425,
-0.0769258216,
-0.0006269142,
-0.2032233924,
-0.037431445,
-0.0765037909,
-0.1397910416,
0.0380024016,
-0.0022657663,
0.3384396434,
-0.0816520825,
0.0173255205,
0.0731510296,
0.1452869475,
0.0633622631,
-0.0883147418,
0.1170082837,
0.2121532261,
-0.1198505908,
-0.0831394792,
-0.0899270624,
0.1223373562,
0.1065191627,
0.2631840706,
-0.2684572935,
-0.1861669719,
-0.0166537799,
0.2970580459,
0.023722006,
-0.156771943,
-0.080401428,
0.193159014,
0.1869922876,
0.1478958726,
-0.1096016467,
0.2880504727,
-0.098183766,
0.1249379665,
0.1941356957,
-0.1072757244,
0.226709038,
-0.0649360791,
-0.0248344243,
0.0245962925,
0.0448240414,
-0.1732371151,
-0.2194976509,
0.3592064679,
0.1631356776,
-0.051830817,
-0.2033406645,
0.0533614643,
0.0576287955,
0.1359999031,
0.1384759694,
0.3363601565,
-0.073691383,
0.6109343171,
-0.2178419232,
-0.4782264829,
-0.0767511353,
-0.0232028179,
0.5032827258,
-0.2337807566,
0.2860401869,
0.0184660032,
-0.1751320064,
-0.0311498605,
0.255907923,
0.1470769346,
0.0428168029,
0.0856799483,
-0.164921999,
0.2176072896,
-0.0642012283,
-0.4341560304,
0.0132348966,
-0.3945586979,
-0.2324022502,
-0.1734165549,
-0.435887754,
-0.0303706229,
-0.2216000259,
0.2527506948,
0.0294940323,
-0.3938132823,
0.2168479115,
0.4286861122,
-0.2361279428,
0.1436593235,
0.0652046353,
0.3111398816,
0.0642907321,
-0.0576627962,
0.0879252329,
-0.1032585129,
-0.1550847292,
-0.0653669238,
-0.124848634,
-0.2228999585,
0.052591905,
0.1372590661,
0.1883080751,
-0.0488751642,
-0.1431198716,
-0.3217916191,
0.0851675421,
-0.212169528,
0.0646225065,
0.1869442761,
0.2090809941,
0.1564133167,
0.1074779704,
-0.2736564279,
0.0798654631,
0.3291400671,
0.044923503,
0.295702219,
0.0472167172,
0.1226034611,
0.1379895359,
-0.2660234571,
-0.3409767151,
0.0906478018,
-0.0949325114,
-0.0699297264,
0.2095035613,
-0.21705167,
0.2317304015,
0.4328904152,
-0.1261777431,
0.2592010498,
-0.020914875,
-0.1793332994,
0.1794490367,
0.130259648,
0.2233304977,
-0.0603437796,
0.0664519146,
0.4383388162,
-0.3042237461,
0.0979853719,
0.1023672968,
-0.120740518,
0.0606674105,
0.1509354562,
0.3464743495,
-0.3671175539,
0.1173055395,
0.0725601763,
0.1537766606,
-0.0700299814,
0.451739192,
0.479398787,
0.0183851533,
0.5429629683,
0.3200550079,
0.4149708748,
0.4656387269,
-0.1855342388,
-0.0896626562,
-0.3549957573,
-0.2353118956,
0.5174940825,
-0.1213665828,
0.0425864458,
-0.1281031668,
0.1910884529,
0.2133025676,
-0.3449544907,
-0.1182982773,
-0.0223556776,
-0.2834679484,
-0.0617321953,
0.3831985891,
0.0348755866,
0.2570712268,
0.3279586732,
-0.039180126,
-0.2646864653,
0.555745542,
-0.0171334781,
-0.2771167457,
-0.2427120656,
0.2536897361,
0.1225281656,
-0.1367987096,
-0.1073721945,
0.2526152432,
-0.312650919,
-0.0337105542,
-0.0128570041,
0.1473041326,
0.1936821043,
0.1900820434,
-0.0141223362,
0.3107690215,
0.0938915312,
-0.0519730151,
-0.2107863724,
0.3367965817,
-0.1680716872,
-0.1389776468,
-0.0691707283,
0.0962318107,
0.0966417864,
0.2490071207,
-0.0509267189,
0.2233561575,
-0.2308835834,
-0.1931045055,
0.1338280439,
-0.0147661194,
-0.2211968899,
0.2575016916,
-0.3979859948,
-0.1736104339,
0.6114830375,
-0.2598449886,
0.0722600073,
-0.1495042443,
0.070110932,
-0.2653841972,
0.5645392537,
0.5225406885,
0.0885150433,
-0.2780852318,
-0.0863829255,
-0.27431041,
0.0643812194,
-0.335290581,
0.5271921754,
-0.1185416728,
0.4859296381,
-0.2018980384,
-0.1149546057,
-0.0428570211,
-0.1220419183,
-0.0005125403,
-0.119847551,
-0.3422715068,
-0.0923210829,
0.125998348,
0.1910203397,
0.0899186209,
-0.1604307592,
0.2859537303,
-0.2384637296,
-0.0576171465,
-0.0538929813,
-0.0453092083,
0.1033481508,
0.3779555857,
0.1462683827,
0.0711487979,
-0.0091991164,
0.1375865489,
0.2291774303,
-0.3952301741,
-0.0752000064,
0.1247663498,
0.0192078762,
0.0290721022,
-0.0575016849,
-0.567225039,
0.2911175191,
0.1137888879,
0.0580924042,
-0.2475101799,
-0.0398726575,
0.0926949307,
-0.4392722845,
-0.1869775057,
0.3382078707,
0.368044138,
0.1801200509,
-0.116571486,
-0.1416971982,
-0.2414920777,
-0.0349173285,
0.3328414559,
0.0415235311,
-0.4040130675,
-0.0493648387,
0.0916974619,
0.415422678,
-0.062009573,
-0.4134925604,
0.1519259959,
0.4260597825,
-0.0775111318,
-0.1994903237,
0.1650744677,
-0.0829910263,
0.2037355006,
-0.0660081133,
0.489936024,
-0.0626696199,
-0.0396935605,
0.2204193175,
-0.1378957927
] |
https://github.com/huggingface/datasets/issues/1973 | Question: what gets stored in the datasets cache and why is it so huge? | Hi ! As Albert said they can sometimes take more space that expected but we'll fix that soon.
Also, to give more details about caching: computations on a dataset are cached by default so that you don't have to recompute them the next time you run them.
So by default the cache files stay on your disk when you job is finished (so that if you re-execute it, it will be reloaded from the cache).
Feel free to clear your cache after your job has finished, or disable caching using
```python
import datasets
datasets.set_caching_enabled(False)
``` | I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you! | 95 | Question: what gets stored in the datasets cache and why is it so huge?
I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you!
Hi ! As Albert said they can sometimes take more space that expected but we'll fix that soon.
Also, to give more details about caching: computations on a dataset are cached by default so that you don't have to recompute them the next time you run them.
So by default the cache files stay on your disk when you job is finished (so that if you re-execute it, it will be reloaded from the cache).
Feel free to clear your cache after your job has finished, or disable caching using
```python
import datasets
datasets.set_caching_enabled(False)
``` | [
-0.0611620881,
-0.0334957242,
-0.1297228634,
0.5127938986,
0.1173944771,
0.2473041117,
-0.0104233697,
0.2427642196,
-0.0756756365,
-0.0384478271,
-0.0576453842,
-0.2856127322,
-0.0628410578,
-0.1385335624,
0.1360731125,
0.0736558512,
0.0571285412,
-0.0924015939,
-0.0630483329,
-0.1605248451,
-0.1804125309,
0.0023383144,
-0.0435124189,
-0.1090048999,
-0.6069411039,
-0.1671299487,
-0.1760504246,
0.013318425,
-0.1266207844,
-0.2256072462,
0.181264326,
-0.0810926408,
0.3298205435,
0.5128859282,
-0.0001129724,
-0.3193743825,
0.3587916791,
0.0623998567,
-0.4379306734,
0.2950621843,
-0.3354640007,
-0.4271190763,
-0.0203633849,
-0.1686971784,
0.3214413226,
-0.0767431185,
-0.1329327673,
-0.4167051315,
-0.0587263852,
0.2930727005,
0.249469921,
-0.1366453916,
-0.3039157391,
0.117889151,
0.1235293746,
-0.0146652013,
-0.0452340953,
-0.2453739047,
0.2158931196,
0.1646103114,
0.0025352687,
0.1477700919,
0.016942054,
0.0003064331,
0.2684414685,
-0.0396773629,
-0.2254714072,
-0.311825186,
0.6033419967,
0.2232897729,
0.8363257051,
-0.2880610824,
-0.2029640228,
-0.2979713678,
-0.1118040457,
-0.2330387682,
0.1009242535,
0.4950152934,
0.0988686159,
-0.0455196649,
-0.4867722392,
-0.3448888361,
-0.0777857751,
-0.046562396,
-0.2821796834,
-0.1423802674,
-0.2524194419,
0.1125408113,
0.1055421531,
-0.0284945611,
0.35754776,
-0.336345613,
-0.0157866981,
0.2345680594,
-0.3509425521,
-0.201149255,
-0.1322092116,
0.5577651858,
0.1247599423,
0.0052213632,
-0.3312216103,
-0.1804829389,
-0.1822400391,
0.0167687126,
0.1956838667,
0.3834739327,
-0.1756946146,
0.1844046861,
0.0359152257,
-0.3828604221,
-0.2349632382,
-0.1346766651,
0.1312394142,
-0.0012336634,
0.4762332141,
-0.4796627164,
-0.0986841321,
-0.3029202521,
0.1707351804,
0.2444287539,
-0.0966220647,
0.0273222513,
0.1169030592,
0.2148655951,
-0.2545851171,
0.2108535916,
-0.2304310352,
-0.1243432313,
-0.0487612337,
-0.1454731226,
-0.2492409199,
-0.0925649852,
-0.0672148988,
0.2685971558,
0.2856830359,
-0.2234884948,
0.3373493552,
0.027776055,
0.106363818,
-0.1411569715,
0.2592742443,
-0.3753346205,
0.371065706,
0.465598464,
-0.0162445977,
0.3749174774,
-0.1396910846,
0.1204946339,
-0.270875901,
0.2553672194,
-0.5015743971,
-0.4222279191,
0.0191154014,
0.1045882627,
-0.2029600441,
-0.0228899065,
-0.3181249797,
-0.0490735285,
0.4451744556,
0.2298893631,
0.2339739949,
-0.2258066535,
-0.3481407166,
-0.2585401237,
-0.17184636,
0.1132034585,
-0.3837774992,
0.2007804066,
0.0188003145,
0.2824503481,
0.0434069,
0.4158512354,
-0.1315063238,
0.1567610055,
-0.0538133234,
-0.1209238321,
0.0304626301,
-0.156662643,
-0.6271010637,
0.1202649027,
0.4168611169,
-0.0274890475,
-0.1223235354,
0.463041991,
0.0704055578,
-0.0158332419,
-0.1096916571,
0.3284004033,
-0.1558928639,
-0.090183422,
-0.2928344011,
-0.0386737585,
0.1061436683,
-0.0216424167,
0.1188711226,
-0.2305829227,
-0.1309049875,
0.1836874485,
0.0527519956,
0.1259610951,
0.0694233775,
0.3437758088,
-0.0101462957,
0.3748395443,
0.3232589662,
-0.0078654736,
-0.3653470278,
0.2573634088,
0.2419637144,
-0.6245026588,
0.1455273479,
-0.1339733005,
0.1272247136,
0.1450664401,
-0.0969083607,
-0.2118735611,
-0.0223336834,
0.0297690146,
0.1151991189,
-0.2108961046,
-0.044968836,
0.5452651978,
-0.0635575652,
-0.1103801429,
-0.4182025194,
-0.1708039641,
0.0938117802,
0.0492023714,
-0.1359283775,
0.0650077611,
0.1010839939,
-0.0895123184,
-0.1746931374,
0.1081862971,
0.2973418832,
-0.1439444721,
0.1077991873,
0.5784064531,
-0.0145125277,
0.1164005473,
0.4233711362,
-0.0106682582,
0.0905219615,
-0.1879725307,
-0.1697005033,
0.0567104928,
-0.1616251022,
0.1676456928,
-0.0029340163,
-0.0588440336,
0.0638453588,
0.1050605327,
0.1870781779,
0.0860390067,
0.1927024126,
-0.0662450194,
0.379272908,
0.4196601212,
-0.0819110572,
0.1654996872,
0.4912530184,
-0.1889421344,
0.0204285439,
0.1143536568,
-0.3488580287,
-0.4260423779,
0.0824613795,
0.3058719635,
0.4276180565,
0.0918692499,
0.2786065936,
-0.018704921,
0.1948916316,
-0.1964646876,
0.1110937148,
0.1457483768,
-0.0842692703,
-0.2639130354,
0.1610331386,
-0.0088863689,
0.0190628991,
0.1868860424,
-0.1532471478,
0.3150745332,
-0.0810950994,
0.4258531928,
-0.2198475897,
-0.066633597,
-0.0421656147,
0.1854019016,
-0.2399908006,
-0.2696066499,
-0.1647604853,
-0.091042459,
0.294296205,
-0.1827509403,
-0.2300926298,
0.469930023,
0.0185605958,
0.0107690748,
0.147135973,
-0.0653237179,
-0.3312749565,
0.0642060116,
0.1377597749,
-0.2550907433,
0.2090716958,
0.0785146505,
0.0315914527,
-0.2303850204,
-0.1585237086,
-0.1949878633,
0.1075049192,
0.1837015301,
-0.2883613706,
0.0521757267,
-0.2645151615,
0.1547861695,
-0.2806057036,
-0.1142954975,
-0.0577978343,
0.1149196699,
-0.050989788,
0.0269434862,
0.2000139207,
-0.0313189328,
-0.2764130235,
0.000523597,
0.2373339683,
-0.0265399329,
0.0129479673,
0.2392880619,
-0.1557642072,
-0.0176597554,
-0.1436422616,
-0.0100038406,
-0.3633742332,
-0.79055655,
0.3151324987,
0.0385606252,
-0.1788747907,
0.2086084783,
0.186661005,
0.0924993008,
0.1387997419,
-0.708563447,
0.0724820942,
-0.3142880797,
0.2180022746,
0.0143603617,
0.0765080899,
0.1385977566,
-0.1808547974,
0.0806726515,
-0.1142865643,
-0.2508400679,
-0.4035723805,
0.202920869,
0.3824837208,
-0.2886652052,
-0.0345723554,
0.2057441324,
0.4963881075,
0.5578779578,
-0.0190251935,
-0.1098964661,
0.3746929169,
0.4645800292,
-0.3987116814,
-0.1185682118,
0.2061815709,
-0.1018317491,
-0.1222380996,
0.3536938727,
0.3558352292,
-0.0725568607,
0.2818090618,
-0.0484955162,
0.0352502465,
0.0698834658,
0.22349599,
-0.2844496071,
0.1933541596,
0.1191816702,
-0.1655665338,
-0.1425038129,
-0.3973286152,
-0.0735084265,
-0.0214168727,
0.045613762,
0.282697767,
-0.1842950583,
-0.127944231,
-0.075401336,
0.142341733,
-0.0231205653,
-0.0497247577,
0.051461935,
0.2763728201,
0.3083383441,
0.3073396981,
0.3131420016,
-0.2744624019,
-0.2138914168,
-0.1058374122,
-0.2411260903,
-0.1186880618,
-0.0318222716,
-0.0057499632,
0.3012588024,
0.1512312591,
0.1316512525,
0.3211078644,
-0.1440430433,
0.0087751653,
0.5344343781,
-0.388807416,
-0.2943492532,
-0.0840831473,
-0.2762684226,
0.2164513767,
-0.1252091527,
-0.1201286167,
-0.1507782787,
0.023403503,
-0.0632727891,
-0.2404501289,
0.1136997044,
0.190148294,
-0.1203188822,
0.1561102718,
0.2085815519,
0.0853590444,
0.184217155,
0.14799577,
0.1042302847,
0.3757859468,
-0.191303134,
0.1126984805,
-0.0072585633,
0.0352625474,
0.2778847516,
0.0563125126,
-0.173789233,
-0.1832462251,
-0.2612352967,
0.0143586826,
-0.2691156268,
0.2237627804,
0.1496375948,
0.2776619196,
-0.2618527412,
-0.6982127428,
0.4616201222,
0.276463151,
-0.0716561079,
0.2424010038,
-0.4183204174,
-0.036516279,
-0.0543631762,
-0.0064189471,
0.6589259505,
-0.2236093581,
0.4306712151,
-0.0395673625,
0.0344163924,
0.4065584242,
-0.4473711252,
-0.0286580659,
-0.0794858783,
-0.1845749319,
0.0040680617,
-0.1019873172,
-0.1771422923,
-0.0575257242,
-0.0451953337,
0.4401684701,
-0.1805891693,
-0.0142155178,
0.0162406079,
0.186169982,
0.043687135,
-0.1095271707,
-0.042176418,
0.2492516637,
-0.1284718513,
-0.0425213687,
-0.0616241172,
0.1496699303,
0.0199740008,
0.1891245246,
-0.3140137196,
-0.2012456954,
-0.0288669877,
0.3245884776,
0.0098946113,
-0.1806678474,
-0.0580399372,
0.1913763285,
0.2203963399,
0.290102303,
-0.0961439386,
0.2670597434,
-0.1876456589,
0.1555692106,
0.2435379773,
-0.0914696455,
0.1821190119,
-0.0270394422,
-0.0338974446,
0.0384039693,
0.0826622546,
-0.2128392011,
-0.1591346413,
0.3341324031,
0.0977140591,
-0.0641100258,
-0.2387067378,
0.0309948996,
0.1164924651,
0.0835457444,
0.1395623386,
0.3734086156,
-0.0597868264,
0.6573854685,
-0.2188909054,
-0.5179988742,
-0.0908967331,
-0.043039564,
0.5688455701,
-0.2462106198,
0.3715307415,
0.0228577182,
-0.2190307528,
-0.02647442,
0.235037908,
0.0635290593,
0.104983516,
0.1324855983,
-0.1333609372,
0.2987749875,
-0.1285154969,
-0.3920567632,
0.0370850191,
-0.387973845,
-0.2860272229,
-0.1671657413,
-0.4105103314,
-0.0266918615,
-0.0878060088,
0.2242878079,
0.0568861514,
-0.297328949,
0.188733235,
0.4064843953,
-0.2433435619,
0.0819339976,
0.029714711,
0.2258061022,
0.0675608292,
-0.0967763066,
0.0996917188,
-0.1272795796,
-0.1318952143,
-0.0946974978,
-0.1185575724,
-0.2398276031,
-0.0179097988,
0.1461030245,
0.1713629961,
-0.0941340253,
-0.1098679453,
-0.3558127582,
0.1286485195,
-0.2562296689,
0.0847608894,
0.2919307947,
0.2346149832,
0.1277680695,
0.0235260092,
-0.2593020499,
0.1167762876,
0.3148726225,
0.0370092019,
0.2348183393,
0.1609543562,
0.1893203557,
0.2360895276,
-0.1818602681,
-0.292329967,
0.0981744975,
-0.1392734945,
-0.1110359877,
0.192197755,
-0.2242860198,
0.2773995101,
0.386836499,
-0.1051514149,
0.247159481,
-0.0514643416,
-0.1714700758,
0.1436867118,
0.1363580078,
0.1666207016,
-0.1629316062,
0.0704920292,
0.4474796951,
-0.2449590564,
0.1808710396,
0.1114564836,
-0.008500427,
-0.021918755,
0.0780232549,
0.3410852551,
-0.3142292798,
0.1521829218,
0.0750652552,
0.1814653873,
-0.1232982576,
0.4441084862,
0.4823722243,
0.0159769207,
0.5705835223,
0.1776808202,
0.3709159791,
0.3553228676,
-0.2679767013,
-0.061626032,
-0.3010143638,
-0.2353514433,
0.3864546418,
-0.172527045,
0.0777340606,
-0.0796902627,
0.1531074345,
0.1485291123,
-0.3190407455,
-0.2061378062,
-0.0086434977,
-0.284135282,
-0.1284306645,
0.4397257566,
0.0606788993,
0.2145417631,
0.2614770532,
-0.0105378125,
-0.4243504107,
0.5057744384,
-0.0288448855,
-0.2109090537,
-0.2632439435,
0.2805337906,
0.1405212134,
-0.1363978088,
-0.1008692756,
0.1854054034,
-0.2370546013,
-0.0349017568,
-0.1282426119,
0.0883429125,
0.2755078077,
0.1468667984,
0.0165943503,
0.3816763163,
0.0619967841,
-0.0660636872,
-0.1962117851,
0.1796406806,
-0.2197113633,
-0.093298085,
-0.0303409696,
0.1074645966,
0.0647101477,
0.2108067274,
-0.0653019845,
0.2227582783,
-0.3473085463,
-0.0111099258,
0.0606897511,
-0.0644291341,
-0.2514471412,
0.3058981597,
-0.4158520699,
-0.1779730022,
0.6756235361,
-0.1745560169,
0.0435053259,
-0.2043659687,
0.0827862173,
-0.2363874614,
0.5724008679,
0.5509408116,
0.1847085357,
-0.2911016345,
0.0006311238,
-0.2649796009,
0.0914620757,
-0.4748072624,
0.3963682652,
-0.0434777997,
0.4691888988,
-0.2120291293,
-0.0628109425,
-0.0016784519,
-0.0164432451,
-0.0396845117,
-0.1721838415,
-0.3049774468,
-0.0741595328,
0.0393010825,
0.1438496262,
0.0957856178,
-0.1642240733,
0.2386149317,
-0.3013192117,
-0.0327399075,
-0.1607341617,
0.0522631183,
0.1634509265,
0.3378364146,
0.168749094,
0.0903687477,
0.0849589556,
0.2190425247,
0.1733669341,
-0.3463304639,
-0.0603383482,
0.0893495008,
0.0750250816,
-0.0569978356,
-0.0385947302,
-0.5262835026,
0.1357944906,
0.1050975174,
0.155525133,
-0.174063921,
-0.185195446,
0.1313710511,
-0.3232077062,
-0.1827082336,
0.3909306526,
0.3479949534,
0.1791635752,
-0.1575752199,
-0.0387179404,
-0.1911492795,
-0.0748232305,
0.3076618314,
-0.0438519791,
-0.3876529634,
-0.0505249463,
0.1258733273,
0.4357413054,
-0.141032964,
-0.4110242426,
0.1245215237,
0.4666087925,
-0.0897401422,
-0.2220545113,
0.2676521838,
-0.1196447611,
0.1683703959,
-0.0717622265,
0.5194827914,
-0.1256664544,
-0.0529368743,
0.2193592787,
-0.1838397086
] |
https://github.com/huggingface/datasets/issues/1973 | Question: what gets stored in the datasets cache and why is it so huge? | Hi @ioana-blue, we have optimized Datasets' disk usage in the latest release v1.5.
Feel free to update your Datasets version
```shell
pip install -U datasets
```
and see if it better suits your needs. | I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you! | 34 | Question: what gets stored in the datasets cache and why is it so huge?
I'm running several training jobs (around 10) with a relatively large dataset (3M samples). The datasets cache reached 178G and it seems really large. What is it stored in there and why is it so large? I don't think I noticed this problem before and seems to be related to the new version of the datasets library. Any insight? Thank you!
Hi @ioana-blue, we have optimized Datasets' disk usage in the latest release v1.5.
Feel free to update your Datasets version
```shell
pip install -U datasets
```
and see if it better suits your needs. | [
-0.1255367398,
-0.0336779952,
-0.1482325792,
0.5212608576,
0.1614956707,
0.2426843345,
-0.0925322175,
0.2584301233,
-0.0802797452,
-0.0226319656,
-0.0793151557,
-0.1993970275,
-0.0890251994,
-0.1937062591,
0.0600060448,
0.1441113204,
0.1028739959,
-0.0664710402,
-0.0705818012,
-0.0968136713,
-0.1211708635,
0.0194599796,
0.0352942497,
-0.1679353267,
-0.5302936435,
-0.1817644835,
-0.1013381481,
-0.0436647646,
-0.1651498824,
-0.1785152555,
0.1776831448,
-0.0290395524,
0.3176941872,
0.4055047035,
-0.0001109538,
-0.3099727035,
0.3714549243,
0.1150019169,
-0.4696368575,
0.3074623048,
-0.3643324375,
-0.3740329146,
-0.0264453646,
-0.14283517,
0.3908943534,
0.0068972223,
-0.0497720093,
-0.4210887551,
-0.06840083,
0.1861254275,
0.2560691833,
-0.1310580075,
-0.3215965033,
0.1046230718,
0.1198370457,
0.0138910096,
-0.1468689144,
-0.2766179442,
0.2572323084,
0.301910907,
-0.1068857983,
0.2367849201,
0.061818555,
0.047615584,
0.2798640132,
-0.0198586658,
-0.1303211451,
-0.1588060856,
0.6128359437,
0.2203997523,
0.9452203512,
-0.325180769,
-0.1090443581,
-0.2270800322,
-0.169943437,
-0.1856395602,
0.2053112984,
0.3774013817,
0.0623727925,
-0.0036441507,
-0.4708914161,
-0.4204159379,
-0.1368879676,
-0.1069784164,
-0.1935636699,
-0.2392665595,
-0.293027699,
0.0738701746,
0.200610742,
-0.035828989,
0.3408769369,
-0.2122138143,
-0.1762357056,
0.1255208105,
-0.3456110656,
-0.2916095257,
-0.1444859803,
0.4545297027,
0.1660124958,
0.0392153487,
-0.3247306943,
-0.1198163331,
-0.1323128045,
0.0058713648,
0.2860296369,
0.3430542648,
-0.0669392049,
0.103803359,
-0.0429319739,
-0.3748684525,
-0.133418411,
-0.1865022629,
0.1570679843,
0.0631487593,
0.4441577792,
-0.5209646821,
-0.0817442611,
-0.3450356722,
0.1652762741,
0.1584218442,
0.0572481602,
-0.0358496234,
0.1435464025,
0.1911142468,
-0.110761106,
0.2182203531,
-0.2996868491,
-0.2145742327,
0.0177829489,
-0.1605705023,
-0.2281498313,
-0.1316340417,
-0.0949162617,
0.2396892011,
0.2744668126,
-0.1647274792,
0.2371568978,
0.0634890795,
-0.0907146856,
-0.1422197223,
0.2349054515,
-0.4032539129,
0.3749279678,
0.4167662263,
-0.0743715093,
0.3681034744,
-0.1970899701,
0.1527628601,
-0.2274257839,
0.244900763,
-0.4916744828,
-0.4740644097,
-0.1170702055,
0.1085805371,
-0.1619236469,
-0.0288123153,
-0.3449947238,
0.0922386721,
0.3634018302,
0.1535674185,
0.160309732,
-0.091957286,
-0.417909503,
-0.2289614975,
-0.1571573466,
0.1740284562,
-0.4291956425,
0.1551632881,
-0.1355956644,
0.186601162,
0.0713288933,
0.3658941388,
-0.1201933175,
0.0764700174,
-0.0518554561,
-0.1532370597,
0.0406179279,
-0.1057318449,
-0.66488415,
0.1029135287,
0.3574776053,
-0.1740622818,
-0.0367922783,
0.4561774433,
-0.0024713986,
-0.0370600149,
-0.2303043157,
0.2907507122,
-0.2054569721,
-0.1454716921,
-0.3278049827,
-0.077661857,
0.0378576182,
-0.0154081918,
0.0659238547,
-0.2320688069,
-0.0371440277,
0.279774636,
-0.0108114518,
0.1934453994,
0.1148357466,
0.3457117677,
0.0314272642,
0.3540060222,
0.3015317321,
-0.021893885,
-0.3611462414,
0.250936985,
0.2351714373,
-0.7048806548,
0.1734371483,
-0.1321181953,
0.1448072195,
0.1432246715,
-0.0207026638,
-0.1719019711,
-0.0045521297,
0.087091282,
0.070075497,
-0.1784891188,
-0.018237114,
0.4595893621,
-0.025414519,
-0.1279369891,
-0.3539277613,
-0.1449450701,
0.0487825796,
0.0091998503,
-0.1538000703,
0.0313860402,
0.0623049997,
-0.07687819,
-0.1596943289,
0.1524796188,
0.2265722156,
-0.1928770989,
0.2952667177,
0.5493301749,
-0.0663551539,
0.1707331836,
0.3852482736,
-0.0702986419,
0.1017338559,
-0.1299524903,
-0.2927913964,
-0.0375095904,
-0.1519738585,
0.1403202564,
-0.0249341577,
-0.0292974152,
0.0416711122,
0.0892700702,
0.3436174691,
0.0104554817,
0.1606140137,
-0.147038877,
0.2927566469,
0.4069282115,
-0.0240143165,
0.2519696951,
0.4782898426,
-0.1528939158,
0.0289916098,
0.174427703,
-0.2387563884,
-0.5282304287,
0.1187322289,
0.3788323998,
0.4037414789,
0.1002400666,
0.3153688312,
-0.1507231295,
0.2983996272,
-0.1419316977,
0.0769499987,
0.080037348,
-0.0390974805,
-0.2749995887,
0.133496508,
-0.0680699497,
-0.0156765804,
0.1987347901,
-0.0829684436,
0.3320788145,
-0.0365246683,
0.3400944173,
-0.2261348963,
-0.1101818308,
0.0874780715,
0.3231964409,
-0.2059070468,
-0.2550525367,
-0.1116030812,
-0.0590510964,
0.2781716883,
-0.1644496471,
-0.1198911965,
0.4901141524,
0.0762804151,
0.1016939953,
0.0700768605,
-0.0708677322,
-0.3164552748,
0.1030517444,
0.1625662148,
-0.2699267864,
0.2397953719,
0.102919966,
0.0852553844,
-0.3361772895,
-0.151069209,
-0.1490524709,
0.119094193,
0.2485793233,
-0.2534004152,
0.0126693323,
-0.3917893767,
0.1810093075,
-0.2936203182,
-0.1614062786,
-0.039520856,
0.1628923267,
-0.1019611806,
-0.040759284,
0.1762523055,
0.0103804134,
-0.249156341,
-0.0023533776,
0.2122998238,
-0.1259873211,
0.0549068525,
0.2655704021,
-0.1126464456,
0.0555636324,
-0.192871049,
-0.0497125834,
-0.366995573,
-0.7784828544,
0.2513560653,
0.1395719647,
-0.1240602732,
0.169546932,
0.2248827368,
-0.0192548037,
0.1598344445,
-0.7960985899,
0.0848228931,
-0.2673852742,
0.2629159093,
0.0260322131,
0.0380784497,
0.236024633,
-0.1774862111,
0.0284533091,
-0.1528527141,
-0.2566244006,
-0.3062722981,
0.136716336,
0.4193833172,
-0.2310950607,
-0.0271235369,
0.3263311386,
0.5199463367,
0.4282686412,
0.0144179026,
-0.057179749,
0.3847503066,
0.5536596775,
-0.3887668252,
-0.0796086341,
0.2006555796,
-0.0727027282,
-0.073162131,
0.3296690583,
0.340688765,
-0.1232188419,
0.2847548425,
-0.0813004002,
0.0937900916,
0.1011780947,
0.1988084912,
-0.267899245,
0.1599435359,
0.1861112416,
-0.1608432978,
-0.0656141192,
-0.3909839988,
-0.0692704245,
-0.0309385508,
0.0783749893,
0.2683599591,
-0.2041403651,
0.0029260889,
-0.0678282604,
0.1011186689,
-0.0287855677,
-0.0061646719,
-0.083876051,
0.3064676225,
0.2835107148,
0.279494822,
0.3214493394,
-0.2633184493,
-0.0617115274,
-0.1646201909,
-0.1650078297,
-0.1018128321,
0.0064615682,
-0.0108330175,
0.2996637523,
0.0837251693,
0.086234048,
0.3078322411,
-0.185203135,
0.0375440121,
0.4310305119,
-0.3830850124,
-0.3002845049,
0.041409485,
-0.2625497878,
0.2326452583,
-0.1294326782,
-0.1187387407,
-0.2153357416,
0.071496807,
0.0059645101,
-0.2001737654,
0.136502564,
0.1608420163,
-0.1547915041,
0.193230778,
0.218178317,
0.1120680124,
0.1524929404,
0.1146081239,
0.0978648961,
0.2603714466,
-0.1108558178,
0.2093981355,
0.0343882032,
-0.0138658769,
0.1596278548,
0.0138397552,
-0.1994663775,
-0.1868091524,
-0.3193315566,
-0.0532792397,
-0.1618551612,
0.2261721641,
0.0621842891,
0.264536202,
-0.2653851211,
-0.5795580149,
0.4742962718,
0.3399648368,
-0.1296412796,
0.2579102814,
-0.2862508893,
0.0161555633,
-0.0667423829,
0.0111747514,
0.5658613443,
-0.1947551668,
0.3254500031,
-0.0794114023,
-0.0616297722,
0.3023921251,
-0.4567988515,
-0.0652407408,
-0.0468614586,
-0.3045466244,
-0.0021181144,
-0.0720990747,
-0.1411137134,
-0.0199837256,
-0.0946819857,
0.3443315625,
-0.1653455645,
-0.0544740707,
0.1015558094,
0.1601575464,
0.0502449349,
-0.125617668,
0.0404087529,
0.2785876691,
-0.1807656735,
-0.1113734171,
-0.0539936647,
0.1307122409,
0.086297825,
0.2149006128,
-0.3181108236,
-0.2407791913,
0.0139594451,
0.3552248776,
-0.0003477484,
-0.1252679825,
-0.0668669418,
0.1332378685,
0.2062505931,
0.2343956232,
-0.144988969,
0.3111361861,
-0.1219342351,
0.1470760405,
0.2458629459,
-0.1360152066,
0.118488729,
-0.0817847922,
-0.0365063325,
-0.0434313975,
0.0705960616,
-0.2296570241,
-0.1445900798,
0.2853673995,
0.1721631885,
-0.0119023435,
-0.2052931637,
0.0354665518,
0.0315031931,
0.0794769377,
0.1419689506,
0.3844062984,
-0.0609967187,
0.5599356294,
-0.2562155426,
-0.4541350007,
-0.0462831706,
0.0224626958,
0.5034425259,
-0.2178083211,
0.2553484738,
-0.0312378556,
-0.1767934114,
-0.0469375923,
0.2533655763,
0.0601818562,
-0.0204425305,
0.156786859,
-0.1809252501,
0.2217563987,
-0.141422689,
-0.4298369288,
0.0265338235,
-0.3845815361,
-0.2996764779,
-0.1156656817,
-0.430551827,
-0.0555205122,
-0.2334783971,
0.196934551,
0.0284529328,
-0.4536215663,
0.1628563553,
0.4174408019,
-0.2616335452,
0.0988546535,
0.1830000728,
0.267996341,
0.0161527507,
-0.143922627,
0.1268777102,
-0.1616188139,
-0.1091143191,
-0.1418681145,
-0.1457023621,
-0.260951966,
0.0503481776,
0.1333000809,
0.1298935264,
-0.0771051943,
-0.1611853838,
-0.3070394099,
0.0476321802,
-0.2632959187,
0.0594224185,
0.3452681303,
0.2517391145,
0.0951140225,
0.1064111516,
-0.3073003888,
0.1278259605,
0.3223635256,
0.0648429841,
0.2507064641,
0.0393301733,
0.2024416327,
0.1677876115,
-0.262960732,
-0.3952641785,
0.1813609898,
-0.0667772591,
-0.1152082086,
0.1839222312,
-0.2410745025,
0.209315747,
0.4576961398,
-0.0930894464,
0.1318604648,
-0.0612673312,
-0.2261503339,
0.0325868577,
0.1872492284,
0.2583865821,
-0.177110061,
0.0614770167,
0.4482592642,
-0.2375609279,
0.0974707007,
0.1723157763,
-0.0793060288,
0.0926528797,
0.1511341631,
0.371278137,
-0.3484844863,
0.1774657965,
0.1149258465,
0.1882869899,
-0.1611397564,
0.4405623376,
0.465023905,
0.0315323137,
0.5609920621,
0.1833840311,
0.3716111183,
0.5011680722,
-0.191582948,
-0.1141740978,
-0.2393338084,
-0.2015490532,
0.4385491014,
-0.1161763966,
0.0592845604,
-0.0714481771,
0.2083077729,
0.1838031113,
-0.3680932224,
-0.1831096411,
-0.1217205301,
-0.3449109197,
-0.1100354642,
0.4735720754,
0.0752968341,
0.1437613368,
0.2953135371,
-0.0625443161,
-0.3859571815,
0.4411939681,
-0.014913857,
-0.1918067336,
-0.287687391,
0.2943648994,
0.2762217522,
-0.1601940542,
-0.1219864041,
0.2891837955,
-0.2347112,
-0.0451047085,
-0.0269340966,
0.133080259,
0.219763577,
0.0857369825,
0.0260295905,
0.3150085807,
0.0979131088,
-0.1002510786,
-0.2087868452,
0.2268521786,
-0.2361114919,
-0.0908071697,
-0.0861887485,
0.1070382819,
0.0798825473,
0.2771604061,
-0.0013056062,
0.2785951197,
-0.2990992665,
-0.1270031631,
0.0565833896,
-0.0288753137,
-0.1554897875,
0.3027584553,
-0.3916991949,
-0.2383485734,
0.6524432302,
-0.2137703896,
0.0422888473,
-0.1606044024,
0.10845007,
-0.2349987328,
0.4478150308,
0.4689293504,
0.0369881317,
-0.3087179661,
-0.0269651115,
-0.2480549812,
0.0657561943,
-0.3029732406,
0.4378801584,
-0.0127350129,
0.5008689761,
-0.2398173958,
-0.0382045768,
-0.0074691847,
-0.0681987256,
-0.0351665728,
-0.1318751574,
-0.3061082959,
-0.0137973754,
0.1377978176,
0.1543894559,
0.0681298077,
-0.2035730928,
0.2141355872,
-0.2355101705,
-0.0289052315,
-0.154102236,
-0.0110404938,
0.1529419124,
0.406737119,
0.086468026,
0.125770092,
-0.0493975244,
0.1526637375,
0.15464288,
-0.391433537,
-0.0762773156,
0.0519484878,
-0.0218605436,
0.0463793352,
0.0035661533,
-0.5463690758,
0.2738309205,
0.1344443262,
0.15156582,
-0.1615917683,
0.0087531321,
0.1073765233,
-0.383451879,
-0.2046896815,
0.3414938152,
0.3123236895,
0.1437022537,
-0.0722698718,
-0.0394231081,
-0.3186281323,
-0.0789776668,
0.3032260537,
0.0869204253,
-0.4065344334,
-0.0037894621,
0.1015674695,
0.5180825591,
-0.1703475267,
-0.3908230662,
0.1476501971,
0.4772285521,
-0.1026957706,
-0.1970788538,
0.2607144713,
-0.1244607568,
0.2009637505,
-0.0636365712,
0.5886346102,
-0.059142381,
-0.0374853089,
0.2227341831,
-0.1312394738
] |
https://github.com/huggingface/datasets/issues/1965 | Can we parallelized the add_faiss_index process over dataset shards ? | Hi !
As far as I know not all faiss indexes can be computed in parallel and then merged.
For example [here](https://github.com/facebookresearch/faiss/wiki/Special-operations-on-indexes#splitting-and-merging-indexes) is is mentioned that only IndexIVF indexes can be merged.
Moreover faiss already works using multithreading to parallelize the workload over your different CPU cores. You can find more info [here](https://github.com/facebookresearch/faiss/wiki/Threads-and-asynchronous-calls#internal-threading)
So I feel like the gains we would get by implementing a parallel `add_faiss_index` would not be that important, but let me know what you think.
| I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
| 79 | Can we parallelized the add_faiss_index process over dataset shards ?
I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
Hi !
As far as I know not all faiss indexes can be computed in parallel and then merged.
For example [here](https://github.com/facebookresearch/faiss/wiki/Special-operations-on-indexes#splitting-and-merging-indexes) is is mentioned that only IndexIVF indexes can be merged.
Moreover faiss already works using multithreading to parallelize the workload over your different CPU cores. You can find more info [here](https://github.com/facebookresearch/faiss/wiki/Threads-and-asynchronous-calls#internal-threading)
So I feel like the gains we would get by implementing a parallel `add_faiss_index` would not be that important, but let me know what you think.
| [
-0.2789407969,
-0.0627322569,
-0.1573459953,
0.1106121317,
-0.3771444559,
0.247851491,
0.2097890675,
0.1062705666,
0.1371022463,
0.171924904,
-0.1766026765,
-0.1353845447,
0.3414870501,
0.1139963046,
-0.3853699863,
0.2495388687,
0.3513413668,
-0.1033919007,
0.2524386942,
0.1125719249,
-0.317397058,
0.0103291739,
0.0291405618,
-0.2674215436,
-0.1780146658,
0.3508557081,
-0.2582990527,
0.0532394908,
-0.0132864472,
-0.2483888716,
-0.2480451167,
0.3926289082,
0.0679369122,
0.3004919589,
-0.0001269445,
-0.0502258763,
0.1069197953,
0.0691177696,
0.1533452868,
0.5967574716,
-0.2711214423,
0.0120971948,
-0.0762948841,
-0.1729466021,
-0.0013870448,
-0.3030719757,
-0.0339292064,
-0.2620636821,
-0.2039723396,
-0.1719794571,
-0.0102490447,
-0.1237888187,
0.0848586932,
0.0481444933,
-0.0428052247,
-0.2038854212,
-0.1553706229,
-0.1455568075,
0.0888654962,
0.2237184644,
0.232163623,
0.1709750593,
0.018992424,
-0.1704600602,
0.0877620876,
-0.0448631272,
0.4525161982,
-0.052651342,
0.0105196871,
-0.1225786358,
0.1086314172,
0.04548366,
-0.4651238918,
-0.2125894427,
0.0684299171,
0.0824428871,
-0.227445364,
-0.1753317118,
0.0390137248,
0.0092718313,
0.0714064389,
-0.4730632901,
0.0494287908,
0.1115151644,
0.3911833763,
0.0608310103,
0.2611268759,
0.1117506251,
0.4002322853,
-0.1047058105,
0.026290644,
-0.1128798351,
-0.096388869,
0.1314449161,
-0.6142693162,
0.0232877433,
-0.0899240226,
-0.2996662855,
0.0209421366,
-0.1793349385,
-0.3022097647,
0.2885175347,
0.0202774294,
0.1198161244,
0.0016667452,
-0.153652519,
0.2066203356,
-0.2501231432,
-0.0125040971,
-0.4881719053,
-0.1144801006,
0.1374481618,
0.1478853077,
-0.1201160625,
-0.6300257444,
-0.143997103,
-0.3549610078,
-0.1606155336,
-0.0620710254,
-0.1819743961,
0.0238425508,
-0.0995016396,
0.1697280258,
0.1614187807,
0.33774966,
-0.0016711829,
-0.1338704973,
0.0112253074,
-0.0672988594,
-0.0540431701,
-0.0022991113,
-0.279543817,
0.2434512377,
0.5302838087,
-0.0042611733,
-0.2697484493,
-0.4165175557,
-0.0278765112,
-0.076470077,
0.1563664824,
0.0336073041,
-0.156120792,
0.033615604,
0.0777927786,
-0.0235537142,
0.0020427927,
-0.0585453063,
0.1687940955,
-0.2429046929,
0.1656703353,
-0.2159647048,
-0.3029149175,
0.0556755736,
0.0054090507,
-0.0320964381,
-0.1371086985,
-0.1657650173,
0.5699170828,
0.176148057,
-0.0432699807,
-0.0384706408,
0.1838903129,
-0.3282158971,
-0.1131457686,
0.1609258056,
0.1564362347,
-0.2961854339,
-0.2146450728,
-0.1436222196,
-0.1857853532,
0.0067236982,
0.570730865,
-0.0211950671,
0.0817072615,
-0.287550211,
0.5438027978,
0.2934737504,
-0.281566143,
0.0505188257,
0.0885652825,
-0.0611448251,
-0.165310964,
0.0520845987,
0.1922506243,
0.117928341,
-0.0568650216,
0.2276798934,
0.294465065,
-0.0839011818,
-0.1194115132,
-0.261570096,
-0.3065007925,
0.4936807454,
0.2430091649,
-0.1071244925,
0.076108858,
0.0439903885,
-0.7627868056,
0.2860036492,
-0.1323407292,
0.1944721341,
0.0383246839,
0.2891152203,
0.4427100122,
0.4458393157,
-0.1140969396,
0.0472652689,
0.2294324934,
-0.1534682214,
-0.0019362019,
0.3032391667,
-0.3450818062,
0.4740671217,
-0.0507176965,
0.0305638965,
0.3955186307,
-0.0676675364,
-0.211754024,
0.1425756514,
-0.3888503313,
-0.3314332962,
0.4658035636,
-0.3955697417,
-0.1659501493,
-0.0724602416,
0.2263809144,
-0.0312095452,
-0.0337084718,
-0.3186896443,
0.0959290564,
0.1591542512,
-0.2405870855,
-0.1059323251,
0.3029584885,
0.0247301385,
0.037686564,
0.9103008509,
0.222501725,
-0.0706847906,
0.2421090603,
0.0582000017,
-0.2595221698,
-0.1585792303,
-0.1893134713,
0.3330983818,
0.0545156971,
0.0887113214,
0.5726434588,
0.002658166,
-0.172075808,
0.1380843967,
0.1619299203,
0.0660991967,
-0.0243275762,
-0.0152290314,
-0.1266033649,
-0.1130961701,
0.0886522159,
-0.0277292356,
0.591016829,
-0.2684079409,
-0.0822855607,
-0.1726531684,
-0.0034399182,
-0.0326990448,
0.0887020305,
-0.0420209207,
-0.1948681176,
0.3121992052,
0.1168775409,
0.0520308726,
-0.1584306359,
0.3804951012,
-0.1136476994,
-0.0397200957,
0.3633241057,
0.2026494741,
0.0750138834,
0.1684558839,
-0.0580856763,
-0.0571017042,
-0.1315047145,
0.0688442588,
0.0319427177,
-0.0110608339,
-0.0850689709,
0.1229411066,
0.0800903887,
-0.149191916,
-0.2043134868,
-0.396910727,
-0.1968984902,
0.2237752974,
0.0126257399,
-0.2535369992,
-0.0580512658,
-0.0775192678,
0.5329557061,
-0.0946547836,
-0.2684037387,
-0.3124782741,
-0.0170469806,
0.1478615254,
-0.1259402037,
0.2970300615,
0.188792482,
-0.0532314181,
0.118851155,
-0.0868266672,
-0.3709445,
-0.093497619,
-0.1448416859,
-0.0173405334,
-0.1443890482,
-0.2989168763,
-0.2838766277,
-0.142950058,
-0.1644758135,
0.1537099183,
-0.0450635627,
-0.0964550376,
-0.2686015368,
0.0291600376,
0.0496851727,
0.1025892794,
0.0748393536,
-0.3937213421,
-0.2396275997,
0.344915241,
-0.061861407,
0.0971539617,
-0.353382647,
-0.2882669568,
-0.1973517388,
-0.2623564601,
-0.2343727946,
-0.3656173348,
0.020294711,
0.1619838923,
-0.1117466986,
-0.1317775249,
0.0018889531,
0.0815020502,
-0.2325873673,
0.3353891373,
0.0072120428,
0.0734023452,
-0.0066112755,
0.1284047961,
-0.2605767846,
0.1623671949,
0.1447144598,
-0.1115815118,
-0.0068715513,
0.0925001055,
0.1554603875,
0.2978828847,
0.1614468098,
0.1604870856,
-0.1416186541,
-0.0257232785,
-0.156770885,
0.7340025902,
0.2577257752,
-0.0254652984,
0.1525012404,
0.1177449226,
0.036469046,
0.064416565,
-0.1403584182,
0.2962309718,
0.0880493373,
-0.1183748096,
0.0249202251,
0.0486862846,
-0.0981475636,
0.044014886,
-0.1993702352,
-0.1111701056,
-0.1187805533,
-0.0568946227,
0.2207625061,
0.5126087666,
0.0025172755,
-0.05494681,
-0.2808341384,
0.009422455,
0.1832814366,
0.362015903,
0.499268055,
-0.1927378625,
0.1517202854,
-0.2025241554,
-0.4655699432,
0.4637283087,
0.0087807439,
0.28163746,
0.0165973455,
-0.0447911061,
0.2494056821,
-0.1914139092,
0.890162766,
0.0034870934,
-0.4772202969,
0.1134793684,
-0.0452142656,
-0.3266535103,
-0.2326373607,
0.3000526726,
0.2172366381,
0.4511128068,
0.2308521867,
-0.2460669875,
-0.3018375337,
0.1756524146,
0.0751907527,
-0.2882428765,
-0.5182566643,
-0.1469024122,
0.0661564842,
-0.0823277161,
0.1156658381,
-0.2963604927,
0.113991946,
0.0361340567,
0.0092627257,
-0.1929906011,
0.0298597813,
-0.0371093899,
0.0602861419,
0.0510644242,
0.1120231748,
0.4447423518,
0.3219394386,
-0.2890281379,
0.2662294805,
0.0822312534,
0.3286916018,
0.0109017231,
-0.0902362019,
-0.0044669732,
0.2150543332,
0.0719433874,
0.0639621317,
0.2853553295,
-0.1218381226,
0.2022845,
-0.2137943208,
0.129850775,
0.1342016011,
0.2930413187,
-0.1915014237,
-0.1733280718,
0.4962486923,
0.0442027971,
-0.1058536395,
-0.0767579004,
0.4734345078,
-0.0099867769,
0.7146953344,
0.045512598,
0.7037386894,
-0.1248853058,
-0.152659446,
-0.2615528107,
0.115844965,
0.04207303,
-0.6474259496,
0.3106972277,
-0.1907191873,
-0.2052682042,
0.1221540645,
-0.1048921496,
0.1149578542,
0.3697928488,
0.2797478437,
0.1120954454,
0.2166123092,
-0.2035364956,
-0.2071073651,
0.3400915861,
0.1826964021,
-0.2283104956,
0.2803764939,
0.0391108841,
0.0797611028,
-0.0306655802,
0.0226752497,
0.0615450218,
-0.1580806077,
0.3526021838,
-0.1197993159,
-0.0483020023,
0.3265137672,
0.2986388803,
-0.2302100062,
-0.0455088317,
0.1209436655,
-0.5679506063,
0.227529496,
-0.2574239373,
-0.1487643272,
0.104451336,
0.1428869963,
-0.1538692862,
-0.0604205728,
-0.5270249844,
-0.0879807994,
0.0198986903,
-0.3047128022,
-0.190198198,
0.2688497901,
-0.8522256017,
-0.3888075352,
0.0930701718,
-0.0728900209,
0.0655892342,
-0.1263833344,
0.5215659142,
-0.191291675,
0.0358547568,
-0.0476521403,
0.2592446208,
-0.4425132871,
0.7070085406,
-0.3910199702,
-0.0085430443,
-0.0221072342,
0.2433185875,
0.2357964367,
-0.3659249246,
0.0845106244,
-0.0970674455,
-0.1915615648,
0.039833948,
-0.5570602417,
0.090536423,
0.4957044721,
-0.0826073885,
0.0287381858,
-0.1235793605,
0.0528356284,
-0.189090848,
-0.1416685283,
-0.0433611646,
-0.60220474,
-0.1440007091,
-0.2501828969,
0.0947243646,
0.2008906454,
0.3670533895,
0.1365481168,
0.1150723919,
-0.3318487406,
0.1516891569,
-0.1088537201,
-0.0053226128,
0.1221116483,
0.0959044248,
-0.0204846226,
-0.0927880853,
0.1320216656,
-0.0813644156,
-0.0900849178,
0.1627067327,
0.1668039262,
-0.0654663816,
0.2371112108,
0.2261743695,
0.3129310608,
0.1566812247,
-0.1954314709,
-0.2620937824,
-0.1907633394,
-0.2348840088,
0.0444746614,
0.0386032611,
-0.2214899659,
-0.0736929253,
0.386351943,
-0.26859954,
0.1082159728,
0.5082783103,
0.4393991828,
0.4513735771,
-0.002762381,
0.5174337029,
-0.2549319267,
0.0965868831,
0.1337030381,
0.049999021,
0.2449611723,
0.0504372604,
0.026285531,
-0.0082347635,
-0.2804997861,
-0.0818449035,
0.4896749556,
0.6631748676,
0.2678459287,
-0.2227075696,
-0.0256316997,
0.007558086,
-0.0481432453,
-0.0765163451,
0.839158237,
-0.1111107767,
0.1397633255,
0.1743853688,
0.3427312374,
0.3477388024,
-0.2052406222,
0.0624891818,
0.0630102381,
0.3767135441,
-0.149923861,
-0.6269668341,
0.2316365689,
0.0486398153,
0.2675713003,
-0.1535963118,
0.1240842938,
0.7165522575,
-0.0639143363,
0.4785978198,
0.0993159935,
-0.0497427583,
0.1411429942,
-0.1296243817,
0.4695599675,
0.1944606006,
0.3647834659,
0.1691024154,
0.1554941237,
0.2924837768,
-0.0987273231,
-0.5738970637,
-0.1505663097,
0.4674294591,
-0.0120793469,
-0.0454110652,
0.4208976626,
0.041887477,
-0.0121289194,
0.0280938298,
0.1323243529,
-0.1338591576,
0.1945985854,
-0.0022146609,
0.1079651639,
0.2100122571,
-0.0201813094,
0.4173364639,
0.4710710943,
-0.0277035907,
-0.2999328673,
-0.0784053952,
-0.0858178288,
0.1237093881,
0.1651204526,
-0.0054092743,
0.1638244241,
-0.0103427954,
0.3054115772,
-0.0628256127,
0.1377596557,
0.0035016239,
-0.0520883612,
-0.3109495044,
-0.1872945577,
0.2082956731,
-0.0313969068,
-0.0078561082,
-0.572486639,
-0.1981352866,
0.4705940485,
-0.1560833752,
-0.0974817127,
0.0979626775,
-0.0830199271,
-0.0531942956,
-0.0803229809,
-0.0383731425,
-0.2392766178,
0.072902143,
-0.4864425063,
-0.2075452656,
0.1342897117,
-0.0296102576,
-0.1673309058,
0.3975628614,
0.231882602,
0.5182492733,
-0.1991004795,
-0.180796206,
-0.2487400174,
0.142068699,
-0.1421350092,
0.3399438858,
0.0479398109,
0.5490871072,
0.0789262578,
-0.0603774488,
0.6871288419,
0.1367731839,
0.1093955189,
0.4636112452,
-0.1878225803,
0.3570120335,
-0.3443447948,
-0.2847063243,
0.1339999437,
-0.0340753347,
0.4974753559,
0.2011548281,
-0.2478418648,
0.2226831317,
-0.2510334253,
0.3260991573,
-0.4909701049,
0.5764993429,
0.0635077059,
0.1297810823,
-0.0798647776,
0.0456419513,
-0.1233187169,
-0.2051603794,
-0.3342982829,
0.0662147403,
-0.3860464692,
-0.2056216896,
-0.2810482085,
-0.159193784,
0.0470135696,
0.2368544489,
-0.1396907866,
-0.199049741,
-0.0603567474,
0.0884798914,
-0.1346511543,
0.2249331474,
-0.0805846825,
0.1380121708,
0.1908539236,
0.2626538873,
-0.2852627635,
0.2808655202,
0.2794203758,
-0.1816103607,
-0.0213589072,
-0.4958339632,
0.1355200112,
0.4050750732,
0.1090613455,
-0.5247434974,
-0.5293172002,
0.1268745661,
0.3336170912,
0.2443784326,
0.1576968729,
-0.0107488334,
-0.3092318475,
-0.2780919671,
-0.3314501941,
0.003730366,
0.1973459423,
0.0342445672,
-0.0339608267
] |
https://github.com/huggingface/datasets/issues/1965 | Can we parallelized the add_faiss_index process over dataset shards ? | Actually, you are right. I also had the same idea. I am trying this in the context of end-ton-end retrieval training in RAG. So far I have parallelized the embedding re-computation within the training loop by using datasets shards.
Then I was thinking of can I calculate the indexes for each shard and combined them with **concatenate** before I save. | I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
| 60 | Can we parallelized the add_faiss_index process over dataset shards ?
I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
Actually, you are right. I also had the same idea. I am trying this in the context of end-ton-end retrieval training in RAG. So far I have parallelized the embedding re-computation within the training loop by using datasets shards.
Then I was thinking of can I calculate the indexes for each shard and combined them with **concatenate** before I save. | [
-0.2849658728,
-0.1696968526,
-0.0912608504,
0.1680026501,
-0.3235112429,
0.3758882284,
0.3347997665,
0.0697202533,
-0.035886284,
0.1546953619,
-0.0876063406,
0.0618726723,
0.382417202,
-0.0200758837,
-0.3032028377,
0.1434957981,
0.2809821367,
-0.058464475,
0.2126169205,
0.0394793004,
-0.3338528275,
-0.1382641941,
-0.0715477318,
-0.2807033062,
-0.2238433808,
0.1862553358,
-0.3562580645,
0.1115738973,
-0.0008238852,
-0.3895473778,
-0.1900934577,
0.3138127923,
0.2559098005,
0.0299704354,
-0.0001277621,
-0.0997195989,
0.1779813319,
0.0183814839,
0.1161236465,
0.6891212463,
-0.2972024679,
-0.0267106555,
-0.0404041111,
-0.2445940077,
0.0822644085,
-0.2505578399,
0.0011063237,
-0.2517291903,
-0.070240438,
-0.2690674663,
-0.0466796719,
-0.0607912354,
0.0575927421,
0.0574584827,
0.0406847186,
-0.3079021275,
-0.1213591397,
0.1398367584,
0.0839519277,
0.2245054841,
0.1343082637,
0.0507924259,
-0.047738038,
-0.160002768,
0.0726076216,
0.1129785925,
0.2452796847,
0.1347254664,
-0.0512391441,
-0.1212150976,
0.2552976608,
-0.0992789268,
-0.534969449,
-0.1582326889,
0.2663682699,
-0.1595628858,
-0.14388071,
-0.1659990698,
0.1047990769,
0.0622216202,
0.0662733912,
-0.5065564513,
-0.0104385614,
0.0325588211,
0.4056908488,
0.0920153037,
0.2955592573,
0.1298530102,
0.3458769917,
0.0445575863,
0.0997550711,
-0.2029112428,
-0.1997180879,
0.1131249517,
-0.6591155529,
-0.0331773236,
-0.2063940465,
-0.222866565,
-0.1241203845,
-0.0351021662,
-0.1060761809,
0.1871547997,
0.2389025986,
0.1609723121,
-0.0058842171,
-0.0895847976,
0.0343843661,
-0.1298586428,
-0.1368225515,
-0.4281221628,
-0.1075875983,
0.0488342494,
0.03502528,
-0.195886746,
-0.5263499022,
-0.0633083731,
-0.4865528345,
-0.0719931424,
-0.1026423648,
-0.173439905,
-0.1677079797,
-0.1291919351,
0.1297935843,
0.0206168033,
0.2822861075,
-0.1592238545,
-0.1701835245,
0.0673739836,
-0.0559868924,
0.0045556277,
-0.0279973634,
-0.2161832601,
0.0174258798,
0.512678206,
-0.0436440706,
-0.2863792181,
-0.3827868402,
0.0505876504,
-0.0481581762,
0.1748678982,
0.0664139017,
-0.2154344916,
0.1209252179,
0.0772427917,
-0.1290804893,
-0.1677449495,
-0.0365751684,
0.2454081774,
-0.1966933906,
0.2098247558,
-0.2311401367,
-0.2099299431,
0.1646125764,
-0.0386469327,
-0.0348178372,
-0.1441637427,
-0.0446504653,
0.3352723122,
0.3445594609,
-0.0064369142,
0.0062948428,
0.1591221243,
-0.2874040604,
-0.1396419704,
0.2494934201,
0.2338721752,
-0.4014921188,
-0.1169342324,
-0.0487823971,
0.0194940567,
0.1672156751,
0.5547193885,
-0.1314323097,
0.1645195782,
-0.3813927174,
0.5477818251,
0.3399351239,
-0.2826058865,
-0.0182867497,
0.032841932,
-0.1953347772,
-0.096134223,
0.2637236416,
0.1584376693,
0.445999831,
0.0293705501,
0.2704941332,
0.2283855826,
-0.0669301897,
-0.306327045,
-0.3454910219,
-0.2478901595,
0.2212373912,
0.1413298845,
-0.0717565566,
0.1985465735,
0.0511374846,
-0.8309345841,
0.3635240793,
-0.0840670913,
0.2159190476,
-0.0107425004,
0.209240064,
0.3811722696,
0.3770608306,
-0.0781181157,
0.0771490037,
0.1599860936,
-0.0280745067,
-0.0409756228,
0.2488017678,
-0.3538299799,
0.4831634164,
-0.075630635,
0.0241907723,
0.468919009,
-0.1199092194,
-0.206251651,
0.062198326,
-0.3314238787,
-0.2941706181,
0.5498024821,
-0.4779761434,
-0.0968442261,
-0.1887804419,
0.2802515924,
0.0779901519,
-0.1488749981,
-0.281652987,
0.0475950614,
0.0733576417,
-0.2442057729,
-0.0596472472,
0.2386412919,
-0.0809266716,
0.1019928455,
0.8778347373,
0.1095693335,
-0.154157564,
0.0711276382,
0.1545339078,
-0.3248101175,
-0.0852529705,
-0.2058343291,
0.3224862814,
0.2268129289,
0.0249039754,
0.360970974,
0.0308593884,
-0.0908336788,
0.0195496902,
0.023493275,
0.0144609883,
0.0800580978,
-0.1024841517,
0.0013244487,
-0.0705541,
0.0450994596,
-0.1120693386,
0.3265724182,
-0.1385029852,
-0.1785278022,
-0.2923911214,
-0.0533653386,
-0.0340082981,
0.1097712815,
-0.0718639642,
-0.153002128,
0.3807484806,
0.1627698988,
0.1704763472,
-0.129546538,
0.2641077936,
-0.194244951,
0.0243813619,
0.2502627075,
0.0645522177,
0.2118135393,
0.0696622655,
-0.0877097696,
-0.0326043777,
-0.0260463543,
0.0928728208,
0.0639977232,
-0.0158801973,
-0.0416263267,
0.0864126757,
0.3875236511,
-0.3104660511,
-0.11043863,
-0.2621285021,
-0.1685070395,
0.1309842169,
0.0351331942,
-0.1759459972,
-0.0140688382,
-0.0472524166,
0.5214402676,
-0.1426889002,
-0.3105711341,
-0.1479192674,
-0.1127291322,
0.1196700484,
-0.1296889186,
0.2577826679,
0.1593426019,
0.1317601055,
0.1190094501,
-0.0478239879,
-0.282017082,
0.0005475581,
-0.170217365,
0.0225035511,
-0.2009868622,
-0.3025680184,
-0.2304278612,
-0.3104278445,
-0.1213894337,
0.0695400015,
0.1559575051,
-0.0151107945,
-0.2252910435,
0.1130051464,
0.0651466027,
0.1385293007,
-0.1371592879,
-0.3187465072,
-0.2540183961,
0.3161886334,
-0.1976725906,
-0.0086672362,
-0.4010063708,
-0.1311910748,
-0.1447111964,
-0.194441244,
-0.2484068274,
-0.3096930087,
-0.1726857573,
0.1525400728,
-0.02631131,
-0.0767534003,
-0.0058417767,
-0.0332635865,
-0.1281507909,
0.6561335325,
-0.0107419267,
0.2173005939,
0.0476655588,
0.0933470279,
-0.2506474257,
0.2637112141,
0.2129896879,
-0.0281586926,
0.0079064295,
0.1556269228,
0.2947205603,
0.3695466518,
0.0974455774,
0.0661058649,
0.0408572592,
0.0288638733,
-0.193277806,
0.9015143514,
0.2490782589,
-0.1850937009,
0.1608606428,
0.259434551,
0.149526298,
0.0079313964,
-0.1160741299,
0.2399918735,
0.1435939968,
-0.1371480376,
0.000988856,
-0.0364588797,
-0.0945906416,
0.1096181273,
-0.201295495,
0.0115332976,
-0.1715347469,
-0.0033562742,
0.0544865653,
0.551669538,
0.0550486669,
0.0214145184,
-0.2025680989,
0.0437314548,
0.1283032149,
0.394572854,
0.586186409,
-0.2756181061,
-0.0562678948,
-0.2580189705,
-0.3013853133,
0.4053606391,
0.0661219954,
0.3546910286,
0.1334128082,
-0.1077655926,
0.3340643048,
-0.1155372411,
0.897454977,
-0.1313095242,
-0.5274404287,
0.1343752444,
0.0533275977,
-0.1458675861,
-0.2606738806,
0.302003026,
0.186032176,
0.3533154726,
0.1942868829,
-0.2180704772,
-0.2082677186,
0.331638664,
0.1343446374,
-0.2656058967,
-0.6529961824,
-0.0941178575,
0.1721437573,
-0.0474796966,
0.2200226337,
-0.3541426659,
0.0357389711,
0.0266457647,
-0.0205888078,
-0.3798433542,
-0.0183669738,
0.0105703548,
-0.0418418571,
0.0875433162,
0.0219415054,
0.515840888,
0.3603673577,
-0.040231429,
0.4263273478,
0.0963511616,
0.0793371871,
0.0130318664,
-0.1868698001,
-0.1968353093,
0.3067004085,
0.0957244188,
0.0654583275,
0.3111416698,
-0.2788874209,
0.063009724,
-0.3485800326,
0.0565941632,
0.1023263261,
0.4389438629,
-0.1225830168,
-0.1930823177,
0.4389567971,
0.0504135862,
-0.1514481157,
-0.0545305051,
0.3455349803,
-0.1894799173,
0.6812266707,
0.0664832518,
0.8806690574,
-0.1594644785,
0.0483082421,
-0.2123775035,
0.0975703225,
-0.0178219751,
-0.615981102,
0.4225198925,
-0.2377066165,
-0.1771939397,
0.1049715206,
-0.2092420757,
0.2017688155,
0.4122804999,
0.1896031797,
0.124285832,
-0.0764312297,
-0.099206537,
-0.1765964925,
0.2509700954,
0.2230387926,
-0.2394246161,
0.3446897268,
-0.0590550229,
0.2382807881,
-0.2323366404,
0.0191418901,
0.0264448151,
-0.1703358889,
0.3796362281,
-0.1086361483,
-0.0167272277,
0.4481121302,
0.1966138184,
-0.1731838286,
-0.2193172872,
0.2574956417,
-0.3695750535,
0.3011194766,
-0.2077326179,
-0.0317819193,
-0.0070876321,
-0.0048745926,
-0.2725214362,
-0.1024234369,
-0.3807837963,
-0.033125218,
0.1110462397,
-0.3711657524,
-0.166139245,
0.263186574,
-0.7294945121,
-0.4386248589,
0.1284068674,
-0.0531709716,
-0.1192603409,
-0.1365409046,
0.4132651985,
-0.1144073978,
0.0019170418,
-0.0689255372,
0.2600789964,
-0.4817503691,
0.8244271278,
-0.3634019494,
0.025421083,
0.0193000156,
0.2881211042,
0.3121021688,
-0.2579222918,
0.2277693748,
-0.1517081261,
-0.2348929644,
-0.0000380427,
-0.5809665918,
0.154198736,
0.4484095573,
-0.0926691368,
0.0182478242,
-0.0281314105,
-0.0096791647,
-0.4759540558,
-0.09658283,
0.0260912813,
-0.5716093183,
-0.1279110014,
-0.3554808199,
0.1013093591,
0.2063256502,
0.429942131,
0.0584561527,
0.1112198159,
-0.3822533488,
0.1869218349,
-0.1091271341,
0.0534862131,
0.1014933139,
0.2128514051,
0.1496152878,
-0.0973928571,
0.1051633209,
-0.0466014408,
-0.0639348626,
0.1806939095,
0.0985822082,
-0.0267343242,
0.2877895236,
0.2327820212,
0.1426940411,
0.1709013134,
-0.0853705257,
-0.3187398314,
-0.2249473482,
-0.2446721941,
-0.0246265084,
-0.0533237904,
-0.1381594688,
-0.0823490769,
0.4531779289,
-0.0884272605,
0.0370726064,
0.4527214468,
0.3983291388,
0.3977434635,
-0.0550010018,
0.4254107773,
-0.217027694,
0.004388161,
0.0188514329,
0.1060651541,
0.328820616,
0.0073819533,
0.1000649333,
-0.0175165124,
-0.3024097681,
-0.2244682312,
0.5101445317,
0.6108434796,
0.342923075,
-0.4377303123,
0.1114995852,
-0.0158239193,
-0.0398657136,
-0.1532559693,
0.7825092077,
0.0298473462,
0.1465705335,
0.1259588599,
0.2123567909,
0.2996018231,
-0.3592543006,
-0.0047609657,
0.1406217515,
0.3161542714,
-0.1753816009,
-0.4867199659,
0.3327390552,
0.0838658363,
0.1554300338,
-0.1178850755,
0.1441019773,
0.6216908693,
-0.1454122663,
0.5556602478,
0.0931108743,
0.0085588498,
-0.0513988025,
-0.3147472739,
0.3127995133,
0.415858686,
0.3478415906,
0.1352379918,
0.146241501,
0.3830259442,
0.2558974624,
-0.5684158802,
-0.1476950645,
0.4132968783,
-0.1333669573,
-0.1575538069,
0.3834910989,
-0.0540320426,
0.1045508534,
0.0788133293,
0.1858241558,
0.0929145813,
0.3751932383,
-0.0736510754,
0.2432723641,
0.2592492104,
0.0974980816,
0.3380098641,
0.2684297562,
-0.135172084,
-0.3136458993,
-0.1853096634,
0.0345402323,
0.1235909611,
0.0863998979,
0.0447230414,
0.1819359958,
0.1531484127,
0.2662103772,
-0.0655694604,
0.1450529993,
-0.0084778592,
-0.0248163268,
-0.2228884101,
-0.1458397061,
0.1872136891,
-0.0870499462,
0.0245116688,
-0.4995046556,
-0.1176993251,
0.3334581554,
-0.219505921,
-0.1744242311,
0.2647452652,
-0.3016048372,
0.0676541775,
-0.0331205614,
-0.1562538445,
-0.0631976351,
0.2823478281,
-0.361515969,
-0.2766747773,
0.2801628113,
-0.0707580894,
-0.3086366057,
0.4806895554,
0.3341833949,
0.5037376881,
-0.2653270364,
-0.1025480181,
-0.2228056788,
0.1971824169,
-0.1594052017,
0.2911323607,
-0.0202686582,
0.4730789661,
0.1639248729,
-0.1333213896,
0.5242289305,
0.0366240181,
0.0648769811,
0.5568503141,
-0.2519762814,
0.187348038,
-0.2975604236,
-0.32768628,
0.0573024452,
-0.1519956887,
0.4205701649,
0.1887060851,
-0.2505376339,
0.1852189898,
-0.0281151012,
0.3232851028,
-0.2908660769,
0.529887557,
-0.020297382,
0.2953536808,
-0.0530071147,
0.0754796937,
-0.1315186173,
-0.1362612545,
-0.3146403432,
0.0852031708,
-0.2904396355,
-0.0889272541,
-0.4956980646,
-0.1259973049,
-0.0269063115,
0.132665351,
-0.0939805135,
-0.1971266419,
-0.0234794207,
0.104743734,
-0.0522617847,
0.2000952661,
0.0460238792,
0.2374134064,
0.1900678277,
0.2453595996,
-0.2602974474,
0.3165749609,
0.289721489,
-0.1897232234,
-0.060159184,
-0.3615054488,
0.0874477923,
0.3728884459,
0.1571474969,
-0.5834989548,
-0.4987612069,
0.2187901735,
0.2269465476,
0.1632259637,
0.2620421648,
0.3271627426,
-0.2740887702,
-0.2698302269,
-0.5132446885,
-0.0430746861,
0.0780404285,
-0.1016155556,
-0.2167234719
] |
https://github.com/huggingface/datasets/issues/1965 | Can we parallelized the add_faiss_index process over dataset shards ? | @lhoestq As you mentioned faiss is already using multiprocessing. I tried to do the add_index with faiss for a dataset object inside a RAY actor and the process became very slow... if fact it takes so much time. It is because a ray actor comes with a single CPU core unless we assign it more. I also tried assigning more cores but still running add_index in the main process is very fast. | I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
| 72 | Can we parallelized the add_faiss_index process over dataset shards ?
I am thinking of making the **add_faiss_index** process faster. What if we run the add_faiss_index process on separate dataset shards and then combine them before (dataset.concatenate) saving the faiss.index file ?
I feel theoretically this will reduce the accuracy of retrieval since it affects the indexing process.
@lhoestq
@lhoestq As you mentioned faiss is already using multiprocessing. I tried to do the add_index with faiss for a dataset object inside a RAY actor and the process became very slow... if fact it takes so much time. It is because a ray actor comes with a single CPU core unless we assign it more. I also tried assigning more cores but still running add_index in the main process is very fast. | [
-0.4054988623,
-0.1552558839,
-0.1257977784,
0.183027491,
-0.305123955,
0.2511135936,
0.3391773105,
0.0727012604,
0.1347902417,
0.1768686175,
-0.1435729563,
0.196087867,
0.350476414,
0.0879318863,
-0.2222460508,
0.0771598443,
0.3349465132,
-0.0737509131,
0.1878646761,
0.1442665756,
-0.3492169976,
-0.0790116638,
-0.0737031996,
-0.3046619594,
-0.2064709514,
0.2073303014,
-0.2225420028,
0.0050881999,
0.1136128306,
-0.3396573067,
-0.2113550901,
0.2969418764,
0.2204648256,
0.2022595704,
-0.0001236163,
-0.0930323005,
0.2447298914,
0.1353629529,
0.2126865685,
0.709495306,
-0.2860055864,
0.0842967108,
0.0118546635,
-0.2131094635,
0.0391237363,
-0.224590078,
0.0313990712,
-0.3076615334,
-0.1945783049,
-0.2164342403,
0.0178864673,
-0.0420156084,
0.1275667995,
0.039708402,
-0.055594489,
-0.2334749252,
-0.1960605085,
-0.1421742439,
0.1881547868,
0.4077610373,
0.0928777009,
0.163182348,
-0.0815116838,
-0.1110179424,
0.0442545302,
0.0191484764,
0.3126094043,
0.1118714064,
0.0620256364,
0.0274289995,
0.1938611716,
-0.1418397725,
-0.5114314556,
-0.1815447807,
0.1477665007,
-0.1431191266,
-0.1886233836,
-0.1851347685,
0.0634021908,
0.0095763002,
0.1191963702,
-0.4742815197,
0.0137951523,
-0.0145702288,
0.3781152964,
0.1848355383,
0.2552728057,
0.1527382433,
0.4181764424,
0.0640506074,
0.138956666,
-0.0919339135,
-0.2201535404,
0.0968007445,
-0.7896098495,
0.0617333576,
-0.1457813829,
-0.1499475092,
-0.0559657589,
-0.1912728995,
-0.3814218342,
0.2947396338,
0.1131916121,
0.1418474317,
0.0255721025,
-0.1043264568,
0.0632382184,
-0.2763237059,
-0.0681231916,
-0.3750478923,
-0.0584924743,
0.0647538304,
0.0459294915,
-0.1101503298,
-0.4853463471,
-0.0342458375,
-0.3445419669,
-0.114913322,
-0.0397003479,
-0.2562004924,
0.1214187741,
-0.1795288175,
0.2044481635,
0.182708919,
0.3225755692,
0.1010049954,
-0.2328572571,
-0.0010055266,
-0.0215394646,
-0.0223838724,
0.0048326142,
-0.1678571403,
0.1890094131,
0.4930637181,
-0.0246062726,
-0.4299056232,
-0.3655105233,
0.0114618093,
-0.0571345985,
0.1619334221,
0.0213699341,
-0.21686472,
-0.0300494246,
0.0381839983,
-0.01989007,
-0.0655275881,
-0.1344529092,
0.2728838623,
-0.2350448072,
0.215194881,
-0.2092874199,
-0.3914924264,
0.0174663123,
-0.0052476693,
-0.0216320939,
-0.1631599665,
-0.0698067397,
0.3777593374,
0.1984348297,
0.0365075767,
-0.0853618383,
0.106201902,
-0.2503992915,
-0.1264424622,
0.231338799,
0.1742164493,
-0.4121522903,
-0.0412243158,
-0.3235963285,
-0.1685478091,
0.1864795536,
0.4758310318,
-0.0292450041,
0.1094246805,
-0.333119601,
0.3905991316,
0.1770892441,
-0.367205739,
0.0471229181,
0.1398413628,
-0.1204542667,
-0.1656447053,
0.331510663,
0.2888587117,
0.3453753293,
-0.0083527863,
0.1355592608,
0.1465896964,
-0.061241284,
-0.1479460597,
-0.2127329409,
-0.253559947,
0.1667657197,
0.2950848043,
-0.0209490489,
0.1681787968,
-0.031033203,
-0.8039102554,
0.3160692155,
-0.1459509134,
0.299919039,
-0.0677223802,
0.2666430473,
0.3727965951,
0.3704648912,
-0.1162355989,
0.0769767612,
0.238286376,
0.0209096372,
-0.0338545702,
0.2227014154,
-0.3855348527,
0.4876060486,
0.0459406003,
0.0490638725,
0.4221504927,
-0.0552330166,
-0.1752130091,
0.0095422119,
-0.4162952304,
-0.348895967,
0.5179665685,
-0.2398486435,
-0.1846757233,
-0.0695812702,
0.0604911149,
0.0085241701,
-0.1463848203,
-0.3137203455,
0.087552011,
0.0763883889,
-0.2518976033,
-0.087073639,
0.3075046241,
-0.0211859196,
0.0460119359,
0.8042544723,
0.0774966031,
-0.1413670629,
0.235442847,
0.0844635963,
-0.2908225656,
-0.060301166,
-0.1021531448,
0.3470939398,
0.1247010604,
0.1034974828,
0.4663468301,
0.0484493375,
-0.0102871023,
0.2089307308,
0.0494255275,
0.1752782613,
0.0404649898,
-0.0039797258,
-0.1105220616,
-0.0594569407,
0.07425455,
-0.2332382798,
0.5649160147,
-0.171733588,
-0.1027771756,
-0.2015344948,
-0.0461472347,
-0.0802184045,
0.1478327364,
0.0841502771,
-0.0952413678,
0.2414818406,
0.2003471851,
0.0308590196,
-0.2746877074,
0.2501146793,
-0.1929259598,
0.0255798921,
0.2707829475,
0.1545012295,
0.0391263589,
-0.0348359495,
-0.1922843754,
0.0270987377,
-0.1999224275,
0.0515192896,
0.1211283281,
-0.0166958496,
-0.1093713641,
0.0189686418,
0.1860828996,
-0.2127862722,
-0.1081466526,
-0.2143589258,
-0.1133391336,
0.176440835,
0.0975931287,
-0.1917556524,
-0.0609496608,
-0.0961271673,
0.3963817656,
-0.1553372443,
-0.4528173804,
-0.2940587997,
-0.1470529288,
0.1323349476,
-0.0661900714,
0.3142541051,
0.0856308937,
0.0921733975,
0.1264201999,
-0.0404540561,
-0.3440806866,
0.0950986892,
-0.15184699,
-0.0034038825,
-0.0941034257,
-0.2936230004,
-0.3489651084,
-0.1819130629,
-0.0824743509,
0.0411470234,
0.0189424902,
0.0167010278,
-0.1476119906,
-0.1336669475,
-0.0711156726,
0.0677141845,
-0.1171518564,
-0.2093515396,
-0.3055464923,
0.3475597799,
-0.1158812791,
0.1410503983,
-0.4759963751,
-0.1276792586,
-0.0629484653,
-0.3044451475,
-0.1956473738,
-0.2405065,
-0.1052337065,
0.1640598923,
0.0499347635,
-0.1538973153,
0.0957068801,
0.0029415265,
-0.0289328936,
0.4696132243,
0.0203416571,
0.0658018142,
-0.0760876238,
0.1825040579,
-0.16493994,
0.1721914411,
0.3239164948,
0.0056632645,
-0.0553416535,
0.1192136332,
0.1172982007,
0.3369154334,
0.0030357577,
0.0329296738,
0.0064068213,
0.0400706902,
-0.201146245,
0.7294511199,
0.1223024577,
-0.1957217902,
0.1946377009,
0.0824974552,
0.0250120386,
0.0861673281,
-0.1782760471,
0.3500838578,
0.1012571305,
-0.0929717571,
0.0026170909,
-0.0438821912,
-0.1264020801,
0.0411883891,
-0.1293792576,
-0.1038941517,
-0.1315621436,
-0.0548463203,
0.2528619766,
0.5113691688,
-0.0436336808,
0.0077448115,
-0.2366497219,
-0.0065446049,
0.1355663389,
0.4619988799,
0.5420900583,
-0.3316130936,
0.0905428603,
-0.3064236045,
-0.3880172372,
0.449092567,
0.0378730074,
0.4197950065,
0.0841394141,
0.0210978985,
0.2919823229,
-0.2219100893,
0.9026582241,
0.0270746369,
-0.5150841475,
0.0960621238,
-0.0259988867,
-0.3318490386,
-0.0938115269,
0.1524904519,
0.2472437024,
0.3681125343,
0.1548550725,
-0.2120522261,
-0.291793555,
0.1343270093,
0.0831084549,
-0.2784323692,
-0.673858285,
-0.1498618722,
-0.0077254847,
-0.0534088984,
0.1820505708,
-0.2881643772,
0.1141939759,
0.0948081464,
-0.0059592724,
-0.3301716447,
-0.0232589021,
0.0441848338,
0.0040749237,
0.0915082097,
-0.0205354746,
0.5346841812,
0.2204900831,
-0.0735490024,
0.2417612225,
-0.0164076574,
0.2224724293,
0.1455624402,
-0.1792442948,
-0.1392754614,
0.2092138082,
0.0098416656,
-0.003031414,
0.3491420746,
-0.2979667187,
0.1423679143,
-0.2083746493,
-0.0092654191,
0.0615615472,
0.4385181367,
-0.0940526575,
-0.2846191525,
0.3209879994,
0.1211754531,
-0.1498609185,
-0.1145359874,
0.3715475798,
-0.0501349047,
0.636487782,
0.1283534467,
0.6783899069,
-0.2733449638,
-0.0964754,
-0.1757934242,
0.135334745,
0.0775395036,
-0.5695589781,
0.4079464972,
-0.2618125081,
-0.1729671359,
0.1151985228,
-0.1011655703,
0.1822108328,
0.267760545,
0.099764213,
0.2494502366,
0.0900090933,
-0.2832881808,
-0.1383253932,
0.3152261376,
0.1780462265,
-0.3854476213,
0.3567739129,
0.0085212961,
0.1185499728,
-0.2574539483,
0.0475543961,
0.0995272398,
-0.0964790732,
0.4011667073,
-0.16188173,
-0.0622569174,
0.472517401,
0.0784944296,
-0.1750520021,
0.0149904191,
0.226522252,
-0.4040832222,
0.2418206632,
-0.2935245633,
-0.1323634982,
0.0944529101,
-0.0516434126,
-0.1664474308,
0.0012743324,
-0.4851129651,
-0.0742491558,
0.0507861525,
-0.4548891187,
-0.1688658595,
0.2750448883,
-0.7210451365,
-0.3166677952,
0.0968268439,
-0.0611468405,
0.1004589945,
-0.145428136,
0.382725656,
-0.1340801418,
-0.0466319695,
-0.0264504142,
0.2520070672,
-0.4767290652,
0.8772205114,
-0.3962701261,
-0.0234911144,
0.0064209308,
0.2088800371,
0.3037534356,
-0.1857493371,
0.0359905101,
0.0318613946,
-0.1825809479,
-0.0297215581,
-0.7237132788,
-0.0615829341,
0.3477553725,
-0.1356629431,
0.0826940387,
-0.2126895189,
0.0430261679,
-0.5923833847,
-0.0986190215,
-0.0355291963,
-0.6348738074,
-0.0941624641,
-0.3619502783,
0.0876315385,
0.1204518825,
0.3645614088,
0.1034474075,
0.1033663899,
-0.3473068178,
0.2193741053,
-0.1451958865,
-0.0548932776,
0.2107296288,
0.1804765314,
0.2133189589,
-0.0586997978,
0.1368573308,
-0.0641075149,
-0.0119603686,
0.1239685938,
0.1416327208,
-0.0910777599,
0.2457412779,
0.2072181255,
0.0595082603,
0.0605389401,
-0.1134943888,
-0.2933692336,
-0.2580487728,
-0.2277740091,
0.0708862841,
-0.0550414734,
-0.2842023075,
-0.0373875909,
0.3796741664,
-0.2199348956,
-0.0367659107,
0.5711753964,
0.3638446629,
0.4119433761,
-0.1304622293,
0.5840530396,
-0.1503131092,
-0.0116172731,
0.1101323962,
0.0380048379,
0.3578720689,
0.1273035705,
0.0257702712,
0.1005513519,
-0.2617843747,
-0.2099003941,
0.5314102173,
0.5130405426,
0.4117484689,
-0.2642946839,
0.0969004482,
0.0436327495,
-0.063516289,
-0.1413559467,
0.760537684,
-0.0652893633,
0.2589678466,
0.2288999856,
0.3385694325,
0.3246195912,
-0.2439424247,
0.0713543668,
0.1840888709,
0.3953975737,
-0.2592018843,
-0.5113338232,
0.3347237706,
0.0319290161,
0.2751270533,
-0.1516730934,
0.1107719094,
0.92304039,
-0.197951749,
0.5783291459,
0.2697030604,
0.0731893629,
0.0477504209,
-0.1127755865,
0.5054673553,
0.2119087875,
0.2594749928,
0.0425579399,
0.1568236351,
0.4555433393,
0.0862916633,
-0.4791409671,
-0.2030678242,
0.363342762,
-0.0651320219,
-0.1690254807,
0.3964125514,
0.0012901127,
0.0932266712,
0.0458045453,
0.2178765982,
0.0357574336,
0.3246454597,
0.0491057038,
0.1516364068,
0.1504166871,
0.0357432775,
0.4240094423,
0.2561798692,
0.0196469873,
-0.2027695477,
-0.1023053974,
-0.033146672,
0.127964288,
0.1681086272,
0.0743427798,
0.1605704129,
0.0622823201,
0.3273926377,
-0.0757855177,
0.1349637657,
-0.0273568369,
-0.0791054219,
-0.3827726245,
-0.0394780077,
0.191684261,
-0.0211960562,
-0.0130573995,
-0.4874782562,
-0.2010093331,
0.3620592952,
-0.0939694569,
-0.0635307729,
0.1349421889,
-0.3062337637,
0.003493296,
0.0126206167,
-0.1436277032,
-0.1309954226,
0.3259011507,
-0.5958310366,
-0.2522016466,
0.1412500739,
-0.0217809863,
-0.2656868994,
0.4139504731,
0.0934203714,
0.5814163089,
-0.2157100439,
-0.0181518942,
-0.2982736528,
0.1779899299,
-0.1417061687,
0.0568242744,
-0.0309728365,
0.473413676,
-0.0127443895,
-0.1003621668,
0.6402201653,
0.0816159397,
0.2973897159,
0.5808255672,
-0.2128430158,
0.188826412,
-0.2031276673,
-0.2613910735,
0.1296495199,
-0.1679708958,
0.4393238723,
0.1654767692,
-0.1756122112,
0.1502399296,
-0.1133834124,
0.425437808,
-0.466451019,
0.4777959287,
0.0649796352,
0.2579562664,
-0.1346856952,
0.1684642136,
-0.1328521371,
-0.117049709,
-0.3317901492,
-0.0999456048,
-0.2547706962,
-0.1374462545,
-0.3246658444,
-0.1422878057,
-0.0342454724,
0.3217970431,
-0.1579920501,
-0.0471626893,
-0.1276910305,
0.1993936002,
-0.1976648867,
0.1830362976,
-0.0548232272,
0.1600774974,
0.2074937075,
0.1919037998,
-0.2801053524,
0.2080817521,
0.2239712775,
-0.223578468,
-0.0168865919,
-0.4630413055,
0.1518837214,
0.3087666631,
0.1494833827,
-0.4369866848,
-0.440608263,
0.0744835585,
0.2929987907,
0.170496285,
0.1940238029,
0.0582561865,
-0.2841020226,
-0.2611773312,
-0.5086884499,
0.0182646886,
0.1092124581,
-0.049623061,
-0.087397106
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | Hi !
To fix 1, an you try to run this code ?
```python
from datasets import load_dataset
load_dataset("squad", download_mode="force_redownload")
```
Maybe the file your downloaded was corrupted, in this case redownloading this way should fix your issue 1.
Regarding your 2nd point, you're right that loading the raw json this way doesn't give you a dataset with the column "context", "question" and "answers". Indeed the squad format is a very nested format so you have to preprocess the data. You can do it this way:
```python
def process_squad(examples):
"""
Process a dataset in the squad format with columns "title" and "paragraphs"
to return the dataset with columns "context", "question" and "answers".
"""
out = {"context": [], "question": [], "answers":[]}
for paragraphs in examples["paragraphs"]:
for paragraph in paragraphs:
for qa in paragraph["qas"]:
answers = [{"answer_start": answer["answer_start"], "text": answer["text"].strip()} for answer in qa["answers"]]
out["context"].append(paragraph["context"].strip())
out["question"].append(qa["question"].strip())
out["answers"].append(answers)
return out
datasets = load_dataset(extension, data_files=data_files, field="data")
column_names = datasets["train"].column_names
if set(column_names) == {"title", "paragraphs"}:
datasets = datasets.map(process_squad, batched=True, remove_columns=column_names)
```
Hope that helps :) | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 170 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
Hi !
To fix 1, an you try to run this code ?
```python
from datasets import load_dataset
load_dataset("squad", download_mode="force_redownload")
```
Maybe the file your downloaded was corrupted, in this case redownloading this way should fix your issue 1.
Regarding your 2nd point, you're right that loading the raw json this way doesn't give you a dataset with the column "context", "question" and "answers". Indeed the squad format is a very nested format so you have to preprocess the data. You can do it this way:
```python
def process_squad(examples):
"""
Process a dataset in the squad format with columns "title" and "paragraphs"
to return the dataset with columns "context", "question" and "answers".
"""
out = {"context": [], "question": [], "answers":[]}
for paragraphs in examples["paragraphs"]:
for paragraph in paragraphs:
for qa in paragraph["qas"]:
answers = [{"answer_start": answer["answer_start"], "text": answer["text"].strip()} for answer in qa["answers"]]
out["context"].append(paragraph["context"].strip())
out["question"].append(qa["question"].strip())
out["answers"].append(answers)
return out
datasets = load_dataset(extension, data_files=data_files, field="data")
column_names = datasets["train"].column_names
if set(column_names) == {"title", "paragraphs"}:
datasets = datasets.map(process_squad, batched=True, remove_columns=column_names)
```
Hope that helps :) | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | Thks for quickly answering！
### 1 I try the first way,but seems not work
```
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 503, in <module>
main()
File "examples/question-answering/run_qa.py", line 218, in main
datasets = load_dataset(data_args.dataset_name, download_mode="force_redownload")
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
### 2 I try the second way,and run the examples/question-answering/run_qa.py,it lead to another bug orz..
```
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 523, in <module>
main()
File "examples/question-answering/run_qa.py", line 379, in main
load_from_cache_file=not data_args.overwrite_cache,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1120, in map
update_data = does_function_return_dict(test_inputs, test_indices)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1091, in does_function_return_dict
function(*fn_args, indices, **fn_kwargs) if with_indices else function(*fn_args, **fn_kwargs)
File "examples/question-answering/run_qa.py", line 339, in prepare_train_features
if len(answers["answer_start"]) == 0:
TypeError: list indices must be integers or slices, not str
```
## may be the function prepare_train_features in run_qa.py need to fix,I think is that the prep
```python
for i, offsets in enumerate(offset_mapping):
# We will label impossible answers with the index of the CLS token.
input_ids = tokenized_examples["input_ids"][i]
cls_index = input_ids.index(tokenizer.cls_token_id)
# Grab the sequence corresponding to that example (to know what is the context and what is the question).
sequence_ids = tokenized_examples.sequence_ids(i)
# One example can give several spans, this is the index of the example containing this span of text.
sample_index = sample_mapping[i]
answers = examples[answer_column_name][sample_index]
print(examples,answers)
# If no answers are given, set the cls_index as answer.
if len(answers["answer_start"]) == 0:
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Start/end character index of the answer in the text.
start_char = answers["answer_start"][0]
end_char = start_char + len(answers["text"][0])
# Start token index of the current span in the text.
token_start_index = 0
while sequence_ids[token_start_index] != (1 if pad_on_right else 0):
token_start_index += 1
# End token index of the current span in the text.
token_end_index = len(input_ids) - 1
while sequence_ids[token_end_index] != (1 if pad_on_right else 0):
token_end_index -= 1
# Detect if the answer is out of the span (in which case this feature is labeled with the CLS index).
if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char):
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Otherwise move the token_start_index and token_end_index to the two ends of the answer.
# Note: we could go after the last offset if the answer is the last word (edge case).
while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char:
token_start_index += 1
tokenized_examples["start_positions"].append(token_start_index - 1)
while offsets[token_end_index][1] >= end_char:
token_end_index -= 1
tokenized_examples["end_positions"].append(token_end_index + 1)
return tokenized_examples
``` | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 434 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
Thks for quickly answering！
### 1 I try the first way,but seems not work
```
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 503, in <module>
main()
File "examples/question-answering/run_qa.py", line 218, in main
datasets = load_dataset(data_args.dataset_name, download_mode="force_redownload")
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
### 2 I try the second way,and run the examples/question-answering/run_qa.py,it lead to another bug orz..
```
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 523, in <module>
main()
File "examples/question-answering/run_qa.py", line 379, in main
load_from_cache_file=not data_args.overwrite_cache,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1120, in map
update_data = does_function_return_dict(test_inputs, test_indices)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/arrow_dataset.py", line 1091, in does_function_return_dict
function(*fn_args, indices, **fn_kwargs) if with_indices else function(*fn_args, **fn_kwargs)
File "examples/question-answering/run_qa.py", line 339, in prepare_train_features
if len(answers["answer_start"]) == 0:
TypeError: list indices must be integers or slices, not str
```
## may be the function prepare_train_features in run_qa.py need to fix,I think is that the prep
```python
for i, offsets in enumerate(offset_mapping):
# We will label impossible answers with the index of the CLS token.
input_ids = tokenized_examples["input_ids"][i]
cls_index = input_ids.index(tokenizer.cls_token_id)
# Grab the sequence corresponding to that example (to know what is the context and what is the question).
sequence_ids = tokenized_examples.sequence_ids(i)
# One example can give several spans, this is the index of the example containing this span of text.
sample_index = sample_mapping[i]
answers = examples[answer_column_name][sample_index]
print(examples,answers)
# If no answers are given, set the cls_index as answer.
if len(answers["answer_start"]) == 0:
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Start/end character index of the answer in the text.
start_char = answers["answer_start"][0]
end_char = start_char + len(answers["text"][0])
# Start token index of the current span in the text.
token_start_index = 0
while sequence_ids[token_start_index] != (1 if pad_on_right else 0):
token_start_index += 1
# End token index of the current span in the text.
token_end_index = len(input_ids) - 1
while sequence_ids[token_end_index] != (1 if pad_on_right else 0):
token_end_index -= 1
# Detect if the answer is out of the span (in which case this feature is labeled with the CLS index).
if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char):
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Otherwise move the token_start_index and token_end_index to the two ends of the answer.
# Note: we could go after the last offset if the answer is the last word (edge case).
while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char:
token_start_index += 1
tokenized_examples["start_positions"].append(token_start_index - 1)
while offsets[token_end_index][1] >= end_char:
token_end_index -= 1
tokenized_examples["end_positions"].append(token_end_index + 1)
return tokenized_examples
``` | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | ## I have fixed it, @lhoestq
### the first section change as you said and add ["id"]
```python
def process_squad(examples):
"""
Process a dataset in the squad format with columns "title" and "paragraphs"
to return the dataset with columns "context", "question" and "answers".
"""
# print(examples)
out = {"context": [], "question": [], "answers":[],"id":[]}
for paragraphs in examples["paragraphs"]:
for paragraph in paragraphs:
for qa in paragraph["qas"]:
answers = [{"answer_start": answer["answer_start"], "text": answer["text"].strip()} for answer in qa["answers"]]
out["context"].append(paragraph["context"].strip())
out["question"].append(qa["question"].strip())
out["answers"].append(answers)
out["id"].append(qa["id"])
return out
column_names = datasets["train"].column_names if training_args.do_train else datasets["validation"].column_names
# print(datasets["train"].column_names)
if set(column_names) == {"title", "paragraphs"}:
datasets = datasets.map(process_squad, batched=True, remove_columns=column_names)
# Preprocessing the datasets.
# Preprocessing is slighlty different for training and evaluation.
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
# print(column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
### the second section
```python
def prepare_train_features(examples):
# Tokenize our examples with truncation and maybe padding, but keep the overflows using a stride. This results
# in one example possible giving several features when a context is long, each of those features having a
# context that overlaps a bit the context of the previous feature.
tokenized_examples = tokenizer(
examples[question_column_name if pad_on_right else context_column_name],
examples[context_column_name if pad_on_right else question_column_name],
truncation="only_second" if pad_on_right else "only_first",
max_length=data_args.max_seq_length,
stride=data_args.doc_stride,
return_overflowing_tokens=True,
return_offsets_mapping=True,
padding="max_length" if data_args.pad_to_max_length else False,
)
# Since one example might give us several features if it has a long context, we need a map from a feature to
# its corresponding example. This key gives us just that.
sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping")
# The offset mappings will give us a map from token to character position in the original context. This will
# help us compute the start_positions and end_positions.
offset_mapping = tokenized_examples.pop("offset_mapping")
# Let's label those examples!
tokenized_examples["start_positions"] = []
tokenized_examples["end_positions"] = []
for i, offsets in enumerate(offset_mapping):
# We will label impossible answers with the index of the CLS token.
input_ids = tokenized_examples["input_ids"][i]
cls_index = input_ids.index(tokenizer.cls_token_id)
# Grab the sequence corresponding to that example (to know what is the context and what is the question).
sequence_ids = tokenized_examples.sequence_ids(i)
# One example can give several spans, this is the index of the example containing this span of text.
sample_index = sample_mapping[i]
answers = examples[answer_column_name][sample_index]
# print(examples,answers,offset_mapping,tokenized_examples)
# If no answers are given, set the cls_index as answer.
if len(answers) == 0:#len(answers["answer_start"]) == 0:
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Start/end character index of the answer in the text.
start_char = answers[0]["answer_start"]
end_char = start_char + len(answers[0]["text"])
# Start token index of the current span in the text.
token_start_index = 0
while sequence_ids[token_start_index] != (1 if pad_on_right else 0):
token_start_index += 1
# End token index of the current span in the text.
token_end_index = len(input_ids) - 1
while sequence_ids[token_end_index] != (1 if pad_on_right else 0):
token_end_index -= 1
# Detect if the answer is out of the span (in which case this feature is labeled with the CLS index).
if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char):
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Otherwise move the token_start_index and token_end_index to the two ends of the answer.
# Note: we could go after the last offset if the answer is the last word (edge case).
while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char:
token_start_index += 1
tokenized_examples["start_positions"].append(token_start_index - 1)
while offsets[token_end_index][1] >= end_char:
token_end_index -= 1
tokenized_examples["end_positions"].append(token_end_index + 1)
return tokenized_examples
``` | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 569 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
## I have fixed it, @lhoestq
### the first section change as you said and add ["id"]
```python
def process_squad(examples):
"""
Process a dataset in the squad format with columns "title" and "paragraphs"
to return the dataset with columns "context", "question" and "answers".
"""
# print(examples)
out = {"context": [], "question": [], "answers":[],"id":[]}
for paragraphs in examples["paragraphs"]:
for paragraph in paragraphs:
for qa in paragraph["qas"]:
answers = [{"answer_start": answer["answer_start"], "text": answer["text"].strip()} for answer in qa["answers"]]
out["context"].append(paragraph["context"].strip())
out["question"].append(qa["question"].strip())
out["answers"].append(answers)
out["id"].append(qa["id"])
return out
column_names = datasets["train"].column_names if training_args.do_train else datasets["validation"].column_names
# print(datasets["train"].column_names)
if set(column_names) == {"title", "paragraphs"}:
datasets = datasets.map(process_squad, batched=True, remove_columns=column_names)
# Preprocessing the datasets.
# Preprocessing is slighlty different for training and evaluation.
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
# print(column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
### the second section
```python
def prepare_train_features(examples):
# Tokenize our examples with truncation and maybe padding, but keep the overflows using a stride. This results
# in one example possible giving several features when a context is long, each of those features having a
# context that overlaps a bit the context of the previous feature.
tokenized_examples = tokenizer(
examples[question_column_name if pad_on_right else context_column_name],
examples[context_column_name if pad_on_right else question_column_name],
truncation="only_second" if pad_on_right else "only_first",
max_length=data_args.max_seq_length,
stride=data_args.doc_stride,
return_overflowing_tokens=True,
return_offsets_mapping=True,
padding="max_length" if data_args.pad_to_max_length else False,
)
# Since one example might give us several features if it has a long context, we need a map from a feature to
# its corresponding example. This key gives us just that.
sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping")
# The offset mappings will give us a map from token to character position in the original context. This will
# help us compute the start_positions and end_positions.
offset_mapping = tokenized_examples.pop("offset_mapping")
# Let's label those examples!
tokenized_examples["start_positions"] = []
tokenized_examples["end_positions"] = []
for i, offsets in enumerate(offset_mapping):
# We will label impossible answers with the index of the CLS token.
input_ids = tokenized_examples["input_ids"][i]
cls_index = input_ids.index(tokenizer.cls_token_id)
# Grab the sequence corresponding to that example (to know what is the context and what is the question).
sequence_ids = tokenized_examples.sequence_ids(i)
# One example can give several spans, this is the index of the example containing this span of text.
sample_index = sample_mapping[i]
answers = examples[answer_column_name][sample_index]
# print(examples,answers,offset_mapping,tokenized_examples)
# If no answers are given, set the cls_index as answer.
if len(answers) == 0:#len(answers["answer_start"]) == 0:
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Start/end character index of the answer in the text.
start_char = answers[0]["answer_start"]
end_char = start_char + len(answers[0]["text"])
# Start token index of the current span in the text.
token_start_index = 0
while sequence_ids[token_start_index] != (1 if pad_on_right else 0):
token_start_index += 1
# End token index of the current span in the text.
token_end_index = len(input_ids) - 1
while sequence_ids[token_end_index] != (1 if pad_on_right else 0):
token_end_index -= 1
# Detect if the answer is out of the span (in which case this feature is labeled with the CLS index).
if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char):
tokenized_examples["start_positions"].append(cls_index)
tokenized_examples["end_positions"].append(cls_index)
else:
# Otherwise move the token_start_index and token_end_index to the two ends of the answer.
# Note: we could go after the last offset if the answer is the last word (edge case).
while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char:
token_start_index += 1
tokenized_examples["start_positions"].append(token_start_index - 1)
while offsets[token_end_index][1] >= end_char:
token_end_index -= 1
tokenized_examples["end_positions"].append(token_end_index + 1)
return tokenized_examples
``` | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | I'm glad you managed to fix run_qa.py for your case :)
Regarding the checksum error, I'm not able to reproduce on my side.
This errors says that the downloaded file doesn't match the expected file.
Could you try running this and let me know if you get the same output as me ?
```python
from datasets.utils.info_utils import get_size_checksum_dict
from datasets import cached_path
get_size_checksum_dict(cached_path("https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json"))
# {'num_bytes': 30288272, 'checksum': '3527663986b8295af4f7fcdff1ba1ff3f72d07d61a20f487cb238a6ef92fd955'}
``` | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 69 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
I'm glad you managed to fix run_qa.py for your case :)
Regarding the checksum error, I'm not able to reproduce on my side.
This errors says that the downloaded file doesn't match the expected file.
Could you try running this and let me know if you get the same output as me ?
```python
from datasets.utils.info_utils import get_size_checksum_dict
from datasets import cached_path
get_size_checksum_dict(cached_path("https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json"))
# {'num_bytes': 30288272, 'checksum': '3527663986b8295af4f7fcdff1ba1ff3f72d07d61a20f487cb238a6ef92fd955'}
``` | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | I run the code,and it show below:
```
>>> from datasets.utils.info_utils import get_size_checksum_dict
>>> from datasets import cached_path
>>> get_size_checksum_dict(cached_path("https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json"))
Downloading: 30.3MB [04:13, 120kB/s]
{'num_bytes': 30288272, 'checksum': '3527663986b8295af4f7fcdff1ba1ff3f72d07d61a20f487cb238a6ef92fd955'}
``` | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 29 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
I run the code,and it show below:
```
>>> from datasets.utils.info_utils import get_size_checksum_dict
>>> from datasets import cached_path
>>> get_size_checksum_dict(cached_path("https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json"))
Downloading: 30.3MB [04:13, 120kB/s]
{'num_bytes': 30288272, 'checksum': '3527663986b8295af4f7fcdff1ba1ff3f72d07d61a20f487cb238a6ef92fd955'}
``` | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1964 | Datasets.py function load_dataset does not match squad dataset | Alright ! So in this case redownloading the file with `download_mode="force_redownload"` should fix it. Can you try using `download_mode="force_redownload"` again ?
Not sure why it didn't work for you the first time though :/ | ### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot! | 34 | Datasets.py function load_dataset does not match squad dataset
### 1 When I try to train lxmert,and follow the code in README that --dataset name:
```shell
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --dataset_name squad --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
the bug is that:
```
Downloading and preparing dataset squad/plain_text (download: 33.51 MiB, generated: 85.75 MiB, post-processed: Unknown size, total: 119.27 MiB) to /home2/zhenggo1/.cache/huggingface/datasets/squad/plain_text/1.0.0/4c81550d83a2ac7c7ce23783bd8ff36642800e6633c1f18417fb58c3ff50cdd7...
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 217, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/load.py", line 746, in load_dataset
use_auth_token=use_auth_token,
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 573, in download_and_prepare
dl_manager=dl_manager, verify_infos=verify_infos, **download_and_prepare_kwargs
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/builder.py", line 633, in _download_and_prepare
self.info.download_checksums, dl_manager.get_recorded_sizes_checksums(), "dataset source files"
File "/home2/zhenggo1/anaconda3/envs/lpot/lib/python3.7/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://rajpurkar.github.io/SQuAD-explorer/dataset/train-v1.1.json']
```
And I try to find the [checksum link](https://github.com/huggingface/datasets/blob/master/datasets/squad/dataset_infos.json)
,is the problem plain_text do not have a checksum?
### 2 When I try to train lxmert,and use local dataset:
```
python examples/question-answering/run_qa.py --model_name_or_path unc-nlp/lxmert-base-uncased --train_file $SQUAD_DIR/train-v1.1.json --validation_file $SQUAD_DIR/dev-v1.1.json --do_train --do_eval --per_device_train_batch_size 12 --learning_rate 3e-5 --num_train_epochs 2 --max_seq_length 384 --doc_stride 128 --output_dir /home2/zhenggo1/checkpoint/lxmert_squad
```
The bug is that
```
['title', 'paragraphs']
Traceback (most recent call last):
File "examples/question-answering/run_qa.py", line 501, in <module>
main()
File "examples/question-answering/run_qa.py", line 273, in main
answer_column_name = "answers" if "answers" in column_names else column_names[2]
IndexError: list index out of range
```
I print the answer_column_name and find that local squad dataset need the package datasets to preprocessing so that the code below can work:
```
if training_args.do_train:
column_names = datasets["train"].column_names
else:
column_names = datasets["validation"].column_names
print(datasets["train"].column_names)
question_column_name = "question" if "question" in column_names else column_names[0]
context_column_name = "context" if "context" in column_names else column_names[1]
answer_column_name = "answers" if "answers" in column_names else column_names[2]
```
## Please tell me how to fix the bug,thks a lot!
Alright ! So in this case redownloading the file with `download_mode="force_redownload"` should fix it. Can you try using `download_mode="force_redownload"` again ?
Not sure why it didn't work for you the first time though :/ | [
-0.3537205458,
0.0625083745,
0.0333653577,
0.3963989317,
0.5554533601,
0.0057207122,
0.5457589626,
0.3404687643,
-0.0983104035,
-0.1216232479,
-0.1650660634,
0.4508374631,
0.108414799,
-0.0774193332,
0.1859292388,
0.2271618098,
-0.1242239475,
-0.0123369023,
-0.1461084336,
-0.2893438041,
-0.1109781936,
0.1282911599,
-0.0823140591,
0.2603986859,
-0.1788993925,
0.0166908,
0.2006343901,
0.2723646164,
-0.0037964508,
-0.2199688852,
0.3428423405,
-0.2606050372,
0.2361015677,
0.5758860707,
-0.0001148581,
0.1132559627,
-0.0115430877,
-0.3397288024,
-0.5327038765,
-0.2002218217,
-0.1021837518,
-0.3757724762,
0.1486450732,
-0.2605553269,
-0.1595319808,
0.0937374309,
0.0203167144,
-0.4645363092,
0.3544550538,
0.4883134961,
0.2139824927,
0.2522568107,
-0.1001009122,
0.0775509849,
0.1142547727,
-0.0009760931,
-0.0695091411,
0.1360990256,
0.3347512484,
-0.0582687184,
0.0025540479,
0.172445789,
0.0828084946,
0.056823425,
0.2703093588,
0.0064036995,
0.1991150528,
-0.3161554933,
0.0564418994,
0.3177841604,
0.4627764821,
-0.4630154073,
-0.334705323,
-0.2631345689,
0.1331045628,
-0.256572783,
0.1320163906,
0.0672776327,
0.0511586145,
-0.0424276032,
0.0477761179,
0.0376647748,
0.11142876,
0.0368288383,
0.0068917796,
0.1999948621,
0.0288672522,
0.238396287,
-0.2015397549,
-0.1135483533,
0.1148920804,
-0.0831490457,
0.0818323493,
0.3924932182,
-0.8285178542,
-0.0814113915,
-0.2042949498,
-0.1421723366,
0.256287396,
0.0819007233,
0.0386923663,
-0.0269714296,
0.1486689001,
0.1044573858,
0.1687796414,
0.3692038357,
0.3052026629,
0.2674021125,
-0.0663634241,
0.0668203235,
-0.3717200458,
0.2254104316,
-0.322581619,
-0.1618057787,
0.0960088968,
0.0835882872,
0.0062305108,
-0.1927383244,
-0.5517168641,
0.1761801243,
0.1544086188,
0.0026421808,
0.0974372327,
0.3247914016,
0.0759165883,
0.3802294433,
0.0263221562,
0.1048556641,
-0.2946237028,
-0.370416224,
-0.2408785224,
0.0330558904,
-0.0880101025,
0.0268360078,
0.0183389783,
-0.0356006511,
0.3227835596,
-0.1957621872,
0.2114951909,
-0.179079622,
0.1150044873,
-0.2556709051,
-0.1285901666,
-0.0065956898,
0.0298098624,
0.0322052166,
0.1742969006,
-0.3730717897,
-0.0482526347,
0.1222925559,
-0.3712678552,
-0.0336521491,
-0.1820588857,
0.1871161163,
-0.1391082108,
0.0208786745,
-0.2373686284,
0.1698871255,
0.3073555231,
-0.187839523,
0.1063960493,
-0.3961226344,
-0.1664005369,
-0.2122272849,
0.3151644468,
0.3882876039,
-0.4074567556,
-0.2117103785,
0.1249399483,
-0.0952139646,
-0.0381637774,
0.1216656119,
-0.1516055465,
0.155561164,
-0.206223771,
-0.0070259422,
0.7145842314,
-0.8443273902,
-0.4243376255,
0.1099728346,
-0.3059236407,
-0.0194992796,
-0.0710132569,
0.066407539,
0.1930088848,
0.0147337317,
0.0827306286,
0.506614089,
-0.0362932384,
0.091403164,
-0.172457844,
-0.2712132335,
0.2526647151,
0.1708516777,
0.0318142995,
-0.0380645581,
0.038837377,
0.4989833236,
0.3322134614,
0.14261657,
0.1112050712,
-0.0299585629,
-0.0359395631,
-0.0821639001,
0.114605695,
-0.1481090188,
-0.6798554063,
0.1638658494,
-0.1644516885,
0.0439889617,
0.2410782427,
-0.1508303285,
-0.2135133445,
-0.0917366818,
-0.4252306223,
-0.102387622,
0.0743926466,
0.1495982409,
-0.0585200116,
-0.1226446852,
-0.159884721,
0.3514788449,
-0.2154508531,
0.2284577787,
-0.4849056602,
-0.096539177,
-0.0637453049,
-0.036215499,
0.0537009686,
0.3318414688,
0.0871924981,
-0.0709581226,
-0.105753459,
0.4979955852,
0.0180257671,
0.1023963243,
0.2883634865,
-0.2016443759,
0.0996336117,
-0.0990447104,
0.0020647347,
0.2259091437,
0.1946656853,
-0.0812617242,
0.022443153,
0.1418920159,
0.1574392468,
0.0746555105,
-0.0048612803,
-0.1536109149,
0.1390526593,
0.0118649527,
0.0111013725,
0.0971624926,
0.0520382859,
0.2940087616,
0.5483119488,
0.1111364961,
-0.1868932843,
0.1310925037,
0.3315555155,
-0.0650449991,
0.275726527,
-0.0043975781,
-0.1713228226,
-0.0783390701,
0.101006031,
0.178946197,
0.5017951131,
-0.0036267415,
-0.1863483489,
-0.1349208057,
-0.0636771619,
-0.046824038,
0.0711336136,
-0.1824287176,
0.2553722858,
0.2577778995,
0.1839803755,
0.0547972396,
-0.1531581581,
-0.0497436151,
0.0207577348,
0.290525347,
-0.3741133213,
0.0461582839,
-0.1272045374,
0.0979507715,
-0.3777188659,
0.1338105798,
-0.350255549,
-0.1833578646,
-0.1357310116,
-0.1108105481,
0.3093044758,
0.2101666778,
-0.3485355377,
0.1555948853,
0.2402639091,
-0.5443776846,
0.1576042622,
0.1009896398,
-0.3830033541,
0.0529526621,
0.1205421761,
-0.069621332,
-0.0969107822,
-0.1528275013,
0.0173261538,
-0.1201019287,
-0.1680039763,
0.1868710071,
-0.052908279,
0.5768342018,
-0.0123323239,
0.1703873277,
-0.0075946525,
-0.0680119693,
0.1667572558,
-0.1843082011,
-0.1167618781,
-0.1657408327,
-0.1121706814,
0.0364880599,
-0.0776865408,
-0.622492671,
-0.4673993886,
-0.2988176346,
-0.1805040538,
0.0007811263,
0.2649472952,
0.0936488509,
0.0102784112,
0.3407397568,
-0.0355760679,
0.036125876,
-0.4035871625,
-0.281840831,
0.2556056678,
0.0628469884,
-0.2947934866,
-0.0994652063,
0.0628203005,
0.2072018087,
-0.0956361219,
-0.395839572,
-0.2739957571,
0.2307887524,
-0.011217162,
-0.0910095647,
0.0472382158,
0.1151575074,
-0.1613556892,
0.1061528027,
-0.2067127675,
-0.5061612725,
0.1450344324,
0.1470153034,
0.6205660105,
-0.1660100222,
0.2980056703,
-0.0526382439,
0.6775237918,
0.1890771389,
-0.2280172259,
0.2107888609,
-0.2105234265,
0.3032869101,
0.0035382733,
-0.2769272327,
0.3039434552,
0.0719062537,
0.0459160507,
0.3248111606,
0.0534218624,
-0.0166757628,
-0.0613192022,
0.3597292006,
-0.1392310858,
-0.1571275592,
-0.0605639368,
-0.0752925575,
-0.2700315118,
0.074131906,
0.3065283,
-0.1772013903,
-0.1673004627,
-0.1517761648,
0.4453923106,
-0.0859647915,
0.2577697337,
-0.6791192293,
-0.004350163,
-0.125990212,
0.2868065238,
-0.1449970007,
0.6176642776,
-0.0544949919,
-0.1857382655,
-0.0650188103,
0.0791008919,
0.4798840582,
-0.045997981,
-0.0093409726,
-0.0361172296,
0.2946525812,
-0.6179638505,
-0.1878315508,
0.1776028425,
0.168530643,
0.0936855748,
0.3528563082,
-0.1821607649,
-0.077110827,
0.1613400429,
0.1991809905,
-0.1662610173,
-0.0170082524,
-0.3835246265,
-0.2073013932,
-0.3429887891,
-0.0511815622,
-0.2764178514,
0.1328060329,
0.1391487569,
0.1173345,
0.2389437407,
-0.1128842235,
0.3738936782,
0.1367885023,
0.3835945427,
0.1973972172,
0.1353373528,
0.233473897,
0.2617392242,
-0.1502474993,
0.4892303944,
-0.4564042687,
-0.3000645041,
0.0343396775,
-0.0737039149,
0.1476984918,
0.3775247633,
-0.1436187625,
0.0837648362,
-0.3207883835,
-0.0862555206,
0.2754157484,
0.1629421115,
0.2510089874,
0.0771016926,
-0.1133952737,
-0.4098260999,
0.1798561513,
-0.1120344773,
-0.042619504,
0.2145650834,
-0.4636403024,
-0.2888948619,
0.0964146107,
-0.1039725468,
0.8939830065,
-0.0649743602,
0.1889763474,
0.3377413154,
0.1632625163,
0.2822514176,
-0.0774447247,
0.1654772758,
-0.4589554667,
-0.0782307088,
0.0744745359,
-0.1392124295,
-0.0647866651,
0.3663627207,
0.0141665637,
0.5146530867,
-0.2688009739,
0.3116571307,
0.0540141463,
0.5210408568,
-0.133799538,
-0.0108748004,
-0.3472366631,
0.1949187219,
-0.1809044033,
0.2802451849,
-0.1203895658,
0.0054679997,
-0.3029421866,
-0.1379413605,
-0.1319406033,
0.2274256349,
-0.5376269817,
0.1669766754,
0.0934279487,
-0.3486425877,
0.0191773474,
0.5643283129,
-0.1002909169,
-0.0820769593,
-0.1084555015,
0.3530827165,
0.0207633302,
0.304631412,
0.0768880844,
0.0457805917,
0.4243014753,
-0.0168560892,
-0.0948247612,
0.0720393509,
-0.0042468794,
-0.3094082177,
-0.2513001859,
0.0408152901,
0.3062188029,
-0.275687933,
-0.1972320825,
-0.1188386083,
0.0159458071,
-0.0596127212,
0.119826071,
0.0611225255,
-0.0966567993,
0.2068725377,
0.1296119094,
-0.4074322879,
0.0401297957,
0.3414312005,
0.0581959896,
0.0030538738,
0.6670226455,
0.0741892308,
0.0704765767,
-0.1949025095,
0.1497738659,
0.3011730909,
-0.2336235046,
-0.0392005146,
-0.2000788152,
-0.2490169406,
0.0004845485,
-0.1210953295,
0.2407468706,
-0.0176704414,
0.0402802899,
-0.4578699768,
-0.4342668355,
0.0315101631,
-0.0630613863,
0.009265596,
0.0782025754,
-0.0439872965,
0.1450645626,
0.045366168,
-0.2561045587,
0.2107705176,
-0.1966143996,
0.2120200247,
0.0993523672,
-0.0401503369,
0.2497246563,
-0.0228080377,
0.0503342152,
-0.0272150673,
-0.3081599772,
-0.1973446012,
-0.1295124739,
0.1182672828,
0.0220859312,
-0.0156385638,
-0.1738799959,
-0.2943678796,
-0.1543196291,
-0.2474770844,
-0.0346162058,
0.0881622881,
0.0274495948,
0.0142341629,
-0.0534726009,
0.0265387129,
0.0657880455,
-0.073973909,
0.0375891738,
0.044000484,
0.2452985644,
0.1012575328,
0.0479807705,
-0.0090395436,
-0.5060154796,
0.1013092548,
0.1878547519,
0.3835133016,
0.0299074426,
-0.1631863266,
-0.0384694263,
0.2729744613,
0.0583594814,
0.4147219956,
-0.4566236734,
-0.0442090221,
-0.0886463076,
0.1434323639,
-0.2171789706,
0.0970658809,
0.3625805974,
-0.1833088696,
-0.0155574754,
0.2735650539,
0.1266646534,
0.1897181422,
0.0217409674,
0.106899336,
0.5871455669,
-0.0874018818,
0.1646403968,
0.2900642753,
-0.2160797715,
0.2205361426,
0.3349786997,
-0.0837548077,
-0.0087306574,
0.3994482756,
-0.1929348558,
0.3431117535,
-0.079034254,
-0.1280986965,
0.2440445572,
-0.2488020062,
0.0799282193,
0.4024312496,
-0.1566052735,
0.1699331403,
0.0469694026,
0.5395813584,
-0.4212684631,
-0.1374428719,
-0.0382391326,
-0.1659310013,
-0.1484350264,
-0.059856195,
0.0648369193,
-0.1156124994,
-0.1001707315,
-0.2388544977,
-0.1484497488,
-0.3574166894,
-0.0503094383,
-0.0039664768,
-0.3478864729,
-0.0714051798,
-0.0686987936,
0.3550385833,
-0.1964205801,
-0.1058574021,
0.3332425654,
-0.1416482627,
0.0370899253,
0.3267583549,
0.2570813596,
0.1953930855,
0.4452639222,
0.0781019181,
-0.1183169782,
0.3155586123,
0.0447257087,
0.0146261156,
0.1478067189,
-0.1330284178,
-0.1975686848,
0.3923157156,
0.1372192204,
-0.0548009984,
-0.2756024003,
0.3441493213,
-0.0665345564,
-0.2472361028,
0.2151200622,
-0.2440623939,
-0.0380830318,
-0.531683743,
0.3191308081,
-0.3130173087,
0.0980823338,
0.4218280911,
0.1543095708,
0.3799421787,
-0.2579368353,
0.0727035627,
-0.0288273245,
0.3897281289,
0.1729900837,
-0.246169135,
-0.1826695949,
0.1881110072,
-0.9209601283,
0.4458392859,
0.0589725897,
-0.0390840285,
0.0334376991,
0.3439893723,
0.0537499934,
0.0531443618,
0.2887405157,
-0.0962833464,
0.1624650061,
-0.0415592641,
-0.3789631724,
-0.3341310918,
-0.3106013238,
0.4049320221,
0.1494322866,
-0.5924780965,
0.2898759544,
0.0384894274,
-0.0501694381,
-0.1049863547,
-0.2132235765,
0.1590006799,
-0.1347796172,
0.417912066,
-0.1402910203,
0.4938714206,
-0.0273793414,
-0.0191060193,
-0.2152630985,
-0.1111692786,
-0.295769453,
0.2690996826,
-0.1461156011,
0.0767934024,
0.1251031756,
0.0760325119,
-0.2987894416,
0.323461473,
0.1469842494,
-0.2812114358,
-0.4066528976,
-0.0215114206,
0.0377544612,
0.0922572464,
0.060123425,
0.3981385231,
-0.1498900503,
0.2334183156,
-0.1931372583,
-0.3373354971,
0.5052194595,
-0.3257141113,
-0.1926415414,
-0.2741821408,
0.252001524,
0.1326467693,
0.0052116271,
-0.6400473118,
0.0537734628,
0.3680357039,
-0.1895976812,
-0.1688363254,
0.1406752169,
-0.2344212085,
0.407517761,
0.0046432912,
0.0375302173,
0.0067866538,
-0.130524829,
0.1636307389,
-0.2550602555
] |
https://github.com/huggingface/datasets/issues/1963 | bug in SNLI dataset | Hi ! The labels -1 correspond to the examples without gold labels in the original snli dataset.
Feel free to remove these examples if you don't need them by using
```python
data = data.filter(lambda x: x["label"] != -1)
``` | Hi
There is label of -1 in train set of SNLI dataset, please find the code below:
```
import numpy as np
import datasets
data = datasets.load_dataset("snli")["train"]
labels = []
for d in data:
labels.append(d["label"])
print(np.unique(labels))
```
and results:
`[-1 0 1 2]`
version of datasets used:
`datasets 1.2.1 <pip>
`
thanks for your help. @lhoestq | 39 | bug in SNLI dataset
Hi
There is label of -1 in train set of SNLI dataset, please find the code below:
```
import numpy as np
import datasets
data = datasets.load_dataset("snli")["train"]
labels = []
for d in data:
labels.append(d["label"])
print(np.unique(labels))
```
and results:
`[-1 0 1 2]`
version of datasets used:
`datasets 1.2.1 <pip>
`
thanks for your help. @lhoestq
Hi ! The labels -1 correspond to the examples without gold labels in the original snli dataset.
Feel free to remove these examples if you don't need them by using
```python
data = data.filter(lambda x: x["label"] != -1)
``` | [
0.1758222878,
-0.2825254202,
-0.1035731584,
0.3496333659,
0.221977666,
0.047198087,
0.3874209523,
0.1183437407,
0.0808187574,
0.296924144,
-0.2027280778,
0.6121094823,
-0.1427202225,
0.0958904549,
0.0018210709,
-0.0009765243,
0.2330548167,
0.3379022777,
0.2097958326,
-0.496683836,
-0.2359666675,
-0.0235045943,
-0.3817801476,
0.2652833164,
-0.2869724035,
0.178448841,
-0.0155611746,
-0.133338064,
0.1334051192,
-0.4898703098,
0.1604640335,
-0.0986121446,
0.1226142049,
0.2325978726,
-0.0001058666,
-0.2743193507,
0.1129734367,
0.0678980649,
-0.369653672,
-0.2605839968,
-0.110146299,
-0.0220086202,
-0.1468922645,
-0.1628938615,
-0.2096337378,
-0.1724423915,
0.014251899,
0.0935409591,
0.1124732047,
0.2662245035,
0.2374286801,
0.0673878789,
-0.1264114678,
0.1347337514,
0.2391381562,
-0.3994210064,
0.1405960023,
0.1139144227,
0.0496282801,
0.0493756235,
0.4056082964,
0.6512388587,
-0.0620472841,
-0.0436956882,
0.1137126163,
0.0903738439,
0.1272949427,
-0.3254039288,
-0.0787117258,
0.3404746652,
0.1091917902,
-0.3263961971,
-0.4188058972,
0.0663315803,
0.1710608602,
-0.462331146,
-0.028323058,
0.2538722157,
0.1406981349,
0.203457728,
0.0178659707,
0.0447483994,
-0.0742186233,
0.3183243871,
0.0500647277,
0.6590714455,
-0.0445928946,
0.181916818,
0.0840542614,
0.0472339094,
-0.1453341246,
0.1597236991,
0.0685572326,
0.3448962569,
-0.5432954431,
-0.0704535097,
-0.0477738604,
-0.0830215216,
-0.1924337149,
0.1565323174,
0.1354234219,
0.0860434026,
-0.1599020809,
0.2858227193,
-0.0184926558,
0.012240041,
0.1436228603,
0.3150086403,
0.14040263,
-0.2455687225,
0.097809121,
-0.0502134003,
-0.1937010586,
-0.2610456347,
0.4902792573,
0.0590683743,
0.031471014,
-0.0605655238,
-0.3164853752,
0.238817662,
-0.3513688445,
0.0303710327,
-0.0477258638,
0.0027171411,
-0.016286632,
-0.012746349,
-0.002686128,
0.0518938601,
0.0969415307,
-0.2098996192,
-0.2651702166,
0.0638556778,
-0.0640472919,
-0.1516255736,
-0.0483410172,
-0.1381339431,
0.1749139726,
0.0200463347,
-0.3853113353,
-0.1688893288,
0.1255014241,
0.0578715019,
0.2358458191,
0.1766266972,
-0.2485657334,
0.4699510634,
-0.1297382563,
-0.1469859928,
0.0430616289,
-0.0063615106,
-0.2545923889,
-0.1627766043,
-0.3255552649,
0.2545196116,
0.0457579046,
-0.039669387,
0.07416372,
-0.1403102279,
0.146074757,
0.023989886,
0.3011032641,
-0.2547022998,
-0.1741402596,
0.0351590067,
0.1350889951,
-0.0714573264,
-0.1860972792,
-0.1540894508,
-0.149803251,
-0.220589608,
0.2482093722,
0.3248288035,
0.0830718651,
-0.0982796103,
-0.1811053753,
0.1455692947,
0.6714297533,
-0.4713215828,
-0.5759910345,
-0.0373700187,
-0.3127925992,
-0.3483294249,
-0.0722916573,
0.5257781744,
-0.2541722655,
0.1841238737,
0.055915691,
-0.028452713,
0.0702485815,
-0.0407906361,
-0.4820292294,
-0.2443045974,
0.4282408059,
0.2879722714,
0.118449077,
0.0484847166,
-0.0653521866,
-0.124533087,
0.3911310136,
0.0004526926,
0.1557040066,
0.2652798295,
0.5162926912,
0.1337869465,
0.1086845621,
-0.2567135692,
-0.0195437744,
0.1996146291,
-0.1025180519,
0.3829932213,
0.1140017211,
-0.0327499509,
-0.2263216525,
-0.2294124067,
-0.371419698,
-0.1633770466,
0.1727941632,
0.1106386185,
0.1249601245,
-0.0002136379,
-0.069063358,
0.2349172235,
-0.1951019466,
0.0574192069,
-0.2534843087,
0.1734799147,
-0.0722107962,
0.0002441742,
-0.2396927327,
0.1996300519,
0.194197163,
0.1150441766,
-0.1352172494,
0.2393587828,
-0.2861434817,
-0.2572815716,
0.128099069,
0.086521104,
0.0745255053,
-0.6160663366,
-0.2318004221,
0.5943588018,
0.1214366332,
0.0328817815,
-0.1761129051,
0.2059392631,
-0.1515800804,
-0.1561627388,
-0.3645669818,
0.2804862559,
0.1399023831,
-0.0466480777,
-0.0844028369,
-0.3972602785,
-0.0248348527,
-0.4928984344,
0.3556154966,
0.1288474798,
-0.3839582205,
-0.0716302395,
0.3216437995,
-0.050756067,
0.1607383788,
-0.0180258621,
-0.1524933875,
0.0303001702,
0.2578043044,
0.118063435,
0.148019433,
0.192257598,
-0.0845913738,
0.1927810013,
-0.1528930664,
-0.1079918146,
-0.0311147571,
0.1146153212,
-0.01901008,
0.0687715709,
0.2076686323,
0.1067496687,
-0.1778412908,
-0.2461327016,
-0.1395558715,
0.268166095,
-0.232405439,
-0.0760475844,
-0.1674964726,
-0.2094909549,
-0.5057201385,
-0.1521825492,
-0.0507109575,
-0.0992071033,
0.1847667545,
-0.0889074057,
-0.0105203986,
0.3142930269,
0.1960568428,
0.1012290418,
0.145154953,
0.1357938945,
-0.1855855584,
-0.3651710153,
-0.2351708859,
0.0717458427,
-0.3185688555,
0.2699110806,
0.362963289,
-0.1586461365,
-0.0961318016,
-0.0891948789,
-0.2441946566,
-0.0196912196,
0.0663826093,
0.0541800037,
0.0247738864,
-0.0128799863,
-0.2053207755,
0.2964855731,
0.2392027378,
0.0107819885,
-0.0804007053,
0.1651144624,
-0.0892624184,
-0.0484051816,
-0.2911716402,
-0.4643013179,
-0.5519577265,
0.0023523718,
0.016096374,
-0.1708484441,
0.1917136014,
0.1217431501,
0.0539651588,
-0.0223812629,
0.1511523575,
0.1787068546,
-0.2152537405,
-0.1564622372,
0.169982031,
0.0465795659,
-0.2690292001,
-0.1365804672,
0.0395991281,
-0.0324103683,
-0.1843266338,
-0.2561137676,
-0.1648022681,
0.1473017484,
0.1070361286,
0.0568778329,
0.0316861346,
-0.0339902602,
0.1185099483,
-0.0793801099,
-0.2076564282,
0.1215778738,
0.1340609938,
-0.3273540139,
0.4176497459,
-0.0254116859,
0.2975834608,
-0.008934699,
0.0571755096,
0.6455109119,
-0.1563931108,
0.0258553103,
-0.159740299,
0.0220549759,
0.1431916952,
-0.257386744,
0.1051397324,
0.4043577313,
-0.037511237,
0.1361240149,
-0.030933762,
0.0562434569,
0.00221885,
0.1778098643,
-0.2866261005,
-0.042188298,
-0.2883037329,
0.0081802951,
0.0121035269,
0.0024795607,
-0.1297769547,
-0.016993776,
-0.1806279421,
0.000050649,
0.1284519583,
-0.1970237345,
0.1025351733,
-0.2461900562,
-0.1150920391,
0.0435955822,
-0.1480238736,
0.1687988192,
0.3643046618,
0.1497537494,
0.164403528,
0.2874141932,
0.2821746469,
0.606143713,
-0.5048745275,
0.2027046084,
0.2420896739,
0.4977802038,
-0.3383361101,
-0.3108196557,
-0.4152237773,
-0.23854509,
0.4774545133,
0.1917116046,
0.0423534513,
-0.1758730561,
0.2739939988,
0.1112618297,
-0.3431811333,
-0.1644871533,
-0.2301605046,
-0.0410894491,
-0.11274001,
0.1340054125,
-0.063691251,
0.1891909391,
-0.097316511,
-0.214648068,
0.0549488589,
0.1077020764,
0.24982813,
0.1666043997,
0.1703021526,
0.1368239075,
0.0883759707,
0.0453658812,
0.124310568,
0.0230896454,
0.3792883158,
-0.1508145779,
-0.3487551212,
-0.0490251742,
-0.4373479784,
0.074811466,
-0.0211913362,
-0.0076099671,
-0.0620713122,
-0.1066272557,
0.3135817349,
0.0335742906,
0.125033021,
0.0776628852,
0.2430592477,
-0.4002364576,
-0.3817455769,
0.564103961,
0.1614263952,
-0.3126892447,
0.1298295707,
-0.1115404218,
-0.232385993,
0.2495946884,
0.1741064787,
0.8710386753,
-0.0439969338,
-0.0777832568,
0.1872612387,
-0.5491982698,
0.4160416424,
-0.0962839127,
-0.156722948,
-0.5003466606,
0.0170134492,
0.0462397784,
0.0295671895,
-0.3262171149,
0.3848620951,
-0.0168425068,
0.3335962594,
-0.0721831173,
0.1315602064,
0.0626657531,
-0.0263761431,
0.3919156194,
0.1101384014,
-0.1359796822,
0.1998816729,
-0.0867555737,
0.36509794,
-0.1807791293,
0.0051469021,
-0.1593275219,
0.0511811301,
-0.1213787571,
-0.0190218501,
0.1898369789,
0.2132985294,
0.0453389548,
-0.1771474034,
-0.3371669054,
0.3609565496,
0.7153918743,
0.0166355819,
-0.0032236725,
0.0695008934,
0.2405068576,
0.4848548472,
0.0438015983,
0.2695525885,
0.4295511246,
-0.0175233632,
-0.3549790382,
0.1668362617,
0.1424149573,
0.152337268,
-0.4034619927,
0.0467116535,
-0.1067343801,
0.1847809851,
0.1023439318,
-0.0712309778,
0.2050368935,
-0.2408464849,
0.1963704228,
0.0711250305,
-0.1680079699,
0.0077840071,
-0.0060903905,
-0.2926674187,
-0.1471736133,
0.3525078297,
0.1104484051,
0.1597245336,
-0.0106634703,
-0.1213547811,
-0.0746301934,
-0.262301445,
-0.0025095493,
0.2081128657,
-0.4890867174,
0.1244212091,
-0.0861644894,
0.019864291,
-0.1201540381,
0.2211391926,
0.0576069579,
0.2330546379,
0.1238746494,
-0.2934542298,
-0.4125736654,
-0.1974736452,
0.0958211273,
0.3416710198,
-0.1268661916,
0.1598140299,
-0.1841473877,
-0.1776506603,
-0.3772060871,
-0.0511798002,
-0.2954211235,
0.2258237898,
0.1912424117,
0.2830272317,
-0.185222134,
0.0047179572,
0.1705357432,
-0.0288295504,
-0.0857433379,
-0.2370088249,
-0.2462336868,
0.0528251007,
0.2717523277,
-0.2514605522,
-0.0944944024,
0.1243265122,
0.060783755,
0.0364015996,
0.2265705168,
0.247709617,
0.2358564138,
0.9074829817,
-0.193711251,
0.0112098381,
-0.0809668079,
-0.0974367708,
0.0123715233,
0.0009788908,
0.0554630756,
0.0699675381,
0.1669417322,
0.0968660116,
-0.2715979815,
-0.1088642031,
-0.3038883805,
-0.0061022616,
0.010286767,
-0.5755165815,
0.4176029265,
0.2746882737,
0.2566926479,
0.0816470981,
-0.3946268559,
-0.2472678721,
0.2157204449,
0.3618086278,
-0.2296124995,
-0.0064820312,
0.2608966827,
0.1352834105,
0.0575638264,
0.0501227379,
-0.0629427433,
-0.129291907,
-0.0882651955,
0.036922507,
0.1134368405,
-0.3506599367,
0.0868880749,
0.4115595222,
0.0686952546,
0.0167494956,
0.1330556124,
-0.0848110169,
-0.1213138252,
0.641063571,
-0.0264559016,
0.4369121194,
-0.0552828982,
0.1763094068,
-0.2318537831,
0.1728785932,
-0.063804388,
0.1332535893,
-0.4932236969,
0.0759486258,
-0.0145826228,
0.2183900326,
-0.0881680474,
-0.2310976684,
0.1531071663,
0.2232790589,
-0.1793227643,
-0.25153929,
-0.3708984256,
0.1215015054,
-0.0856519789,
0.2131280601,
0.008108452,
-0.0478744544,
-0.0732748881,
0.198565051,
-0.2597442865,
-0.4831647277,
-0.4174052477,
0.4844006896,
0.3321519494,
-0.202314496,
-0.0913057551,
0.1243589967,
-0.0094579346,
-0.0002586874,
-0.1325835586,
0.1982094496,
0.180758208,
0.3041765094,
0.0009502694,
-0.0164919421,
-0.1214538664,
-0.0175774321,
-0.0014497545,
0.2395069748,
-0.0714898556,
0.101791054,
0.1085180566,
-0.2647215426,
0.4198532701,
0.0897680074,
0.5018247366,
-0.3040900528,
0.2026568353,
-0.2344893664,
-0.2914104164,
-0.4042403996,
-0.2128189951,
-0.1952726841,
0.2986360192,
0.1747798473,
0.4102655351,
0.242488578,
-0.2421735823,
0.1144187078,
0.0731976107,
0.1749114841,
0.1492751837,
0.0461607836,
-0.0892123282,
-0.0291164741,
-0.6194485426,
0.1014878601,
0.0836123824,
0.0044274628,
-0.0240105651,
-0.0721827149,
-0.5134979486,
-0.1284177303,
0.1638018489,
0.2362140566,
0.0536449775,
0.417386204,
-0.1874000281,
-0.0113129737,
-0.0348636806,
-0.0060417168,
0.2730380297,
-0.1915695071,
-0.0835699737,
-0.2481810153,
0.1709205657,
-0.1321861595,
0.0614206642,
-0.1478130519,
0.2032230645,
0.4639740586,
-0.0967972502,
0.3414816856,
0.04732164,
0.2152623236,
-0.1106108278,
-0.084546715,
-0.3863245547,
0.0817852467,
-0.138954863,
0.49321872,
0.1644179821,
-0.1385681033,
-0.204620868,
0.1893076152,
0.0041489378,
-0.1906442791,
-0.283673197,
0.3273252249,
-0.0022105351,
0.138322264,
0.0360358581,
0.0244636126,
0.208770588,
0.2192705721,
-0.2801861763,
-0.5413847566,
0.3990260661,
-0.2757388949,
-0.0321879387,
-0.2142977118,
0.3366419673,
0.393022716,
-0.2063712925,
-0.6328854561,
-0.0117066801,
0.3883119822,
-0.1035702974,
-0.4395234287,
0.0690474659,
0.2488585711,
0.2722865045,
-0.0126968846,
-0.1619984508,
0.2330732644,
-0.0640471205,
0.4385648966,
-0.2267674059
] |
https://github.com/huggingface/datasets/issues/1959 | Bug in skip_rows argument of load_dataset function ? | Hi,
try `skiprows` instead. This part is not properly documented in the docs it seems.
@lhoestq I'll fix this as part of a bigger PR that fixes typos in the docs. | Hello everyone,
I'm quite new to Git so sorry in advance if I'm breaking some ground rules of issues posting... :/
I tried to use the load_dataset function, from Huggingface datasets library, on a csv file using the skip_rows argument described on Huggingface page to skip the first row containing column names
`test_dataset = load_dataset('csv', data_files=['test_wLabel.tsv'], delimiter='\t', column_names=["id", "sentence", "label"], skip_rows=1)`
But I got the following error message
`__init__() got an unexpected keyword argument 'skip_rows'`
Have I used the wrong argument ? Am I missing something or is this a bug ?
Thank you very much for your time,
Best regards,
Arthur | 31 | Bug in skip_rows argument of load_dataset function ?
Hello everyone,
I'm quite new to Git so sorry in advance if I'm breaking some ground rules of issues posting... :/
I tried to use the load_dataset function, from Huggingface datasets library, on a csv file using the skip_rows argument described on Huggingface page to skip the first row containing column names
`test_dataset = load_dataset('csv', data_files=['test_wLabel.tsv'], delimiter='\t', column_names=["id", "sentence", "label"], skip_rows=1)`
But I got the following error message
`__init__() got an unexpected keyword argument 'skip_rows'`
Have I used the wrong argument ? Am I missing something or is this a bug ?
Thank you very much for your time,
Best regards,
Arthur
Hi,
try `skiprows` instead. This part is not properly documented in the docs it seems.
@lhoestq I'll fix this as part of a bigger PR that fixes typos in the docs. | [
0.1184593439,
-0.5157737136,
-0.0002606697,
0.1341338754,
0.2405334562,
0.3162594438,
0.3996222019,
0.1163899973,
0.1897607744,
0.2767885923,
0.3576666713,
0.3101139665,
0.0523399152,
0.1391007602,
0.2122828215,
0.0097792428,
0.3222461045,
-0.09710899,
-0.2278699577,
-0.142526269,
-0.4665182233,
0.2059865296,
-0.4954794943,
-0.0062194318,
0.0046918504,
0.2700181603,
-0.1468361914,
0.3250584304,
0.1460926831,
-0.0820217133,
0.4361982644,
0.0927888453,
0.0622522533,
0.4919845462,
-0.0001279238,
-0.0485201031,
0.0909044221,
-0.1174368933,
-0.25088346,
-0.2423612922,
-0.0103978962,
0.1511082649,
-0.1586595923,
-0.3461463451,
-0.079708524,
0.2208736241,
-0.1931835413,
-0.3198305666,
0.0172094218,
0.2279164195,
0.1132520735,
0.0995626152,
-0.2322336733,
0.013390705,
0.3679782748,
0.1688150764,
-0.202867955,
0.5136338472,
0.074979268,
-0.131292969,
0.217003271,
0.3002196252,
-0.2162893116,
0.1819369346,
0.6969505548,
0.2346368134,
-0.0080312807,
0.067085892,
0.1547786891,
0.1427003592,
0.3508743644,
-0.2583888471,
-0.1454956234,
-0.5243524313,
0.114811644,
-0.4422194958,
0.3829134405,
-0.0757528618,
-0.1903963387,
0.1892316341,
-0.1961866319,
-0.3662990034,
-0.2399331033,
0.1452040374,
0.2731885016,
-0.3786745071,
-0.1079847664,
0.2083943337,
0.659031868,
0.0167084858,
-0.0740461871,
-0.145029977,
0.1362964958,
0.411984086,
-0.5192497969,
0.0457274765,
0.2632990777,
0.5026041865,
0.1499709785,
0.1526492536,
-0.0932037383,
-0.4410411716,
0.4102215171,
0.1674042642,
0.1500698924,
-0.004112076,
0.1651002765,
0.2374446392,
0.1703531444,
-0.0576959141,
0.1190756336,
-0.0006030058,
-0.0619616807,
-0.2427826524,
-0.0386946239,
-0.2457967997,
0.3429153264,
-0.2678746581,
-0.237525776,
0.1017668843,
-0.1411717236,
-0.134310618,
-0.2298790514,
0.3612948656,
0.3145488501,
-0.1731252074,
0.0057554767,
0.2405340672,
-0.1080144867,
-0.1394805759,
-0.1983849555,
0.1062841564,
-0.0351235755,
0.2528304458,
0.0619632229,
-0.4568334222,
0.0769030228,
0.0545501336,
0.3502004743,
-0.0496461429,
-0.0791886374,
-0.038057588,
-0.0190582015,
0.2549312115,
-0.13630189,
-0.1448656321,
0.297691226,
0.1185017824,
-0.0435421243,
0.5042290688,
-0.2794050574,
-0.2222584933,
0.0607969239,
-0.0945081413,
-0.1695892513,
0.0945609212,
-0.6603602171,
0.2222949266,
0.1386061013,
-0.201570183,
0.0344229117,
-0.006375608,
-0.3262870014,
-0.2263972908,
0.2093480974,
0.537298739,
-0.5187488794,
-0.3551662266,
0.1033999622,
0.0400551744,
0.0190912634,
-0.011484162,
-0.0296289641,
0.1079791784,
-0.3077189326,
0.2494835556,
-0.0521682054,
-0.2833322287,
-0.1313026994,
0.0714628994,
-0.0621325485,
0.2180784643,
0.0261953268,
-0.2533246577,
-0.020431377,
-0.1669731438,
0.2533343434,
0.1141674519,
-0.2502170801,
0.0942133814,
-0.1150973663,
-0.1965747774,
0.3166657984,
0.0250526667,
-0.296420753,
0.114484787,
-0.2064063996,
-0.5897095799,
0.3778427541,
-0.0728812814,
-0.1749579906,
0.1832054704,
0.0438153408,
0.1798685044,
0.2096752971,
0.0379983485,
-0.6307150126,
0.108689189,
0.4949718416,
-0.0462615788,
-0.092327185,
-0.2249065191,
-0.2521228194,
-0.054898452,
0.1884771287,
0.2854734361,
-0.132773608,
0.0781118944,
-0.1239917874,
-0.0686810762,
-0.1948325038,
0.1549472213,
-0.3675633669,
0.1537379473,
-0.4071646929,
0.2684522867,
0.2598710954,
-0.0959239453,
0.0065903664,
-0.0984928906,
0.2952401936,
-0.3557018638,
-0.132019937,
0.450560838,
0.2342851013,
-0.0091307424,
0.0868651718,
0.0611075163,
-0.1808294356,
0.0192997605,
-0.1605569273,
-0.1600668132,
0.1238658652,
-0.0958672762,
-0.1965262145,
0.3076045215,
-0.3604245186,
0.3255198598,
-0.2628008127,
-0.0352185741,
-0.0754592791,
-0.0341865867,
-0.0842180476,
-0.197915405,
0.5240817666,
-0.0061131176,
0.1118531749,
0.0035340097,
-0.521261394,
-0.2066821456,
0.3830851018,
-0.0621212572,
-0.2070294023,
0.3191438019,
0.1557872593,
-0.0262889415,
0.3411842287,
-0.1251763403,
0.5859798789,
0.0449354723,
-0.3043901324,
-0.066645585,
0.1478591263,
-0.0423214361,
0.2223358899,
0.0385462604,
0.1551992744,
-0.2557176352,
-0.1896425486,
0.0230878927,
-0.0583357811,
0.1618838608,
0.0453285873,
0.1793261617,
-0.6254635453,
0.2487066537,
-0.0588402823,
-0.1682410538,
0.1682817191,
0.0879585817,
-0.3060453832,
-0.2525308132,
-0.084233135,
0.0649742931,
-0.1292808503,
0.2338947356,
-0.2523552477,
0.6201031208,
-0.1122031882,
-0.2239249945,
-0.3348615766,
-0.0328213088,
-0.0588840358,
-0.0667504147,
0.0557749234,
0.1014679223,
0.3207905293,
-0.0589229725,
-0.13965258,
-0.0037372261,
-0.0733496994,
0.1602298766,
-0.3428145647,
0.3066673875,
0.2471903861,
0.3997299075,
-0.1542535424,
0.1513815671,
0.0548964627,
-0.0476549789,
0.1007011458,
0.033522591,
0.253308177,
0.0797876641,
-0.29209584,
0.0357822739,
-0.1299383044,
-0.0000387207,
0.2213921249,
0.0198711827,
-0.0376497433,
-0.0669297576,
0.1179809347,
0.0733958781,
-0.0719337314,
-0.0484875254,
-0.3079356551,
-0.6267730594,
0.2409047484,
-0.0264317114,
-0.1951417029,
-0.0236928761,
0.3533587456,
-0.0311959367,
-0.1339343637,
-0.4656541348,
-0.2469316125,
0.1305473298,
0.151763007,
-0.1969884336,
0.0388504192,
0.0503356531,
-0.2450461686,
0.1536441743,
-0.4438824058,
-0.4812512994,
0.08252877,
0.1209162921,
0.1065653265,
0.4717404246,
0.2879651189,
-0.2733363807,
0.3163411915,
0.1464788616,
0.3638207912,
0.3352498114,
-0.1675288379,
0.2650375962,
0.0734613687,
-0.3172515035,
0.1638059765,
0.0464832671,
-0.2111839801,
0.1785170734,
0.0041285902,
0.1407556832,
-0.2208393067,
0.0856052786,
-0.0104810148,
-0.389654547,
0.0177371893,
-0.2655602992,
-0.0000429116,
0.0667146742,
0.268460989,
-0.4285846353,
-0.0578811727,
-0.0875502974,
0.1557599604,
0.6310109496,
0.0573400334,
-0.073030211,
-0.0804760009,
-0.3710909486,
0.3203974366,
0.0968664438,
-0.0332949422,
0.0873861536,
-0.0594171919,
0.12381161,
0.1924567223,
0.9702677131,
0.1808312684,
0.1366526932,
0.1105831861,
0.2175791562,
-0.3055568933,
-0.0134121254,
0.0733852535,
0.2672161758,
0.3817239702,
0.2656690776,
-0.4033344388,
-0.2226453424,
0.2769917548,
0.2830925584,
-0.0478490666,
-0.3223298192,
-0.3425840139,
-0.1045977995,
-0.3616629541,
0.1127391905,
0.2983872294,
0.0953748822,
-0.377410084,
0.364978075,
0.2887409925,
-0.1708687097,
0.0014056973,
0.2205380052,
0.2984917164,
-0.3236279488,
-0.0663358271,
0.4265283644,
-0.2523587644,
0.4026011229,
0.6713854074,
-0.3897367716,
-0.4563657045,
-0.0650465339,
-0.5007568002,
0.5561572909,
0.2915156484,
-0.3470169306,
0.2494366765,
-0.0746388882,
0.0408954099,
-0.1202683896,
-0.3268797398,
0.4433286488,
0.0008113943,
-0.1251004338,
-0.357588172,
0.2981075943,
-0.085907191,
-0.1844008416,
0.4269385636,
0.7055801749,
-0.2070474923,
0.074295111,
-0.4285740852,
0.9876384139,
0.4479187429,
0.1335485727,
0.1885948032,
0.2559273839,
0.5449032784,
0.4156981111,
-0.0394430608,
-0.116679661,
-0.3400417566,
0.0358402953,
-0.0436227918,
0.0801741034,
0.1187169105,
-0.1670794785,
0.2674461603,
-0.0833326802,
0.0977663845,
-0.0778848231,
-0.3814985156,
0.3286144733,
-0.0183285438,
-0.2122381032,
-0.0213965159,
0.1279188693,
0.0291783437,
0.0027432516,
0.0765293241,
0.0225500464,
-0.4613628983,
-0.3341674209,
-0.0798491165,
-0.1575939655,
-0.3594914675,
0.2929362655,
-0.271004498,
0.2584755123,
0.1882322133,
0.0791061893,
-0.1033837795,
-0.2842766643,
-0.0838429779,
0.0818977728,
0.1955945194,
-0.3937596679,
0.0037561543,
0.1803209484,
0.0484675169,
-0.2307668477,
0.0925994366,
-0.1555214226,
-0.4176952839,
0.0138240308,
0.0713842958,
-0.3471813202,
-0.3193705082,
-0.4029920101,
-0.0582876801,
0.3188605011,
-0.2722457647,
-0.0319174752,
0.1824083924,
0.0771119893,
0.025236737,
-0.0104834083,
-0.341566205,
-0.0579077005,
0.2334397733,
-0.1890172064,
-0.1578361839,
0.229095459,
0.3940143585,
-0.1259292364,
0.0110726878,
-0.0159359798,
0.1251060218,
-0.4891855121,
0.5048874021,
0.4031615853,
0.4566210508,
0.1132586077,
0.1569547653,
0.0025208765,
-0.131683737,
0.0398230851,
-0.1272896826,
0.0206687152,
0.1495549083,
0.22389099,
0.150229156,
0.1870160699,
0.0887275636,
0.0358880162,
-0.0848182663,
-0.1349641085,
0.0610038042,
-0.1800457239,
-0.1442871094,
0.734351933,
0.4233570695,
0.2425296456,
-0.1966067553,
-0.0783286393,
0.0659481287,
0.0952813178,
-0.041514501,
-0.3675373197,
0.2147177309,
-0.0072648991,
-0.0344186388,
0.0490242057,
-0.2043952644,
-0.1229929551,
-0.2577932775,
0.2782375216,
0.08526057,
0.0525207072,
0.2377353311,
0.3915508091,
0.103222549,
-0.0592512935,
0.1123412997,
0.0507622249,
0.2624181807,
-0.181173712,
0.1087928489,
-0.1437574923,
0.0702597797,
-0.0417995267,
-0.0027775094,
0.0403459966,
0.171027571,
-0.0305337496,
-0.1313511133,
0.1698530465,
0.0017845184,
0.0675756633,
0.3770520389,
-0.3807774782,
0.0984960794,
0.0769971833,
-0.0105652921,
-0.3598611951,
-0.1977273822,
0.2949621677,
0.1285373271,
-0.0150994286,
0.2488168478,
-0.2123206854,
0.0362749696,
-0.1478244811,
-0.1595071405,
0.5118710399,
0.068044588,
0.008655455,
0.5517325401,
-0.1172772348,
0.2473681271,
0.0751563311,
0.3716818392,
0.2079108208,
0.3752680719,
-0.0928772986,
-0.1525135934,
0.0226768367,
0.0332648158,
-0.3393267095,
-0.6127128601,
0.1138675809,
0.4662081599,
-0.4114248157,
0.0233356729,
0.012315847,
0.3759468496,
0.0782102793,
-0.1980501264,
-0.1014161706,
0.0741260424,
0.0208277162,
-0.1253558695,
-0.2443071306,
0.0554778129,
-0.04717917,
-0.0114518031,
-0.2020471543,
-0.0929235816,
0.6958203316,
-0.1093802154,
0.2166820467,
-0.264477253,
-0.06233114,
-0.200268954,
0.1111037731,
0.0200598389,
-0.0145323696,
-0.0366860181,
0.0475838482,
0.337798804,
-0.0703435391,
0.3302781284,
-0.0045206333,
0.3747520447,
-0.3745862544,
-0.2005813569,
0.0350319296,
0.0564237311,
0.2205988616,
-0.0823414326,
0.290530473,
0.1375498772,
0.0074117146,
-0.0456091687,
0.4056409895,
0.2595394254,
0.1267183721,
-0.5624562502,
0.2345647514,
-0.4512850642,
-0.2537252307,
-0.4652950168,
0.1222000644,
0.0452434272,
0.1385638863,
0.1585768759,
0.1255732775,
-0.0869323015,
0.0617628135,
0.0072453097,
0.1048959717,
0.4281701446,
0.2795760036,
0.2275765538,
-0.4540309906,
-0.1966571808,
-0.4089964032,
0.10941495,
-0.0585409291,
0.1503302306,
0.4332965314,
-0.1026706994,
0.215516746,
0.0754045844,
0.2592981458,
0.0064721182,
-0.0435533747,
0.1727090478,
-0.1886401922,
-0.5312672853,
-0.2145920545,
0.0049194377,
0.1421577185,
-0.1503656209,
0.1773812473,
-0.0768594146,
-0.1850156784,
-0.2282262444,
0.1627286226,
0.1208617538,
0.1198450625,
0.2785859108,
-0.1667266637,
0.2345070243,
-0.0160356462,
-0.063096568,
-0.0092435181,
-0.0470899865,
-0.3542246521,
-0.0529597923,
-0.2087928504,
-0.000917159,
-0.381829083,
-0.3481459022,
0.0208477825,
0.0150352232,
0.1303403229,
0.2017200142,
-0.4272089899,
0.2923904955,
-0.074792698,
0.0261828676,
-0.14066571,
-0.0978341848,
-0.1842039824,
0.3183945119,
-0.0623557717,
0.0342245176,
0.8977257013,
-0.2795377076,
-0.1409466416,
-0.3316088915,
0.1682139337,
0.1322269142,
-0.5405361652,
-0.6190088391,
-0.0539385602,
0.2388799787,
0.3874327838,
-0.2610633075,
0.5420950651,
-0.3744543791,
0.1921135187,
0.0887778103,
0.4205941856,
0.1564857215,
-0.0732058585,
0.2644814849,
-0.2102508992
] |
https://github.com/huggingface/datasets/issues/1956 | [distributed env] potentially unsafe parallel execution | You can pass the same `experiment_id` for all the metrics of the same group, and use another `experiment_id` for the other groups.
Maybe we can add an environment variable that sets the default value for `experiment_id` ? What do you think ? | ```
metric = load_metric('glue', 'mrpc', num_process=num_process, process_id=rank)
```
presumes that there is only one set of parallel processes running - and will intermittently fail if you have multiple sets running as they will surely overwrite each other. Similar to https://github.com/huggingface/datasets/issues/1942 (but for a different reason).
That's why dist environments use some unique to a group identifier so that each group is dealt with separately.
e.g. the env-way of pytorch dist syncing is done with a unique per set `MASTER_ADDRESS+MASTER_PORT`
So ideally this interface should ask for a shared secret to do the right thing.
I'm not reporting an immediate need, but am only flagging that this will hit someone down the road.
This problem can be remedied by adding a new optional `shared_secret` option, which can then be used to differentiate different groups of processes. and this secret should be part of the file lock name and the experiment.
Thank you | 42 | [distributed env] potentially unsafe parallel execution
```
metric = load_metric('glue', 'mrpc', num_process=num_process, process_id=rank)
```
presumes that there is only one set of parallel processes running - and will intermittently fail if you have multiple sets running as they will surely overwrite each other. Similar to https://github.com/huggingface/datasets/issues/1942 (but for a different reason).
That's why dist environments use some unique to a group identifier so that each group is dealt with separately.
e.g. the env-way of pytorch dist syncing is done with a unique per set `MASTER_ADDRESS+MASTER_PORT`
So ideally this interface should ask for a shared secret to do the right thing.
I'm not reporting an immediate need, but am only flagging that this will hit someone down the road.
This problem can be remedied by adding a new optional `shared_secret` option, which can then be used to differentiate different groups of processes. and this secret should be part of the file lock name and the experiment.
Thank you
You can pass the same `experiment_id` for all the metrics of the same group, and use another `experiment_id` for the other groups.
Maybe we can add an environment variable that sets the default value for `experiment_id` ? What do you think ? | [
-0.3388577104,
-0.4435220063,
-0.012153469,
0.0259140879,
-0.0003200229,
-0.1188516989,
0.4191743731,
-0.0220881347,
0.6945429444,
0.3347569704,
-0.0704572797,
0.2575861812,
0.0208578892,
0.0921889022,
-0.132985875,
-0.029275272,
-0.0619417727,
-0.1000038534,
-0.1330745369,
-0.1348368973,
-0.3230049014,
0.1404808462,
-0.1086489037,
0.0332358964,
-0.1115665212,
-0.05100815,
-0.2452467084,
0.2668409944,
-0.2192959189,
-0.3962018788,
-0.0470971391,
0.6610614061,
-0.0401508957,
0.3781708479,
-0.0001052099,
-0.0733862147,
0.1403620988,
-0.1786129177,
0.1255245656,
-0.0316812471,
-0.0004163608,
-0.2217449248,
0.1787691116,
-0.584471941,
0.1618597507,
-0.374935627,
0.1602334082,
-0.6200057864,
0.5229046345,
-0.0920470729,
0.1828145683,
0.4089612365,
-0.3570384979,
-0.1778224707,
0.0755405575,
-0.0741305798,
0.0622762069,
0.4285986722,
0.438211143,
-0.1648913622,
-0.3774005175,
0.047923062,
-0.0095326975,
0.41122666,
0.523832798,
0.1457094848,
0.1289580166,
-0.2262365371,
-0.3590705693,
0.543749094,
0.0621518716,
-0.0059187687,
-0.2175003737,
-0.0875972211,
-0.0445221439,
-0.201862514,
0.0316545777,
0.0730778649,
-0.275506407,
0.0701167583,
-0.0908006653,
0.1095008701,
-0.2334239781,
-0.0166135244,
0.1179820746,
0.4391714931,
0.1565183252,
0.3073608279,
0.4072219729,
0.174400419,
-0.5871927142,
0.3229157031,
-0.1573163718,
-0.137491256,
-0.2163518816,
-0.0715377182,
-0.13450104,
-0.2474487275,
0.2707518041,
0.2180186808,
-0.1847602427,
-0.034735471,
0.5012087226,
0.2264459729,
0.1733481139,
0.2225487381,
-0.0145002753,
0.2927634418,
0.3371620178,
0.0463934131,
0.0130741149,
0.1034734249,
0.2154529691,
-0.0866064876,
0.040712975,
0.0291973166,
0.0415123999,
-0.0102339089,
-0.4665406346,
0.1799821258,
0.129678309,
-0.0897995606,
0.1274872869,
0.2349567264,
0.2761308849,
-0.1984883547,
0.2240096033,
0.0891116709,
-0.2850963175,
0.2397985458,
-0.155806452,
-0.0192007087,
-0.392246753,
0.0479355603,
-0.0023124069,
0.2354629636,
0.3477460146,
0.0911862776,
0.3725244999,
0.0711540952,
0.1579837501,
0.1498430967,
0.1748496741,
0.1164056212,
-0.0279611312,
-0.1533218324,
0.1613994092,
-0.0924463123,
-0.1918703914,
-0.2538034618,
-0.299000591,
-0.1078140959,
-0.0957442522,
0.2104452103,
-0.4288645387,
0.1794065982,
0.3127838373,
-0.0441896245,
-0.0737615526,
-0.135402143,
0.4412112236,
-0.1397589445,
-0.0608982779,
-0.1054958105,
0.060392186,
0.5278874636,
0.1167065203,
-0.3138533533,
0.1733955741,
-0.213150546,
-0.0626811907,
0.2512880564,
-0.073137641,
-0.1633245945,
-0.1726193428,
0.0451324433,
0.1839678288,
-0.5704572201,
-0.4551756978,
0.2267525494,
-0.4629926682,
-0.1833775789,
0.5246549249,
0.3992156982,
0.4919352531,
-0.1816743314,
0.1842802167,
0.1284999847,
-0.0225712694,
-0.0210046247,
-0.2199541628,
-0.2132083178,
0.2055386603,
0.1335552931,
-0.0407901108,
-0.0299865752,
0.1842959225,
0.0449085906,
0.3806377649,
0.0599329993,
0.0494305938,
0.0249998085,
0.2068835199,
0.0262045488,
0.1731615812,
0.0735589862,
-0.0773378536,
0.1987993717,
-0.2551989555,
0.078685075,
0.1668830067,
-0.2616325617,
0.0597445965,
-0.1807134151,
-0.3017634749,
-0.3694490194,
0.133928299,
0.1007094681,
-0.4253525436,
-0.2251141071,
-0.176315546,
0.4126907289,
-0.0026025474,
-0.0783253759,
-0.1774894297,
-0.1190526113,
-0.2866606712,
0.0046238638,
-0.0351303816,
0.1059896946,
0.1628840715,
-0.1626183987,
-0.1835404038,
0.3546157777,
0.3039297462,
0.1092464328,
0.3355322182,
0.1854080558,
0.1129208356,
-0.1525606811,
0.1541460752,
0.0319652073,
0.0296031386,
0.0468521863,
-0.1756986678,
0.3792438507,
-0.1660758406,
-0.0425317623,
-0.1739492714,
0.2052102238,
-0.040757928,
-0.175989002,
-0.3004975021,
-0.1131734848,
0.5896609426,
0.2008984387,
0.187855348,
0.0096998736,
-0.0725956708,
0.0389043167,
0.0934053957,
0.1792058647,
0.0130480733,
-0.1290573031,
0.2640553713,
0.0338387564,
0.0100354459,
0.3682820797,
0.5299091339,
0.2942912877,
-0.0035121311,
0.2998713255,
-0.0220678914,
-0.2826882005,
-0.0080960393,
0.0303523354,
-0.1510361284,
0.1437176466,
-0.042793829,
0.0623151548,
-0.1387621611,
-0.2496974766,
0.1067960858,
0.0581836104,
-0.3008230925,
0.2420987785,
-0.0928125009,
0.2503349483,
-0.0295380652,
-0.1750534773,
-0.1717952788,
-0.3238007724,
0.2914565206,
0.0818373784,
-0.4521722198,
0.1862104237,
0.0592271835,
0.3545639515,
-0.1497488916,
0.0780102387,
-0.0674722344,
-0.3504399061,
-0.0287413467,
0.0338268802,
0.0693551302,
0.2826059163,
0.7376534939,
-0.089564383,
-0.1716842949,
-0.3883165419,
-0.1675385237,
-0.0548319519,
-0.29273808,
0.5755819678,
0.3770565391,
0.1170967072,
-0.1565520167,
-0.2018890381,
0.1285162121,
-0.4124997556,
-0.1544498503,
-0.1241047829,
0.0733179152,
-0.2878054082,
-0.3132178187,
-0.425173521,
-0.3864904642,
-0.3381044269,
0.1782754213,
-0.1811826527,
0.1448942721,
0.0601225831,
-0.3515413404,
-0.0152209401,
0.2928806543,
-0.1243956685,
-0.4747362137,
-0.6132941842,
0.1247876063,
-0.3289965987,
-0.2105765045,
-0.0699183345,
0.2135722935,
-0.1581747681,
0.2809073925,
-0.2412757874,
-0.7833878398,
0.1669144183,
0.1982199252,
-0.1628027856,
0.0633635148,
0.2734371126,
-0.0901560485,
-0.1465457678,
-0.1769039333,
0.0182818994,
0.5482853055,
-0.0525148772,
0.1894310415,
-0.1148563474,
0.0894955695,
0.0083573684,
0.5683792233,
0.3369275928,
0.0129040182,
0.4058046043,
-0.1494743526,
0.4014565349,
0.1018191129,
-0.2884703875,
0.3126340806,
0.2290748656,
-0.2266397178,
0.0509812087,
-0.2824379802,
0.214992404,
-0.0292828418,
-0.1109165922,
0.1569551826,
-0.388176918,
-0.1111626998,
0.0122718159,
0.0890299827,
-0.2054054737,
0.1362082809,
0.0220757797,
-0.0265205782,
0.0318309627,
0.5858515501,
-0.0124932118,
0.1177796125,
-0.079720594,
-0.1668654382,
-0.3578244746,
0.4702191651,
-0.0713933706,
0.4873971045,
-0.3471037745,
0.0088338405,
0.1524146497,
-0.0133304037,
0.5068030357,
0.0612411126,
0.1467698663,
-0.0510436669,
0.0666102991,
-0.0926595107,
-0.2801856101,
-0.2568100989,
-0.0583577156,
0.2295811176,
0.1282974184,
-0.4007204175,
-0.3233204782,
0.0673332363,
-0.1687119305,
-0.3320307434,
-0.4579871595,
-0.4411282837,
-0.2853358984,
-0.3242385089,
0.2526415586,
0.0713431761,
-0.1816559881,
-0.0214646235,
0.0082045868,
0.147040084,
0.2471082211,
0.0978485644,
0.0833588019,
0.1141774654,
-0.1949959695,
0.131614089,
0.4105846584,
0.1862627864,
0.0891106576,
0.2787952423,
-0.2423748523,
0.0301725194,
0.1517166346,
-0.1628012508,
0.085158959,
0.3966115415,
0.0636258572,
-0.0428082459,
0.0545016602,
0.3373205066,
-0.3178878427,
0.0252184719,
0.3418808281,
0.0325719528,
-0.1114490181,
-0.0825198591,
0.1174813807,
0.1140533611,
-0.2343798578,
0.3709225059,
0.0198114179,
-0.1443083286,
0.304214716,
0.1867208481,
0.9030945897,
0.0311159231,
0.2280942649,
0.2173770368,
0.0732169673,
0.1633869708,
-0.0176223665,
0.0419674516,
-0.396194607,
-0.3439331353,
-0.1805068254,
-0.134769693,
0.2987571359,
-0.2991400063,
-0.3444775045,
0.3026093543,
-0.0014979094,
-0.2954223454,
-0.4523963332,
0.4312303364,
-0.3331824541,
-0.1938415766,
0.2419091314,
0.1094018966,
-0.0754102767,
0.2564892769,
-0.0621123537,
-0.243682608,
0.0105549768,
-0.3213074803,
-0.389061451,
-0.2424219549,
0.0219552331,
0.2903731167,
-0.0039328304,
-0.158026427,
-0.10698241,
-0.0413960703,
0.0664161071,
0.2461608052,
-0.132876426,
0.2183260173,
-0.2534838021,
0.0385649279,
0.0761466473,
-0.3603297174,
0.3276643753,
-0.1171879247,
-0.5082360506,
-0.1755793989,
0.1454010308,
-0.2381664813,
0.0332745463,
0.1399684548,
0.3557229638,
-0.0296966005,
0.1975513697,
-0.1510795355,
-0.2765430212,
-0.1418729424,
0.1220529303,
0.2428268492,
-0.2791323066,
0.3559330404,
-0.1460068524,
-0.3269099295,
0.1301742494,
0.103587389,
-0.0482688583,
-0.0570160076,
0.2107544392,
-0.2135887593,
-0.0768896937,
-0.2594357729,
0.0709309876,
-0.0081099086,
0.0050344691,
0.0122504234,
0.130884856,
-0.2050807178,
0.0797919482,
0.1393649876,
0.3301214874,
0.117008239,
0.0605256259,
-0.4716094732,
-0.2771269381,
0.0881798267,
-0.4826113582,
0.4530036747,
0.0873674527,
0.0094038136,
0.0200226456,
0.2794269025,
-0.3884775937,
0.0165666565,
-0.3601018786,
-0.1482971013,
0.3039810359,
-0.0586705953,
0.0161727369,
-0.0914910883,
0.0403942242,
0.1112427711,
-0.5826278925,
-0.1632361561,
-0.3143565655,
0.1723042578,
-0.1240758598,
-0.0353658684,
0.0486164205,
-0.4985814095,
-0.0422139727,
-0.0605109781,
0.1797477007,
0.4235542119,
-0.1383515745,
0.0699021369,
0.0984756798,
0.2840331793,
0.1419793069,
0.2620129585,
0.1505043209,
-0.1604101658,
-0.2000433356,
0.1210674569,
-0.2418046743,
-0.097705029,
-0.1702809334,
0.1415649951,
0.4924519658,
0.2272937596,
-0.1555067897,
0.2823543549,
-0.1449465752,
-0.0772123486,
0.414052248,
0.3748760819,
-0.2612325251,
-0.13078399,
0.0538160913,
0.2483971417,
-0.25286448,
-0.0649077669,
0.1949884295,
0.3159150481,
0.1384044886,
0.3598709404,
0.3957141042,
-0.0030835718,
-0.1721450835,
0.0037229434,
0.3735212684,
-0.1775872111,
0.2339084744,
-0.2443972379,
-0.1510741711,
0.1608308256,
0.3821860552,
0.0139731485,
0.3264164925,
0.0110325273,
-0.2467038184,
0.3727155924,
0.360727042,
0.0910368934,
0.0966886729,
-0.0279840901,
0.0726910233,
-0.147321403,
-0.0689114332,
0.0848787352,
0.238933444,
0.6212899089,
-0.5725151896,
-0.1553124785,
-0.2476174831,
0.3540138304,
-0.184091717,
-0.0106130503,
0.1967913806,
-0.1030417979,
-0.1526865214,
-0.1799849868,
0.2436831445,
-0.0352679975,
0.2975537777,
0.2469140738,
0.1061947644,
-0.2136849016,
-0.1666729748,
0.1843448281,
0.1577886194,
-0.4586445391,
-0.1551934332,
0.314216584,
-0.144261077,
0.4445005953,
0.3467243016,
0.2973740995,
0.129078716,
-0.2095438093,
-0.028506875,
0.0848276317,
0.1700477004,
-0.1163836718,
0.1652919203,
-0.0131248347,
-0.3018895686,
0.2105092704,
0.144544065,
-0.1336323619,
0.0790036321,
0.10948392,
-0.0485222489,
-0.0589643233,
0.5235517621,
0.0981118679,
0.0814315528,
0.0804445297,
0.1009468287,
-0.6264944673,
0.1366875917,
0.0718841255,
0.0941769481,
0.2018310875,
-0.3069220185,
0.1171847433,
-0.0118222665,
0.3767246902,
0.1390981227,
0.0868160874,
-0.4619700611,
0.1721450984,
-0.7986825109,
0.5472496748,
0.053406395,
-0.4315549135,
-0.0160555895,
0.0442313775,
0.0583800636,
0.1264324635,
0.3685826659,
0.3651025593,
-0.037274316,
0.3127579689,
-0.5110433102,
0.2197711617,
0.1067203283,
-0.165602684,
0.2129305303,
-0.2172721624,
0.1808339059,
-0.0124268755,
0.0676133782,
0.070223242,
0.0709807277,
0.2476883084,
0.0499051251,
0.2937147021,
0.0575812533,
0.5136569142,
-0.0202783458,
-0.0191466324,
-0.1074812636,
-0.0272797253,
-0.2270097435,
-0.3902620077,
0.1994392127,
0.0170684382,
-0.208676219,
-0.1834440231,
-0.4254196286,
0.1746607572,
-0.2887329757,
-0.0300676227,
-0.300660193,
0.2215654254,
0.0388802364,
0.1303007007,
0.0145916119,
0.327349633,
0.2044259459,
0.1432476044,
-0.1279268116,
-0.0846336782,
0.5487345457,
-0.2959804237,
-0.1867412776,
-0.2512252927,
0.4572094083,
0.1084875613,
0.1357613206,
-0.4568594098,
0.0080925971,
0.0233044289,
-0.089466013,
-0.3460147381,
0.1537268609,
-0.2121057361,
-0.1914071739,
-0.044337865,
0.0226718578,
0.0970134884,
-0.419960171,
0.2008335292,
-0.1824046969
] |
https://github.com/huggingface/datasets/issues/1956 | [distributed env] potentially unsafe parallel execution | Ah, you're absolutely correct, @lhoestq - it's exactly the equivalent of the shared secret. Thank you! | ```
metric = load_metric('glue', 'mrpc', num_process=num_process, process_id=rank)
```
presumes that there is only one set of parallel processes running - and will intermittently fail if you have multiple sets running as they will surely overwrite each other. Similar to https://github.com/huggingface/datasets/issues/1942 (but for a different reason).
That's why dist environments use some unique to a group identifier so that each group is dealt with separately.
e.g. the env-way of pytorch dist syncing is done with a unique per set `MASTER_ADDRESS+MASTER_PORT`
So ideally this interface should ask for a shared secret to do the right thing.
I'm not reporting an immediate need, but am only flagging that this will hit someone down the road.
This problem can be remedied by adding a new optional `shared_secret` option, which can then be used to differentiate different groups of processes. and this secret should be part of the file lock name and the experiment.
Thank you | 16 | [distributed env] potentially unsafe parallel execution
```
metric = load_metric('glue', 'mrpc', num_process=num_process, process_id=rank)
```
presumes that there is only one set of parallel processes running - and will intermittently fail if you have multiple sets running as they will surely overwrite each other. Similar to https://github.com/huggingface/datasets/issues/1942 (but for a different reason).
That's why dist environments use some unique to a group identifier so that each group is dealt with separately.
e.g. the env-way of pytorch dist syncing is done with a unique per set `MASTER_ADDRESS+MASTER_PORT`
So ideally this interface should ask for a shared secret to do the right thing.
I'm not reporting an immediate need, but am only flagging that this will hit someone down the road.
This problem can be remedied by adding a new optional `shared_secret` option, which can then be used to differentiate different groups of processes. and this secret should be part of the file lock name and the experiment.
Thank you
Ah, you're absolutely correct, @lhoestq - it's exactly the equivalent of the shared secret. Thank you! | [
-0.2438626885,
-0.5634171963,
-0.0252402518,
-0.078923583,
-0.0883313715,
-0.046587579,
0.4310413897,
-0.094735384,
0.6884126067,
0.3382450938,
-0.032446634,
0.2416057438,
0.0294990093,
0.0651078671,
-0.0662357882,
0.0515498817,
-0.013719216,
-0.0775891691,
-0.2323584259,
-0.1388343573,
-0.2581668794,
0.1642914563,
-0.1347394735,
0.1182623506,
-0.131055057,
-0.0675268695,
-0.2938540876,
0.351873219,
-0.2069535255,
-0.3734537065,
-0.0511100926,
0.5885390639,
-0.1115159616,
0.3470315039,
-0.0001025435,
-0.0290896222,
0.2390750945,
-0.1840211451,
0.093775481,
-0.0225166678,
-0.0063964874,
-0.214779824,
0.2390901148,
-0.6090909243,
0.2548007667,
-0.3178235292,
0.346506387,
-0.5903226137,
0.4716383219,
-0.1917704195,
0.2114200443,
0.4353040457,
-0.2383695543,
-0.2103888243,
0.0767695159,
-0.1481730938,
0.0649376363,
0.3659193814,
0.4074766338,
-0.1650005579,
-0.3139853179,
-0.0395432562,
0.043598406,
0.3771087825,
0.5423665047,
0.2160796225,
-0.0078244079,
-0.3159536421,
-0.3685431778,
0.5449936986,
0.0689926744,
-0.0175683666,
-0.155213207,
-0.0432235375,
-0.1245838553,
-0.0624333471,
0.0405219458,
0.0801770762,
-0.2795662582,
0.0957417041,
-0.0633940846,
0.2075609267,
-0.1429090202,
-0.045265425,
0.0962148905,
0.4674391747,
0.1451298147,
0.3589617908,
0.3896106482,
0.2626072764,
-0.5951404572,
0.2887126803,
-0.0776977614,
-0.1011505574,
-0.1418188065,
-0.0618655719,
-0.0439788774,
-0.1890522987,
0.2088949829,
0.1767124087,
-0.2218022197,
-0.0964418203,
0.4521385431,
0.2948688269,
0.1431005299,
0.1408369243,
-0.0696249455,
0.268387109,
0.3005172908,
0.0206617527,
0.0306793749,
0.1363247782,
0.2108893096,
0.0027401336,
0.0445301533,
-0.0351567492,
0.0238798037,
0.0576750785,
-0.4421530664,
0.1751862466,
0.0797957182,
-0.1001361459,
0.0160643011,
0.1829921901,
0.3118673563,
-0.2796682119,
0.1751924455,
-0.1004704908,
-0.274445951,
0.2910326719,
-0.1642244756,
0.0376682021,
-0.3985894024,
-0.0958595574,
-0.0917955637,
0.0960876122,
0.3988890052,
0.0868711621,
0.4171566367,
0.1293654889,
0.2241772711,
0.1347909123,
0.2759169042,
0.178447783,
-0.0017333925,
-0.2029621154,
0.2175433338,
-0.0810276568,
-0.2209540457,
-0.2160371542,
-0.2931859791,
-0.0774590969,
-0.1473875195,
0.2421828508,
-0.3560351133,
0.2010424435,
0.380035907,
-0.1475384831,
-0.1068172604,
-0.0983295962,
0.5739684701,
-0.1218079701,
0.0830889121,
-0.0840711147,
0.0201893896,
0.4858987033,
0.0717011988,
-0.2658755183,
0.2152181119,
-0.2768615186,
-0.0522134379,
0.2739186287,
-0.0850303471,
-0.2639817894,
-0.245792836,
0.160736382,
0.1307990104,
-0.4818877578,
-0.405736208,
0.2260940522,
-0.4426644146,
-0.1356584579,
0.5243065357,
0.3977624774,
0.3573782146,
-0.1086715013,
0.1821477115,
0.0534404628,
0.01588233,
-0.0299051292,
-0.1290168166,
-0.1947336793,
0.0986650959,
0.0532402918,
-0.1217257157,
-0.0514579937,
0.2830916047,
0.0329830162,
0.3958415985,
0.0993597358,
0.0005165413,
-0.0211904515,
0.0979679525,
0.1156113446,
0.2123179734,
0.1543364227,
-0.0907533243,
0.2483418882,
-0.3218516707,
0.1370418966,
0.1254310608,
-0.2333320379,
0.0656791031,
-0.1063251793,
-0.265973717,
-0.3408949673,
0.1597789079,
0.1032041162,
-0.418020457,
-0.2754877806,
-0.2062260807,
0.4432170391,
-0.0138524398,
-0.1089848578,
-0.2169676274,
-0.1268144846,
-0.3346158266,
0.0727747828,
-0.1406826675,
0.0365248024,
0.0747133568,
-0.1410054415,
-0.1647875309,
0.3544576168,
0.319593668,
0.1061794758,
0.302287817,
0.1927320063,
0.0328775644,
-0.1241670325,
0.1496137828,
0.0756307542,
0.1013346761,
0.0801177248,
-0.0873714313,
0.2563194633,
-0.1455116868,
-0.109331727,
-0.1375806928,
0.1314449012,
-0.0765735209,
-0.136586532,
-0.2957277,
-0.1035108417,
0.610616684,
0.3171095848,
0.2319965959,
0.0240163561,
-0.0837851018,
-0.0262041651,
0.1193376482,
0.1129374653,
0.037723124,
-0.1472761184,
0.2499713153,
-0.0389148518,
-0.0289271381,
0.3769490123,
0.5130673051,
0.2503075004,
0.0716995001,
0.3051805794,
0.0016123876,
-0.2875824273,
0.035873197,
0.0561418533,
-0.1126171947,
0.0684217364,
-0.0435018949,
0.0493566021,
-0.1518484801,
-0.3390763998,
0.1330270618,
-0.0304417219,
-0.1809449196,
0.2108366787,
-0.0941147581,
0.2258037776,
-0.049672503,
-0.1246382222,
-0.2088016123,
-0.4025353193,
0.2573354542,
0.0917731375,
-0.4607611001,
0.095169574,
-0.0385734215,
0.2664900422,
-0.1746658832,
0.0735592768,
-0.1235506833,
-0.3214639723,
-0.0437072739,
0.0795471892,
-0.0880629793,
0.2526986897,
0.7304583788,
-0.1311332434,
-0.2009437382,
-0.3724404573,
-0.16435422,
0.0462006442,
-0.3329817951,
0.4429242611,
0.4895550609,
0.0533525348,
-0.1120606363,
-0.1844490618,
0.084066391,
-0.3830344677,
-0.2660087347,
-0.1845015585,
0.1068548486,
-0.3374785781,
-0.3387934566,
-0.3685581982,
-0.4663521349,
-0.3014016151,
0.2419754267,
-0.1762252748,
0.1574787498,
0.0984286815,
-0.4044068456,
0.0433088169,
0.2788300216,
-0.1425389051,
-0.4990938008,
-0.5869212151,
0.0771414489,
-0.2953597903,
-0.1439427435,
-0.0453419983,
0.1576923728,
-0.2012325227,
0.1635485888,
-0.2414558083,
-0.7453755736,
0.0937655941,
0.235443905,
-0.0996864438,
0.0434936211,
0.3776494563,
-0.0950520337,
-0.2161880434,
-0.2172877789,
0.061105445,
0.4402586222,
-0.0263331663,
0.1234300509,
-0.0219241045,
0.0885736346,
-0.0025028288,
0.5244193077,
0.2314628661,
0.041503571,
0.3437654078,
-0.1912547499,
0.4083037972,
-0.073866114,
-0.2043544352,
0.2984808981,
0.1865424812,
-0.3711796999,
0.0471344516,
-0.3468581438,
0.1800032854,
0.0091125593,
-0.0582054369,
0.2116907984,
-0.285746932,
-0.1605327278,
0.0355201401,
0.0172331259,
-0.1656158864,
0.0695132762,
0.1110110134,
-0.0095531195,
0.0143052526,
0.5648733377,
-0.0347728394,
0.1002478302,
-0.1071099117,
-0.2136799395,
-0.3458735347,
0.4169011712,
-0.023503188,
0.4492664337,
-0.3281522393,
-0.1347950995,
0.2345646024,
0.0092921108,
0.4790711403,
0.1126457676,
0.1230678707,
-0.0280253217,
0.100786835,
0.0051924735,
-0.3153503239,
-0.3267295361,
-0.0421087816,
0.1992311776,
0.0815792754,
-0.3377790153,
-0.261932224,
-0.0124278758,
-0.2160035074,
-0.3102510571,
-0.4012082815,
-0.5420235991,
-0.2751144767,
-0.360060364,
0.2392568737,
0.1309222877,
-0.2203641832,
-0.0489163324,
0.0770780742,
0.1968030483,
0.303155303,
0.0091499537,
0.1745313406,
0.0539991781,
-0.163809076,
0.0084143057,
0.3750195801,
0.2809545994,
-0.0584354401,
0.2942311764,
-0.1721452326,
0.1452236921,
0.2947832644,
-0.1858272254,
0.0139186084,
0.2652874887,
0.0881236494,
-0.0277722552,
0.0551488511,
0.3172251284,
-0.2777992189,
-0.0882018209,
0.3498714864,
0.0299355499,
-0.079906486,
-0.0434261262,
0.0550971478,
0.0568045005,
-0.256783694,
0.2959793508,
0.0565923899,
-0.0883458555,
0.2275805175,
0.1434838921,
0.8623200655,
0.0624682978,
0.4144766927,
0.1813770682,
0.0522226021,
0.2632082403,
0.0515544415,
0.0629201978,
-0.4102434516,
-0.3862263858,
-0.2433964908,
-0.1412846595,
0.2726682425,
-0.302634269,
-0.4003546834,
0.3024196327,
0.0394350439,
-0.207344085,
-0.4099217653,
0.3298552632,
-0.2978416681,
-0.0898287594,
0.2729329169,
0.1163660437,
0.0611588247,
0.392424345,
-0.0717784315,
-0.1201844811,
-0.0320482627,
-0.3505809307,
-0.3782682717,
-0.2039822638,
0.0113131292,
0.3753529489,
0.035637185,
-0.0861977488,
-0.197911337,
-0.0643183887,
-0.0133303739,
0.2368096709,
-0.1093848571,
0.222795397,
-0.0976636633,
0.0239580758,
0.1493283957,
-0.3480621576,
0.355784893,
-0.0891307592,
-0.480922997,
-0.0553524718,
0.0374373458,
-0.2613353729,
-0.0033930242,
0.2813750803,
0.3861353397,
0.0169831887,
0.0992798209,
-0.137686789,
-0.3311129808,
-0.0852764174,
0.1279796362,
0.1529993564,
-0.2878355682,
0.2079715729,
-0.1581048667,
-0.3879678249,
0.1795874834,
0.1025427431,
-0.0463124663,
-0.031726867,
0.2740698159,
-0.2099537402,
-0.0563411936,
-0.2196568251,
0.078994751,
-0.0370821469,
0.0329973847,
0.0156206153,
0.1447080523,
-0.2545281053,
0.0552962385,
0.1285775155,
0.3506360054,
0.2885817587,
-0.0284397379,
-0.4474081099,
-0.3208756447,
0.0160211772,
-0.4509328604,
0.4844257236,
0.2211308926,
-0.0322230794,
0.1585772783,
0.1567704976,
-0.3978053927,
0.0319295228,
-0.4641880095,
-0.1248670071,
0.3408112824,
-0.0752540678,
0.0603978708,
-0.0641894639,
0.090621531,
0.13479954,
-0.5935945511,
-0.1863763034,
-0.3097167909,
0.1394762844,
-0.1753056347,
-0.0386484005,
0.0729880482,
-0.4741797447,
-0.0042618383,
0.0098958388,
0.2145985216,
0.4166872501,
-0.1314973235,
0.0742179081,
0.1948981881,
0.2095609307,
-0.0010700673,
0.227753669,
0.1270659864,
-0.1274963915,
-0.1741636395,
0.0659342557,
-0.3609364927,
-0.1272587627,
-0.1918004304,
0.2360035181,
0.4632053375,
0.2804305553,
-0.1466751099,
0.3265671432,
-0.1062845141,
-0.2187989652,
0.4266742468,
0.329438597,
-0.2642707825,
-0.2751121521,
0.0969030112,
0.2293528765,
-0.2391541451,
-0.0873810798,
0.2002389729,
0.2963558137,
0.171322003,
0.2899864316,
0.3527078629,
0.066103518,
-0.211689204,
-0.0075862221,
0.3411163092,
-0.1817128956,
0.1577338874,
-0.1876578927,
-0.1835861206,
0.2591576874,
0.3726929724,
-0.115864411,
0.3025061488,
-0.001233584,
-0.3071973026,
0.385319978,
0.4048243165,
0.0581485368,
0.176349327,
0.0141606927,
0.1094285846,
-0.1097696573,
-0.0881031975,
0.1642715931,
0.1753253788,
0.5363360047,
-0.5529486537,
-0.015297249,
-0.233634755,
0.3381833434,
-0.1672063768,
-0.1116960496,
0.2486128658,
-0.0818463862,
-0.1377280653,
-0.1703469455,
0.2336868048,
0.022230655,
0.3978546858,
0.2234878391,
0.150752455,
-0.1550403833,
-0.152281478,
0.0944678336,
0.1013661176,
-0.4949467778,
-0.1825045496,
0.3110488653,
-0.1196640581,
0.4652065635,
0.3376262784,
0.3382672369,
0.1054527536,
-0.1465232372,
0.0138811246,
0.0766863748,
0.1530277282,
-0.1337038428,
0.1422641128,
-0.004886061,
-0.2454774827,
0.1626683027,
0.1714868248,
-0.1359951198,
0.0845953226,
0.1394142359,
-0.0188517831,
-0.0488451794,
0.4834090471,
0.0646334887,
0.0362899303,
0.0416611359,
0.3051946461,
-0.6858343482,
0.1099763885,
0.094408676,
0.1007443368,
0.1709933281,
-0.2672558427,
0.125939399,
-0.0470304415,
0.490487963,
0.2188588679,
0.1342143416,
-0.3599074185,
0.2000602782,
-0.6501299143,
0.5871244073,
0.0781985447,
-0.5412564874,
-0.0465626754,
0.0264714509,
0.0851542428,
0.0336237885,
0.3250839412,
0.3061599135,
-0.1271301508,
0.3344298303,
-0.3145222664,
0.1504699737,
0.044455871,
-0.1469232142,
0.1582151651,
-0.2173132598,
0.1303637922,
0.0323744379,
0.0812970996,
0.0324244238,
0.0689332634,
0.2904609442,
-0.0080101537,
0.2579010129,
0.0819114,
0.4832309484,
-0.0748589411,
-0.0566987619,
-0.0253481157,
-0.0559693947,
-0.1712685972,
-0.3650261164,
0.2243225276,
-0.0869929045,
-0.1901639998,
-0.1799834073,
-0.3332695365,
0.2587375343,
-0.2860460281,
-0.0420760848,
-0.3193358779,
0.1278974712,
0.0068239197,
0.1146001518,
0.0743902996,
0.3767511845,
0.2553516328,
0.0016884655,
-0.0977996513,
-0.090539135,
0.467768997,
-0.3282396197,
-0.1754851341,
-0.3249819875,
0.3709000945,
0.037672013,
0.1435145736,
-0.4403662682,
0.0035194159,
0.0317308977,
-0.0757507682,
-0.3692752719,
0.1333638579,
-0.1457456499,
-0.1425707042,
-0.115592286,
0.1194954067,
0.1037469506,
-0.4305737913,
0.1731839776,
-0.1977107823
] |
https://github.com/huggingface/datasets/issues/1954 | add a new column | Hi
not sure how change the lable after creation, but this is an issue not dataset request. thanks | Hi
I'd need to add a new column to the dataset, I was wondering how this can be done? thanks
@lhoestq | 18 | add a new column
Hi
I'd need to add a new column to the dataset, I was wondering how this can be done? thanks
@lhoestq
Hi
not sure how change the lable after creation, but this is an issue not dataset request. thanks | [
-0.2342519164,
-0.0549111515,
-0.1812908202,
-0.052450832,
0.0165327005,
0.0354573131,
0.3398598731,
-0.0366016068,
0.0798501223,
0.144506976,
0.0107777715,
0.1978343427,
0.0306360424,
0.3719207644,
0.1402907819,
0.1070635319,
-0.2483593971,
0.4518116713,
-0.1740532219,
-0.0729186237,
-0.3456431627,
-0.0778269246,
0.1791185439,
-0.2770398259,
-0.3205129802,
-0.220287025,
0.0320803076,
0.0979111567,
-0.3490825295,
-0.407387048,
-0.0071716607,
0.319930315,
-0.0588899367,
0.1259934008,
-0.0001025004,
-0.1651873142,
-0.0742467344,
-0.1672581434,
-0.2324925959,
0.1484257132,
-0.2225051969,
-0.1204708964,
-0.0649787113,
-0.2121858001,
-0.4337715507,
0.0666611791,
-0.090705432,
-0.4884948134,
-0.002818197,
0.3639041185,
0.3648744524,
0.1379204094,
-0.1298271418,
-0.351028055,
0.2751465738,
0.0080140103,
-0.1131886914,
0.1160289645,
0.1404921561,
-0.0684449524,
-0.1430791318,
0.0452410541,
0.5009627938,
0.1005611494,
0.0736441165,
0.007634785,
-0.0122203566,
-0.0688312054,
0.2871031165,
0.2532126904,
0.5526140332,
-0.1473832726,
-0.2127428502,
0.0277346671,
0.2715099454,
-0.3950330615,
-0.1023691669,
0.0229896288,
0.1035381854,
0.0007531173,
0.3265816867,
-0.3593198657,
-0.3072423637,
0.1033373475,
-0.1509903669,
0.3630670011,
-0.0937534422,
-0.0232562162,
-0.1438334584,
-0.045735728,
0.0405792035,
0.5270215869,
-0.04868754,
0.0868930966,
-0.1699164212,
-0.1534495801,
0.1258374453,
0.0771982819,
-0.0013117269,
0.0071970392,
0.013060648,
-0.1483257711,
-0.1018103063,
-0.0491406731,
-0.098541379,
-0.2586114705,
0.2241180688,
0.0814240426,
0.0712921619,
-0.1250396967,
-0.1409029365,
-0.3387120962,
0.090477258,
-0.059112981,
0.0375071988,
-0.0255633201,
0.1908127218,
-0.1325503588,
-0.1077481136,
-0.0947954804,
0.1682733744,
-0.0805520788,
0.0298455395,
0.4504489005,
0.1969479769,
-0.0125309909,
-0.0428344496,
0.1681354642,
0.0000442639,
-0.36502105,
-0.2425491214,
0.0406484716,
0.0234599933,
0.0446028113,
0.0642386004,
0.1984072626,
0.359516263,
0.0659382492,
-0.0261974856,
0.1453186274,
0.1730487645,
-0.0396583192,
0.2590417862,
0.163030386,
-0.0337747298,
0.0389917269,
-0.064385213,
-0.0070229471,
0.0199396983,
0.0474147275,
0.0771664903,
-0.0852074027,
-0.6027118564,
0.296312958,
0.3269844949,
-0.2982278764,
0.1507371813,
0.187655434,
-0.1407330632,
-0.2708616555,
0.0805226788,
0.1842940152,
-0.1445275992,
-0.1888030171,
-0.0908173919,
-0.0513372645,
-0.3442673385,
-0.0079603195,
-0.2085406482,
-0.1071881354,
0.1412756145,
-0.1655448079,
-0.0701745152,
0.1713635325,
-0.0367522314,
0.087622188,
0.5381996632,
-0.0310597755,
-0.3604516387,
-0.2498096824,
-0.1576159,
-0.3682190776,
0.2996691465,
0.5366458893,
-0.0117448084,
0.0206747018,
-0.204498902,
0.0247422792,
-0.2616028786,
-0.0983521566,
-0.1163444668,
-0.0286874212,
0.1365666986,
0.0614789873,
0.124801755,
0.1944217086,
0.3542416096,
0.1688766181,
0.1422252059,
-0.2761055231,
-0.0772150531,
0.114511326,
0.5605570674,
0.0042300913,
-0.0603506193,
-0.0865193233,
-0.3809378445,
-0.131901443,
-0.2419222891,
0.0342394263,
0.2410074919,
-0.3780499697,
-0.3833189607,
-0.1641273499,
-0.011692863,
-0.047061421,
0.2636747956,
0.1956167519,
-0.0568315089,
-0.2675727308,
-0.0989278927,
0.1625025123,
-0.0958890095,
-0.1135599166,
-0.2682311833,
0.0426566154,
-0.0947998688,
-0.1658969074,
0.2144952863,
0.1803324819,
0.2492696345,
0.1912119985,
-0.0645205677,
0.1831040084,
-0.0789256841,
0.2383157164,
0.2969736159,
0.1007826477,
0.0447557643,
0.1680232733,
0.0345777869,
-0.2406421155,
0.0310393125,
0.2692002058,
-0.2842738628,
0.1433782578,
0.0153189451,
-0.1771542579,
0.0303519592,
-0.1171382889,
0.2075390816,
-0.3155143857,
0.1554965675,
-0.5195200443,
0.0813988,
0.0396898463,
-0.0432150066,
-0.0339905545,
-0.3423888385,
0.2929728925,
0.1848214865,
0.1485245824,
0.2735671103,
-0.0245732628,
-0.1060428917,
-0.0345940888,
-0.0283581205,
0.375040561,
0.1763316095,
0.3945909441,
0.0966873765,
-0.0656397939,
-0.1075385213,
-0.2565464973,
-0.0951364636,
-0.3713630736,
0.0835186318,
0.2370496243,
0.3434163332,
-0.3087706864,
-0.4307763577,
0.0973554552,
0.0629801378,
0.3789814115,
-0.2094774544,
-0.218433097,
-0.095888637,
0.0126110092,
0.0270038694,
0.0428646244,
0.1422493756,
-0.2052132785,
0.2989899814,
0.2755518258,
-0.0869217664,
-0.0169358924,
-0.2405189574,
0.3863027692,
0.1515254676,
0.2785093486,
-0.0031321198,
-0.3506254554,
0.1799197644,
0.289672941,
-0.2606050074,
0.1840183437,
0.8060863614,
-0.1483342648,
0.2207279354,
-0.4004446268,
-0.3926117718,
-0.0963894278,
0.2622621357,
0.396368593,
0.1159330755,
0.2579714656,
-0.0414652899,
-0.1163628697,
0.0850420296,
-0.1183883846,
-0.1783896685,
-0.2418839335,
-0.2830155492,
0.3030080199,
0.0470835492,
-0.6182642579,
0.1425397247,
-0.2114255428,
0.0926957875,
0.0451596528,
0.0243388377,
-0.2476559728,
-0.1087787002,
-0.0807372853,
-0.1766067594,
0.0192313679,
-0.5788751245,
-0.3256312013,
0.2002627403,
-0.4118052423,
-0.2977394462,
-0.1887418032,
-0.1371979713,
0.3828651607,
-0.1759367138,
-0.0962097943,
-0.2526670098,
-0.1949558854,
0.2247779369,
-0.1720817685,
0.0294296872,
0.5308919549,
-0.1651631594,
-0.2386846542,
-0.2381103933,
-0.1847108006,
0.1396242082,
0.2987615764,
0.1010990217,
-0.2928365171,
0.3094173968,
-0.2106477469,
0.1787433922,
-0.2057984322,
-0.2192952931,
0.4171489775,
-0.4435090721,
0.4400139153,
0.0062043071,
-0.2436642051,
0.0464618057,
0.3540068269,
0.2039540708,
0.1661487222,
0.0458549671,
0.0117863249,
0.100876011,
-0.021409519,
-0.3324847519,
-0.1376768798,
0.0581159219,
-0.1252358258,
0.288102746,
0.0094225407,
-0.2280371189,
-0.4572377801,
-0.1159291118,
-0.1513101906,
-0.0643424392,
0.0590968393,
-0.3631084859,
-0.4082353115,
-0.0058832169,
-0.3175398409,
0.1657297015,
0.0785016567,
0.2407677621,
-0.1516014934,
0.0132911429,
0.0675241351,
-0.0793043673,
0.4309128523,
-0.4250863194,
-0.0646547526,
0.129569754,
0.1728387624,
-0.2631373405,
0.3178608716,
-0.2250891775,
0.2126707584,
0.0132097434,
0.3453509808,
-0.0616824999,
0.0251384862,
0.4632743001,
0.2184382975,
-0.3159710765,
-0.2171535194,
-0.184207499,
-0.2916347384,
-0.2276747823,
0.0814896822,
-0.0870786682,
0.1912451535,
0.30892542,
-0.1160626113,
-0.140740037,
-0.2207109928,
0.1488070786,
0.2851305902,
0.0713689104,
0.388225466,
0.1160018742,
-0.1079921424,
0.0300683845,
0.1671794504,
0.2251333147,
-0.2639664114,
0.1820217967,
0.0570058972,
-0.3748995066,
0.4564006329,
0.0895123631,
0.1582808644,
0.4401007891,
-0.1284203976,
0.0947440863,
-0.1609490514,
0.2963099182,
0.3025246263,
-0.1867974401,
-0.0204953104,
-0.2033822536,
0.3991321921,
-0.0663110465,
-0.203661859,
0.2702281773,
0.2597520351,
-0.1118681207,
0.1192702204,
0.1386225373,
0.8258501887,
-0.1651408076,
-0.1446481496,
-0.0164527074,
-0.4330641627,
0.2019694149,
-0.121813789,
0.1498009712,
-0.3629226089,
-0.1199737191,
0.0581466332,
0.0594906211,
0.2272732556,
0.0732948631,
-0.0408145525,
-0.0277750175,
-0.2128305435,
-0.1205979735,
-0.0634454861,
0.3888411224,
-0.1706169099,
-0.0385124162,
0.0300865546,
0.2427951694,
0.0967734158,
-0.3202816546,
-0.0654102936,
0.1378673762,
-0.0022843555,
0.0278225392,
-0.0952257961,
-0.339186877,
-0.0943234563,
-0.1221598089,
-0.0982920378,
-0.3562054634,
-0.0801301152,
0.1126286834,
0.4637309611,
0.22258991,
-0.1460873932,
0.3929460049,
-0.0290305726,
-0.0106849633,
-0.0356688499,
0.1002593637,
-0.0205145665,
-0.2906255126,
-0.3704013228,
0.0385581926,
0.1290340871,
-0.4437743127,
-0.0630923659,
-0.0641659349,
0.1369270235,
-0.5037251711,
-0.0970796943,
0.1996073127,
0.176892221,
-0.4096497297,
0.2604003251,
0.2085071206,
-0.0506003238,
0.1573242545,
0.2987824678,
-0.1548346281,
-0.1539896131,
0.153716445,
0.2887461185,
0.3733950853,
0.2711695433,
-0.1032876447,
-0.0903463811,
-0.5066464543,
0.1425128877,
0.3526087403,
-0.449600935,
0.1566506624,
0.2619934082,
-0.1789031178,
0.1585229188,
0.2552878857,
-0.1255610287,
0.2080293149,
-0.0764537454,
0.0193244517,
-0.1772871315,
-0.1674602181,
-0.101203531,
0.3335440159,
-0.0086053163,
0.1660341471,
-0.1989147514,
0.2513757348,
-0.4863964319,
0.0772575811,
-0.2586610317,
0.3740481734,
0.2822450399,
0.0298457965,
0.0526983067,
-0.2013112605,
0.3311570883,
-0.1693148762,
0.0260724239,
-0.3804883957,
-0.1417732835,
0.0272809118,
-0.0407106429,
0.2253507376,
-0.1291323602,
-0.0980428979,
-0.0999182016,
0.1576020718,
0.0955436379,
-0.0132001936,
0.1009495184,
0.271495223,
0.0844439045,
-0.0818280131,
0.0909770727,
0.0607519113,
-0.0926823914,
-0.0806735978,
-0.2555573583,
0.1211445108,
-0.0813703686,
-0.0933391005,
0.2188301235,
0.0702442378,
0.1460734308,
0.0843618959,
-0.0059982594,
0.0182248447,
0.1019706428,
0.3345038891,
-0.0408118144,
-0.1155474037,
0.068617776,
0.1410686076,
0.246254921,
0.4119706452,
-0.527520895,
0.1997169852,
0.1063260734,
-0.207549423,
0.4894815981,
0.2371239215,
0.0950020477,
0.1738277078,
0.0527153276,
0.0670531988,
-0.1209325939,
-0.1525199413,
-0.0445639938,
0.2598796785,
0.0038235933,
-0.1028247327,
0.4838430882,
0.1744036674,
0.040119648,
0.2454651445,
-0.0153557286,
0.3443308473,
0.1854995787,
-0.0901880413,
0.1924192309,
0.0522948131,
0.3868965805,
0.1154328883,
0.1420224309,
0.0033467598,
0.2396482527,
0.3484845757,
0.0982985198,
-0.1710908711,
-0.0935467035,
0.1353799999,
-0.1843031943,
-0.2212317884,
-0.3129015863,
-0.1616815776,
-0.020254232,
-0.1022179276,
-0.0152747538,
0.0196007937,
0.0218563471,
-0.1224046275,
0.0537976585,
-0.1761558354,
0.2304535359,
-0.022475129,
0.0916208625,
0.0089871213,
0.1721441895,
0.1660164148,
0.0933439806,
0.3902436495,
0.5117958784,
0.0263725221,
-0.1082324162,
0.1417185366,
0.3036314845,
-0.0487565622,
-0.3072279692,
-0.0955747962,
-0.1135287881,
0.2628571391,
0.2950420082,
0.0611840598,
0.3272846341,
-0.3016422689,
0.2575120926,
0.0694975257,
0.0337027833,
-0.0859599262,
0.0288314614,
0.1409157217,
-0.2790927887,
-0.23509413,
-0.0579072908,
-0.1712063551,
0.008579988,
-0.0225511603,
-0.3474243879,
0.0604505055,
-0.1142212003,
0.1349506527,
0.2431202233,
0.317786932,
-0.036428839,
-0.075440824,
0.0554303266,
-0.3573135138,
-0.4765751958,
0.1471115649,
-0.1192049906,
-0.3870981634,
-0.2098789662,
0.1780356467,
0.1869481206,
0.0780629814,
0.5429638028,
-0.0980799571,
-0.079296872,
0.0778163224,
-0.2451913953,
-0.057221815,
0.2500935793,
0.0704836398,
-0.0130245276,
-0.093049407,
0.1171160042,
0.0759093761,
0.2199333459,
-0.0563219376,
-0.0838341564,
0.0421147794,
-0.0226465836,
0.3171391189,
0.0366461128,
0.3963229954,
-0.3592708707,
0.0693522692,
-0.1080048829,
-0.1451851577,
-0.2400188446,
-0.0097663067,
0.4331318438,
0.3607578278,
-0.0265032444,
-0.1312754005,
0.07545764,
0.244697243,
-0.0946353972,
0.1830010116,
-0.0927543938,
0.540756166,
0.2331821024,
-0.1764007807,
-0.0008145235,
-0.0152570624,
0.0184418932,
0.0915623531,
-0.0991520882,
-0.4116052985,
0.3260042071,
-0.0473124795,
-0.1467531621,
-0.1849723905,
0.2483226955,
0.3317495286,
-0.028645996,
-0.3332258761,
0.2292402387,
0.2574436069,
0.0949164629,
0.1930336058,
0.4474112689,
0.0441587456,
0.1114349663,
-0.0722741634,
0.2434657663,
0.1745957732,
-0.1794504076,
-0.164686799,
-0.2129770666
] |
https://github.com/huggingface/datasets/issues/1954 | add a new column | Hi ! Currently you have to use `map` . You can see an example of how to do it in this comment: https://github.com/huggingface/datasets/issues/853#issuecomment-727872188
In the future we'll add support for a more native way of adding a new column ;) | Hi
I'd need to add a new column to the dataset, I was wondering how this can be done? thanks
@lhoestq | 40 | add a new column
Hi
I'd need to add a new column to the dataset, I was wondering how this can be done? thanks
@lhoestq
Hi ! Currently you have to use `map` . You can see an example of how to do it in this comment: https://github.com/huggingface/datasets/issues/853#issuecomment-727872188
In the future we'll add support for a more native way of adding a new column ;) | [
-0.2675052285,
-0.3267517686,
-0.2293752134,
-0.0306970477,
0.071279563,
0.1932585537,
0.2296766639,
0.0986113027,
0.2282396406,
0.1450357437,
-0.1557602733,
0.1634852886,
0.0027357191,
0.5119231939,
0.222769022,
-0.1781962961,
-0.062553063,
0.198813051,
-0.053424105,
0.0027412325,
-0.3495627642,
0.0484528542,
0.2379430532,
-0.1702800691,
-0.2531000972,
-0.2114899457,
-0.0195247419,
0.0866773427,
-0.2067594379,
-0.3567587733,
0.0423740968,
0.2098608166,
-0.074900426,
0.185972929,
-0.0000973422,
-0.0641717017,
-0.0359070338,
-0.1346301436,
-0.10335318,
-0.034768194,
-0.2414605469,
-0.1699302047,
-0.0692434832,
-0.0861013681,
-0.4185827076,
-0.1042872667,
-0.0217463113,
-0.1531261504,
0.1256463379,
0.1134686098,
0.4063778818,
0.2553827763,
-0.0764758959,
-0.3638367355,
0.1776878983,
0.146154508,
-0.2223313749,
0.0608456023,
0.0292492621,
-0.0947451741,
-0.0867012739,
0.2065437436,
0.374355644,
0.1480083317,
0.1906537116,
0.2405008823,
-0.1616062373,
0.0865646377,
0.1893064827,
0.3273337483,
0.4445173442,
-0.1060114205,
-0.3236676455,
-0.2550572753,
0.030813776,
-0.3803779185,
-0.0968920738,
-0.0229367912,
0.0951693952,
0.1734919399,
0.0142115951,
-0.1790037751,
-0.1423600614,
0.1836368591,
-0.2275917679,
0.211358428,
-0.266908437,
-0.0052920356,
0.1706733406,
-0.0678177774,
-0.2897090614,
0.374371767,
-0.0731748194,
0.1141526252,
-0.1223932952,
-0.1932971478,
0.1971587837,
0.1913181245,
0.2126956284,
0.0485497341,
0.0913346559,
-0.0541549213,
-0.1107606217,
0.1634423584,
0.0828429013,
-0.1297964454,
0.2090186179,
-0.0888696909,
0.2225172967,
0.0488690846,
0.0825484097,
-0.3772424459,
0.2139921039,
-0.015441291,
-0.093818076,
-0.1949006915,
0.2368555069,
-0.1049164981,
-0.1222468913,
-0.1332406998,
0.2561585605,
-0.091819495,
0.0656533539,
0.4747042358,
0.0947093666,
-0.1584171057,
-0.0456205495,
0.1981343031,
0.0805895478,
-0.1449044943,
-0.3359636664,
-0.0149344746,
0.0481397808,
0.1808346957,
0.0345311984,
-0.0946614146,
0.4725375772,
-0.0024810843,
-0.0302328877,
0.129701823,
0.0783945173,
-0.0043348409,
0.1934851557,
0.1643795073,
0.0044706464,
-0.0576050095,
0.0403552242,
-0.1187350824,
-0.135262087,
-0.0327267461,
-0.0106871165,
-0.1250885576,
-0.3687202036,
0.3041080236,
0.104108721,
-0.2859867811,
-0.0987826735,
0.4664130807,
-0.0598142073,
-0.2486291379,
0.0581966862,
0.3222052753,
-0.3405730426,
-0.0693768486,
0.0408707857,
-0.0870447755,
-0.2525869012,
-0.2556757331,
-0.034922868,
-0.2431699038,
-0.1837899834,
0.0473113023,
-0.0754371658,
0.0797239766,
-0.0735879391,
0.2277760208,
0.258200109,
0.1087662727,
-0.2271645963,
-0.1417521685,
-0.2780778408,
-0.0971004516,
0.1476488113,
0.3498809636,
0.1436211169,
0.0415609591,
-0.0008692704,
0.0429346189,
-0.1848500371,
0.0126023814,
0.0297187939,
-0.0185102783,
0.0225342661,
0.1027056873,
-0.0757716596,
0.1558426023,
0.3176922202,
0.023015473,
0.1711734533,
-0.1810259074,
-0.0701376647,
0.2003708184,
0.4800210297,
0.0673366636,
-0.121924296,
-0.2546572685,
-0.528920114,
-0.0148890391,
0.0389763936,
-0.0887864456,
0.1192277521,
-0.3768485785,
-0.3831667304,
-0.0615022108,
-0.1084754765,
-0.0569792986,
0.2907109857,
0.1430885792,
0.0679389685,
-0.1913000494,
-0.1594550908,
0.1520783454,
-0.0369494334,
0.0091192797,
-0.2650404871,
0.1284198165,
0.0173632856,
-0.1359552443,
0.0963118374,
0.1147398204,
0.1042969823,
0.0309455581,
0.1269077063,
0.2909140885,
-0.1034535766,
0.3359622955,
0.2832550406,
0.1434998363,
0.1118826345,
0.0508649349,
0.0449769646,
-0.1548964679,
0.017394416,
0.0894267559,
-0.4237242043,
0.2668479979,
0.142869398,
-0.0632410049,
-0.041942291,
-0.0038182884,
0.2277862132,
-0.1741272807,
0.0426245108,
-0.4620883763,
0.0182925127,
-0.0694315061,
0.0898769349,
-0.0262678172,
-0.4064921737,
0.2828009725,
0.337998867,
0.0339881405,
0.0409410261,
0.1030980274,
-0.2593681514,
0.0518030301,
0.1649905145,
0.2050794512,
0.222869277,
0.4660231173,
0.259660095,
-0.0010430021,
-0.1270842999,
-0.1808647811,
-0.0776131824,
-0.2090060711,
-0.0364618935,
0.1165699586,
0.2806752026,
-0.3271844983,
-0.5827217102,
0.0823417306,
-0.081048131,
0.1986764073,
-0.1886475682,
-0.2168994844,
-0.0906996131,
0.0099928007,
-0.0009747185,
-0.2157728821,
-0.1718830615,
-0.2708534598,
0.2800835073,
0.3527165949,
-0.208255589,
0.0128023755,
-0.0188148245,
0.2867795825,
0.0408829153,
0.1994551122,
-0.0749720708,
-0.2418559045,
0.1457320154,
0.3013760149,
-0.227024287,
0.0597375259,
0.6364293098,
-0.1969309896,
0.162260443,
-0.4014016092,
-0.4579340219,
0.0456496775,
-0.0933279768,
0.2280400246,
0.1930569857,
0.4515185058,
-0.1139104664,
-0.178826198,
0.2495380789,
-0.0848383829,
-0.3495542109,
-0.2163495272,
-0.1054065153,
0.1298487782,
-0.0546005666,
-0.3900423646,
0.0308278967,
-0.3032820821,
0.5029953122,
-0.0068597421,
0.0459586754,
0.0246299431,
0.0617440417,
0.0616712868,
-0.3267000914,
-0.0044152085,
-0.5509763956,
-0.3817511797,
0.334666729,
-0.4371897578,
-0.2686751187,
-0.05914478,
-0.0356202796,
0.1367811859,
-0.1566546559,
-0.095357269,
-0.3855730891,
-0.18488729,
0.2939255238,
-0.0176989604,
0.1278984845,
0.484179616,
-0.1433972716,
-0.24567011,
-0.2272548079,
-0.3126323819,
0.152545765,
0.3414379954,
0.1171693951,
-0.2383913994,
0.2780503333,
-0.003025189,
0.2944168746,
-0.0667824596,
-0.1000754684,
0.3630544543,
-0.4525367916,
0.4308474064,
-0.0333026573,
-0.2724323869,
0.1023564488,
0.1703724861,
0.1682207286,
0.3327379823,
0.2434002459,
0.0377823338,
0.1129043847,
-0.0568234846,
-0.1521761715,
-0.1492475569,
0.0324966311,
0.19751513,
0.1152204648,
-0.0478089117,
-0.243474111,
-0.4107985497,
-0.02371783,
0.0472184457,
0.1210914254,
0.0176356472,
-0.1858106405,
-0.4173740745,
0.0242550485,
-0.4365699291,
0.2714169621,
0.0817384794,
-0.1642329246,
-0.2187956274,
-0.1694921404,
-0.0801294148,
0.0094458349,
0.5412450433,
-0.1944127083,
-0.1243829802,
-0.1509479284,
-0.1719125509,
-0.214743793,
0.3027730882,
-0.136138469,
0.0351208448,
0.0809632093,
0.388494432,
-0.1651663631,
-0.0014950074,
0.4199512005,
0.2146368921,
-0.2711300254,
-0.2036133558,
-0.2595191598,
-0.2660928071,
-0.2679540515,
0.1394277513,
0.03624814,
0.0924400389,
0.410651803,
-0.2905563116,
0.0161659624,
-0.1776151508,
0.1379459798,
0.4061157107,
0.1916995794,
0.4186702371,
-0.0198620632,
0.24233374,
0.1439828575,
0.1078941226,
0.3881812394,
-0.2044385821,
-0.1244413704,
-0.0786234587,
-0.2945460975,
0.5106816292,
0.132892862,
0.0991695225,
0.3325242996,
0.0147333667,
0.1614188701,
-0.2695329785,
0.3139100373,
0.1731737405,
0.1405242831,
0.0327327736,
-0.1346593797,
0.4818722606,
-0.0100157335,
-0.1039175168,
0.2979793549,
0.3700315654,
-0.2019834816,
0.0462893397,
-0.0542659424,
0.7508457303,
-0.1673709154,
-0.0357275866,
0.0385229699,
-0.1774647683,
0.222601831,
-0.1097051054,
0.1153544709,
-0.1414432079,
-0.3266515434,
-0.0257296301,
0.1210596785,
0.2492420673,
0.0732877702,
-0.0535690933,
0.0528167151,
0.0574188828,
0.153706938,
-0.1922419965,
0.0704624876,
-0.0091986293,
-0.292491138,
-0.2694923282,
0.3007314503,
-0.010421427,
-0.0957230404,
-0.0762970075,
0.1155964732,
0.0195111856,
-0.2082203925,
-0.1271172017,
-0.3215968013,
-0.0585809276,
-0.1984927952,
0.1279902756,
-0.2477899194,
-0.2515178621,
0.170794338,
0.4153107703,
0.0573320985,
-0.2216246128,
0.205568701,
-0.2225222737,
0.0856783316,
-0.2475015819,
0.0399666838,
0.1094514877,
-0.1779586375,
-0.2346382439,
-0.0101336949,
0.1052415222,
-0.4131994545,
-0.0338256061,
0.0321349055,
-0.2425576746,
-0.4690614939,
-0.1052206233,
0.0798240378,
0.177329734,
-0.3729858696,
0.265660733,
0.1998649836,
-0.1658198833,
-0.0440210998,
0.3708614707,
-0.1272995621,
-0.1514553279,
0.3287454844,
-0.0637821555,
0.0696693435,
0.2002502531,
0.0091879517,
-0.1790072322,
-0.3545345962,
-0.0733048767,
0.4246880114,
-0.1535009444,
0.1729313731,
0.2348184288,
-0.230989635,
-0.0066765323,
0.3626253307,
0.0038940643,
0.1581722796,
-0.0481380373,
-0.1246740371,
-0.4298340976,
0.0749492496,
0.0049049994,
0.33726421,
-0.0327980965,
0.2446380556,
0.0093611181,
0.1696312726,
-0.522657752,
-0.0651670024,
-0.2895471752,
0.2796230018,
0.3717902005,
-0.0287876129,
0.2083657831,
-0.1875352859,
0.3329310417,
-0.0352034867,
-0.1669902056,
-0.383159399,
-0.2214455307,
0.0608603731,
0.0196373239,
0.1678805351,
0.0087600835,
0.1811011136,
-0.0178548433,
-0.1832393259,
0.3455119133,
0.1026546881,
0.0202969015,
0.2028229684,
0.238294214,
-0.0692553446,
0.1074718684,
-0.1394744515,
0.0720089152,
-0.0467457995,
-0.1000200361,
-0.073156096,
-0.3047171533,
-0.0495123118,
0.3234215677,
0.1798159182,
0.2336515486,
-0.0063009644,
0.1078687161,
0.1271709502,
-0.0252149776,
0.32400769,
0.2929457128,
-0.0965292752,
0.0720095038,
0.2770683467,
0.1118193418,
0.4103550017,
-0.4879572392,
0.1754482388,
0.3631242812,
-0.1352906972,
0.5029090643,
0.2272491455,
0.0658406094,
0.2021547556,
0.0913498476,
0.0503363907,
0.0012865551,
-0.2114714682,
0.0581785329,
0.136584118,
-0.0296760947,
-0.0812328011,
0.4753046632,
0.2625533938,
0.1824615002,
0.1179975271,
-0.2786807716,
0.2116144449,
-0.0583929121,
0.042012617,
0.1983388215,
-0.1137420088,
0.3527482152,
0.1639517844,
0.1773965508,
-0.0448388867,
0.1969462186,
0.4129961431,
-0.0848412067,
-0.1146607399,
-0.1688687801,
0.2937826216,
-0.1005146652,
-0.2533902824,
-0.1573496461,
-0.1492339671,
-0.0493522286,
-0.0913517773,
0.1106574833,
-0.199267745,
0.3730929494,
-0.1738865674,
0.1257233024,
-0.3164146245,
0.0824535489,
-0.0605838411,
0.0700605139,
0.0404518694,
0.1876730621,
0.1593519747,
-0.059955813,
0.3969539106,
0.3891628087,
-0.0284736268,
-0.3357330263,
0.2878657579,
0.4870364964,
-0.267085433,
-0.3102597892,
-0.1628082246,
-0.0846296996,
0.0545230284,
0.2395136207,
0.122345008,
0.3468785882,
-0.3152140677,
0.2867633402,
-0.0915713161,
0.1400621384,
-0.1972034723,
0.0750787258,
0.0684301853,
-0.3613373041,
-0.1958662421,
-0.1309575886,
-0.2448278666,
-0.1584970504,
-0.0164301805,
-0.2886988819,
0.1080750749,
0.0048587285,
0.1596199274,
0.199192822,
0.2223719656,
0.0157016926,
-0.1388694644,
0.0664108098,
-0.2817450762,
-0.4920037389,
0.0352380984,
-0.0334374458,
-0.270737648,
-0.1485719681,
0.0685995147,
0.2237578183,
0.156960845,
0.3889604211,
-0.1579191983,
0.0279866979,
-0.2279443741,
-0.2847085297,
-0.0991772562,
0.0880910084,
0.0599071309,
0.1327790469,
0.1024203151,
0.0482949391,
0.1003048122,
0.1921441406,
-0.2107076049,
0.1916915923,
-0.2847425342,
-0.2442820668,
0.1931994706,
-0.0159529746,
0.2518630624,
-0.3791727722,
0.1583209783,
-0.0179832652,
-0.1998199821,
-0.3959466219,
0.0735316575,
0.1107607931,
0.5224798918,
-0.0584190376,
-0.2186153829,
0.0750894696,
0.2099727839,
0.0515120029,
-0.002999045,
-0.1215343922,
0.5358170867,
0.0955800563,
-0.154790774,
0.0824589133,
0.2031577826,
-0.0403055511,
0.0380425155,
-0.0200887248,
-0.473479569,
0.4227637947,
-0.1249491274,
-0.2019238472,
0.0458446667,
0.0944092423,
0.2825829983,
-0.1984641999,
-0.5064718723,
0.1899572909,
0.2214782834,
0.0548426397,
0.0497553498,
0.3494915664,
-0.1279301196,
0.0390068665,
-0.0718113109,
0.4237105846,
0.2931395471,
-0.1638946831,
-0.0361735672,
-0.245703578
] |
https://github.com/huggingface/datasets/issues/1949 | Enable Fast Filtering using Arrow Dataset | Hi @gchhablani :)
Thanks for proposing your help !
I'll be doing a refactor of some parts related to filtering in the scope of https://github.com/huggingface/datasets/issues/1877
So I would first wait for this refactor to be done before working on the filtering. In particular because I plan to make things simpler to manipulate.
Your feedback on this refactor would also be appreciated since it also aims at making the core code more accessible (basically my goal is that no one's ever "having troubles getting started" ^^)
This will be available in a few days, I will be able to give you more details at that time if you don't mind waiting a bit ! | Hi @lhoestq,
As mentioned in Issue #1796, I would love to work on enabling fast filtering/mapping. Can you please share the expectations? It would be great if you could point me to the relevant methods/files involved. Or the docs or maybe an overview of `arrow_dataset.py`. I only ask this because I am having trouble getting started ;-;
Any help would be appreciated.
Thanks,
Gunjan | 113 | Enable Fast Filtering using Arrow Dataset
Hi @lhoestq,
As mentioned in Issue #1796, I would love to work on enabling fast filtering/mapping. Can you please share the expectations? It would be great if you could point me to the relevant methods/files involved. Or the docs or maybe an overview of `arrow_dataset.py`. I only ask this because I am having trouble getting started ;-;
Any help would be appreciated.
Thanks,
Gunjan
Hi @gchhablani :)
Thanks for proposing your help !
I'll be doing a refactor of some parts related to filtering in the scope of https://github.com/huggingface/datasets/issues/1877
So I would first wait for this refactor to be done before working on the filtering. In particular because I plan to make things simpler to manipulate.
Your feedback on this refactor would also be appreciated since it also aims at making the core code more accessible (basically my goal is that no one's ever "having troubles getting started" ^^)
This will be available in a few days, I will be able to give you more details at that time if you don't mind waiting a bit ! | [
-0.0609653145,
-0.2426419556,
-0.176455617,
-0.0540158786,
0.0382517911,
-0.2388271093,
-0.1074122712,
0.2039253116,
0.2231045663,
-0.1845398694,
-0.2857308984,
0.4401738644,
-0.111110501,
0.3817582726,
0.0005040839,
-0.2247797251,
-0.1491404325,
-0.1010932848,
-0.1609328091,
-0.0210788548,
0.2228640616,
-0.0318533629,
-0.1598430723,
0.0115776546,
0.0991677865,
-0.0908594579,
0.4640899897,
-0.0603518002,
-0.3861956596,
-0.4257188141,
0.2159524411,
0.4622388482,
-0.2772482932,
0.3484130204,
-0.0001159239,
-0.1035580784,
0.3350033462,
0.0698488206,
-0.1678131372,
0.1456147581,
-0.2793068886,
-0.464890331,
0.3460852504,
0.0931736976,
-0.2401726246,
-0.3023800552,
-0.4887271821,
-0.5661332607,
0.4217898846,
0.1795722544,
0.1517729312,
0.1879254729,
0.1759022474,
0.0575222895,
0.1689592898,
0.3293294311,
-0.2442489862,
-0.0968679786,
0.7156171203,
-0.1107311547,
-0.0707975626,
0.414040029,
-0.0608406141,
-0.1602593958,
0.3494176865,
-0.265789628,
-0.1116122603,
-0.51766783,
0.2471854985,
0.3313285708,
0.4620247781,
-0.1396790594,
-0.3891332746,
-0.2872940898,
-0.2797903419,
-0.130030036,
0.0210037921,
-0.0488145947,
-0.1954259276,
0.193659246,
-0.1636243761,
-0.2893835604,
-0.3083901107,
0.0227004178,
0.2393587828,
0.263418138,
-0.2676102519,
-0.0376471207,
0.2280461937,
-0.3477034271,
0.0239021815,
-0.1890485287,
0.0998354852,
0.3675711453,
-0.2276188582,
-0.0771989971,
0.168614924,
0.4670671821,
0.3032640815,
0.190650925,
-0.1145788953,
0.3745816052,
0.0616543964,
0.0119902128,
-0.0012413952,
0.0475071333,
-0.0682884753,
0.2453007102,
0.2181311995,
-0.0863904804,
-0.0993872508,
0.060477905,
0.0828245878,
-0.2747675776,
-0.1462862194,
-0.3215172887,
-0.0542431474,
-0.4708521962,
-0.0017120466,
-0.1667614579,
-0.0813659951,
-0.2673491538,
0.4737506807,
0.2197334915,
-0.1195392907,
-0.0570732318,
-0.2483888716,
-0.1011036113,
-0.1768929958,
-0.0029422678,
0.0027062446,
-0.1255300939,
-0.1310472488,
-0.1374053061,
0.1485550255,
-0.194898501,
-0.0996299982,
0.0176742859,
-0.0331113562,
0.2060007006,
0.2575746477,
0.0909351408,
0.3814993203,
0.3757704198,
-0.4013730884,
-0.0247619823,
0.1775442362,
0.0400691628,
-0.2153910398,
-0.0036564935,
-0.2171406299,
-0.5327578783,
0.0032878313,
0.0952743143,
-0.2179058343,
-0.1383995116,
-0.1729110032,
0.6192803383,
-0.2132289708,
-0.0004943982,
0.1261021644,
0.0303854831,
-0.0746981502,
-0.2937880158,
0.3139738739,
0.2567205429,
-0.4902034998,
-0.06085971,
-0.2886764407,
-0.1347983032,
0.2083830982,
0.1620915234,
-0.4143373668,
-0.213201791,
0.162150532,
-0.0254685208,
0.5568775535,
-0.1211442798,
-0.3929443061,
0.0143117048,
-0.0648195446,
0.3086544275,
0.0995057225,
0.0061431229,
0.2001387775,
-0.1530794203,
0.0048772134,
0.6350788474,
-0.1657500267,
-0.1484430432,
-0.3304504752,
-0.0937000364,
0.0300595649,
0.0449816361,
-0.0954458416,
-0.0510531999,
0.1706356257,
-0.5012771487,
0.1360242069,
0.0712876469,
0.175914526,
-0.0128360046,
0.4707919955,
0.2733282447,
0.3618062139,
0.2172623575,
-0.2060660571,
0.2344363481,
0.03011325,
-0.1215240955,
-0.4391415715,
-0.0571953133,
-0.0998982713,
0.1922445297,
-0.1285839826,
-0.0275911726,
0.0785058886,
-0.2916639745,
0.2678439617,
-0.1258751154,
-0.3642390966,
0.2730606496,
-0.123047784,
0.0722343177,
-0.1536019295,
0.0954931825,
0.237028569,
0.1554510295,
-0.0102560669,
-0.090042159,
-0.0304582678,
-0.1076328903,
-0.100215815,
0.006186422,
0.1352827847,
0.2469386458,
0.4057639241,
0.5706697702,
0.1249561384,
-0.4567496181,
0.4129876196,
0.08315216,
0.1664699018,
0.0793100148,
-0.2738562822,
0.7109122276,
-0.496045351,
0.1653508544,
0.1060896143,
0.0411677584,
0.0584858283,
0.1099967808,
-0.3593263924,
0.1287244111,
0.2998857498,
-0.3171957135,
0.2916972935,
-0.0638212115,
-0.1837733984,
0.4710711837,
-0.1567670852,
0.1237304062,
-0.2450218648,
0.3046878576,
0.0002446063,
-0.1571687311,
0.0060962141,
0.3676767051,
0.0476041399,
0.3104171157,
0.074371472,
-0.1966024041,
0.1642756611,
0.0204417296,
0.3753753901,
0.0974472463,
0.3182746768,
-0.0777938738,
0.2552367151,
-0.0056446511,
-0.2419348359,
-0.3230295181,
-0.1648765802,
0.1371054798,
-0.1429524124,
-0.0491913855,
0.1000470147,
-0.3030160964,
0.0799468011,
-0.4713404179,
-0.0805591568,
-0.0791647434,
0.1630922258,
-0.148185432,
-0.0590707213,
0.1847270429,
-0.3384577632,
0.2940372527,
-0.0116719157,
-0.243003726,
-0.3440415859,
-0.1907552481,
0.1860197335,
0.0658499748,
0.0843952,
0.2708714604,
0.3436901569,
0.3919385076,
0.0891069621,
-0.1109164208,
-0.6208747029,
0.037650764,
-0.0244238786,
0.3294962943,
-0.0550893918,
0.1703157425,
0.0902502984,
0.0219804645,
0.2207977474,
-0.1830914617,
-0.0867154077,
-0.2048916221,
-0.2004704475,
0.1766237617,
0.1549229026,
0.0479066074,
-0.4060259461,
-0.3476779759,
0.2588636577,
0.0216843039,
0.1105114967,
0.1468439698,
0.1031277031,
0.110096544,
-0.3070746958,
0.206784457,
0.0702999532,
-0.0732145756,
0.2750935256,
-0.1457663774,
-0.2073545158,
-0.1615955979,
-0.2363075912,
-0.0982560664,
0.31429106,
-0.1593428254,
0.0102769583,
-0.4495509863,
0.2834903002,
0.08643426,
0.0271236766,
0.334581852,
0.0846129879,
0.0097950026,
-0.2262041271,
-0.2326358259,
0.0154641233,
-0.1563066691,
0.1400000602,
0.3474456072,
0.3566596806,
-0.0228051022,
0.7490891218,
0.2729086876,
0.3102783561,
-0.0216605999,
0.2651239932,
-0.0104619972,
-0.0047168881,
-0.3897789717,
-0.2076573074,
-0.1832050681,
0.0713236332,
0.0647227392,
0.4708168507,
-0.2996088862,
-0.2581759691,
-0.1289062351,
-0.230077967,
-0.1261658669,
0.1658199579,
-0.0819253027,
0.2298602015,
0.0215348974,
-0.248134613,
-0.3278767169,
0.021204602,
0.0939891636,
0.1197006404,
0.369430244,
-0.1155928969,
-0.3070029318,
-0.0110327192,
-0.4549898207,
0.3740398586,
0.0163292289,
-0.0884355009,
-0.0765634924,
-0.1763864756,
0.1765042692,
0.0622373708,
0.4188172221,
-0.1716432571,
-0.2019243836,
0.0084137321,
-0.2102385163,
-0.3476476669,
0.3733867705,
-0.1467582881,
0.1551925242,
-0.1636205018,
0.1940008551,
-0.2316624969,
-0.43525967,
0.0153543241,
0.0025667921,
0.1551006734,
0.1351791322,
0.1571703106,
-0.216516912,
-0.0974029154,
0.2244460583,
-0.0569238998,
-0.1157418787,
-0.0190442111,
0.1077831089,
0.0715730786,
0.2140096724,
-0.0328813866,
0.030570399,
0.1167550236,
0.1374804676,
0.1845524609,
0.2986932099,
0.0114757521,
-0.0831664428,
0.4986405671,
0.0181760378,
-0.1970024556,
-0.4065936804,
0.3767670393,
0.4093341529,
0.6419486403,
0.0083806291,
0.4672681093,
0.1627710015,
0.1252395809,
0.0951696187,
-0.2514252365,
0.0303575471,
0.1789841354,
-0.4732808769,
-0.4895400107,
0.5944273472,
0.0797561705,
-0.2069955319,
0.6742947698,
0.4888079166,
-0.0213641673,
0.2731508017,
0.0317812935,
0.9381296635,
0.1353499144,
-0.2550571859,
-0.0551715568,
-0.4572877884,
0.2276632041,
0.1907259971,
0.0677271485,
-0.1280805469,
-0.0280982405,
0.0904291943,
-0.1635538787,
0.2674816549,
-0.0624960773,
-0.1392668784,
-0.1324778497,
-0.014885433,
-0.3627672493,
-0.1140646636,
0.1413879097,
0.1496997029,
-0.3225052953,
-0.4596540332,
0.1375408471,
0.029719878,
0.2756156027,
0.1242970377,
-0.3360020816,
-0.1396518499,
0.0116645992,
-0.1610211134,
-0.2538186908,
-0.4586699605,
0.0900086313,
0.3469236195,
-0.689470768,
0.4541455209,
-0.218952626,
-0.0276081637,
0.024076974,
-0.2672508955,
0.3191042542,
0.3087884486,
0.0923415124,
0.1026989669,
-0.0448272936,
0.4351015389,
-0.1017229855,
-0.1091322005,
-0.1032359302,
-0.2462423146,
-0.3320295811,
-0.1245474517,
-0.118307963,
0.4498852193,
0.0405547023,
-0.1221790612,
0.1863029897,
-0.2323573232,
-0.0750323236,
0.145091325,
0.2975462377,
-0.137802735,
0.3352732658,
0.2150523812,
-0.01837942,
-0.0883292407,
0.1541642398,
0.0119426772,
-0.2380980551,
0.1957039684,
0.4382340908,
-0.3694272339,
-0.072626777,
-0.04237701,
0.413772285,
-0.3821431994,
0.0919388011,
0.0623243116,
0.0000770241,
0.3333820701,
0.2797306478,
0.0447165854,
-0.1187899411,
-0.2067809105,
-0.2249267846,
-0.2354302704,
0.2200055271,
-0.2416063398,
-0.0036624111,
-0.6351543069,
-0.3681564331,
0.1231978834,
0.2730609775,
-0.3414291739,
0.1466784179,
0.1788664162,
0.0029467214,
0.1062560827,
0.1206169575,
0.2502304018,
0.087568678,
0.0558763705,
-0.1611801684,
-0.4421407282,
-0.1634099185,
0.0420130342,
0.1547056288,
-0.227391094,
-0.3869336843,
0.1276440173,
-0.2015355825,
-0.4394989014,
-0.0248725079,
0.0112369508,
0.1066530049,
-0.0087062642,
-0.418230176,
-0.1489038318,
-0.3942815065,
0.1087362319,
0.1185575724,
0.1818752736,
-0.0532112904,
0.2519567013,
0.0725564137,
-0.0979094431,
-0.288905561,
0.068379432,
0.2930140197,
0.0831353664,
-0.1490925252,
0.3740787506,
-0.3148295283,
-0.0208638981,
0.0578758977,
0.4791161418,
0.1315646172,
0.0495789573,
0.2891439199,
0.2545942962,
0.1510763615,
-0.2919921577,
-0.0689228103,
0.2296594381,
0.0560883805,
0.1602949947,
0.0991427675,
0.1550760567,
0.3570217788,
0.1556581855,
0.1796747595,
0.1728062332,
-0.1825159788,
0.0343783237,
0.0758159161,
0.2442067564,
-0.2312828898,
0.5170530081,
0.4599156976,
0.1527623832,
0.2112316191,
0.0077613778,
0.3438524604,
0.1532794684,
0.3685178161,
-0.0573917143,
-0.3248310983,
0.1044099331,
0.1820528358,
-0.3671522737,
0.1109516248,
0.0719504654,
0.4479798675,
0.2105860114,
-0.0966869816,
0.1776971817,
0.229419753,
0.1842699647,
-0.1984754801,
0.2682888508,
-0.4554391205,
-0.0567956641,
-0.0700843111,
-0.1246257126,
0.1792713553,
0.5062456727,
-0.0336424373,
0.492392689,
-0.3771257102,
0.0177506767,
0.1347044855,
0.5880489945,
0.0054553077,
0.1343384683,
-0.0756399184,
-0.0395637043,
0.0040606363,
0.1912316829,
0.2396374047,
0.0882498994,
-0.2215314507,
0.1217187345,
-0.1278242022,
-0.3407156169,
0.0758999139,
0.5048363209,
0.3018204868,
-0.3505642116,
0.006039042,
0.0432858765,
-0.1411370486,
-0.0027774789,
-0.1935842931,
-0.0184469447,
-0.01771608,
-0.0776138157,
0.0754836798,
-0.1306012273,
-0.4775843322,
-0.1475536227,
-0.1269413084,
-0.2714188993,
0.2402141243,
-0.2049629986,
-0.1174961776,
-0.0226438846,
0.0711855218,
-0.0046638772,
0.500295043,
0.2632890642,
0.3848915696,
-0.084428221,
-0.3183429539,
-0.2290224135,
-0.3159653544,
-0.115521118,
0.1433748007,
0.239930898,
0.0960697904,
0.3070606589,
0.4269803762,
0.1642242074,
-0.3078039885,
0.2154695839,
0.1345228702,
-0.2297240049,
0.2803105414,
0.2107986957,
0.3605069816,
-0.1337547302,
0.0139036998,
0.0631632358,
-0.3434785008,
0.0417122543,
-0.1177060455,
0.1252274662,
0.273296535,
0.042407278,
0.0395916142,
0.1712906212,
0.2689296901,
0.135490492,
-0.0002496466,
-0.2420799583,
-0.189033851,
-0.0756739974,
0.163183257,
0.1734725088,
0.3652727604,
0.0811430961,
-0.4510499239,
-0.0391219333,
-0.0669403374,
-0.2234857231,
-0.071023494,
0.0192717537,
-0.0203313641,
0.0477411002,
0.0642385781,
-0.2072569877,
0.2440381497,
0.1739782393,
0.011301443,
-0.4065945446,
-0.2832551599,
0.6696617007,
-0.4198071063,
-0.2190532833,
-0.1257772893,
0.0656546578,
0.3203108907,
-0.1362008303,
-0.2478349209,
-0.1276458651,
0.0048933066,
-0.2941199839,
-0.0921112001,
0.3316380084,
0.0929633379,
0.0761641413,
-0.0724546462,
0.1414703429,
0.1727241576,
0.2042236477,
-0.4060375392,
0.0155453384
] |
https://github.com/huggingface/datasets/issues/1949 | Enable Fast Filtering using Arrow Dataset | Sure! I don't mind waiting. I'll check the refactor and try to understand what you're trying to do :) | Hi @lhoestq,
As mentioned in Issue #1796, I would love to work on enabling fast filtering/mapping. Can you please share the expectations? It would be great if you could point me to the relevant methods/files involved. Or the docs or maybe an overview of `arrow_dataset.py`. I only ask this because I am having trouble getting started ;-;
Any help would be appreciated.
Thanks,
Gunjan | 19 | Enable Fast Filtering using Arrow Dataset
Hi @lhoestq,
As mentioned in Issue #1796, I would love to work on enabling fast filtering/mapping. Can you please share the expectations? It would be great if you could point me to the relevant methods/files involved. Or the docs or maybe an overview of `arrow_dataset.py`. I only ask this because I am having trouble getting started ;-;
Any help would be appreciated.
Thanks,
Gunjan
Sure! I don't mind waiting. I'll check the refactor and try to understand what you're trying to do :) | [
-0.1475261003,
-0.0889888257,
-0.2098283768,
-0.0918973833,
0.0624384992,
-0.2331445068,
-0.0470117144,
0.2564979792,
0.1409168839,
-0.1652853787,
-0.2117006034,
0.5572012663,
-0.1610824764,
0.2340772599,
-0.0721703619,
-0.1115354747,
-0.1630254984,
-0.0334035158,
-0.1529834569,
-0.0570407808,
0.1605862528,
-0.0926775709,
-0.1693582982,
-0.0569528937,
0.0641343221,
-0.0823348761,
0.4899253845,
-0.1223515272,
-0.3783580363,
-0.5257971883,
0.1804878712,
0.412571609,
-0.2424083799,
0.2819817662,
-0.0001122101,
-0.0977285802,
0.3756687939,
0.0460016057,
-0.1633186489,
0.2018046379,
-0.3833834529,
-0.5163573027,
0.3023418486,
-0.0402290598,
-0.2140229493,
-0.3167766333,
-0.4475964308,
-0.6233637929,
0.3333076239,
0.278131932,
0.1977688521,
0.0779646188,
0.1296117604,
0.0343555063,
0.2299921513,
0.1702781916,
-0.2757144868,
-0.1611310095,
0.6908954382,
-0.1348017305,
-0.1131286919,
0.4136620164,
-0.0534186997,
-0.1600668579,
0.2612384558,
-0.3271077275,
0.0970111042,
-0.5140105486,
0.2721890807,
0.2440675795,
0.5241882205,
-0.0686454922,
-0.2503685951,
-0.1154989749,
-0.2959459424,
-0.0541290045,
-0.0665446371,
-0.0593074486,
-0.1523035765,
0.1953791678,
-0.141002655,
-0.3106046021,
-0.3834900558,
0.0470866375,
0.1975130737,
0.3864696622,
-0.1185575798,
-0.0790662915,
0.1377027929,
-0.3078044057,
0.2436963022,
-0.1139412671,
0.0077825282,
0.3805399537,
-0.2030075938,
-0.0627390444,
0.1559076905,
0.4094283283,
0.2585118115,
0.1290133595,
-0.0537549853,
0.3599945903,
0.1209414005,
0.03245496,
-0.0365003422,
0.0077360347,
-0.0455789343,
0.4321995676,
0.1505797803,
-0.207146138,
-0.2090794444,
0.0376982316,
0.0157882459,
-0.2357437611,
-0.0105470419,
-0.2417125404,
-0.0886404216,
-0.4509319663,
-0.0292540528,
-0.1328239292,
-0.1488552988,
-0.2649077475,
0.3727762699,
0.2461597621,
-0.0908475667,
-0.0422482714,
-0.277698487,
-0.131328553,
-0.205775708,
0.0612548925,
0.0159738213,
-0.1020883769,
-0.1534935832,
-0.2902712822,
0.1802890599,
-0.0157663282,
-0.1032728925,
-0.0190819688,
-0.0743510872,
0.2151860148,
0.3460686505,
0.0992514342,
0.411331892,
0.3538830578,
-0.4210653901,
-0.018423032,
0.1159556955,
0.065464437,
-0.2028471529,
0.113978371,
-0.2340937555,
-0.5071297288,
-0.012211673,
0.1530790925,
-0.166732192,
-0.1694043577,
0.0689222962,
0.5219019055,
-0.1484539509,
-0.0539499149,
0.1301654279,
-0.0499785468,
-0.0169827715,
-0.354962945,
0.1976924837,
0.1616173834,
-0.5772002339,
0.0571639538,
-0.3905583024,
-0.0376619659,
0.2938014865,
0.100963667,
-0.4094144702,
-0.2548655272,
0.1447504163,
-0.0853997469,
0.662193954,
-0.1478879452,
-0.5032473207,
0.0212403983,
-0.1166827381,
0.2362808585,
0.1395221949,
0.0708173215,
0.2376326919,
-0.1459854692,
0.0106070526,
0.5886855721,
-0.1571557075,
-0.1358805746,
-0.3851483464,
-0.0330217183,
0.0929223597,
0.0147235766,
-0.0122916745,
-0.0489021949,
0.2204796225,
-0.3853615224,
0.1980083436,
0.0757929832,
0.168153137,
-0.046123907,
0.5657215714,
0.1406770349,
0.3982256353,
0.2781224847,
-0.081304796,
0.24658297,
-0.0564611852,
-0.1005461067,
-0.3765895963,
-0.0060678199,
-0.1080690771,
0.1723632812,
-0.1178416908,
-0.0534225143,
0.1456339955,
-0.2586216629,
0.1876115799,
-0.1119328141,
-0.2767265439,
0.1864237189,
-0.1401676238,
-0.0221049022,
0.0777283758,
0.0473793894,
0.1509615779,
0.1477296352,
-0.0362887383,
-0.2032633424,
-0.0335150473,
-0.0979412496,
-0.1415887326,
-0.0482922383,
0.1992946863,
0.2076705098,
0.3779621422,
0.5796351433,
0.0671272278,
-0.4598428011,
0.4000549912,
0.1309712976,
0.1228168905,
0.0600437298,
-0.2230876684,
0.6486539841,
-0.4913370311,
0.1301897764,
0.0743035674,
0.0571307316,
0.1209029555,
0.0916390121,
-0.3350030482,
0.1061153039,
0.2322736681,
-0.2754028141,
0.2704344988,
-0.0466506891,
-0.1053303927,
0.4582203031,
-0.1308725625,
0.2124437094,
-0.2018691003,
0.2182037234,
0.0384849831,
-0.1373000741,
0.0497141108,
0.4776943624,
0.0552188158,
0.3242242932,
0.1072229743,
-0.2351409197,
0.0779879689,
0.0317963324,
0.4003838003,
0.0332390554,
0.3895940483,
-0.0226085931,
0.2586918771,
-0.0754908174,
-0.1984188259,
-0.3335971534,
-0.0943611711,
0.1433905661,
-0.0976476222,
-0.1084435433,
0.1243606806,
-0.2786762714,
0.0896420777,
-0.482486397,
0.0609074533,
-0.1182475537,
0.1991486847,
-0.2700442672,
-0.025898017,
0.2583578229,
-0.476493448,
0.2215436399,
0.0698768422,
-0.2258893698,
-0.2692491412,
-0.2329248339,
0.1807035208,
0.1188022867,
0.0096918978,
0.2544654608,
0.3076719642,
0.4234210849,
0.185579434,
-0.060786739,
-0.5851409435,
0.0178727992,
0.0574479811,
0.2471530735,
-0.1005004048,
0.0984463096,
0.0665506423,
0.0384894051,
0.1948616505,
-0.177629292,
0.003501974,
-0.2151723653,
-0.1799063236,
0.1618775427,
0.181491375,
-0.0074705593,
-0.4025436044,
-0.384893328,
0.0839819759,
0.0953596458,
0.1507593095,
0.0864572525,
0.1575749069,
0.049817346,
-0.2199694663,
0.1554566622,
0.101988256,
0.0315356851,
0.3705919087,
-0.1950341612,
-0.1830603182,
-0.1638776064,
-0.2358260453,
-0.0301396027,
0.3896689713,
-0.117368035,
0.1325737238,
-0.3845602274,
0.2650755942,
0.0392277054,
-0.0429851487,
0.2871624827,
0.1398596466,
-0.0358020812,
-0.2890631855,
-0.077852346,
0.0468317643,
-0.1745108068,
0.15893659,
0.2671410441,
0.251563251,
-0.0135227293,
0.6781638861,
0.238544777,
0.3053129613,
-0.0278059952,
0.2382784486,
-0.0804035962,
0.005087778,
-0.3636898994,
-0.2951538563,
-0.1504166424,
-0.0515608341,
0.0314595625,
0.392291069,
-0.5023053885,
-0.2534084618,
-0.107609503,
-0.1740991175,
-0.0939804316,
0.2185048163,
-0.2374888957,
0.2750198841,
0.017765522,
-0.1958892941,
-0.3312812448,
0.0534340218,
0.0893672854,
-0.0237317793,
0.3159024417,
-0.1551209241,
-0.2472105026,
-0.0224771742,
-0.3784748614,
0.3182934225,
-0.007368397,
-0.013121631,
-0.0167229325,
-0.175975889,
0.1313332617,
0.0682751387,
0.3394862115,
-0.2564071119,
-0.1299344599,
0.0996433198,
-0.0654612482,
-0.2634763718,
0.3052337468,
-0.1720010638,
0.2245872468,
-0.1720324159,
0.1267336905,
-0.0689732432,
-0.2498940676,
-0.0183737539,
-0.0665834323,
0.1111675352,
0.1037799567,
0.2244229764,
-0.1516263485,
-0.0731590614,
0.2071253508,
-0.0555613935,
-0.1221339554,
-0.020314645,
0.057015948,
0.0557577386,
0.2013000548,
-0.0699209571,
-0.0022004135,
0.1236115247,
0.09784659,
0.1559081972,
0.1842034608,
0.001070709,
-0.0571026579,
0.4396475852,
-0.0187202487,
-0.0930866301,
-0.4215839207,
0.3052021861,
0.417478621,
0.5442762375,
0.0049808882,
0.3545248806,
0.0523049422,
0.0445030257,
0.2079079747,
-0.2735947371,
0.0200397,
0.081699118,
-0.4799914956,
-0.5529646873,
0.5566237569,
0.0445524231,
-0.2231654376,
0.6911586523,
0.2764503956,
-0.0742664039,
0.3161837757,
0.1092560142,
0.8662736416,
0.1756531,
-0.3162345588,
-0.1896564364,
-0.4516644776,
0.1645267606,
0.1649387479,
0.0966423824,
-0.1555037946,
-0.0197429843,
0.100792177,
-0.1575295627,
0.2383475602,
0.0757960156,
-0.1866053194,
-0.1635293961,
-0.0692259967,
-0.4089564383,
-0.1013879254,
0.1564006507,
0.1919611841,
-0.2369520217,
-0.3374944925,
0.1550122499,
0.0376736,
0.1941002905,
0.1098065972,
-0.2859630883,
-0.1854336411,
0.0339030921,
-0.1372964978,
-0.2098489702,
-0.4218372703,
0.2017422616,
0.1767583787,
-0.6858206391,
0.4032756686,
-0.259016037,
-0.1760684401,
0.0660738051,
-0.263384968,
0.3766174912,
0.4508700967,
0.0541924201,
0.1525779217,
-0.0431579314,
0.3689365983,
-0.113098897,
-0.1333815902,
-0.0969930142,
-0.3404680192,
-0.3378995657,
-0.2013331503,
-0.0738955885,
0.5246475935,
0.1063144058,
-0.1258165985,
0.1729201078,
-0.3183770776,
-0.0435613282,
0.2024709433,
0.279201746,
-0.108126916,
0.4356890321,
0.2355087101,
-0.0234058276,
-0.0776554197,
0.1419812441,
0.1190920323,
-0.1750060469,
0.1872938275,
0.3225567937,
-0.3461344838,
-0.116336152,
-0.0046139807,
0.329977423,
-0.2947354913,
0.0382047519,
0.0107052065,
0.0373412371,
0.3760464787,
0.231084466,
0.0844667032,
-0.0195481516,
-0.3427930176,
-0.1993969679,
-0.2802715898,
0.1333852261,
-0.2288029194,
0.0179556422,
-0.6987080574,
-0.3938127756,
0.1250788122,
0.2507770956,
-0.3786206841,
0.1363942176,
0.1850529164,
0.0663275793,
-0.0127424542,
0.1230915412,
0.1247163862,
0.1519988626,
0.1075396985,
-0.139976114,
-0.3695691824,
-0.192740351,
0.1210193336,
0.1453265697,
-0.191930294,
-0.345367372,
0.0982166603,
-0.225456506,
-0.4630486965,
0.064600125,
-0.1146151721,
0.093168959,
-0.0822217762,
-0.3298391104,
-0.1761526465,
-0.4999930859,
0.1140101403,
0.1175866425,
0.1143505275,
-0.0884713233,
0.2574736476,
0.1593894809,
-0.0012377091,
-0.292288959,
0.0625134408,
0.2208312303,
-0.0397204384,
-0.0318805873,
0.4271566868,
-0.2684900165,
0.0735093206,
0.0119832493,
0.4289993644,
0.0997652784,
-0.0149689,
0.2136553079,
0.2370222956,
0.2167618871,
-0.2880710959,
-0.1528852135,
0.2142186463,
-0.0141334981,
0.1783150733,
0.0957961455,
0.1697106063,
0.2545419335,
0.122724548,
0.1712328345,
0.1429407448,
-0.2657507956,
-0.0098023117,
0.2091663033,
0.1680660546,
-0.2541532516,
0.4537974298,
0.3679918349,
0.1609520316,
0.3294352293,
0.0872916579,
0.4186420739,
0.2954574227,
0.3099860847,
-0.0988233835,
-0.2240899205,
0.1348630339,
0.1933920532,
-0.3778967857,
0.1893370599,
0.1609066427,
0.277222693,
0.2905341685,
0.0100823976,
0.1538795531,
0.1100172848,
0.1400521696,
-0.1770973504,
0.2699902058,
-0.4617326856,
-0.0446579903,
-0.0875977129,
-0.1264297962,
0.3221152425,
0.4303368032,
-0.0477118678,
0.456476897,
-0.2363161296,
0.1498159915,
0.0908818096,
0.5641852617,
-0.1071924195,
0.0874127969,
-0.1085973233,
-0.0462371148,
0.0002309806,
0.0986824781,
0.2250639945,
0.226462692,
-0.1615537256,
0.0607656576,
-0.0552968904,
-0.3745657206,
-0.0088105723,
0.5079407692,
0.3447353244,
-0.3521853685,
0.0400465056,
0.0511242598,
-0.1720166802,
0.0784048811,
-0.1419505477,
-0.0302248374,
0.0869739205,
-0.1116787046,
0.1773301661,
-0.2569563389,
-0.4046735466,
-0.102426216,
-0.0818589926,
-0.3288161457,
0.2311630547,
-0.2292255461,
-0.0891654119,
-0.0596245751,
0.0937023461,
-0.0756822079,
0.5075571537,
0.1621735394,
0.3664552569,
-0.0580712408,
-0.3708943427,
-0.2053265274,
-0.1831037551,
-0.2422367185,
0.1734281927,
0.2796364427,
0.1507984996,
0.2834038734,
0.4105998576,
0.2246816307,
-0.2500523627,
0.1204197556,
0.288400948,
-0.2290094048,
0.2530551553,
0.2458891571,
0.3859765828,
-0.1710738987,
-0.0497931056,
0.0808863863,
-0.327778399,
0.1301947385,
-0.1378733516,
0.0346539915,
0.4280060232,
0.1171982586,
0.1686578989,
0.1115587577,
0.319814682,
0.1346818209,
0.0524809137,
-0.2619461715,
-0.2163188457,
-0.0217668042,
0.1465343088,
0.2841084599,
0.3330192566,
0.1076075584,
-0.4194907546,
0.0117203817,
-0.114737317,
-0.3243843913,
-0.0019169711,
0.0411206856,
-0.0516377203,
0.0358680487,
0.0699701309,
-0.3306320012,
0.1953561604,
0.1989712566,
-0.0378248096,
-0.4084856212,
-0.2806211412,
0.5808031559,
-0.3440419436,
-0.2982539833,
-0.0936835408,
0.0427402481,
0.1809695661,
-0.0274207499,
-0.1848160923,
-0.1665576994,
0.1074380353,
-0.3471878767,
-0.0620346889,
0.27253443,
0.1391054988,
0.1319848597,
-0.0350429006,
0.0918885022,
0.1254365742,
0.2822115421,
-0.5548896194,
-0.0043712854
] |
https://github.com/huggingface/datasets/issues/1948 | dataset loading logger level | These warnings are showed when there's a call to `.map` to say to the user that a dataset is reloaded from the cache instead of being recomputed.
They are warnings since we want to make sure the users know that it's not recomputed. | on master I get this with `--dataset_name wmt16 --dataset_config ro-en`:
```
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-2e01bead8cf42e26.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-ac3bebaf4f91f776.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-810c3e61259d73a9.arrow
```
why are those WARNINGs? Should be INFO, no?
warnings should only be used when a user needs to pay attention to something, this is just informative - I'd even say it should be DEBUG, but definitely not WARNING.
Thank you.
| 43 | dataset loading logger level
on master I get this with `--dataset_name wmt16 --dataset_config ro-en`:
```
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-2e01bead8cf42e26.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-ac3bebaf4f91f776.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-810c3e61259d73a9.arrow
```
why are those WARNINGs? Should be INFO, no?
warnings should only be used when a user needs to pay attention to something, this is just informative - I'd even say it should be DEBUG, but definitely not WARNING.
Thank you.
These warnings are showed when there's a call to `.map` to say to the user that a dataset is reloaded from the cache instead of being recomputed.
They are warnings since we want to make sure the users know that it's not recomputed. | [
-0.1510902941,
-0.3636068404,
-0.0170671567,
0.3475190103,
0.4003514349,
0.4622994363,
0.4706438184,
0.1432850361,
0.2523235083,
-0.0047667027,
0.0243788064,
-0.016878346,
-0.2481702268,
-0.1975331157,
-0.3672414422,
0.1944211572,
-0.1171081066,
-0.0153788626,
-0.3518668115,
-0.0757234246,
-0.1171180457,
0.0906819999,
-0.0266269371,
0.5081022382,
-0.7216523886,
-0.0836911201,
0.1400511265,
0.1188434511,
0.0099195912,
-0.7821522355,
0.2707004547,
-0.1280098706,
0.1496112496,
0.0640892833,
-0.000122115,
0.2060289383,
0.6381807923,
0.0515550524,
-0.4871642292,
0.0292479843,
-0.4754453003,
-0.4789412022,
0.2521519363,
-0.1042922437,
-0.0297282599,
-0.3696422577,
0.192222625,
-0.4077892601,
0.3286341429,
0.1021778882,
0.1139924154,
0.1136899516,
-0.1395549923,
0.032682009,
0.3232572377,
0.2458926886,
0.0373669416,
0.0479511581,
-0.0644855052,
0.0700881109,
-0.2977601588,
0.440929085,
-0.2956631482,
0.052874662,
0.2401666343,
-0.0134788044,
0.2989348173,
-0.230270043,
0.1710722446,
0.3026789725,
0.5181795955,
-0.0112025812,
0.113106966,
-0.4032348394,
-0.2267823815,
-0.0801946968,
0.167365998,
0.1750007272,
0.071756959,
0.1927926838,
-0.377902478,
0.1259965152,
0.1591121405,
-0.0451120362,
0.2188044786,
0.3621397018,
-0.3730173409,
0.2935866416,
0.1711705923,
-0.0791648701,
0.1143634096,
-0.2527512908,
-0.3305522203,
0.1376242489,
-0.2099728286,
0.0138350651,
-0.1173723787,
0.3929843009,
-0.103616409,
0.0236905441,
0.1074152589,
-0.0235768352,
0.1798297167,
0.1378366351,
0.4080687761,
0.0065898336,
0.4469652474,
0.053667061,
-0.1491610855,
-0.1865979433,
0.1863410771,
-0.0538990162,
0.0364209972,
-0.1401541084,
0.5328086615,
0.1328555942,
0.1173178107,
-0.1514059603,
-0.0866407081,
0.0324661359,
-0.106475614,
-0.2918037176,
0.0128358826,
0.1817290038,
-0.046955701,
-0.0455382913,
-0.1222740635,
-0.064284645,
0.0570406467,
-0.1738080084,
-0.0831569433,
-0.3341276646,
-0.3643179834,
0.1104172915,
-0.0247188881,
-0.2077739835,
0.3294540644,
0.1876463294,
-0.1365976781,
0.0210159943,
0.1092820987,
-0.1964776218,
0.1147877127,
0.6276437044,
-0.228298828,
0.2196656466,
0.3440072536,
0.1959388256,
-0.1997302771,
0.4353929162,
-0.410954088,
-0.4699915648,
0.0184686314,
0.0884319842,
-0.1389924884,
0.0448095016,
-0.3083497584,
0.1366654187,
0.3585318029,
0.1251951903,
0.2424764037,
-0.1067552119,
-0.3246655464,
-0.1552182287,
-0.1979090869,
0.5811362267,
-0.1848371625,
-0.3025491834,
-0.3280837834,
-0.4322118163,
0.2125729024,
0.121833615,
-0.2136900723,
0.2800592482,
-0.3107713461,
-0.0972536579,
0.3082107902,
-0.1050787941,
0.0822919607,
0.4469822347,
-0.2502986789,
0.168002829,
0.2125221193,
0.0836042911,
-0.215675503,
-0.1224222928,
0.013581533,
-0.4536337852,
0.0406979024,
-0.1546108723,
-0.1473660469,
-0.1895030737,
0.2960462272,
0.0144897178,
-0.2710362673,
-0.0140962303,
0.0367450938,
-0.1554631591,
0.2397704422,
0.127126947,
0.1040887386,
0.0431242883,
-0.1647825092,
0.0198249035,
-0.045083385,
0.3280400634,
-0.6091819406,
0.0961714387,
0.1165383533,
-0.0305935442,
0.0045086704,
-0.0909323171,
-0.0490079969,
-0.0273274221,
-0.3792022467,
-0.1640517563,
-0.0221963394,
0.0016260687,
-0.0596201383,
-0.0850861818,
-0.0751338825,
0.2245910764,
-0.5344880819,
0.0863712057,
0.0056905411,
-0.166102469,
0.0117708668,
0.3507809341,
-0.2235101014,
-0.1054347381,
0.1355482638,
0.079268083,
-0.1547687799,
0.217343241,
0.1086673141,
0.2987656593,
-0.0428355597,
0.394744873,
0.0307756122,
-0.1995356381,
0.4566541612,
-0.1920306683,
-0.1919047832,
-0.1184579879,
0.1470094323,
0.1488821507,
-0.0293640383,
0.2458748817,
-0.3589743674,
0.0014750436,
0.0724731907,
0.1763547063,
-0.1827460229,
-0.5318389535,
-0.0535601191,
0.1382285208,
0.2236330509,
0.3524543941,
0.0328618884,
0.0614071414,
0.5036966205,
-0.022522714,
-0.0350023434,
0.1725948751,
-0.1172599345,
-0.1267616451,
0.2107432336,
0.0995999426,
0.4006417096,
0.14335756,
0.0964038074,
0.2135555744,
0.0398832932,
-0.0287606381,
0.2845731378,
0.0011582077,
0.291947037,
0.0412641838,
-0.4688799679,
-0.053495504,
-0.2464309782,
0.0426364541,
-0.1107658967,
-0.1338512152,
-0.5484199524,
-0.1352428496,
-0.1794121563,
-0.2473459244,
-0.3247791529,
-0.2180157304,
-0.2057369202,
-0.4174847901,
0.0159911327,
-0.1967618167,
-0.355907619,
0.1864649653,
-0.1796559095,
0.2185804844,
0.1087457314,
0.2702389657,
-0.3594065905,
-0.2073976398,
-0.1868628561,
0.0056629665,
-0.1753743887,
-0.1987685263,
0.0720699653,
-0.3009378612,
0.2761228681,
-0.2969667912,
-0.1735819429,
0.1628869474,
-0.3161107004,
-0.0158542283,
0.154159978,
0.0007794909,
0.0341844708,
-0.1384245157,
0.0944355354,
0.120767273,
-0.0642792061,
-0.3195790648,
0.1604952961,
0.1289813519,
-0.1551725417,
-0.2121983767,
-0.3349227607,
-0.1513200998,
-0.1797146201,
-0.3641606569,
0.1897163689,
0.3864741623,
0.0024575628,
0.1253540814,
-0.277235806,
0.2045512497,
-0.3516031802,
-0.6407561302,
0.0419359021,
-0.0941981971,
0.0447053723,
0.0233053267,
0.1199380457,
0.1616580784,
-0.0717072189,
-0.8328556418,
0.0837570652,
-0.1131228358,
-0.2497929186,
0.2192755193,
-0.0311908443,
-0.0892970935,
0.2955491245,
-0.0040109977,
-0.0155310854,
-0.2937011123,
0.0342545062,
-0.4827653766,
0.2337225229,
0.0650282875,
0.4955636263,
-0.1274766326,
0.6671243906,
0.1786765158,
-0.0275562517,
0.1970892847,
0.0470818356,
0.6038403511,
-0.1499833614,
-0.1548582911,
-0.1112251654,
0.0743584186,
-0.1259294897,
-0.0157108493,
-0.0841408968,
0.0476630852,
0.1504991055,
0.1654735208,
0.0786558837,
-0.242595911,
0.0733596012,
-0.4269168079,
-0.0912574753,
-0.1446543038,
0.0739910305,
0.1281629205,
-0.0123728253,
0.0110634081,
0.4801493585,
0.3755214512,
0.0773745775,
-0.276647985,
-0.1589406282,
-0.1690038145,
0.1122552976,
0.0917742625,
0.0670725629,
-0.3882782161,
-0.306481719,
-0.0721232817,
0.3549329042,
0.1371837854,
-0.1570129246,
0.2949316204,
0.1318045855,
-0.1216840968,
-0.1347219348,
-0.2034213245,
0.0879287869,
0.154179424,
0.1276380718,
0.1717557609,
-0.0838698968,
-0.3960536718,
0.1100465581,
-0.1419894248,
-0.1392141879,
-0.1915452778,
-0.2848085165,
0.299253881,
-0.203809455,
-0.0032045171,
-0.0412967727,
-0.1334179491,
-0.4892790914,
-0.1212310344,
0.1779166609,
0.2980811596,
-0.0719583184,
-0.1004739553,
-0.018829871,
0.3366834819,
-0.0939930454,
0.363850534,
0.647228539,
0.2555455267,
0.7740372419,
0.0363736041,
-0.1040381417,
-0.0139204189,
-0.0484956466,
0.2307560146,
0.1523724198,
-0.0274607297,
0.0000711493,
0.3212536275,
0.3440133631,
-0.4421431124,
0.133749485,
0.2313802689,
0.1668225825,
-0.4083358645,
-0.2868031859,
0.5116583705,
0.214274317,
-0.1043916047,
0.5442044139,
0.0711139068,
-0.1615945101,
-0.2187596858,
0.3741801977,
0.8942172527,
-0.1471620351,
0.1002800912,
-0.0160290767,
-0.0071231425,
0.1717579663,
0.1440545619,
-0.2678993642,
0.0073181465,
0.1029145867,
-0.09251073,
-0.0707774311,
0.1881528646,
0.5290837288,
-0.1787945628,
0.0160600543,
-0.2433731854,
0.3730193377,
0.0617844388,
-0.0121213794,
-0.0502512902,
-0.0936079547,
-0.1660182476,
0.0052701347,
0.1045098305,
0.2041027695,
-0.2394819558,
0.0460052826,
0.3798737228,
-0.2689359784,
-0.1060881019,
-0.2011123449,
-0.2168130279,
0.1850456297,
-0.057273943,
-0.0470171347,
0.1191348657,
0.0937592685,
-0.0234394521,
0.1446027309,
0.0119470563,
-0.1187983304,
0.0615762323,
0.0238210745,
-0.364318192,
0.0179007202,
0.2325511873,
-0.0072844997,
-0.2646381259,
0.2048376799,
-0.384092927,
-0.5043622851,
0.0228942633,
0.0711932778,
-0.1922393441,
-0.2036357671,
0.053313002,
0.1699158847,
-0.2309154272,
0.2873204052,
0.0293757766,
0.1502043903,
0.1965527236,
0.0601567999,
0.2314127684,
-0.6483149529,
-0.180545181,
0.2899328768,
0.0787511915,
-0.0787152946,
0.4794632792,
-0.2117161155,
-0.0789495409,
-0.1393014789,
0.0622330606,
0.3695784211,
-0.0275952443,
-0.1137569845,
0.3495696783,
0.0395734459,
0.112949796,
-0.025881689,
0.2895484865,
0.2128801048,
-0.2625930309,
-0.3203587234,
-0.626681149,
0.2629420757,
-0.5137579441,
0.1910708547,
-0.1408806592,
-0.245164305,
0.3036810756,
0.1079547703,
-0.1916730404,
-0.0999563187,
0.0556899607,
-0.0071932822,
0.1709100008,
0.1493298113,
0.3320843875,
0.2791548967,
-0.0922303945,
-0.3216715455,
-0.4433019757,
-0.0400056466,
0.168495357,
0.2169547528,
0.009491032,
-0.0795173645,
-0.2575776875,
-0.1280806512,
0.1019526869,
-0.0778397694,
0.3905064464,
0.1049132496,
-0.009597335,
0.3817891479,
-0.1426328868,
-0.0086786989,
-0.2456882298,
-0.0357542709,
0.1109168679,
0.0550183952,
-0.151347518,
-0.1801466346,
-0.0491040349,
-0.0539739281,
0.2511942983,
0.0999223888,
0.0972090513,
-0.0785997137,
0.1991381496,
0.1278102398,
0.1732059717,
0.2066395134,
0.1552710086,
0.0559154004,
0.0019715019,
-0.3858051002,
-0.2215704769,
0.0646759868,
0.0356073007,
0.0378214046,
-0.1296637356,
0.4016623199,
0.0059828982,
0.0710973069,
0.1350277364,
0.1085010916,
0.5503321886,
0.1401060671,
0.1703459471,
-0.4319109619,
0.2125663459,
0.3312865198,
-0.280924648,
0.3473608494,
0.535461545,
0.132288307,
0.2809296548,
0.0803073272,
0.2377756536,
0.5244956017,
0.3298660815,
0.0640480369,
-0.1033366174,
-0.552082181,
0.1015585437,
0.6678974032,
-0.5076916814,
-0.1573924571,
0.1145322174,
0.0160784405,
0.1352503151,
-0.1314967722,
-0.1689447463,
-0.1418941915,
-0.1557695866,
-0.0667338595,
0.0940258205,
-0.0192206949,
-0.0575185716,
-0.0362390727,
0.1004227698,
-0.0434549637,
0.2444920987,
-0.3040904701,
-0.2544709444,
-0.0272780638,
-0.0317717306,
0.278442353,
0.1199143231,
-0.2305402756,
-0.0807027966,
0.1579214036,
-0.0214180835,
0.4265592396,
0.5706605911,
0.5717047453,
0.6063531637,
0.1830430925,
0.1278856695,
0.0131119676,
0.0590675548,
-0.0073182918,
0.6166630387,
-0.0045633949,
-0.543900311,
0.278452903,
0.0439494625,
0.0041392781,
0.3585561216,
-0.2119915187,
0.322488755,
0.1105728373,
0.1054095775,
0.0642698258,
-0.0584098063,
0.1374062002,
0.0827970505,
-0.2589651644,
0.072637096,
0.9143496156,
-0.1347180009,
-0.0051502623,
-0.3627816141,
0.0297408104,
0.0700255781,
0.7529669404,
0.6106792688,
0.0658988655,
-0.310523361,
0.0288971364,
-0.579603374,
0.0823533088,
-0.0141900629,
0.3293285072,
0.1985840648,
-0.2157305032,
0.1154162362,
0.3421982527,
0.2262532711,
-0.3068056107,
-0.0698385686,
-0.4682079256,
-0.2701198161,
-0.0169811696,
0.345425576,
0.1913653016,
-0.0666048229,
-0.109635964,
0.333304286,
-0.0088536553,
-0.0584820881,
-0.1018453985,
0.2546591163,
0.0168319046,
0.4844091237,
0.3636512458,
0.1958845556,
0.3726078272,
0.1436445415,
0.0181048736,
-0.3456308842,
-0.2510225773,
0.0901650116,
0.2232736051,
0.1879868656,
0.1458601356,
-0.1542795002,
-0.0382019319,
-0.2758578658,
0.0652720034,
0.0706899688,
0.0428656302,
0.1385472715,
-0.0437364243,
-0.1341779828,
-0.1813514978,
-0.0378306285,
0.1209158674,
0.4951341748,
0.1649114192,
-0.0142053068,
-0.2301945984,
0.3458405137,
-0.1777029335,
0.0022799224,
-0.0202442408,
0.1730801463,
0.1003086418,
-0.2233655751,
-0.8071295619,
0.0816919953,
0.4822102189,
-0.2035747766,
-0.1766884774,
0.0255614929,
-0.0607889779,
-0.0086673871,
0.0344971791,
0.4376495183,
-0.1060258374,
-0.0757306963,
-0.2217494696,
-0.5205209851
] |
https://github.com/huggingface/datasets/issues/1948 | dataset loading logger level | Thank you for explaining the intention, @lhoestq
1. Could it be then made more human-friendly? Currently the hex gibberish tells me nothing of what's really going on. e.g. the following is instructive, IMHO:
```
WARNING: wmt16/ro-en/train dataset was loaded from cache instead of being recomputed
WARNING: wmt16/ro-en/validation dataset was loaded from cache instead of being recomputed
WARNING: wmt16/ro-en/test dataset was loaded from cache instead of being recomputed
```
note that it removes the not so useful hex info and tells the user instead which split it's referring to - but probably no harm in keeping the path if it helps the debug. But the key is that now the warning is telling me what it is it's warning me about.
```
Warning:Loading cache path
```
on the other hand isn't telling what it is warning about.
And I still suggest this is INFO level, otherwise you need to turn all 'using cache' statements to WARNING to be consistent. The user is most likely well aware the cache is used for models, etc. So this feels very similar.
2. Should there be a way for a user to void warranty by having a flag - `I know I'm expecting the cached version to load if it's available - please do not warn me about it=True`
To explain the need: Warnings are a problem, they constantly take attention away because they could be the harbinger of a problem. Therefore I prefer not to have any warnings in the log, and if I get any I usually try to deal with those so that my log is clean.
It's less of an issue for somebody doing long runs. It's a huge issue for someone who does a new run every few minutes and on the lookout for any potential problems which is what I have been doing a lot of integrating DeepSpeed and other things. And since there are already problems to deal with during the integration it's nice to have a clean log to start with.
I hope my need is not unreasonable and I was able to explain it adequately.
Thank you. | on master I get this with `--dataset_name wmt16 --dataset_config ro-en`:
```
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-2e01bead8cf42e26.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-ac3bebaf4f91f776.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-810c3e61259d73a9.arrow
```
why are those WARNINGs? Should be INFO, no?
warnings should only be used when a user needs to pay attention to something, this is just informative - I'd even say it should be DEBUG, but definitely not WARNING.
Thank you.
| 351 | dataset loading logger level
on master I get this with `--dataset_name wmt16 --dataset_config ro-en`:
```
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-2e01bead8cf42e26.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-ac3bebaf4f91f776.arrow
WARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/stas/.cache/huggingface/datasets/wmt16/ro-en/1.0.0/9dc00622c30446e99c4c63d12a484ea4fb653f2f37c867d6edcec839d7eae50f/cache-810c3e61259d73a9.arrow
```
why are those WARNINGs? Should be INFO, no?
warnings should only be used when a user needs to pay attention to something, this is just informative - I'd even say it should be DEBUG, but definitely not WARNING.
Thank you.
Thank you for explaining the intention, @lhoestq
1. Could it be then made more human-friendly? Currently the hex gibberish tells me nothing of what's really going on. e.g. the following is instructive, IMHO:
```
WARNING: wmt16/ro-en/train dataset was loaded from cache instead of being recomputed
WARNING: wmt16/ro-en/validation dataset was loaded from cache instead of being recomputed
WARNING: wmt16/ro-en/test dataset was loaded from cache instead of being recomputed
```
note that it removes the not so useful hex info and tells the user instead which split it's referring to - but probably no harm in keeping the path if it helps the debug. But the key is that now the warning is telling me what it is it's warning me about.
```
Warning:Loading cache path
```
on the other hand isn't telling what it is warning about.
And I still suggest this is INFO level, otherwise you need to turn all 'using cache' statements to WARNING to be consistent. The user is most likely well aware the cache is used for models, etc. So this feels very similar.
2. Should there be a way for a user to void warranty by having a flag - `I know I'm expecting the cached version to load if it's available - please do not warn me about it=True`
To explain the need: Warnings are a problem, they constantly take attention away because they could be the harbinger of a problem. Therefore I prefer not to have any warnings in the log, and if I get any I usually try to deal with those so that my log is clean.
It's less of an issue for somebody doing long runs. It's a huge issue for someone who does a new run every few minutes and on the lookout for any potential problems which is what I have been doing a lot of integrating DeepSpeed and other things. And since there are already problems to deal with during the integration it's nice to have a clean log to start with.
I hope my need is not unreasonable and I was able to explain it adequately.
Thank you. | [
-0.1465868801,
-0.2024428099,
0.0129190609,
0.3102096021,
0.4003249109,
0.3999481797,
0.4905205071,
0.2311710417,
0.1243602335,
0.0282331109,
0.0093347281,
-0.1472629458,
-0.2483211011,
-0.1882982254,
-0.2508574128,
0.0609391853,
-0.1264947355,
-0.0315522701,
-0.385997504,
-0.1280374825,
0.0110194236,
-0.0093770688,
0.0819810629,
0.461540848,
-0.6953532696,
-0.0577310771,
0.0907772481,
0.269282937,
0.0484173223,
-0.7044392824,
0.268795222,
-0.0610607192,
0.1792586148,
0.1538813561,
-0.0001220934,
0.1301615536,
0.6268299818,
0.0067864135,
-0.6243700981,
0.0161630884,
-0.3979125917,
-0.3823044896,
0.1909533739,
-0.0893726349,
-0.0878834724,
-0.3329227269,
0.3174686134,
-0.3982447088,
0.3191030324,
0.1539317369,
0.0972732157,
0.1309636831,
-0.1569447815,
0.1304769516,
0.3219232857,
0.2671608627,
-0.0668996722,
0.0625432879,
-0.0271568671,
0.1869475543,
-0.3544816077,
0.3654178977,
-0.189787209,
-0.0355946049,
0.1682559848,
-0.0251545031,
0.2707449794,
-0.0829858929,
0.098023355,
0.4570212364,
0.4693351388,
-0.1381198466,
0.096009016,
-0.3951647282,
-0.1142802685,
-0.3058124781,
0.1722686887,
0.1817943901,
-0.1236339658,
0.1642875224,
-0.2786458135,
0.0962312669,
0.0633185506,
-0.0412001349,
0.2753942311,
0.5356910825,
-0.2549028993,
0.1994910985,
0.2321581841,
-0.1086489186,
0.3180997968,
-0.0647457093,
-0.292390734,
0.0557423085,
-0.3284544945,
0.096123822,
-0.1569818556,
0.2279086262,
-0.1270326525,
0.0550275296,
0.0303926133,
0.0645744354,
0.1766624898,
0.0465368815,
0.3021320105,
0.0782381073,
0.3390363753,
-0.0437265038,
-0.203904599,
-0.0232509002,
0.1906762719,
-0.0522891432,
-0.1306728721,
-0.0884444043,
0.4426143169,
0.2055934966,
0.0061229691,
-0.0329379067,
-0.1966414601,
0.0259854272,
-0.1524895132,
-0.3309821188,
0.0248685479,
0.1910765618,
-0.079017207,
-0.0314244777,
-0.1334364712,
-0.0265854448,
-0.0438803062,
-0.2736991644,
-0.0795496851,
-0.2813803256,
-0.3722063005,
0.1771147251,
-0.1128015444,
-0.1633091718,
0.1749126911,
0.1784884781,
-0.056344986,
0.0769143999,
-0.0438769385,
-0.147441119,
0.125185892,
0.4888393879,
-0.1188969612,
0.2109149396,
0.315344125,
0.1144105196,
-0.250860244,
0.321914196,
-0.3088694513,
-0.4429711699,
-0.1388503015,
0.0568972677,
-0.1872850955,
-0.0427054763,
-0.2263441384,
0.2049742937,
0.2080990076,
0.2616478801,
0.3043171465,
-0.0870516598,
-0.2893236578,
-0.1400417536,
-0.1068024933,
0.6045224071,
-0.230040729,
-0.3448929787,
-0.2233899832,
-0.3908490539,
0.3220080733,
0.1940955073,
-0.1252971888,
0.3429296613,
-0.2669637501,
-0.1275546998,
0.2065175772,
-0.1993078887,
0.1538255513,
0.5722751021,
-0.2666300535,
0.3573940396,
0.2487148345,
0.0757365823,
-0.3795432448,
-0.1587416232,
0.0931626037,
-0.4061234593,
0.0346497297,
-0.2195474207,
-0.1422884017,
-0.3101750016,
0.3162405491,
0.0956732631,
-0.1243338436,
-0.0407623313,
-0.027005624,
-0.0596589595,
0.5715069771,
0.0659451634,
0.0881295055,
0.0159739815,
-0.0474754944,
0.0137910228,
-0.1247669905,
0.3047452867,
-0.4733554721,
0.0033312738,
0.2965785861,
0.15414235,
0.0607186407,
-0.0349126533,
-0.1101763099,
-0.0416644178,
-0.4651914835,
-0.1592129022,
-0.0148240067,
0.0448273793,
-0.2315163761,
-0.1339939982,
-0.1228308529,
0.3283226192,
-0.5877948999,
0.059179917,
-0.1173667312,
-0.3510160148,
0.0371701047,
0.3243013918,
-0.2366800457,
0.011676223,
0.1059908271,
0.0001264215,
-0.1354234964,
0.2674869299,
-0.0427941643,
0.3116444945,
-0.082394816,
0.5417582989,
0.0388028175,
-0.1560543627,
0.3681598008,
-0.2089682072,
-0.2602339387,
-0.0510699712,
0.1137241051,
0.0264502056,
0.0294056311,
0.2619310915,
-0.2610862255,
-0.1066247299,
0.0779918879,
0.0519059151,
-0.1736052334,
-0.3629188538,
0.1012514457,
0.1045230925,
0.2224512398,
0.3565239608,
-0.0605902635,
0.1401379108,
0.3738937676,
-0.0226566344,
-0.0009534433,
0.2139961123,
-0.0660067946,
-0.1223923266,
0.1966650486,
0.089625597,
0.1997384429,
0.1502247155,
0.1785116047,
0.2778918743,
-0.0631349385,
-0.0572558716,
0.2577067614,
0.0119200032,
0.4047442675,
0.0357543752,
-0.4561926723,
0.0109884962,
-0.2117693722,
-0.084150292,
-0.1447161883,
-0.0529184379,
-0.6462906599,
0.0031947866,
-0.2683839202,
-0.343390882,
-0.3749729395,
-0.2526741326,
-0.1563916802,
-0.3785585761,
0.0086876824,
-0.3024218976,
-0.3954232633,
0.1360270679,
-0.3061368465,
0.2309263647,
0.1001673341,
0.2513029575,
-0.2852824032,
-0.066187039,
-0.2680496275,
-0.0152587593,
-0.0835174695,
-0.2519969344,
0.0985284895,
-0.4905314445,
0.162921384,
-0.1752012372,
-0.2026418447,
0.3088509142,
-0.197500512,
0.0879950151,
0.1897616684,
-0.0893781632,
-0.0432894044,
-0.2671485841,
0.0250471793,
0.1185781658,
0.0083141029,
-0.3542826772,
0.0778483301,
0.1457301974,
-0.1193104088,
-0.0913557708,
-0.3239559531,
-0.1959174275,
-0.1272506118,
-0.3672881722,
0.295948863,
0.4755598009,
0.031113755,
0.0013715737,
-0.1817461848,
0.2356294096,
-0.2804740369,
-0.7693065405,
-0.032526549,
-0.1196924001,
0.1096620932,
-0.0435090885,
0.1002776325,
0.162879169,
0.0031550452,
-0.8037424684,
0.1855158508,
-0.1419698894,
-0.3532152772,
0.160369426,
-0.003666563,
-0.1876030713,
0.2181836963,
-0.0146572627,
0.0057895631,
-0.2862211466,
0.0531044081,
-0.4676642716,
0.4226563573,
0.1411725283,
0.4266451597,
0.1103249565,
0.479778558,
0.1754453778,
-0.0096645551,
0.2206353992,
-0.0405064076,
0.6518884897,
-0.0414567813,
-0.0822483674,
0.0396546423,
0.0212174952,
-0.2239888608,
-0.0989486724,
-0.0745001882,
0.2100462615,
0.2413644791,
0.0937358886,
0.035237506,
-0.401629746,
0.2715890408,
-0.4244948626,
-0.139444083,
-0.1997301579,
0.1225217283,
0.2278075367,
-0.0328871794,
0.1564407349,
0.4967174232,
0.2978892624,
0.171515733,
-0.2740747929,
-0.1311539263,
-0.3372913897,
-0.0180857256,
0.0550625026,
0.1616723239,
-0.3739036024,
-0.3055794537,
-0.110557884,
0.305177927,
0.1360486597,
-0.1391137242,
0.2344474792,
0.1739257127,
-0.0696469843,
-0.0882598609,
-0.1640912294,
0.0512593016,
0.0295520835,
0.1467778981,
0.0859368145,
-0.1525841951,
-0.4390389323,
0.1397967637,
-0.0961281583,
-0.2378575802,
-0.250474453,
-0.205490768,
0.1677539945,
-0.0954988152,
-0.024551034,
-0.0470243692,
-0.0282835662,
-0.5195003152,
-0.0082284212,
0.1157224178,
0.3064212799,
0.0755931288,
-0.1433748603,
-0.0902584195,
0.3614209592,
-0.1069243029,
0.4327397048,
0.7041025758,
0.2550818026,
0.7299550176,
0.0712158382,
-0.0811748505,
-0.0565038547,
-0.0520723239,
0.2179765552,
0.3498032987,
0.1193919927,
-0.0570035726,
0.3241609335,
0.4321884811,
-0.4970781505,
0.3054850996,
0.2432096899,
0.100424692,
-0.4775583744,
-0.3319886327,
0.4986372888,
0.1059210896,
-0.040184252,
0.4706761241,
0.2508615255,
-0.1733489931,
-0.1791165918,
0.4485978186,
1.0204644203,
-0.141329363,
0.2379664779,
0.0887254924,
-0.0671919435,
0.2032370418,
0.0498398095,
-0.2287024707,
-0.0339878723,
0.0954980403,
-0.1102619618,
-0.0079032481,
0.0217751749,
0.4259313941,
-0.081496954,
0.0552999303,
-0.3799418211,
0.1997650266,
0.2381652892,
0.0111748204,
-0.1948653311,
-0.1852945387,
-0.1854828,
-0.0109014772,
0.1929552555,
0.1259744763,
-0.2549643219,
-0.0789040476,
0.3553371429,
-0.1292883754,
-0.2059783041,
-0.1289222687,
-0.218099758,
-0.0243698545,
-0.1056511551,
-0.0971039534,
0.0242865384,
-0.06893076,
0.0443329476,
0.1664552987,
0.0667944923,
0.005808752,
0.0035966989,
0.0768847913,
-0.3164058924,
-0.0807073712,
0.3426612914,
0.0499627292,
-0.3006896973,
0.2033223361,
-0.2766219079,
-0.5778279901,
0.0519155264,
-0.1594391167,
-0.2073710263,
-0.0885657147,
0.1399163902,
0.1879903972,
-0.1414651126,
0.2217410654,
0.0167124309,
0.1536527872,
0.2042518407,
0.0143994596,
0.2500071824,
-0.5792986751,
-0.16281192,
0.0786971301,
0.0129219759,
-0.1310414672,
0.4648005962,
-0.0984342769,
-0.0492203757,
-0.1144692972,
-0.0070031509,
0.5963888764,
-0.1043779701,
-0.1934036463,
0.272580266,
-0.0487122387,
0.0769602656,
-0.1061722636,
0.3125026226,
0.0999433994,
-0.2827931046,
-0.2267075777,
-0.540958941,
0.2808478475,
-0.4361154437,
0.2125407606,
-0.1690122783,
-0.1476346403,
0.225999862,
0.0833063275,
-0.178397432,
-0.2111173123,
-0.0559467524,
-0.1625299156,
0.1657119691,
0.2391399443,
0.3696462214,
0.2524446249,
-0.0903136879,
-0.3158458769,
-0.3205567896,
-0.0216380116,
-0.0654083639,
0.224413529,
-0.0104352878,
-0.1175984368,
-0.0966226459,
-0.0900589004,
0.1384567469,
-0.1828059852,
0.3072930574,
0.2030452192,
0.0739671439,
0.4574134946,
-0.2511945367,
0.19564417,
-0.3103450537,
-0.0566112697,
0.3130634427,
0.2139312625,
-0.1051529497,
-0.1277214736,
-0.0128356889,
-0.0179276764,
0.3755545616,
0.124396354,
0.2455987483,
-0.0501771048,
0.0954123512,
0.2405618876,
0.152650401,
0.1736777127,
0.252951026,
0.1828892529,
-0.015157789,
-0.4614329636,
-0.2064890563,
0.0549772605,
0.0874795467,
0.0284565017,
-0.3894967139,
0.3267395794,
-0.001632683,
0.163383007,
0.1570383757,
0.1106403023,
0.4825287163,
-0.0165398195,
0.0794721916,
-0.4653490186,
0.1203657165,
0.3244591951,
-0.3549874127,
0.2751416564,
0.4656182528,
-0.0184545945,
0.3258893192,
0.2565747201,
0.3138429523,
0.5164310336,
0.3635669947,
0.1383249313,
-0.1878062338,
-0.4322498441,
0.0178328361,
0.5528191924,
-0.3737697005,
-0.0973557085,
0.1278449446,
0.0378065854,
0.2741699219,
-0.1087611467,
-0.1625675559,
-0.2601648271,
-0.0821955055,
-0.1250253618,
0.0548075736,
-0.0606840104,
-0.2160615325,
-0.0553317964,
0.0889469981,
-0.1751570404,
0.1543829739,
-0.2759080231,
-0.1503823847,
-0.0826845765,
-0.0239454098,
0.3113578856,
0.1572371423,
-0.2301957011,
-0.1721219718,
0.331058532,
-0.1155634075,
0.4097522497,
0.4752765298,
0.6177266836,
0.591073513,
0.197744295,
0.141023621,
-0.0959319174,
0.0992012471,
-0.0008839667,
0.466473788,
0.0532715768,
-0.5392205715,
0.3291275203,
0.0190834366,
-0.0180720035,
0.4304615855,
-0.1307460815,
0.3070872128,
0.0956974626,
0.1636206508,
0.0522704385,
0.0454728603,
0.0786593482,
0.1035526395,
-0.2113203108,
0.0828140527,
0.9799096584,
-0.2187815756,
0.0082387887,
-0.3598150015,
0.0376449116,
0.1337390691,
0.7504998446,
0.5301433802,
0.0191350318,
-0.2689059079,
0.109871611,
-0.6785556078,
0.1615492553,
-0.0100112064,
0.1195642427,
0.2431478053,
-0.1740594506,
0.1362670362,
0.2347295433,
0.2257972956,
-0.3438854814,
0.0795258954,
-0.3624969423,
-0.2689085901,
0.0313593969,
0.4597538114,
0.2654235065,
-0.0453866422,
-0.1311107576,
0.2973898649,
0.0306785479,
-0.0711345151,
-0.1218308583,
0.1676906645,
0.15737921,
0.3829617202,
0.3445103765,
0.1535736173,
0.3686422408,
0.0324400812,
-0.1236779392,
-0.4942080975,
-0.1948025823,
-0.0650608018,
0.2983180285,
0.1826139092,
-0.0177901164,
-0.096800223,
0.0101351887,
-0.3805669844,
0.1060069203,
0.0610489622,
0.0909553915,
-0.0067865998,
0.1168817803,
-0.1358055919,
-0.0998944491,
0.0673669204,
0.1554570049,
0.5111299753,
0.1131735817,
0.0115883052,
-0.2730745971,
0.2586716413,
-0.0700674579,
-0.0500632897,
-0.1905662268,
0.1469298303,
0.226237148,
-0.0769284889,
-0.8204523921,
0.1293432117,
0.3578501642,
-0.1564284861,
-0.1171126291,
0.1346117407,
-0.0860339254,
0.0065339655,
0.0855474472,
0.4466445744,
-0.0737442523,
-0.1286571473,
-0.1838977486,
-0.517406106
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | Hi !
The cache at `~/.cache/huggingface/metrics` stores the users data for metrics computations (hence the arrow files).
However python modules (i.e. dataset scripts, metric scripts) are stored in `~/.cache/huggingface/modules/datasets_modules`.
In particular the metrics are cached in `~/.cache/huggingface/modules/datasets_modules/metrics/`
Feel free to take a look at your cache and let me know if you find any issue that would help explaining why you had an issue with `rouge` with no connection. I'm doing some tests on my side to try to reproduce the issue you have
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 84 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
Hi !
The cache at `~/.cache/huggingface/metrics` stores the users data for metrics computations (hence the arrow files).
However python modules (i.e. dataset scripts, metric scripts) are stored in `~/.cache/huggingface/modules/datasets_modules`.
In particular the metrics are cached in `~/.cache/huggingface/modules/datasets_modules/metrics/`
Feel free to take a look at your cache and let me know if you find any issue that would help explaining why you had an issue with `rouge` with no connection. I'm doing some tests on my side to try to reproduce the issue you have
| [
-0.0358739235,
-0.0009468794,
0.0741712973,
0.1184606254,
0.0837399736,
-0.1097530797,
0.1720350534,
0.2317183912,
0.263871938,
0.1391704977,
0.0678789169,
0.1733182073,
-0.2854743898,
-0.0121779889,
0.1369283795,
0.0551351048,
-0.0628577247,
-0.0301496275,
-0.3259329796,
-0.1252781451,
-0.2200646251,
0.5307021141,
-0.0628678054,
0.0645003542,
-0.456232816,
0.175228402,
-0.1284712106,
0.3620725572,
-0.0833549127,
-0.5840672851,
0.3293273449,
0.1146460027,
0.0539535135,
0.4142378569,
-0.000119533,
-0.1037964001,
0.2400082797,
-0.1888288558,
-0.2738632858,
-0.2778948545,
-0.0753612369,
-0.2522023916,
0.5798795223,
-0.0966042876,
-0.1344480813,
0.0916112661,
0.010013938,
-0.4268661737,
0.4767813981,
0.1211743802,
0.1166026369,
0.4053644538,
0.0331893787,
-0.3333913386,
0.2213086486,
-0.1046053171,
0.0415681899,
0.4594830275,
0.0687563419,
0.0936746299,
-0.1464715749,
0.2114187032,
0.347068429,
-0.1844199747,
0.3098644912,
0.0112121031,
0.0013161208,
-0.1081184,
-0.028411746,
0.3066345453,
0.2365588397,
-0.1305870712,
-0.2608056962,
0.1283789575,
-0.0879924148,
-0.4039273858,
0.1538639963,
-0.0939178243,
0.0763647407,
0.0891127139,
0.0873678848,
-0.198005408,
-0.0160592273,
0.1474249661,
-0.1284566671,
-0.1398057491,
-0.1876243651,
0.0414875969,
-0.0495054387,
0.0219503269,
-0.5056308508,
0.4271096587,
-0.1831524074,
0.1320413053,
-0.4020822942,
0.1370481253,
0.279212594,
0.1743565351,
0.0485690907,
0.5940818191,
0.054407049,
0.0976384133,
0.324097693,
0.1152184308,
-0.1439851522,
0.463716954,
0.1471066475,
-0.0318709761,
0.2290679514,
0.2400331199,
-0.0997859538,
-0.2906396687,
0.220740363,
-0.2854759991,
0.1802100986,
0.2259307653,
0.2352601886,
-0.3656278849,
0.0109471902,
0.2513978779,
-0.1731026769,
-0.0880762935,
0.0778518617,
0.3774323463,
0.0209158547,
-0.0662799552,
0.3336493969,
0.353848964,
-0.2321900427,
-0.04830008,
-0.4542720914,
-0.1579870731,
-0.1675679982,
0.3084727824,
0.1213164032,
-0.0429604426,
0.522587657,
-0.1377247572,
0.3675760031,
-0.0634157583,
0.2979547381,
0.1495725065,
-0.0578730702,
0.2460674047,
0.0045494735,
0.1730863154,
0.3100202382,
-0.0152879581,
-0.1046914831,
-0.3136596084,
-0.2143964767,
-0.4901868403,
0.0863237679,
0.0288708732,
-0.3260273337,
0.1841581464,
0.3849878609,
0.0545697659,
-0.2149498612,
-0.258146733,
0.120428659,
0.0444045961,
-0.0732085258,
-0.1320747435,
0.5114515424,
0.7114236951,
-0.25737378,
-0.4081453681,
0.2153406441,
0.0144802928,
-0.1083215699,
0.1975733787,
-0.0630954057,
0.1563923657,
-0.3403096795,
-0.1043980792,
0.4113847017,
-0.517152071,
-0.4667462707,
-0.0168587267,
-0.3854105473,
-0.1010326892,
0.1008024216,
-0.0357387103,
0.0745938867,
0.2175189704,
0.1196306571,
-0.172498405,
0.241447255,
-0.3219717145,
-0.260199964,
-0.2055729032,
0.1193031222,
-0.0603428259,
0.2362509966,
0.1523995697,
-0.0052174553,
-0.0732076988,
0.1255850345,
0.1020375341,
0.1981782168,
0.3781460524,
0.2921809256,
0.1836387217,
0.1978507191,
0.1566253603,
-0.1678200513,
0.2583166361,
-0.4689188004,
0.1527782083,
-0.0503248572,
-0.0147543475,
-0.2192663103,
-0.2279685587,
-0.0850849003,
-0.4767321944,
0.0177706853,
0.0121324901,
0.2931830883,
0.3102689385,
-0.1623550504,
0.2714128494,
0.0924601108,
0.2920161486,
-0.6656669974,
0.0322582498,
-0.111372143,
-0.3356735706,
-0.0986839682,
0.1647352725,
0.0611554533,
-0.008172147,
-0.1790616065,
0.3883352876,
-0.0169808343,
0.1749451756,
0.3794829845,
0.0891619027,
-0.0574566275,
-0.4079514146,
-0.0328811035,
0.1797970682,
0.0542467833,
-0.0016230643,
-0.200784862,
0.3237921,
-0.0825716257,
0.1331497282,
-0.2450237572,
0.3666040599,
-0.0399599224,
-0.0688232929,
-0.5726246238,
-0.0933299363,
0.3485018313,
-0.1906703115,
0.4904287457,
-0.1709935665,
-0.032026507,
0.0218924601,
0.0991402641,
0.1401266009,
0.2501838207,
-0.0037822211,
-0.2171488553,
-0.0424433015,
0.0269457493,
-0.14596048,
0.4352151155,
0.1362626255,
0.1439883113,
0.1243448555,
0.2096872181,
-0.1694744825,
0.0424717292,
-0.0767042935,
-0.1214770228,
0.0951080322,
0.1849319786,
0.1680720598,
-0.2335238159,
-0.1956406534,
-0.3971342742,
0.0322262421,
-0.3484332561,
0.0946539417,
-0.2037184834,
0.3463998437,
0.3030620813,
-0.0559966564,
-0.3411490321,
-0.2638902366,
0.1385771632,
-0.1877672374,
0.0301666558,
0.2229875326,
-0.2555916905,
0.4870505929,
0.1352508664,
-0.1863757074,
-0.3071426153,
-0.200461328,
-0.1395705491,
-0.1502980739,
-0.287016809,
0.08714737,
0.0249529202,
-0.213568151,
0.1292312145,
-0.0343739726,
-0.3483101428,
0.0483235344,
0.051832296,
0.5172299147,
0.3936596811,
0.0021315254,
-0.2833368182,
0.2200521678,
0.5371549129,
-0.3732430339,
-0.309099257,
0.0990037471,
0.0583363958,
0.2021278739,
-0.2552039325,
-0.3111088872,
-0.0462788343,
-0.3856251836,
0.6103233695,
0.1012191921,
-0.1091395915,
-0.0645338595,
0.0123222992,
0.1742912978,
-0.2101204097,
0.362780869,
-0.3324127793,
-0.6493900418,
0.3374321759,
-0.1905546188,
-0.3684261143,
0.0284810513,
0.033761628,
0.3815253079,
-0.0200923029,
-0.5352693796,
-0.3893999755,
0.0445667505,
0.2492183447,
-0.1060669422,
0.062319383,
0.5808557272,
-0.0551421605,
-0.0207335316,
-0.0587934367,
-0.1301970929,
0.434307009,
-0.1079052687,
0.1379552931,
-0.1012511626,
0.4365172386,
0.0528160408,
0.7605788708,
0.4823166132,
0.2873233855,
0.2661587298,
0.0358388647,
0.4865825176,
-0.2143176049,
-0.1602762938,
0.1947723329,
0.1949832141,
-0.0684700087,
0.1258613169,
0.1112140566,
0.2847435474,
-0.2191254497,
-0.0268843397,
-0.3354533911,
-0.3300557733,
-0.0206895154,
-0.2873110175,
0.2177420408,
-0.0690013915,
0.0413903072,
-0.2042778879,
0.0539590791,
0.3671606481,
0.3646478355,
0.275891006,
0.2595493793,
-0.277564615,
0.0905445144,
-0.586440444,
0.3170909584,
-0.518406868,
0.1670908034,
-0.2591428757,
-0.2926998436,
0.0993678421,
-0.0163175352,
0.5642930269,
0.0010581426,
-0.1702981144,
0.0686847419,
-0.2607214153,
-0.5911592245,
-0.0195032284,
-0.0470785312,
-0.0120293424,
0.2525355816,
0.2193756253,
-0.3621322513,
-0.1756043732,
-0.1045861542,
-0.1455087662,
-0.216161266,
-0.1501788497,
-0.3854457736,
0.0320449099,
-0.1962868869,
0.103323251,
-0.29601264,
0.0171899162,
0.096318081,
0.0973909497,
-0.2283384502,
-0.0517566651,
-0.0071312599,
0.0747467652,
0.3057206273,
-0.3038360476,
0.1706122905,
0.4674376249,
0.0272764713,
-0.0002299361,
0.5265594721,
0.0551013425,
-0.2797265947,
0.2987191081,
-0.0631923079,
0.2123972178,
0.5377953053,
-0.2559631169,
0.1114688069,
-0.0735080838,
0.3654129803,
-0.1908652633,
-0.1126984358,
0.3272384703,
-0.177387327,
0.3741318285,
0.1637037098,
0.0672193468,
0.1586406231,
-0.1380213797,
0.1574545205,
0.0404477678,
-0.1430931985,
0.3999297321,
-0.1042417362,
0.9887385368,
0.2283456326,
0.2489139438,
0.1314040124,
0.0513560623,
0.454508841,
0.0168129653,
0.0221086442,
-0.2404449582,
-0.2034008503,
-0.0113503896,
-0.118537195,
0.0253494419,
-0.2338594496,
-0.2589438856,
0.4140236378,
-0.3903061748,
-0.0517868847,
-0.2063389421,
-0.1757089049,
-0.2596440315,
-0.1164496467,
0.0121733062,
0.086863108,
0.0721113011,
0.6109398007,
-0.2037318945,
-0.1589172184,
-0.4563351572,
-0.3336653113,
-0.0674572885,
0.0130487755,
-0.3301524818,
0.08816313,
0.2974458933,
-0.1824517548,
-0.1842074245,
0.3558036685,
0.3632244766,
0.1726474464,
-0.2810918391,
0.0607843511,
0.0395956226,
0.0463072062,
0.1197914034,
0.0193877481,
0.2255721539,
0.0099451914,
-0.1676052809,
0.1641508937,
-0.2100358307,
0.0512366593,
0.0139882863,
-0.0534162,
-0.1955826133,
-0.113757506,
-0.0202233456,
0.1057807654,
0.0967818126,
-0.1549053192,
0.0860570595,
0.3396830261,
-0.1765404493,
-0.0151248053,
0.1682720929,
-0.2041359097,
-0.0545960665,
0.5798852444,
-0.3019781411,
-0.1088525578,
0.3987400532,
0.033434093,
-0.0022550002,
-0.1496373415,
0.1192859337,
0.4620959461,
-0.7651283145,
0.0641345084,
0.0818872601,
0.0467191041,
0.0998906642,
0.1696166843,
0.3279042244,
-0.17383717,
0.0599948913,
-0.2481648177,
-0.4355518818,
0.3564048409,
-0.1347399503,
0.1367651224,
-0.3409818113,
-0.0871528536,
-0.0929878578,
0.2171244472,
-0.2840719819,
0.1434050798,
-0.2832881212,
-0.1181832552,
-0.0723365098,
-0.1273000985,
0.3977067471,
-0.0696902573,
-0.0448787287,
0.0667283237,
-0.4174092114,
-0.1143616587,
-0.1739067435,
0.128813982,
-0.1347729117,
0.1333720088,
-0.0616761483,
0.2145869583,
0.0732681975,
-0.1694598943,
0.1197838113,
0.2172563672,
0.0872647688,
0.0729129463,
-0.0300871991,
0.0965834558,
-0.122650221,
0.2455373108,
0.1651060879,
-0.1488167197,
-0.0761011317,
0.0207684971,
-0.3127220869,
-0.0915167034,
0.1688803434,
0.1058797538,
-0.1790581048,
0.1402076632,
0.1360845566,
-0.0907965004,
-0.3047741652,
-0.218834877,
0.4004143775,
0.1239716783,
-0.1787183732,
0.069637157,
-0.0565711856,
0.1327506602,
-0.0852432996,
0.3196440637,
0.0953553021,
0.3395170569,
0.135705173,
0.2389124483,
0.1446430385,
-0.0525930226,
0.0786915049,
0.1679390669,
-0.064635098,
-0.1023806483,
0.3450415432,
-0.1604189575,
0.0050430298,
0.2077206075,
0.5121743083,
0.1764219403,
0.0708958805,
0.043559745,
-0.0611323565,
0.226265654,
-0.1423234791,
0.2956943512,
-0.1136340797,
-0.2759337127,
-0.1424783021,
0.0311711133,
-0.0090723634,
-0.2327331752,
-0.4316948652,
0.702419281,
-0.2806921303,
-0.3212974668,
-0.166773349,
0.1408961415,
-0.1320924908,
-0.2029650956,
-0.1144986972,
-0.1392984986,
-0.1908463389,
-0.1848253608,
0.2290683091,
-0.0132929087,
0.0634706914,
0.143985033,
0.0835379437,
-0.2804710865,
-0.4855870903,
0.1773282886,
0.1425660104,
-0.0415458791,
0.0629002899,
0.1436044425,
-0.1707990468,
0.1758074611,
0.6104491949,
0.2904425263,
0.0435970537,
-0.1393881589,
0.2393371761,
0.2593468726,
-0.049691584,
-0.1322290152,
0.1564970315,
0.2695863843,
-0.4744091928,
0.0466505438,
0.0854167491,
-0.1447212994,
0.3480510414,
-0.3163054287,
0.359174937,
-0.1538484693,
0.3500992656,
-0.2364439219,
0.1475691795,
-0.1308515072,
-0.1009435505,
-0.7179358602,
0.1895138919,
-0.1232710034,
0.0928877443,
0.0719829798,
-0.3484206796,
0.0471336655,
-0.1072723195,
0.2144947797,
0.4300643206,
0.1629474759,
-0.2852114439,
-0.3108716607,
-0.3476510644,
0.246307224,
0.0730974823,
0.2230472267,
-0.1770625412,
0.444093883,
-0.103121601,
0.0922275931,
0.2410820723,
0.1612481922,
-0.0969934911,
0.0524557754,
-0.4813612103,
0.1660901308,
-0.1789523214,
-0.0562411994,
0.0814189315,
0.253574878,
-0.1228969693,
-0.1291749477,
0.0204993486,
0.1910955161,
-0.2418767214,
-0.2137032449,
0.1613460332,
0.5954047441,
0.172268182,
0.1285201609,
0.0685033947,
-0.1416751146,
-0.3088626862,
-0.2423847914,
0.006213598,
0.0871113837,
-0.1713268757,
0.34399122,
-0.2581758201,
-0.42274791,
-0.3795549273,
0.3892646432,
0.1324906647,
-0.1534837186,
-0.1188777387,
0.311021328,
-0.0710716471,
0.0651841462,
0.181398049,
0.2737932205,
-0.0974208117,
0.3329164982,
-0.3157680631,
-0.1971640587,
0.6771652102,
-0.3671153486,
0.28056252,
-0.2694847882,
0.5114609599,
0.3368268013,
-0.3409713507,
-0.3263531327,
0.1121219099,
0.0807694942,
0.0419809632,
-0.1084787101,
0.1534228474,
-0.3455207944,
0.03237205,
-0.1215448231,
0.3072325885,
0.1546482295,
0.1746728718,
0.0677530617,
-0.1550116688
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | Thank you for clarifying that the metrics files are to be found elsewhere, @lhoestq
> The cache at ~/.cache/huggingface/metrics stores the users data for metrics computations (hence the arrow files).
could it be renamed to reflect that? otherwise it misleadingly suggests that it's the metrics. Perhaps `~/.cache/huggingface/metrics-user-data`?
And there are so many `.lock` files w/o corresponding files under `~/.cache/huggingface/metrics/`. Why are they there?
for example after I wipe out the dir completely and do one training I end up with:
```
~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock
```
what is that lock file locking when nothing is running? | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 93 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
Thank you for clarifying that the metrics files are to be found elsewhere, @lhoestq
> The cache at ~/.cache/huggingface/metrics stores the users data for metrics computations (hence the arrow files).
could it be renamed to reflect that? otherwise it misleadingly suggests that it's the metrics. Perhaps `~/.cache/huggingface/metrics-user-data`?
And there are so many `.lock` files w/o corresponding files under `~/.cache/huggingface/metrics/`. Why are they there?
for example after I wipe out the dir completely and do one training I end up with:
```
~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock
```
what is that lock file locking when nothing is running? | [
0.2474487722,
0.0509264618,
0.0548085347,
0.2140229642,
0.1152904108,
0.1876896024,
0.2276508212,
0.3381587565,
0.2461998314,
0.1004050225,
0.1633259952,
0.1200300902,
-0.3513065279,
-0.1723563373,
0.0993683785,
-0.0273564551,
0.108128503,
-0.0378125533,
-0.2317160964,
-0.1649742872,
-0.2525991797,
0.5168861151,
-0.080842711,
0.0938808918,
-0.564119339,
0.2315035015,
-0.1601385474,
0.2439410985,
-0.043449048,
-0.5871257782,
0.3680898845,
0.1198715121,
-0.063629128,
0.5067223907,
-0.0001260311,
-0.1185866147,
0.2515876889,
-0.1634840369,
-0.2404907346,
-0.1307276189,
-0.0129824877,
-0.27894485,
0.4915620089,
-0.2036509067,
0.1352379918,
-0.0842312425,
0.1214472055,
-0.425988853,
0.3512769639,
0.0431349725,
0.0366243497,
0.2477322221,
-0.1989924163,
-0.2882103324,
0.152717039,
-0.0924046263,
0.0252721608,
0.3664670885,
-0.083611086,
-0.0094819628,
-0.2560973763,
0.3051267862,
0.2019369006,
-0.137405023,
0.3521671295,
0.0150464587,
-0.0426548086,
-0.2301338911,
0.0013804976,
0.2264509946,
0.237470299,
-0.1862582862,
-0.2228962481,
-0.0533667579,
-0.0899376571,
-0.2590978444,
0.2228358388,
-0.0521905497,
0.0696588606,
0.14248088,
0.0115557201,
-0.1976008862,
0.0148121417,
-0.0037732311,
-0.1053488404,
-0.005746197,
-0.1575381756,
-0.0152486302,
-0.051495675,
0.198974669,
-0.3450109959,
0.3006660044,
-0.0482683405,
0.2071721852,
-0.2239118367,
0.0468743891,
0.1979002506,
0.1199530661,
0.0460010394,
0.4275060296,
0.0228464007,
0.0263505019,
0.4165911973,
0.090977475,
-0.1048303246,
0.469540745,
0.3464438319,
-0.0673513263,
0.0851757675,
0.1602272093,
-0.1593470275,
-0.25618729,
0.1269488782,
-0.223104924,
0.2775916755,
0.0905743688,
0.0148454979,
-0.4099763334,
0.0961337015,
0.3147746921,
-0.2490137368,
-0.2345290184,
0.0893958434,
0.3153008223,
0.251827091,
-0.0041997964,
0.1808400154,
0.2091758847,
-0.242661804,
-0.0500774905,
-0.3978197873,
-0.1054063588,
-0.2283218205,
0.3502308726,
0.0507635623,
-0.1394720674,
0.7004554272,
-0.2605068982,
0.3893996775,
-0.1397275329,
0.2826372087,
0.2533380389,
0.0007687882,
0.4032485187,
-0.0881250426,
0.2115939558,
0.3516607583,
0.0514884293,
-0.1892667115,
-0.2525211871,
-0.2373768538,
-0.5181611776,
0.0860113055,
-0.0284100547,
-0.5039561391,
0.2390990555,
0.4408905804,
0.1475693285,
-0.0292735584,
-0.1374205202,
0.2848836482,
0.0987271369,
0.017406106,
-0.124729611,
0.3923792839,
0.7586102486,
-0.2284234911,
-0.475679934,
0.1044251323,
-0.011790744,
-0.0543392152,
0.1477326304,
-0.0291077998,
0.232129544,
-0.2840010524,
0.1421152055,
0.3668578267,
-0.3609600365,
-0.3620802164,
-0.0106191635,
-0.3919944167,
-0.1379082203,
0.1821038723,
0.0160971247,
0.0728415847,
0.1438128799,
0.1258187443,
-0.385025233,
0.1877110302,
-0.4215299785,
-0.3467741013,
-0.1896640211,
-0.0304800533,
-0.1046682894,
0.2787398398,
0.0370404907,
0.0111545138,
0.1009100676,
0.1036302298,
0.1744167209,
0.174862951,
0.2626432776,
0.2613893449,
0.1620635092,
0.2222010791,
0.1930807978,
-0.3110489547,
0.2377900034,
-0.3628159165,
-0.0176313054,
0.1102191433,
-0.0732133761,
-0.1398178041,
-0.2352029383,
-0.1244361103,
-0.527023077,
-0.0462952554,
-0.0360659137,
0.1183855236,
0.1227987632,
-0.2726368308,
0.1473128796,
-0.0587884709,
0.327455014,
-0.5532151461,
0.0320396125,
0.0278662983,
-0.1970102191,
-0.2257992476,
0.1093444228,
-0.0025993753,
0.023267895,
-0.0975063518,
0.3807028532,
-0.0499368161,
0.183205992,
0.558686018,
0.3541171551,
-0.1420739293,
-0.3481736183,
0.0158951264,
0.0016873293,
-0.1003443152,
-0.001289539,
-0.2387942672,
0.3765331805,
0.1348476261,
0.1348783374,
-0.3614092469,
0.2761768103,
-0.2093582451,
-0.1233223602,
-0.5444700718,
-0.1353485882,
0.4068960547,
-0.2476062626,
0.3068440557,
-0.0123208612,
0.0024222285,
0.0304552428,
0.1419314146,
0.0483443066,
0.0429756418,
0.0081627099,
-0.2377897799,
-0.0533848852,
-0.0278108642,
0.0037661791,
0.4933541119,
0.1662805378,
0.2230533957,
0.0144515131,
0.1677997112,
-0.2444669902,
0.0901627541,
0.1739690155,
-0.2210896611,
0.1015079916,
0.1378708333,
0.208347559,
-0.2946523428,
-0.0503392816,
-0.2348015606,
0.0318769738,
-0.3016397357,
0.1512338519,
-0.1539792418,
0.3830916882,
0.0986965001,
-0.0210698768,
-0.2364770919,
-0.3332991302,
0.2775134742,
-0.3183352947,
-0.027029071,
0.2005966008,
-0.3842666745,
0.638785243,
0.0182664916,
0.0167174023,
-0.2889689803,
-0.0855252966,
0.0128286406,
-0.2142504901,
-0.1412099302,
0.0268612467,
0.0817522332,
-0.266895473,
0.0615633875,
-0.1273284107,
-0.4955728352,
0.0512228049,
-0.0161660276,
0.6132104993,
0.2828388214,
0.0391496457,
-0.3447295427,
0.0620121174,
0.2875091434,
-0.2682444453,
-0.3798854947,
0.0450981483,
0.0512688197,
0.2367152274,
-0.3826582432,
-0.3334690034,
0.1506505758,
-0.3048090637,
0.5795778632,
-0.0784141868,
-0.0105536189,
-0.1236328334,
-0.1348556429,
0.2229510397,
-0.3493733406,
0.2714139819,
-0.323159337,
-0.7829338312,
0.4090439975,
-0.0861364901,
-0.0492392071,
0.0389751047,
0.0890195221,
0.3141539395,
-0.1025538072,
-0.7185193896,
-0.4109574854,
0.0587697253,
0.1837126017,
-0.0423481427,
0.0645666048,
0.4620448351,
-0.0869351476,
-0.0113906972,
-0.1259305775,
-0.1665930748,
0.4881488085,
-0.1238276809,
0.1869422197,
-0.1846834868,
0.3211716115,
0.1527524143,
1.0194029808,
0.5859875679,
0.3531094491,
0.1362785995,
0.2488947958,
0.4330224693,
-0.1152089536,
-0.0847560465,
0.2251206487,
0.2476466894,
-0.1349826604,
0.306597501,
0.2231491655,
0.1870114207,
-0.0333176218,
0.0093368217,
-0.317751646,
-0.2941358685,
-0.0049864445,
-0.1403056234,
0.227934286,
-0.0769745335,
0.1744735986,
-0.1848293543,
0.0769240707,
0.393335253,
0.5813072324,
0.3059708774,
0.261300534,
-0.1821737587,
-0.130993247,
-0.5380687714,
0.1992969364,
-0.4458009899,
0.0942804366,
-0.2295224965,
-0.3128385842,
0.2112234831,
0.1381263733,
0.5694954991,
0.1833451539,
-0.2201474905,
0.0576128513,
-0.1568572074,
-0.2315966636,
-0.1577762067,
-0.0046726316,
0.0672965869,
0.279584527,
0.3941031992,
-0.3906000853,
-0.3081783652,
-0.124281615,
-0.105490014,
-0.3238486648,
-0.2797124982,
-0.2802337408,
0.0235746577,
-0.0901428163,
0.0684604496,
-0.2081898749,
-0.0828766748,
0.017026484,
0.1113159209,
-0.037993148,
-0.0876009762,
0.1582578421,
0.0745304525,
0.2948907614,
-0.233771354,
0.1438097358,
0.562640667,
-0.0551483594,
0.0503552556,
0.4550113082,
-0.0244705863,
-0.0061207376,
0.3603447974,
-0.1528990418,
0.3670762777,
0.6862483025,
-0.197018519,
0.1456227005,
0.0650579929,
0.3557294607,
-0.0740638152,
-0.0981432498,
0.3586219549,
-0.1224846095,
0.3750063181,
0.1454332173,
0.2473687232,
0.2430823594,
-0.1910925657,
0.3061974347,
0.1086228043,
-0.1157654673,
0.335449636,
0.0150604043,
1.0408430099,
0.244695738,
0.2499575615,
0.0714624226,
0.1312486082,
0.3983025551,
0.0646646544,
0.0467553921,
-0.1198778376,
-0.1944515556,
-0.0132673681,
-0.1940737069,
-0.0282721519,
-0.2923018932,
-0.1856508851,
0.4099214673,
-0.4929341674,
-0.1550749838,
-0.119569093,
-0.302126348,
-0.2614346743,
-0.0750592127,
0.1547850668,
-0.0412603617,
0.1292812675,
0.5553105474,
-0.2514210343,
-0.223262623,
-0.3335103393,
-0.2589262724,
-0.1489760429,
-0.2111894488,
-0.189193517,
-0.0236804895,
0.3486911952,
-0.1595157087,
-0.1513005644,
0.2677080631,
0.3971248567,
0.0755349845,
-0.3020040393,
0.0638313293,
0.168032974,
0.0085617825,
0.0653499514,
0.0340458862,
0.350212127,
0.0003313422,
0.0511180013,
0.0249728393,
-0.3353208303,
0.0492205918,
-0.0794572979,
-0.0835642666,
-0.0526296571,
-0.0755615979,
0.0267639793,
0.204214707,
-0.0968359113,
-0.1874866188,
0.0331466421,
0.3781019449,
-0.1960802078,
-0.0299337506,
0.1211581901,
-0.3434657454,
-0.0560232066,
0.4342229664,
-0.1925756335,
-0.1195447221,
0.5066857338,
-0.0645438433,
-0.1223666668,
-0.0741554648,
0.1144796312,
0.4430146813,
-0.7368452549,
0.0767197311,
0.0242279135,
0.0279223174,
0.0765299052,
-0.0071539115,
0.5015129447,
-0.1989104003,
-0.2411553413,
-0.2157033086,
-0.3882723749,
0.3828385174,
-0.3040483892,
0.1324844211,
-0.6194871068,
0.0594518259,
-0.0539757833,
0.129678756,
-0.2124767005,
0.2109171599,
-0.2437586188,
-0.1170602217,
-0.0746396184,
0.034661781,
0.2913800478,
-0.0565143861,
-0.1333194524,
0.0286501795,
-0.3847768009,
-0.0517194159,
-0.1915149689,
0.1899383962,
-0.1081531942,
0.2442275435,
0.0066734254,
0.3041709065,
0.0198465213,
-0.0583772138,
0.0101102889,
0.2571416497,
0.127961114,
0.1851984859,
0.1432302892,
0.2480724603,
-0.0729077309,
0.2049414963,
0.1763912141,
-0.1000487804,
-0.1000213549,
-0.1039726287,
-0.428586036,
-0.0848951638,
0.1566679478,
0.1853017956,
-0.3170948923,
0.0999711528,
0.06185399,
0.0173471291,
-0.2353606522,
0.0403747782,
0.211148262,
0.2245328426,
-0.2050519288,
0.064470388,
-0.0084530301,
0.0521704219,
0.0229843631,
0.2303787321,
-0.0552917495,
0.4240313768,
-0.0268419534,
0.2237039208,
0.1670222878,
-0.1279318631,
0.142768532,
0.1989019662,
-0.0372640267,
-0.0433573127,
0.3380185664,
-0.0348778702,
0.1034378856,
0.3432772458,
0.5569770336,
0.1540990472,
0.0934892893,
-0.0954650491,
0.0071916264,
0.2507782578,
-0.0900278687,
0.2582258582,
-0.2793654203,
-0.1373146623,
-0.2210544944,
0.1737849414,
0.0515689105,
-0.1402485669,
-0.3882139623,
0.8121159673,
-0.1609301567,
-0.2608855963,
-0.0700841993,
0.1249642372,
-0.1619193852,
-0.1566227823,
-0.1197467148,
-0.1842712462,
-0.1253319979,
-0.1607531309,
0.2283629924,
-0.0062959623,
0.1427664757,
0.0652463138,
-0.0398541018,
-0.107496269,
-0.4368211627,
0.2725351453,
0.2284077555,
-0.2035957575,
0.0297513027,
0.1274764836,
-0.252805084,
0.2495366037,
0.5566772223,
0.2099296451,
-0.0059624454,
-0.1163635775,
-0.003518261,
0.2883097231,
-0.0202122927,
-0.0466434434,
0.2402108461,
0.183072865,
-0.6153120995,
-0.0621648729,
0.0086090378,
-0.1076653749,
0.388800621,
-0.2084250748,
0.3229672909,
-0.1921547949,
0.353045404,
-0.2691181898,
0.0926731601,
0.0585446134,
-0.0918288901,
-0.746835053,
0.2137385905,
-0.1359364986,
0.0162212178,
0.033040002,
-0.2474635541,
0.0152216442,
-0.095531337,
0.1805479825,
0.5530500412,
0.0532688946,
-0.3335349262,
-0.3067727685,
-0.4826695919,
0.2709025443,
0.1446656883,
0.4335972369,
-0.022839034,
0.3380472362,
-0.0630272478,
-0.0161664747,
0.2461426556,
-0.0855208486,
-0.1611038148,
0.0040016142,
-0.5652292371,
0.1719987839,
0.0195853692,
-0.0025381707,
0.1001021713,
0.2926968038,
0.0207594968,
-0.2334725559,
-0.091447264,
0.2584429383,
-0.2112587392,
-0.05533785,
0.3531960547,
0.4249579608,
0.0721090063,
0.0359956846,
0.030497171,
-0.1994165778,
-0.2580091655,
-0.1420318186,
-0.0411227047,
-0.1456903219,
-0.2933095098,
0.341289252,
-0.4688858986,
-0.4392545223,
-0.273902297,
0.2539844215,
0.0947208554,
-0.1432618797,
-0.3065037727,
0.3340903223,
-0.2120443881,
0.0159966126,
0.3520704508,
0.3827581406,
0.0340379477,
0.2756769657,
-0.3647691607,
-0.1141461283,
0.7994844913,
-0.5541844368,
0.2426103354,
-0.1927601099,
0.3955243826,
0.5046326518,
-0.3508094549,
-0.4224117994,
-0.0967805684,
0.13825275,
0.0550413132,
-0.2188417166,
0.144193694,
-0.4663904607,
-0.0743136182,
-0.0288973413,
0.2651972473,
0.1610631794,
0.0547092482,
0.037573278,
-0.1937295794
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | The lock files come from an issue with filelock (see comment in the code [here](https://github.com/benediktschmitt/py-filelock/blob/master/filelock.py#L394-L398)). Basically on unix there're always .lock files left behind. I haven't dove into this issue | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 30 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
The lock files come from an issue with filelock (see comment in the code [here](https://github.com/benediktschmitt/py-filelock/blob/master/filelock.py#L394-L398)). Basically on unix there're always .lock files left behind. I haven't dove into this issue | [
0.097865358,
0.0017486513,
0.107012555,
0.1418970376,
-0.0015063416,
0.0359287113,
0.2448603958,
0.2713266909,
0.2091094255,
0.1213857457,
0.1688693911,
0.1361202598,
-0.3981264532,
-0.0413277932,
0.0657574683,
0.0775554106,
0.0212285966,
-0.0395919643,
-0.3166528046,
-0.1807786822,
-0.2565912008,
0.5491794348,
-0.1192009524,
0.0093764514,
-0.5109115839,
0.2307157665,
-0.230617255,
0.2772119343,
-0.0274776872,
-0.5815621018,
0.3894341588,
0.1215367466,
0.0478552803,
0.4788637161,
-0.0001274045,
-0.1663456261,
0.1572808325,
-0.1869962215,
-0.1530368775,
-0.1362831891,
0.0305327252,
-0.296792984,
0.5233107805,
-0.2339408994,
0.0798513442,
-0.0288463179,
0.1023539603,
-0.2997954488,
0.4383220673,
0.0624675006,
0.0467605442,
0.3075243831,
-0.0342120007,
-0.2570322752,
0.0702651888,
-0.0924271494,
0.001526989,
0.3444682956,
-0.2050891668,
0.0921002999,
-0.2054678798,
0.2364490479,
0.2473953515,
-0.1546232402,
0.2657470703,
0.0201909319,
-0.0640985295,
-0.1387816668,
-0.0949206948,
0.3408108652,
0.1928484142,
-0.152159512,
-0.2342792153,
0.0592424124,
-0.1122278348,
-0.3646725118,
0.2042398453,
-0.0701379478,
0.1202763319,
0.1426340789,
0.0680515692,
-0.182393685,
-0.0471961647,
0.0707626045,
-0.1521101594,
0.0407558568,
-0.140402779,
0.0206550024,
-0.040888533,
0.1827443242,
-0.4113307893,
0.354447633,
0.0115477294,
0.1267993003,
-0.3444468975,
0.1168115437,
0.2434753627,
0.0282619204,
0.0463364795,
0.4056188762,
-0.0580374859,
0.0199389681,
0.3959122598,
0.1504532993,
-0.0784414187,
0.4660253227,
0.2361450791,
0.0168703273,
0.1764599383,
0.1849363148,
-0.2115628123,
-0.2638478279,
0.1365807205,
-0.2362145931,
0.4281726778,
0.1122877523,
0.1105808914,
-0.4319956601,
0.1628947556,
0.2799923122,
-0.1437161267,
-0.187398538,
0.096094653,
0.3751058578,
0.1744755805,
0.0033344682,
0.2302560508,
0.3915506899,
-0.2153752446,
0.0076532625,
-0.3904958963,
-0.2105623484,
-0.0540186502,
0.3026130795,
0.1356307864,
-0.1125971228,
0.6726793647,
-0.2044870555,
0.4261628091,
-0.0901328027,
0.328532964,
0.2119804323,
0.0645421296,
0.3416905105,
-0.0264650844,
0.2123864293,
0.2896439731,
0.1480874717,
-0.2393878251,
-0.2758550644,
-0.1812233478,
-0.5380879045,
0.079878822,
-0.0424433388,
-0.465970248,
0.1792516559,
0.4528219104,
0.1464155763,
-0.101521641,
-0.303516537,
0.2009050995,
-0.0098100342,
-0.0084777176,
-0.1502698064,
0.5102134943,
0.6958428621,
-0.3465328217,
-0.3804929852,
0.0367819034,
0.0076955874,
-0.0054188147,
0.2611055374,
0.0363227502,
0.1693777144,
-0.3158251047,
0.1305027604,
0.3877373636,
-0.3373550475,
-0.4568698108,
-0.0679742396,
-0.368324995,
-0.1110465899,
0.1929323226,
0.0133989565,
0.1307078004,
0.1547205448,
0.0300891642,
-0.2190199345,
0.2606110275,
-0.3739933372,
-0.3569112122,
-0.2306004316,
0.0067295656,
-0.0920922607,
0.3358701169,
0.0176149216,
0.0747951791,
0.0229974985,
0.0826776028,
0.0752031431,
0.162635982,
0.2784366012,
0.337235868,
0.100065507,
0.2661373913,
0.1485668421,
-0.2092299759,
0.2349851727,
-0.4146920145,
0.0708451122,
0.1174753159,
0.0061449856,
-0.1154768765,
-0.2034729123,
-0.0905112773,
-0.4177540243,
-0.0547221527,
-0.0495193079,
0.210953474,
0.2493547797,
-0.176796034,
0.0389826,
0.0869970024,
0.2594944537,
-0.4751551151,
0.0103413537,
-0.0947172344,
-0.3357909322,
-0.0653164834,
0.1485401839,
0.0108669326,
0.0354320183,
-0.1438514739,
0.3710699081,
-0.1724668592,
0.2730034888,
0.4506232738,
0.2415378094,
-0.1196843386,
-0.3655081391,
0.0325866044,
0.1424409896,
-0.108857289,
0.0363418013,
-0.2451750934,
0.3830002844,
0.2120264769,
0.0713481456,
-0.3359490633,
0.3146031499,
-0.0740360096,
-0.0987765267,
-0.4815731049,
-0.1280822009,
0.3917379975,
-0.2381813824,
0.2410223931,
-0.1055558622,
0.0105511919,
0.00256183,
0.2497155517,
0.0799554139,
0.2332527488,
0.0780289918,
-0.2227703333,
-0.0949149132,
0.0184848886,
-0.1300747097,
0.4450002909,
0.1906785667,
0.2278553993,
0.0538471043,
0.105255045,
-0.1888127327,
0.1054110527,
-0.0117076412,
-0.160594523,
0.1285530627,
0.2171351612,
0.2108034045,
-0.2892761528,
-0.1394125819,
-0.4038011432,
-0.0391596481,
-0.3721092939,
0.0799338594,
-0.2586630583,
0.4104161859,
0.1836583465,
0.1734470725,
-0.2760924101,
-0.3279845715,
0.2431276739,
-0.2901172936,
-0.0558714494,
0.2372848094,
-0.3406095803,
0.5479474068,
0.1450074911,
-0.2062621564,
-0.2959773242,
-0.1340128332,
-0.0942154974,
-0.2200795412,
-0.2425171435,
0.0989218503,
0.1037224084,
-0.1725021303,
0.0372335166,
-0.0137784313,
-0.4823436141,
0.0696526021,
0.0511413589,
0.6196221709,
0.3521494567,
0.0815300494,
-0.3057149649,
0.0834967047,
0.4136028886,
-0.4095363617,
-0.2688109279,
0.1824415624,
0.0434780754,
0.2113899291,
-0.26208359,
-0.361744523,
0.0265497975,
-0.2976145148,
0.6853889227,
0.0961090624,
-0.034079846,
-0.1620344222,
-0.0633250922,
0.1737493724,
-0.3233644366,
0.3805913031,
-0.2930787206,
-0.6595553756,
0.5047425032,
-0.1025176644,
-0.2188172042,
-0.003696166,
0.058648169,
0.3344171643,
-0.0894468278,
-0.6739206314,
-0.4296786189,
0.0067403875,
0.2661027908,
-0.0254285652,
0.1197580174,
0.5252694488,
-0.0839201659,
0.0137430578,
-0.0674874857,
-0.2023649961,
0.5829994082,
-0.0621243045,
0.1375797987,
-0.0825310647,
0.3729032874,
0.1241424754,
0.9117971063,
0.4719630182,
0.3536292613,
0.2457000911,
0.0920004472,
0.3813194335,
-0.1479357779,
-0.1129257232,
0.3361158371,
0.2327179909,
-0.1589667946,
0.2437699139,
0.1992545724,
0.1921726018,
-0.0608861968,
-0.0730623975,
-0.3863716125,
-0.3076614141,
-0.0325330459,
-0.0471555814,
0.2321034074,
-0.0069960579,
0.0424786583,
-0.2280811518,
0.0112631656,
0.4220137298,
0.4312683642,
0.2322887033,
0.2378756851,
-0.1441173255,
0.058937002,
-0.5730270147,
0.2734694183,
-0.5298243165,
0.0832479745,
-0.2206547409,
-0.3305587769,
0.1510059536,
0.0413566791,
0.5898938179,
0.1778954566,
-0.1399522722,
-0.0725888088,
-0.2002185881,
-0.425773859,
-0.152362287,
-0.128887713,
-0.1171435863,
0.3681681454,
0.3341635764,
-0.3990163207,
-0.1482420266,
-0.1328411698,
-0.0242101848,
-0.2809097767,
-0.3642935753,
-0.2995693088,
0.0558414906,
-0.2109501064,
0.0794525221,
-0.2678173184,
-0.0450140983,
0.0515149683,
0.0376875028,
-0.1677217185,
-0.1312996447,
0.0200152732,
0.0744418576,
0.354090631,
-0.367410928,
0.1572969109,
0.4705129862,
-0.0993365049,
0.0339574739,
0.3758523166,
0.0386761203,
0.0075478666,
0.3231614232,
-0.1632545888,
0.2435813844,
0.5914889574,
-0.3155204952,
0.0485775024,
0.0017068982,
0.4769370556,
-0.154553771,
-0.1676848382,
0.3743361533,
-0.1851759553,
0.3994725645,
0.217222929,
0.1465721279,
0.2307448387,
-0.1425456554,
0.2136998177,
0.0921450108,
-0.1486647427,
0.4187478423,
-0.019174289,
1.0156929493,
0.2073533684,
0.2826999426,
0.1329191923,
0.1772160083,
0.4591485262,
0.1079505831,
0.0556263253,
-0.2091247141,
-0.2929944396,
0.0010580942,
-0.149810344,
-0.0807723999,
-0.2313542366,
-0.276281178,
0.4570977688,
-0.3854295611,
-0.1603198647,
-0.196232602,
-0.2233335972,
-0.3173441887,
-0.1051386148,
0.0847922787,
-0.0389762484,
0.1487230957,
0.5471584797,
-0.2226289511,
-0.1768110245,
-0.3493128419,
-0.3877288103,
-0.1265833974,
-0.0337558426,
-0.2098588198,
0.0311238058,
0.3574779332,
-0.1190102994,
-0.2099214196,
0.279209882,
0.3486990035,
0.1478073597,
-0.3286240697,
-0.0536577404,
0.1387415528,
0.0693157241,
0.1134879962,
-0.016283337,
0.2863364518,
-0.0592633039,
-0.0206090659,
0.0139441136,
-0.2617379725,
0.062509805,
0.0623126477,
-0.1763905585,
-0.2089602351,
-0.046376802,
-0.0278084092,
0.0901820511,
-0.0493520461,
-0.2061767876,
0.0301747173,
0.3500427008,
-0.2381433994,
0.0757842362,
0.194808647,
-0.2536817789,
-0.1046509072,
0.5353070498,
-0.2839657664,
-0.040486332,
0.5257987976,
0.0521444827,
-0.0739262775,
-0.0879664719,
0.1231452078,
0.3842421174,
-0.8295775652,
0.0623213425,
-0.014783252,
0.020648405,
0.0930654705,
0.0981902033,
0.4433171153,
-0.1498869359,
-0.1371235549,
-0.2625293136,
-0.2995175123,
0.2950275838,
-0.3477451503,
0.1463387609,
-0.4390572608,
0.0171230882,
-0.1422360986,
0.1940520555,
-0.2237336934,
0.1365992725,
-0.2280648649,
-0.1690292209,
-0.1042844951,
-0.0442445725,
0.3885757327,
-0.0331423357,
-0.0840897709,
0.0814568475,
-0.3113954365,
-0.0289358329,
-0.2361382097,
0.1762452573,
-0.1520954072,
0.1542674303,
-0.0213324502,
0.2418332845,
-0.0256442502,
-0.135126397,
0.019991301,
0.1746340543,
0.0819576681,
0.1829497963,
0.1014146656,
0.1096102893,
-0.1310427785,
0.2293000519,
0.0827876031,
-0.200168252,
-0.0907209814,
0.0441640541,
-0.4078600407,
-0.0517293215,
0.1881528199,
0.1526675373,
-0.2726954818,
0.1853770614,
0.0912859142,
-0.002634964,
-0.3078108132,
0.0038789436,
0.4702509642,
0.1285603195,
-0.204887554,
0.0074319616,
-0.0171013754,
0.052266553,
-0.1066369116,
0.2567212284,
0.0800272226,
0.4084861279,
-0.0540619344,
0.2842749953,
0.138012141,
-0.0800255537,
0.0952326506,
0.2558291554,
-0.0737555772,
-0.1462481171,
0.3656683862,
-0.0703356117,
0.0768480301,
0.3500857949,
0.4774138331,
0.1101925224,
0.1348934323,
-0.0343455933,
-0.0622374006,
0.1586264521,
-0.0870702714,
0.2827892005,
-0.1443685293,
-0.2000662386,
-0.1374967247,
0.0946217328,
0.0407012217,
-0.1743671894,
-0.4197375774,
0.7717676759,
-0.2196438909,
-0.2426418215,
-0.186988771,
0.2187613398,
-0.1312939227,
-0.1028858125,
-0.1330423504,
-0.1627593786,
-0.1881947964,
-0.1369979978,
0.2413048148,
0.0748142377,
0.0135706998,
0.1831949055,
0.0779203922,
-0.2401615828,
-0.4097250104,
0.1471297443,
0.2348366231,
-0.1723954082,
0.0443610325,
0.2696972787,
-0.2988671362,
0.1826587766,
0.6049938202,
0.3143848181,
0.0188653804,
-0.1037894487,
0.1164323688,
0.2726609707,
0.010430254,
-0.1231351718,
0.144232899,
0.2494458854,
-0.4571598768,
0.0450881198,
0.0002661273,
-0.1097637862,
0.4263655543,
-0.2372506112,
0.3066158295,
-0.1923082322,
0.4076455235,
-0.1710708737,
0.1075956225,
0.0466248579,
-0.1395915598,
-0.7387021184,
0.2041075081,
-0.101489611,
0.0151313879,
0.0688965023,
-0.2466745377,
0.0107894056,
-0.1486221254,
0.1965498924,
0.50742805,
0.063034907,
-0.2921726406,
-0.1924870461,
-0.4941078424,
0.312924087,
0.1277846098,
0.2551855445,
-0.0394152366,
0.3570332527,
-0.0903415605,
0.0310945399,
0.2803322077,
-0.0243288688,
-0.1102451533,
0.0334420726,
-0.4978342056,
0.1731982231,
-0.0518372729,
-0.0960946009,
0.0645285696,
0.2707466781,
-0.044634603,
-0.2914807498,
-0.0436214097,
0.2637949586,
-0.2875214517,
-0.1453558207,
0.192730397,
0.4304184616,
0.1727125496,
0.1035241857,
-0.0241544358,
-0.1447628736,
-0.1757724434,
-0.1006722897,
-0.0385696217,
-0.1139796674,
-0.2793583274,
0.423961103,
-0.3997513652,
-0.4653258026,
-0.3727267981,
0.313313365,
0.0486242026,
-0.1338500679,
-0.2125532925,
0.3537253141,
-0.1586042941,
0.0163043588,
0.2923788726,
0.2744101286,
-0.0121015552,
0.3496614695,
-0.2905238271,
-0.1873505414,
0.7223867178,
-0.5237224102,
0.3236787319,
-0.2664924264,
0.4198945165,
0.3544082344,
-0.3350406885,
-0.3600740135,
-0.0202661753,
0.1277983785,
0.0886094421,
-0.1551573873,
0.089078486,
-0.3894799948,
-0.0616634972,
0.016400449,
0.2397495955,
0.1686127633,
-0.0078572929,
-0.0522726402,
-0.1597505212
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | are you sure you need an external lock file? if it's a single purpose locking in the same scope you can lock the caller `__file__` instead, e.g. here is how one can `flock` the script file itself to ensure atomic printing:
```
import fcntl
def printflock(*msgs):
""" print in multiprocess env so that the outputs from different processes don't get interleaved """
with open(__file__, "r") as fh:
fcntl.flock(fh, fcntl.LOCK_EX)
try:
print(*msgs)
finally:
fcntl.flock(fh, fcntl.LOCK_UN)
```
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 75 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
are you sure you need an external lock file? if it's a single purpose locking in the same scope you can lock the caller `__file__` instead, e.g. here is how one can `flock` the script file itself to ensure atomic printing:
```
import fcntl
def printflock(*msgs):
""" print in multiprocess env so that the outputs from different processes don't get interleaved """
with open(__file__, "r") as fh:
fcntl.flock(fh, fcntl.LOCK_EX)
try:
print(*msgs)
finally:
fcntl.flock(fh, fcntl.LOCK_UN)
```
| [
-0.0958078206,
-0.0942352936,
0.1255573928,
0.0927402601,
-0.0868550688,
-0.0296322852,
0.2317223102,
0.2375226766,
0.2348069251,
0.1782365143,
0.0034630075,
0.147409752,
-0.2378921062,
-0.0178874806,
-0.0204141214,
0.1772462279,
0.0075171068,
-0.0802014172,
-0.3421197534,
-0.1448379606,
-0.2741202116,
0.5315681696,
-0.1308237612,
-0.0719342232,
-0.4290015399,
0.2123154104,
-0.2263912857,
0.1211146712,
-0.0081585292,
-0.3395677507,
0.1049850062,
0.3507530689,
0.1030159071,
0.4625876248,
-0.0001223227,
-0.2479466498,
0.1706747711,
-0.1718239486,
-0.1011757702,
-0.0645023212,
0.1405717731,
-0.310826093,
0.4765368104,
-0.2978203893,
0.1028476059,
0.084108986,
0.1408371627,
-0.3395663202,
0.4358518124,
-0.0378725044,
0.0893844292,
0.2628900707,
0.1200163811,
-0.1775285006,
-0.0331900977,
-0.1780334711,
0.0121775717,
0.2705991268,
-0.1119370908,
0.0844491422,
-0.2585886121,
0.2509042919,
0.2237700373,
-0.20246315,
0.1423098594,
0.0610954501,
-0.1400076002,
-0.2547032535,
-0.1634259969,
0.2697404623,
0.136480391,
-0.3204960525,
-0.3008385897,
0.0492608622,
-0.1172188148,
-0.3431066275,
0.0262462385,
-0.0340272821,
-0.0005135909,
0.0618580729,
0.1232939661,
-0.199305132,
-0.0104409605,
0.0003196876,
-0.0139875636,
0.0958359987,
-0.1606515795,
0.0895802677,
-0.132229358,
0.2888586819,
-0.3445695639,
0.3877013922,
0.0099183368,
0.1042268872,
-0.2629309297,
0.1574041843,
0.2871654034,
-0.1044998765,
0.1025783271,
0.3618665636,
-0.0343257599,
0.0118682217,
0.5354159474,
0.1580043435,
-0.0143519221,
0.3216002584,
0.2761747837,
0.0193978995,
0.2049494982,
0.4004884958,
-0.2367347628,
-0.2947700918,
0.2996626198,
-0.20983091,
0.3451396525,
0.1116673425,
0.1468068659,
-0.4740606546,
0.3149008751,
0.3492852151,
-0.0679861754,
-0.0676056743,
-0.0274081845,
0.3686688542,
0.2322145402,
0.0972456336,
0.3592588305,
0.2219019532,
-0.2242743373,
-0.0752860308,
-0.3204059601,
-0.170639798,
-0.0248742551,
0.4072386324,
-0.0777525604,
-0.0594259277,
0.5295736194,
-0.0586508326,
0.5135625005,
-0.0885758996,
0.0644366071,
0.1758890599,
0.0725544617,
0.2391676009,
0.0254027098,
0.2357243299,
0.3454736769,
0.2439839989,
-0.2318499684,
-0.3414819241,
0.063952595,
-0.4619211853,
0.1058227643,
-0.0182387568,
-0.3980554342,
0.1951521784,
0.6058907509,
0.1725614071,
-0.1805834025,
-0.2601393461,
0.3114192188,
0.0852640197,
-0.0760378465,
-0.087410748,
0.4565058947,
0.7431334257,
-0.3583597243,
-0.3069778681,
0.1075278148,
-0.1093928739,
0.0635074154,
0.1491246521,
0.0973033607,
0.0965094268,
-0.2766648233,
-0.1178216711,
0.3571414351,
-0.4726912081,
-0.4725746512,
0.0852726176,
-0.4719013572,
-0.0274124704,
0.2066821307,
0.0076041371,
0.2238253653,
0.2170169055,
0.1575409174,
-0.1981642246,
0.2113956064,
-0.2767272294,
-0.3929209709,
-0.1785396636,
-0.1324966848,
-0.0649427772,
0.2047933787,
0.0115921069,
0.0840648115,
0.0313538834,
0.2286919057,
0.089783676,
0.1888237894,
0.2832200229,
0.3775963187,
0.1742910445,
0.2464331985,
0.1135071814,
-0.2764571011,
0.2135853767,
-0.4569102526,
0.117864415,
0.0813401565,
-0.0489258282,
-0.0218003318,
-0.1058619171,
-0.0418673195,
-0.4147402048,
0.0085284859,
-0.0434774123,
0.3030487895,
0.2495977581,
-0.1796234846,
0.2385916114,
-0.0130308121,
0.2911897004,
-0.4203339219,
-0.1497298926,
-0.0477628708,
-0.3882253766,
0.0575979501,
0.1718963385,
0.0015681582,
0.0748586953,
-0.1377816796,
0.4197936952,
-0.05710347,
0.2876296639,
0.2521412075,
0.1678889245,
-0.1971279979,
-0.2420719564,
-0.012515272,
0.065709427,
-0.0356097817,
0.0254986808,
-0.1123305261,
0.490195334,
0.0970410034,
0.0857906938,
-0.337297678,
0.2576319873,
-0.1290616542,
-0.0623214059,
-0.5252735615,
-0.1358706355,
0.3787256479,
-0.2251192927,
0.2366429865,
-0.1315060258,
0.0301115364,
0.0798022747,
0.1785757393,
0.2459711581,
0.1715171486,
-0.0005884751,
-0.1394903362,
-0.0125993565,
0.0347564444,
-0.0848474652,
0.258292228,
0.2357061058,
0.1050707102,
0.088127628,
0.1087510809,
-0.1732837409,
0.1077509597,
-0.0638255924,
-0.1016963869,
0.169249922,
0.1857383102,
0.1188378185,
-0.2927374542,
-0.1776985526,
-0.3255644441,
-0.032076288,
-0.2366853058,
-0.0204164088,
0.0345231816,
0.3458483219,
0.0091696717,
0.0864998028,
-0.2900828719,
-0.4045253992,
0.2174942493,
-0.179891035,
-0.0688813329,
0.1823816597,
-0.4417116642,
0.4122616649,
0.1074883118,
-0.2418731451,
-0.329423219,
-0.1597182006,
-0.078658551,
-0.2074675262,
-0.2837306261,
0.1251960397,
0.2220122665,
-0.0486109108,
0.0800971091,
-0.1392988563,
-0.3460690379,
0.0014419006,
-0.0510541797,
0.628565371,
0.2911646068,
-0.0305410177,
-0.1663740277,
0.0547244139,
0.4496111572,
-0.559389472,
-0.1918304265,
0.1030515879,
-0.0518235974,
0.1852311194,
-0.2551891208,
-0.268727839,
0.0448061004,
-0.414847672,
0.6636124849,
0.023417905,
-0.0101746991,
-0.2235938907,
-0.1076810509,
0.1781019419,
-0.2553525567,
0.4349592328,
-0.1497937143,
-0.5847464204,
0.424423486,
-0.1281865239,
-0.2230436802,
-0.1408720016,
0.0822280645,
0.1759923995,
-0.144905597,
-0.5932174325,
-0.500626266,
-0.101444371,
0.2448945343,
-0.0263819918,
0.1439135075,
0.4210920334,
-0.1415072531,
-0.0179312751,
-0.0129034817,
-0.1239337623,
0.5327207446,
-0.1963027269,
0.0532472171,
-0.0795465931,
0.4616313875,
0.0842699111,
0.8287322521,
0.5620366335,
0.3317269087,
0.183295548,
0.0219698567,
0.4228360057,
-0.1368257254,
-0.1904711723,
0.3991364837,
0.3047937453,
-0.2714652419,
0.2112303227,
0.0435506441,
0.2230719328,
-0.1163756996,
0.0106747895,
-0.4465951324,
-0.4564571977,
-0.0719127506,
0.0920201465,
0.2174007148,
-0.0941519886,
0.0339869708,
-0.0975616127,
0.0202381313,
0.5146849751,
0.3764461279,
0.163100034,
0.1277968138,
-0.1811329722,
-0.0057471618,
-0.6595854163,
0.3577420115,
-0.4646303654,
0.1572018266,
-0.1583261788,
-0.3413119912,
0.1660345048,
-0.0564945415,
0.6355432272,
0.1735784113,
-0.1956253499,
-0.0046776459,
-0.2547478676,
-0.4369534254,
-0.1027991176,
-0.2421707958,
-0.1507626176,
0.362005651,
0.091847904,
-0.4735631943,
-0.1324451864,
-0.1301801354,
-0.1474507451,
-0.3104040027,
-0.3958632052,
-0.2934834659,
0.0456175283,
-0.3154401481,
0.0738438219,
-0.3079229593,
0.0215214305,
0.0181092545,
0.0927209482,
0.0271866899,
-0.1077790186,
0.0773516148,
0.1022580415,
0.2887766659,
-0.2136119455,
0.0983377844,
0.4372391701,
-0.1915368438,
-0.0192963257,
0.3132374287,
0.1018752009,
-0.0999717042,
0.175930202,
-0.0794540793,
0.279961884,
0.603248477,
-0.1729382575,
0.0549091697,
-0.0586393587,
0.4818811417,
-0.1178967282,
-0.285554111,
0.2373218685,
-0.1547115147,
0.3927035332,
0.2192185968,
0.0485757068,
0.3099291027,
-0.1057200208,
0.3555973768,
-0.0205975957,
-0.141152218,
0.4236706793,
0.1191990525,
0.9506549835,
0.0976579189,
0.1860411465,
0.2422059476,
-0.0376854911,
0.4492133856,
0.0296560302,
0.0142839588,
-0.3872035146,
-0.275667578,
0.0917525887,
-0.1032311022,
-0.0816480368,
-0.3444141746,
-0.2914732695,
0.315526396,
-0.3911912143,
-0.3037448823,
-0.3794368505,
-0.196008563,
-0.4446934164,
-0.2635653615,
-0.0477044098,
-0.0090241618,
0.1433115304,
0.5026949644,
-0.3078815937,
-0.1107046828,
-0.2886676788,
-0.3712603152,
-0.2273339927,
-0.0222585015,
-0.202735588,
-0.0640974268,
0.2401204258,
0.0918193907,
-0.3199107349,
0.3298313618,
0.381529063,
0.1273181289,
-0.2358826846,
-0.178047955,
0.2861910462,
0.0382463112,
0.1124107391,
0.0014292747,
0.2514291704,
0.0359149203,
-0.0307595134,
0.1709095687,
-0.1664932817,
0.087222591,
0.1005961895,
-0.1597256362,
-0.1423913091,
-0.1958401054,
-0.0657536089,
0.1554858088,
-0.066339612,
-0.2104572356,
0.048237443,
0.2457895726,
-0.3146841526,
0.1481902748,
0.0379268527,
-0.2771348953,
-0.0750763789,
0.5247074366,
-0.2640320063,
-0.0248466432,
0.3970451653,
0.2828390598,
-0.0860160738,
-0.1441111565,
0.129332602,
0.2986588776,
-0.7690415978,
0.2018032968,
0.1477665156,
-0.0192423612,
0.0434803851,
0.2532593906,
0.3413791358,
-0.1568591446,
-0.0955138728,
-0.22066392,
-0.3461403251,
0.3158045709,
-0.2180738151,
0.2099336088,
-0.2942484021,
0.1017865986,
-0.1563839912,
0.2841732502,
-0.2742086053,
0.0098688416,
-0.2385792136,
-0.1825029552,
-0.0759446919,
-0.0353176259,
0.3743843138,
0.1295117736,
-0.0323033705,
0.2220454812,
-0.2932638526,
-0.089410685,
-0.2077839077,
0.1688522846,
-0.1008317173,
0.2274568677,
-0.0931681544,
0.1616182625,
-0.1409089267,
-0.1505562216,
0.1215028316,
0.3491713703,
0.0348307565,
0.1579635292,
0.2235540897,
0.0524990633,
-0.1588915586,
0.1782958955,
0.0967275351,
-0.1647046059,
-0.0673633367,
0.0238148756,
-0.4683919251,
-0.075499475,
0.1976671219,
0.1554303765,
-0.1301892698,
0.1547017694,
0.0438502245,
0.0733111799,
-0.3498418927,
0.0681978688,
0.5667786598,
0.2293313146,
-0.1947151721,
-0.0845656767,
-0.0761869997,
0.0748457611,
-0.079774797,
0.4089385569,
0.0196243078,
0.222022146,
0.0884396583,
0.3918867707,
0.0069044009,
-0.154984504,
0.0887216628,
0.1567452103,
-0.1267599761,
-0.1499547213,
0.27385813,
-0.097753942,
0.1879409552,
0.3699199557,
0.501770854,
0.185787648,
0.1736930907,
-0.1599158943,
-0.0085368007,
-0.0041764379,
-0.183228761,
0.3578148782,
-0.06316448,
-0.1351080388,
-0.061901588,
0.0247356817,
0.099988617,
-0.1619315296,
-0.4995587766,
0.7212769985,
-0.1596978456,
-0.0727415457,
-0.1729461104,
0.420239687,
-0.1257973462,
-0.1201534867,
-0.080722928,
-0.1388189346,
-0.0810435116,
-0.1081052795,
0.3586573899,
0.1451633126,
0.0638322085,
0.2073589712,
0.1183852553,
-0.278842479,
-0.4896352887,
0.2174433768,
0.2856248915,
-0.2475679517,
0.0128942747,
0.2346801609,
-0.2322445363,
0.1432933658,
0.6515592933,
0.2464316934,
0.0117483847,
-0.0120476019,
0.2025608271,
0.2649373412,
0.0501919985,
-0.122369878,
0.1640230417,
0.2685832977,
-0.4571159482,
0.21599783,
0.0437235013,
-0.1576764286,
0.2495298684,
-0.2333384454,
0.2681142092,
-0.002987843,
0.5281677246,
-0.0881280676,
0.2892680764,
-0.0906341523,
-0.1573890448,
-0.6749098301,
0.109842658,
-0.1533485651,
-0.1440504789,
0.0397293158,
-0.3130022883,
0.0267414078,
-0.0766395852,
0.0271470733,
0.5202210546,
0.1198401451,
-0.2506420612,
-0.2004326433,
-0.409830749,
0.3709430099,
0.1497473121,
0.1613407433,
-0.0727334991,
0.41484797,
-0.016673632,
0.078647241,
0.2746059895,
-0.0772276223,
-0.2901549935,
0.2360419482,
-0.4297747314,
0.2515312135,
-0.2711081803,
-0.072457701,
0.0736431032,
0.3046462834,
-0.0841524974,
-0.281391114,
-0.0065454841,
0.2192516923,
-0.2696211338,
-0.1493307948,
0.0387558341,
0.367033571,
0.1418909281,
0.0027358606,
-0.1245593876,
-0.1818845868,
-0.1920180023,
-0.0750166178,
-0.1639817804,
-0.1211032271,
-0.2894136012,
0.4088882208,
-0.3357613087,
-0.5149752498,
-0.3786816895,
0.3133418262,
0.1088767797,
-0.1906794906,
-0.1274162978,
0.3311882615,
-0.1567313671,
0.0528099723,
0.2602563202,
0.4714582264,
0.0496161543,
0.4661207795,
-0.2932116091,
-0.1922185719,
0.6927309036,
-0.464907825,
0.2622734904,
-0.1952013522,
0.4434478879,
0.1342340857,
-0.3789181113,
-0.3714778423,
0.0214506537,
0.0577610061,
0.1420639455,
-0.0396044552,
0.1356398612,
-0.3098530769,
-0.1109305695,
-0.0667541027,
0.1820857078,
0.1642805338,
0.1262946427,
-0.1123878807,
-0.0309418514
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | OK, this issue is not about caching but some internal conflict/race condition it seems, I have just run into it on my normal env:
```
Traceback (most recent call last):
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 356, in _finalize
self.data = Dataset(**reader.read_files([{"filename": f} for f in file_paths]))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 236, in read_files
pa_table = self._read_files(files, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 171, in _read_files
pa_table: pa.Table = self._get_dataset_from_filename(f_dict, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 302, in _get_dataset_from_filename
pa_table = ArrowReader.read_table(filename, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 97, in pyarrow.lib.check_status
FileNotFoundError: [Errno 2] Failed to open local file '/home/stas/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow'. Detail: [errno 2] No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "examples/seq2seq/run_seq2seq.py", line 655, in <module>
main()
File "examples/seq2seq/run_seq2seq.py", line 619, in main
test_results = trainer.predict(
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer_seq2seq.py", line 121, in predict
return super().predict(test_dataset, ignore_keys=ignore_keys, metric_key_prefix=metric_key_prefix)
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer.py", line 1706, in predict
output = self.prediction_loop(
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer.py", line 1813, in prediction_loop
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
File "examples/seq2seq/run_seq2seq.py", line 556, in compute_metrics
result = metric.compute(predictions=decoded_preds, references=decoded_labels)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 388, in compute
self._finalize()
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 358, in _finalize
raise ValueError(
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
```
I'm just running `run_seq2seq.py` under DeepSpeed:
```
export BS=16; rm -r output_dir; PYTHONPATH=src USE_TF=0 CUDA_VISIBLE_DEVICES=0,1 deepspeed --num_gpus=2 examples/seq2seq/run_seq2seq.py --model_name_or_path t5-small --output_dir output_dir --adam_eps 1e-06 --do_eval --do_train --do_predict --evaluation_strategy=steps --label_smoothing 0.1 --learning_rate 3e-5 --logging_first_step --logging_steps 1000 --max_source_length 128 --max_target_length 128 --num_train_epochs 1 --overwrite_output_dir --per_device_eval_batch_size $BS --per_device_train_batch_size $BS --predict_with_generate --eval_steps 25000 --sortish_sampler --task translation_en_to_ro --val_max_target_length 128 --warmup_steps 500 --max_train_samples 100 --max_val_samples 100 --max_test_samples 100 --dataset_name wmt16 --dataset_config ro-en --source_prefix "translate English to Romanian: " --deepspeed examples/tests/deepspeed/ds_config.json
```
It finished the evaluation OK and crashed on the prediction part of the Trainer. But the eval / predict parts no longer run under Deepspeed, it's just plain ddp.
Is this some kind of race condition? It happens intermittently - there is nothing else running at the same time.
But if 2 independent instances of the same script were to run at the same time it's clear to see that this problem would happen. Perhaps it'd help to create a unique hash which is shared between all processes in the group and use that as the default experiment id?
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 409 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
OK, this issue is not about caching but some internal conflict/race condition it seems, I have just run into it on my normal env:
```
Traceback (most recent call last):
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 356, in _finalize
self.data = Dataset(**reader.read_files([{"filename": f} for f in file_paths]))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 236, in read_files
pa_table = self._read_files(files, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 171, in _read_files
pa_table: pa.Table = self._get_dataset_from_filename(f_dict, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 302, in _get_dataset_from_filename
pa_table = ArrowReader.read_table(filename, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 97, in pyarrow.lib.check_status
FileNotFoundError: [Errno 2] Failed to open local file '/home/stas/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow'. Detail: [errno 2] No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "examples/seq2seq/run_seq2seq.py", line 655, in <module>
main()
File "examples/seq2seq/run_seq2seq.py", line 619, in main
test_results = trainer.predict(
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer_seq2seq.py", line 121, in predict
return super().predict(test_dataset, ignore_keys=ignore_keys, metric_key_prefix=metric_key_prefix)
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer.py", line 1706, in predict
output = self.prediction_loop(
File "/mnt/nvme1/code/huggingface/transformers-master/src/transformers/trainer.py", line 1813, in prediction_loop
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
File "examples/seq2seq/run_seq2seq.py", line 556, in compute_metrics
result = metric.compute(predictions=decoded_preds, references=decoded_labels)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 388, in compute
self._finalize()
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 358, in _finalize
raise ValueError(
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
```
I'm just running `run_seq2seq.py` under DeepSpeed:
```
export BS=16; rm -r output_dir; PYTHONPATH=src USE_TF=0 CUDA_VISIBLE_DEVICES=0,1 deepspeed --num_gpus=2 examples/seq2seq/run_seq2seq.py --model_name_or_path t5-small --output_dir output_dir --adam_eps 1e-06 --do_eval --do_train --do_predict --evaluation_strategy=steps --label_smoothing 0.1 --learning_rate 3e-5 --logging_first_step --logging_steps 1000 --max_source_length 128 --max_target_length 128 --num_train_epochs 1 --overwrite_output_dir --per_device_eval_batch_size $BS --per_device_train_batch_size $BS --predict_with_generate --eval_steps 25000 --sortish_sampler --task translation_en_to_ro --val_max_target_length 128 --warmup_steps 500 --max_train_samples 100 --max_val_samples 100 --max_test_samples 100 --dataset_name wmt16 --dataset_config ro-en --source_prefix "translate English to Romanian: " --deepspeed examples/tests/deepspeed/ds_config.json
```
It finished the evaluation OK and crashed on the prediction part of the Trainer. But the eval / predict parts no longer run under Deepspeed, it's just plain ddp.
Is this some kind of race condition? It happens intermittently - there is nothing else running at the same time.
But if 2 independent instances of the same script were to run at the same time it's clear to see that this problem would happen. Perhaps it'd help to create a unique hash which is shared between all processes in the group and use that as the default experiment id?
| [
-0.0576908551,
-0.0025047362,
0.1063461602,
0.3295959234,
-0.0730612054,
0.0291782022,
0.1474630535,
0.2418279946,
0.3056668043,
0.1599035859,
0.204788059,
0.0719601661,
-0.3148396611,
-0.1378427595,
-0.0102979895,
0.2024060488,
0.0270131379,
0.0037931763,
-0.2088182271,
-0.1798239052,
-0.3237954378,
0.5144025683,
-0.0927537233,
-0.0305793434,
-0.3576313555,
0.3367958963,
-0.1782271713,
0.3024883866,
0.0206067935,
-0.6091864705,
0.4519517422,
0.1594462693,
0.1245751977,
0.487296313,
-0.0001303917,
-0.162183851,
0.2800581157,
-0.1859595478,
-0.1424786001,
-0.2011274993,
-0.0963118076,
-0.2803776264,
0.4510984421,
-0.2343735695,
-0.023727458,
-0.0773366839,
0.175815776,
-0.5915820003,
0.35426718,
0.1136427596,
-0.0048830397,
0.3083102703,
-0.114485234,
-0.2425564378,
0.1217082143,
-0.042546533,
0.0690075085,
0.2845216095,
-0.0860332549,
-0.1388478577,
-0.2688542902,
0.2733644545,
0.1988230348,
-0.0540878624,
0.4352083504,
0.0169220008,
0.0313856825,
-0.049776908,
-0.1019932106,
0.3329289854,
0.1740119457,
-0.1938282549,
-0.2087355554,
0.0363135636,
-0.1447735131,
-0.4275952876,
0.2937575877,
-0.0938258618,
0.0557498336,
0.1103090644,
0.074509345,
-0.0691118538,
0.0193979442,
0.1733285189,
-0.1224159747,
0.1237846315,
-0.0248195268,
0.0248816311,
-0.0011233373,
0.2460561544,
-0.1643095762,
0.1366795301,
-0.0891159624,
0.2079619765,
-0.3737040162,
0.132975325,
0.1770112216,
-0.0221960545,
-0.0012673065,
0.2791177928,
-0.0404294617,
0.0242974684,
0.3617474139,
0.1396157444,
-0.0119775105,
0.4315190613,
0.2799910009,
0.1480380446,
0.2010218799,
0.2454796135,
-0.0930147395,
-0.1489388943,
0.1177477092,
-0.2112676948,
0.384324193,
0.1562825143,
-0.0006589293,
-0.4661144912,
0.0681592673,
0.3652067184,
-0.2267411053,
-0.1440906674,
0.2043336481,
0.3926240504,
0.1960395873,
-0.0364984944,
0.1631371081,
0.2825303078,
-0.2508921325,
-0.0268498063,
-0.4424970746,
-0.2588559985,
-0.1373984218,
0.2259522676,
0.1951782256,
-0.0326906778,
0.5749141574,
-0.2132208347,
0.3801065087,
-0.0748554617,
0.2046854347,
0.1003377736,
0.0419857018,
0.2588790059,
-0.0775680691,
0.2467141747,
0.4501028359,
0.085313797,
-0.2225542068,
-0.1783109903,
-0.1147635058,
-0.5404673815,
0.1880509555,
-0.0822354779,
-0.4932708144,
0.2703613043,
0.4431529641,
0.1586529762,
0.058016561,
-0.3409630358,
0.2449739277,
-0.0816509128,
-0.2438093126,
-0.0823499411,
0.4338892698,
0.7405846715,
-0.3019001186,
-0.391982317,
0.0709850639,
0.1103328541,
-0.0848063976,
0.2470319718,
-0.1676897258,
-0.0589926541,
-0.3711161613,
0.029759869,
0.3290179074,
-0.4905433953,
-0.6476852298,
-0.1112072021,
-0.3637508154,
-0.0950267836,
0.2161109895,
0.1401778907,
0.0239367709,
-0.0563680902,
0.1230909154,
-0.1741122901,
0.2157103866,
-0.3362262845,
-0.2547807693,
-0.3030841947,
0.0921703875,
0.0116519853,
0.3164175451,
-0.0692321807,
-0.0339652523,
-0.1159879863,
0.104760997,
0.1246333718,
0.1933835447,
0.3314184546,
0.2837236524,
-0.0269380882,
0.3410245776,
0.2112782151,
-0.3488748074,
0.3641500473,
-0.5019806623,
0.0062013362,
0.1006615907,
0.0107668489,
-0.0791024566,
-0.236192584,
-0.1411095113,
-0.4627399147,
-0.0980776995,
-0.1100678965,
0.0589429885,
0.1902162135,
-0.1977373958,
0.1868090034,
-0.0470206365,
0.2514114976,
-0.4648360014,
0.0201666951,
-0.0818684772,
-0.2810688019,
-0.0736853257,
0.1205331236,
0.0165863372,
0.052100677,
-0.2455531955,
0.3516985774,
-0.088623181,
0.2942870557,
0.2905910313,
0.2052251101,
-0.0483783409,
-0.4059664011,
-0.0110340482,
0.1969523281,
-0.107371524,
0.0252720565,
-0.3456169367,
0.4294167757,
0.0680319369,
0.1740294695,
-0.3788429499,
0.1567652524,
-0.1176288724,
-0.1660419405,
-0.4288588762,
-0.1451913863,
0.5433690548,
-0.1508425176,
0.3000461459,
-0.0514610186,
0.0512115508,
-0.0064015463,
0.2151769698,
0.2164016366,
0.1246333271,
0.0957660973,
-0.2813660204,
-0.1090494543,
0.0874327794,
0.0135742426,
0.5847154856,
0.1977364719,
0.2524628341,
0.147108987,
0.0353332162,
-0.1976372898,
0.171787262,
-0.0318981484,
-0.0335671008,
0.2177738994,
0.2405482829,
0.1597722471,
-0.2981685996,
-0.1236057058,
-0.2751498222,
0.0664379001,
-0.3651403785,
0.1122678369,
-0.2560161352,
0.3395062089,
0.1566369534,
0.0014964566,
-0.2696928084,
-0.239630267,
0.1786826551,
-0.1410635263,
0.0234253481,
0.2518822849,
-0.4109796286,
0.4310765266,
0.0332369432,
-0.2168579549,
-0.249537766,
-0.3102409244,
-0.1903652549,
-0.2834860682,
-0.1112969369,
0.0557816811,
0.0091719739,
-0.0543885604,
-0.0524968132,
0.0212189108,
-0.3649506569,
-0.0103319548,
0.1506589353,
0.6111649275,
0.2543147802,
0.0139045902,
-0.3684001565,
0.0578796826,
0.3914338052,
-0.2630780339,
-0.3664730489,
0.0817095265,
0.0431454591,
0.2250203937,
-0.2592799067,
-0.3096569777,
0.0280252211,
-0.3285586238,
0.4363333285,
-0.0327877738,
0.0186584219,
-0.0697610453,
-0.0786050633,
0.0785866678,
-0.2438888848,
0.17522493,
-0.3726034164,
-0.5516346693,
0.3954832852,
-0.0112834238,
-0.1737853736,
-0.1263962537,
0.0514000878,
0.3583474159,
0.0741555318,
-0.7520509958,
-0.4106814861,
0.0507170483,
0.2256347239,
-0.1302639246,
-0.0699438229,
0.6061976552,
-0.0339116529,
0.0616026446,
-0.0909516588,
-0.1484256983,
0.611536324,
-0.1508138627,
0.1900128573,
-0.0960244983,
0.4599850476,
0.2075084895,
0.9944365621,
0.5102086067,
0.1887706667,
0.2738809288,
0.2238534987,
0.4091358483,
-0.0956356153,
-0.1983336508,
0.2562210858,
0.2452373803,
-0.0865917653,
0.2047561705,
0.0359233692,
0.1220121533,
-0.0613909736,
-0.0637565553,
-0.2538513839,
-0.3101133704,
-0.0132151395,
-0.1616047472,
0.3300981224,
0.0614189431,
0.0212093741,
-0.2757086158,
-0.0623923317,
0.3666173816,
0.4859734178,
0.1757146865,
0.2632333636,
-0.2251615524,
0.0044158511,
-0.443444252,
0.2522541583,
-0.3854749501,
0.2858230472,
-0.2330494374,
-0.1856445074,
0.0981950909,
-0.0833996087,
0.6718572378,
0.1430876702,
-0.0848101005,
-0.0781539008,
-0.0773268417,
-0.3918817639,
-0.158213377,
0.0495567173,
0.01075463,
0.2382261902,
0.4341593385,
-0.373568505,
-0.2426628321,
-0.0838243961,
-0.1122094095,
-0.2599931061,
-0.4525397122,
-0.2815352976,
0.1446146071,
-0.1928816438,
0.183347106,
-0.211340934,
-0.0374571048,
0.0612426549,
-0.0035394207,
-0.2129250616,
-0.1019912735,
0.0182372071,
-0.0392021611,
0.3360673785,
-0.4024116099,
0.2283703834,
0.5376234055,
-0.1064088345,
0.0376134366,
0.3673104048,
0.1663350165,
-0.1827935427,
0.2683010995,
-0.0340066254,
0.2161350846,
0.6718819141,
-0.1646299958,
0.138886258,
0.1991298795,
0.208063513,
-0.1390982866,
-0.088497445,
0.3013364673,
-0.1133782864,
0.2461325973,
0.1548599601,
0.2360220402,
0.2951619327,
-0.1596688032,
0.4172233343,
-0.0116957128,
-0.1624191552,
0.294678092,
0.039343521,
1.0834730864,
0.2922219038,
0.3179517984,
0.1359239817,
-0.0182779282,
0.3813753724,
0.095042035,
0.0987547934,
-0.2544328272,
-0.3131207824,
-0.0655984357,
-0.2473011911,
-0.0368868075,
-0.2755256891,
-0.113683641,
0.4412727952,
-0.2594438791,
-0.0031396982,
-0.160179019,
-0.2166808248,
-0.2920578122,
-0.1391659677,
0.1713824272,
-0.0610484965,
0.0194269046,
0.549574852,
-0.1914758682,
-0.1742052883,
-0.4280659258,
-0.2743672431,
-0.1341755837,
-0.0437693074,
-0.3516483605,
0.0898430794,
0.2500087917,
-0.3409357071,
-0.1055031866,
0.3168903887,
0.4565447271,
0.1514368057,
-0.3181405365,
0.1140344441,
0.2022185773,
0.1065969989,
0.0955068916,
-0.1106806695,
0.1808838248,
-0.0044127852,
-0.0609439686,
0.02494766,
-0.2868686616,
0.0445365831,
0.0080902427,
-0.1284432113,
-0.0748336315,
0.0710113645,
-0.1148346663,
0.0865434781,
0.0567271411,
-0.2192519754,
-0.0001807995,
0.3327896595,
-0.1402466744,
-0.0548495017,
0.0633068383,
-0.3442893028,
-0.1063786596,
0.7176711559,
-0.0840654671,
-0.1729817986,
0.6388205886,
-0.0108446628,
-0.0114884302,
-0.0330061913,
0.0686271638,
0.425932169,
-0.8050134182,
-0.0211883448,
-0.0666623712,
-0.0668367445,
0.1689563543,
0.0528478324,
0.4529182911,
-0.1656749547,
-0.0827276111,
-0.2919699848,
-0.4071457088,
0.2054392397,
-0.4431145191,
0.2485616207,
-0.3803114891,
-0.0345147438,
-0.3471583128,
0.1759300381,
-0.1712401956,
0.2049072385,
-0.1429180205,
-0.1194587499,
-0.107310921,
-0.0242532659,
0.2813020647,
-0.1077293307,
-0.1590749323,
0.0560279749,
-0.4286457002,
-0.0027017966,
-0.2546129823,
0.2196523696,
-0.0806879848,
0.1286475509,
-0.0928507373,
0.0388122424,
0.057488896,
-0.1133802533,
-0.0858350396,
0.1680558473,
0.0827875137,
0.0316740274,
0.065155603,
0.2205144316,
-0.1086528152,
0.3121111095,
0.0699835569,
-0.2269551158,
-0.0963365436,
0.0597955845,
-0.3726488948,
-0.0195326731,
0.09808667,
0.0736281574,
-0.2188917547,
0.1540913433,
0.1128558367,
0.0042944066,
-0.3112642169,
0.0228291079,
0.3397363424,
0.0601584315,
-0.1754276156,
-0.0453231931,
0.0236441083,
0.01852341,
-0.0767243803,
0.2651642263,
0.1048921049,
0.407500416,
0.0353326797,
0.2018359005,
0.2429069281,
-0.0319001228,
0.1388361454,
0.2965881824,
0.0931645557,
-0.2163159847,
0.3850468695,
-0.1120214462,
0.1703717113,
0.3353476822,
0.5294115543,
0.1533006132,
0.3033119142,
0.055889599,
-0.002909191,
0.1036747247,
0.0792404115,
0.2830418348,
-0.2751308978,
-0.2732944787,
-0.210527271,
0.1802258492,
0.0099642947,
-0.1593450457,
-0.2688835561,
0.8103466034,
-0.2867062092,
-0.1412341595,
-0.1363693178,
0.1648871154,
-0.1185227185,
-0.1748319268,
-0.1090281531,
-0.1780403107,
-0.0861643404,
-0.0919656232,
0.1850931346,
0.149837181,
-0.0308987759,
0.2067727,
-0.031886667,
-0.2017470896,
-0.3404053152,
0.3621703386,
0.1624514461,
-0.1593183577,
0.0908348113,
0.292106837,
-0.3177046776,
0.148734659,
0.5308626294,
0.2546333373,
0.0195456073,
-0.2139033675,
0.0672455877,
0.4024709165,
0.0209688693,
-0.1070492342,
0.109600313,
0.268801868,
-0.4691241086,
-0.009823475,
-0.0169434398,
-0.0519172288,
0.39326334,
-0.1417685151,
0.2539665103,
-0.1621344537,
0.3582158089,
-0.1994123012,
0.1481001377,
0.1612876207,
-0.0726235583,
-0.6035970449,
0.2233772427,
-0.0262329131,
0.0818555206,
0.1133687943,
-0.1424364746,
-0.0047132988,
-0.0867537335,
0.1382464916,
0.5117950439,
-0.0139975194,
-0.4154542685,
-0.274556756,
-0.6609954238,
0.388194859,
0.033392895,
0.3113515377,
0.0009760037,
0.2808102667,
-0.02230547,
-0.0369717106,
0.2481423169,
0.078900069,
-0.0846414268,
0.1082659215,
-0.4920294881,
0.09828078,
0.0012782384,
-0.1708384007,
0.0341424271,
0.2196374536,
0.0084803384,
-0.1435926557,
-0.0773816109,
0.2027600706,
-0.1723476648,
-0.0357822292,
0.2105976939,
0.4662000537,
0.0523875356,
0.0963562131,
0.0934222639,
-0.133904025,
-0.2228318304,
-0.0566933304,
-0.0889867544,
-0.0846365616,
-0.3063414395,
0.4797862768,
-0.5571390986,
-0.5037090182,
-0.3182573915,
0.2528632283,
0.020078972,
-0.26037395,
-0.2814151645,
0.3865295053,
-0.1927476674,
0.1288116425,
0.2811797261,
0.3434962332,
0.1242490858,
0.3990004063,
-0.4151054621,
-0.1510778666,
0.7318443656,
-0.4763930738,
0.3372198939,
-0.1236958802,
0.3626725674,
0.4212174416,
-0.3103902936,
-0.2815277576,
-0.0851891935,
0.1817667335,
-0.051208511,
-0.1511385143,
0.0933477432,
-0.2861336172,
0.0352644399,
-0.0865419805,
0.1908095479,
0.2411008328,
-0.1176286638,
0.0448563397,
-0.0937294066
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | When you're using metrics in a distributed setup, there are two cases:
1. you're doing two completely different experiments (two evaluations) and the 2 metrics jobs have nothing to do with each other
2. you're doing one experiment (one evaluation) but use multiple processes to feed the data to the metric.
In case 1. you just need to provide two different `experiment_id` so that the metrics don't collide.
In case 2. they must have the same experiment_id (or use the default one), but in this case you also need to provide the `num_processes` and `process_id`
If understand correctly you're in situation 2.
If so, you make sure that you instantiate the metrics with both the right `num_processes` and `process_id` parameters ?
If they're not set, then the cache files of the two metrics collide it can cause issues. For example if one metric finishes before the other, then the cache file is deleted and the other metric gets a FileNotFoundError
There's more information in the [documentation](https://huggingface.co/docs/datasets/loading_metrics.html#distributed-setups) if you want
Hope that helps ! | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 173 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
When you're using metrics in a distributed setup, there are two cases:
1. you're doing two completely different experiments (two evaluations) and the 2 metrics jobs have nothing to do with each other
2. you're doing one experiment (one evaluation) but use multiple processes to feed the data to the metric.
In case 1. you just need to provide two different `experiment_id` so that the metrics don't collide.
In case 2. they must have the same experiment_id (or use the default one), but in this case you also need to provide the `num_processes` and `process_id`
If understand correctly you're in situation 2.
If so, you make sure that you instantiate the metrics with both the right `num_processes` and `process_id` parameters ?
If they're not set, then the cache files of the two metrics collide it can cause issues. For example if one metric finishes before the other, then the cache file is deleted and the other metric gets a FileNotFoundError
There's more information in the [documentation](https://huggingface.co/docs/datasets/loading_metrics.html#distributed-setups) if you want
Hope that helps ! | [
-0.0598984696,
0.0068144351,
0.0844843984,
0.1580616087,
0.0199412405,
-0.0320760235,
0.2774338424,
0.2207646221,
0.3088777065,
0.160241276,
0.0177518949,
0.1031416655,
-0.2257238775,
-0.0206123032,
-0.0592385493,
-0.0310597233,
-0.0435304902,
-0.066776365,
-0.3691176772,
-0.1325116754,
-0.3067208827,
0.5019857287,
-0.0801182538,
0.0445252731,
-0.480796665,
0.1278131455,
-0.1756109446,
0.2225744575,
-0.078924939,
-0.5587316751,
0.3777539432,
0.18311885,
0.0421273857,
0.6000637412,
-0.0001176384,
-0.1433922499,
0.1927300692,
-0.1245241314,
-0.0832812116,
-0.193016842,
0.0741739571,
-0.3329312503,
0.4511275589,
-0.2442205995,
0.0344926864,
-0.051198028,
0.0247722249,
-0.3721565604,
0.4274138808,
-0.0064344034,
0.1307449341,
0.3117667139,
-0.0682267249,
-0.3215267658,
-0.0999849066,
-0.1143519431,
0.0776160657,
0.4171809852,
-0.1312236637,
0.0894627571,
-0.2742328644,
0.1245765463,
0.2123534381,
0.0292279795,
0.2388047278,
0.0246073343,
-0.0142711885,
-0.1256715953,
-0.0685603097,
0.2887806892,
0.1676783413,
-0.1594264209,
-0.2014689445,
0.0085322261,
-0.2289426327,
-0.4411579072,
0.1714061052,
-0.1245804206,
0.0746695846,
0.1032147482,
-0.0762279332,
-0.1549825221,
-0.0232048929,
0.1175038368,
-0.1456238776,
0.0324373916,
-0.1697128266,
0.1212584823,
-0.0279212482,
0.1497501731,
-0.3855248988,
0.3516911268,
-0.0194257982,
0.0714745596,
-0.3793978393,
0.1364801079,
0.2671203911,
0.0217038412,
0.1065052673,
0.4761021137,
-0.0047422461,
-0.0132050645,
0.389166683,
0.1737164259,
-0.0644987822,
0.5233269334,
0.2027724832,
0.0692861825,
0.2185039371,
0.2179415971,
-0.1748854816,
-0.3413077593,
0.283627063,
-0.3860957921,
0.3315581679,
0.184278816,
0.0930345654,
-0.4868882298,
0.0585249513,
0.1844702959,
-0.0363665149,
-0.1577915251,
0.0875827521,
0.3153524697,
0.1501620412,
0.0135479067,
0.2400891185,
0.39811185,
-0.2120676786,
0.0780818984,
-0.3880886734,
-0.1634977013,
-0.1441084146,
0.4293540716,
0.051711455,
-0.1105699688,
0.6856903434,
-0.1622718424,
0.4642172456,
-0.0613955855,
0.1908003092,
0.1755946875,
0.0188386999,
0.2674922049,
0.0719171464,
0.1758924723,
0.2669177949,
0.1048621684,
-0.2120068073,
-0.3701925576,
-0.1887371391,
-0.5549443364,
0.1263834685,
0.0321530662,
-0.4145854712,
0.2830302119,
0.3431293666,
0.2120161951,
-0.0794250444,
-0.2337052077,
0.1338793039,
0.0511644073,
-0.0858735293,
-0.1279808879,
0.4915287495,
0.7377229333,
-0.22204566,
-0.344671607,
0.0168371461,
-0.048713021,
-0.0864720792,
0.3050105572,
-0.0926363543,
0.2284961343,
-0.2772684693,
0.1222092211,
0.351583451,
-0.4234384596,
-0.4113767445,
0.0080032721,
-0.4272919297,
-0.1334394068,
0.1083740294,
-0.0730171725,
0.2581317425,
0.130847007,
0.054811392,
-0.1267483383,
0.1568039507,
-0.2836751938,
-0.3865197003,
-0.1858032346,
-0.0045477753,
-0.1014781594,
0.3006925583,
0.0563223958,
0.0986363441,
0.0875675976,
0.0597039871,
0.1062115729,
0.260607779,
0.2477435768,
0.226208359,
0.0003544614,
0.289906919,
0.1102009714,
-0.1867312044,
0.2562482357,
-0.3439136744,
0.1210741848,
0.1683987826,
-0.0155255273,
-0.2000710964,
-0.2549752593,
-0.1508954167,
-0.4215327501,
0.0472918749,
-0.0709690675,
0.3166151643,
0.1520330161,
-0.1438019127,
0.09164913,
0.1583403945,
0.1830411404,
-0.4161323309,
0.0770385563,
-0.0725938827,
-0.3368438184,
-0.0147532001,
0.0729849488,
0.0808981135,
0.0420652218,
-0.1817387491,
0.4776446819,
-0.2574471831,
0.2789970636,
0.3836796284,
0.1822804511,
-0.0755050629,
-0.3026513755,
0.0588993654,
0.0483252704,
-0.0166239403,
0.0011704564,
-0.172883451,
0.3591368794,
0.0592461377,
0.0952931195,
-0.3097134233,
0.3031128645,
-0.0461055897,
-0.1465948075,
-0.495123893,
-0.1190968156,
0.3968293071,
-0.3081522584,
0.2047472447,
-0.1520290226,
0.0571072698,
0.0319926552,
0.1416762024,
0.1343101561,
0.1265489757,
-0.0061655268,
-0.1617846787,
-0.0100209564,
-0.0072227493,
-0.139654249,
0.5526064634,
0.193341434,
0.2127412409,
0.1210232452,
0.0394825898,
-0.2031922936,
0.0009546131,
0.0007295832,
-0.1463172585,
0.1554468274,
0.2229654044,
0.1337161511,
-0.2336521149,
-0.0772270262,
-0.3163040876,
0.0065661538,
-0.3755757213,
0.0749984831,
-0.1550009698,
0.4108362794,
0.0599399731,
0.0543512292,
-0.2262233496,
-0.2606705129,
0.2092503011,
-0.1175967306,
-0.0695594773,
0.3137814403,
-0.1835401803,
0.5583069324,
0.0173082873,
-0.0202200655,
-0.2246251404,
-0.2112014145,
-0.0854689926,
-0.1483571529,
-0.1897706091,
0.2323058844,
0.1300983876,
-0.1664834619,
0.0585788414,
-0.0695102364,
-0.3264134228,
0.0377150513,
0.0303661451,
0.6761791706,
0.2347933799,
0.0001519956,
-0.2518475652,
0.0283684023,
0.4868872166,
-0.4140204191,
-0.2675405443,
0.0591723397,
0.019523561,
0.0564182661,
-0.2959037125,
-0.3545252681,
0.0704139695,
-0.3689334989,
0.5647744536,
0.0072405487,
-0.0687927157,
-0.2838416696,
-0.0883291364,
0.1034516916,
-0.2374042273,
0.2703515291,
-0.3315574825,
-0.703283906,
0.3768455982,
-0.3030436039,
-0.2822327614,
-0.0451205038,
0.0785691887,
0.428989619,
-0.0625868812,
-0.5393746495,
-0.4895907044,
0.0535352603,
0.1895357221,
-0.010536043,
0.1493694782,
0.5207619071,
-0.1010601744,
-0.0189073719,
-0.0783481598,
-0.2318596542,
0.6520734429,
-0.0872168243,
0.1458658576,
-0.1006412879,
0.416723758,
0.0971856266,
0.876550436,
0.5474236012,
0.2963706255,
0.2949183583,
0.1211512387,
0.3381681144,
-0.026252009,
-0.1907773614,
0.4884670079,
0.2340402305,
-0.1446775496,
0.2166245878,
0.1231644899,
0.1313861758,
-0.173493728,
-0.0926732644,
-0.3810845912,
-0.3277295232,
-0.1055307463,
-0.1176436096,
0.2585914135,
-0.0223613344,
-0.0434568375,
-0.227435708,
0.0631560236,
0.4233259857,
0.3457271457,
0.2006039023,
0.2726465464,
-0.1781518012,
0.0196407512,
-0.5355626345,
0.3346012533,
-0.4541780353,
0.0489896461,
-0.2220095694,
-0.2432730198,
0.1035630554,
0.0479664356,
0.5427533984,
0.0469461232,
-0.1899239719,
-0.0557614863,
-0.2372996956,
-0.4443705976,
-0.1240645647,
-0.0472362787,
-0.062691085,
0.2470573783,
0.3423605859,
-0.5057764649,
-0.1438985169,
-0.1255825758,
-0.1004825681,
-0.306489259,
-0.3933877647,
-0.3682079911,
-0.0740959942,
-0.2561746836,
0.1216928065,
-0.2381001711,
0.035732884,
-0.0201440454,
0.0438736677,
-0.1827232838,
-0.0074397624,
-0.0261026807,
0.1901349723,
0.3149790764,
-0.3515064716,
0.2579290867,
0.5438997149,
-0.0235523544,
0.0557704866,
0.4056893587,
0.0491004586,
-0.0081696548,
0.2634592652,
-0.0709715113,
0.2871201634,
0.5680273175,
-0.2031902671,
0.0198511016,
-0.0444086827,
0.4116784632,
-0.1413416117,
-0.1564926207,
0.2634868324,
-0.1303110421,
0.3417814672,
0.2045197785,
0.0988922566,
0.2688887119,
-0.175383091,
0.2153656781,
-0.1017573923,
-0.1341550797,
0.3826417625,
0.016426336,
0.9436199665,
0.2054094821,
0.0592286475,
0.1934204102,
0.1771511883,
0.3641619682,
0.087619096,
0.0328881629,
-0.205556035,
-0.2237048298,
0.0278854072,
-0.120230481,
-0.0501212515,
-0.2236299515,
-0.2545050085,
0.4385680556,
-0.2676753104,
-0.2376934886,
-0.3451877534,
-0.152562663,
-0.2159320563,
-0.2082404196,
0.0347566567,
0.0756387115,
0.0542348325,
0.534883678,
-0.2138372511,
-0.1893494725,
-0.3338548839,
-0.308246851,
-0.071499072,
-0.0873033479,
-0.10677699,
0.0069393329,
0.3344351351,
-0.0596457645,
-0.2383834422,
0.2095322758,
0.4598331451,
0.0625680089,
-0.2789724767,
-0.0697871447,
0.0656181127,
0.0188158639,
0.0810706764,
-0.0724338591,
0.240883708,
-0.1024617776,
-0.1515272558,
0.0943487734,
-0.1460157633,
0.0819114298,
0.0776295513,
-0.1399122775,
-0.1492321193,
-0.0896359533,
0.0508770645,
0.1414060891,
0.0980151147,
-0.2040064037,
0.1001998186,
0.266670078,
-0.1922360808,
0.1376901865,
0.1894952655,
-0.2131112218,
-0.1155995429,
0.5188004971,
-0.2788772583,
-0.0439730063,
0.3037275076,
-0.0226098299,
-0.0551310703,
-0.0629001856,
0.0001366511,
0.4646104276,
-0.8037574291,
0.0456760675,
0.022014983,
0.0338211209,
0.0662963614,
0.2331279069,
0.4397281706,
-0.119704321,
-0.0598191172,
-0.2463519573,
-0.3375788927,
0.307607919,
-0.3285467029,
0.1088334993,
-0.3543598354,
0.1391607523,
-0.1507173181,
0.352865696,
-0.3035137057,
0.0882373452,
-0.2990914583,
-0.0992690697,
-0.044594273,
-0.0650520995,
0.2502366602,
0.0224684998,
-0.0775945559,
0.0911416113,
-0.3674953282,
-0.1012230739,
-0.221525833,
0.140322417,
-0.0527468435,
0.0371609032,
-0.0516503826,
0.17350474,
-0.0011845957,
-0.1697571725,
0.0934672803,
0.2170433104,
-0.0037537254,
0.0508840494,
0.1405668259,
0.0473800525,
-0.113700375,
0.2166004181,
0.1527902633,
-0.2643917799,
-0.171489507,
0.0890153274,
-0.3603462577,
-0.0957002118,
0.2447440177,
0.0772196352,
-0.1432611793,
0.2082004547,
0.0476755053,
0.0165244564,
-0.2506272793,
0.0097152665,
0.4669169486,
0.1286634803,
-0.1391873658,
0.0958867669,
0.0040331259,
0.1172939092,
-0.0484069325,
0.3132459223,
0.0363496542,
0.336686343,
0.0685615391,
0.2757846713,
0.2371046692,
-0.1068969145,
0.1015348211,
0.2575312257,
0.0356711484,
-0.1651239097,
0.2285100967,
-0.1312102973,
0.082456477,
0.3728132546,
0.6446863413,
0.1242337972,
0.214391917,
-0.1020228416,
-0.0754538625,
0.2085906118,
-0.1400793791,
0.3282518089,
-0.1582532227,
-0.1204885095,
-0.125169307,
-0.0651012957,
0.0011578947,
-0.2720156908,
-0.3687114716,
0.6771074533,
-0.3241756558,
-0.3469701707,
-0.1790501773,
0.3376559615,
-0.1456153989,
-0.106958285,
-0.2152795494,
-0.1884308606,
-0.1496013701,
-0.1087997556,
0.2406615317,
0.0445789397,
0.2287882864,
0.0562944934,
0.0164168552,
-0.2694602609,
-0.4721547365,
0.2357537448,
0.1594559103,
-0.1011682898,
0.053424485,
0.234387219,
-0.3595523834,
0.1360554397,
0.7672566772,
0.2582758367,
0.075595364,
-0.0845423937,
0.0233775489,
0.267013967,
0.047937654,
-0.1087183505,
0.1379795671,
0.1853433698,
-0.4984887838,
0.0855315551,
0.1060677022,
-0.1725194603,
0.3995276392,
-0.3463377059,
0.3304549456,
-0.193364799,
0.4167948961,
-0.198838681,
0.23099567,
0.0298630483,
-0.2229298949,
-0.7316350341,
0.3039992154,
-0.117372945,
-0.0747350305,
0.1165497452,
-0.3806371391,
0.0477004424,
-0.0271399785,
0.1896597147,
0.3733299971,
0.049493771,
-0.3342291415,
-0.2568193078,
-0.5867596865,
0.3230748475,
0.1014963984,
0.3235939443,
-0.0898159593,
0.3570632041,
-0.0974762365,
0.1376841366,
0.1949174404,
0.1135402694,
-0.1029614359,
0.071287699,
-0.4963091612,
0.1593197584,
-0.1107428372,
-0.1466664076,
0.1599317044,
0.2845370173,
-0.048319865,
-0.2016776949,
0.027036719,
0.2672986686,
-0.1220198274,
-0.1107843369,
0.1535865217,
0.3878782392,
0.2212018967,
0.1139355674,
0.0036388785,
0.0192444697,
-0.2425034642,
-0.1357336491,
-0.1199789345,
-0.1062790304,
-0.2875248194,
0.3848075271,
-0.3711641729,
-0.391548574,
-0.3861626685,
0.2148723453,
0.0746119469,
-0.1054603234,
-0.1129575223,
0.3017436862,
-0.0959619209,
-0.0575610213,
0.2854053974,
0.3148619831,
0.001485764,
0.4434119463,
-0.24270612,
-0.159681052,
0.7152581215,
-0.4597455859,
0.2641887069,
-0.2539714873,
0.393712908,
0.3925568461,
-0.4172932804,
-0.3282020092,
0.0121975243,
0.0605191886,
0.0799088925,
-0.161070466,
0.0920376927,
-0.3906289339,
-0.1147466227,
-0.0314645059,
0.1399295926,
0.2236700207,
-0.0557532161,
0.0957489535,
-0.0571784303
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | Thank you for explaining that in a great way, @lhoestq
So the bottom line is that the `transformers` examples are broken since they don't do any of that. At least `run_seq2seq.py` just does `metric = load_metric(metric_name)`
What test would you recommend to reliably reproduce this bug in `examples/seq2seq/run_seq2seq.py`? | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 48 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
Thank you for explaining that in a great way, @lhoestq
So the bottom line is that the `transformers` examples are broken since they don't do any of that. At least `run_seq2seq.py` just does `metric = load_metric(metric_name)`
What test would you recommend to reliably reproduce this bug in `examples/seq2seq/run_seq2seq.py`? | [
0.0750928074,
-0.1511461586,
0.156808883,
0.1809434593,
-0.0645124316,
-0.0121278092,
0.3743320405,
0.2074542642,
0.2033593804,
0.0637144372,
0.2323808074,
0.1092874855,
-0.3682082891,
-0.2463030219,
0.119103983,
-0.1598071456,
0.0618002228,
0.0787237883,
-0.3983022571,
-0.1459853053,
-0.2288623005,
0.597700417,
-0.2141058147,
0.0435035229,
-0.5726748109,
0.2785583138,
-0.2246072441,
0.1813752949,
0.1039818674,
-0.5100975633,
0.4914011359,
0.107530117,
-0.0393090956,
0.6184604764,
-0.0001310607,
-0.1159780622,
0.2905943394,
-0.2518159449,
-0.1454315633,
-0.2269470841,
-0.1584107876,
-0.2461211383,
0.449108988,
-0.205394417,
0.0618128851,
0.0542067923,
0.1260380894,
-0.1677696556,
0.4823169708,
0.0367281064,
0.0278774984,
0.48978284,
-0.1625123173,
-0.3433549702,
0.0686693043,
0.1456460208,
0.0055331141,
0.2566852868,
-0.0682887137,
0.0118007585,
-0.2210910469,
0.2391412258,
0.2916743457,
-0.0231560078,
0.2404491156,
0.0032987036,
-0.0292705446,
-0.0683815032,
-0.1375893801,
0.3957452178,
0.2470198274,
-0.1506506652,
-0.3314956129,
-0.0105178133,
-0.3707768917,
-0.1938935518,
0.3266065121,
-0.1777873188,
0.0734511837,
0.1111944914,
-0.0882412046,
-0.119377777,
-0.0133480132,
0.0675632805,
-0.2427283823,
0.0775503144,
-0.0889168009,
0.0202898476,
-0.0826742202,
0.0575079061,
-0.2245979458,
0.2535657883,
-0.0392754823,
0.1483735144,
-0.3013090789,
-0.0075877085,
0.156876877,
-0.0926357433,
0.0377058536,
0.399567157,
-0.0197045282,
0.1337977648,
0.3218589127,
0.1455344409,
-0.0883822739,
0.563924551,
0.2473169863,
0.195664838,
0.142361179,
0.2767656446,
-0.0769011676,
-0.2230590135,
0.1054572687,
-0.1692460179,
0.3477951586,
0.0648900121,
0.0919377059,
-0.4264973104,
0.1201482192,
0.2431018502,
-0.0429435149,
-0.147258088,
0.1478132904,
0.4604138136,
0.1981287748,
0.0761528537,
0.1781303585,
0.3603520393,
-0.1025500521,
-0.0604585893,
-0.3435695767,
-0.1752094626,
-0.2113114744,
0.2515895963,
0.2330031097,
0.1246437132,
0.6372713447,
-0.2465672046,
0.3610939085,
-0.1116094589,
0.3067769706,
0.2852132916,
0.1577808708,
0.4128850102,
-0.1351287216,
0.2474035472,
0.3334743381,
0.0782880485,
-0.2614816129,
-0.1910321414,
-0.1088007092,
-0.4993450344,
0.2095870227,
-0.059375979,
-0.6320044398,
0.2099976838,
0.3444993794,
0.1252349317,
-0.0880123526,
-0.290630132,
0.1899116337,
0.0113038644,
0.0194319785,
0.0363645963,
0.5021125078,
0.8084045649,
-0.1071284413,
-0.5354681611,
-0.0404285751,
0.0006154682,
-0.0503949225,
0.1723443568,
-0.0216382146,
0.1558130682,
-0.2820472717,
0.0270244628,
0.3047232628,
-0.3449213505,
-0.3754403591,
-0.042483665,
-0.2123759389,
-0.0130924135,
0.1726594269,
-0.0992883369,
0.1596010178,
0.0654158667,
0.0390033238,
-0.1574853212,
0.1784495264,
-0.1477155089,
-0.2838846743,
-0.3624427617,
0.0209675021,
-0.0074380711,
0.3962644041,
0.1487718523,
-0.0420768484,
0.2471258044,
-0.0140903629,
0.154695034,
0.120830968,
0.2546758354,
0.1531658769,
-0.2394434363,
0.3301900029,
-0.0040533096,
-0.3316981494,
0.3137978315,
-0.412864387,
-0.0016719704,
0.1717616469,
0.0299322382,
-0.3266586661,
-0.1333886087,
-0.0559239388,
-0.4057342112,
-0.0640612245,
-0.06816753,
0.118989706,
0.0956795141,
-0.1376006156,
0.0342140123,
0.2938392162,
0.246194303,
-0.499899447,
0.0841351077,
0.0144147445,
-0.2128109634,
-0.2256053537,
0.0138927707,
-0.011057945,
-0.1588862985,
-0.2033102214,
0.3690860271,
-0.066735357,
0.371739924,
0.2778249681,
0.3218528628,
0.0117867757,
-0.4765182734,
-0.0597843379,
0.274915725,
-0.0537500232,
-0.0168479681,
-0.456501931,
0.3429723084,
0.1099391058,
0.0574720502,
-0.2757980824,
0.3847723603,
-0.036701303,
-0.2347461134,
-0.5471621752,
-0.1155283898,
0.4180025458,
-0.2595297098,
0.2554792166,
-0.1064639241,
0.1143486351,
-0.207110405,
0.1366579533,
-0.0392937325,
0.2872011065,
0.152462557,
-0.3554169536,
-0.0619290769,
0.0335067511,
-0.0804223716,
0.3691936731,
0.0703882426,
0.2867378891,
-0.0314375013,
0.0858654827,
-0.1533984542,
0.1686101258,
0.1308804601,
-0.1188943684,
0.1236363426,
0.2123599201,
0.1744195819,
-0.2839015722,
-0.1667080373,
-0.2365745008,
0.0640980676,
-0.3988455534,
0.1490959525,
-0.0114228353,
0.5295213461,
0.0435039401,
-0.0604519881,
-0.2244169265,
-0.2906615436,
0.157187134,
-0.2209062874,
-0.0536588728,
0.3262693882,
-0.2227447629,
0.6643279195,
0.052576825,
-0.1865572482,
-0.4059666395,
-0.0892626196,
-0.1126827896,
-0.1945125908,
-0.2627269626,
0.0734461695,
-0.1226370707,
-0.1925714612,
0.080547981,
0.1065049842,
-0.4834923744,
0.1850721538,
0.2666202188,
0.6751851439,
0.3146786094,
-0.0235575587,
-0.1379268765,
-0.0140278637,
0.3056757152,
-0.4094473422,
-0.2437089682,
0.3083648086,
-0.0299441256,
0.157907933,
-0.297554791,
-0.2534156442,
0.163084507,
-0.259406805,
0.5369827747,
0.0357930139,
-0.0390029699,
0.048677031,
-0.048249159,
0.1972008049,
-0.2278415859,
0.2416341305,
-0.0970919952,
-0.4955851436,
0.493900001,
-0.0899772048,
-0.2325497568,
-0.2330012172,
-0.2596037984,
0.2908658087,
-0.0499923043,
-0.4904943705,
-0.2846316099,
-0.1554354131,
0.1077006459,
0.0514275953,
0.0468778014,
0.5173735619,
-0.0740087181,
0.0814280882,
-0.1966249943,
-0.1054343134,
0.4829235077,
-0.0403454155,
0.2072460353,
-0.1482867151,
0.2523771822,
0.1457010359,
0.9683127999,
0.4021626115,
0.210883826,
0.1378674656,
0.1212738231,
0.2691400647,
-0.0327656306,
-0.1612856388,
0.129048273,
0.1736800075,
-0.2389680892,
0.1625223309,
0.1619471014,
0.0524047837,
-0.0170535073,
0.0480593741,
-0.3372256458,
-0.3736347258,
0.0258929338,
-0.286744386,
0.2674738765,
0.0084654838,
0.1298522502,
-0.2105976939,
0.0496193729,
0.3940190673,
0.3825775087,
0.2130386084,
0.1041271538,
-0.1515607536,
0.1164904833,
-0.434096396,
0.1125115901,
-0.4435678422,
0.2958863974,
-0.1493602097,
-0.347263366,
-0.0203619264,
-0.0741199329,
0.7864437103,
0.129467845,
-0.0121836364,
-0.0056255758,
-0.1805306971,
-0.5279908776,
-0.1744871438,
-0.2424480617,
-0.0621801428,
0.3286204636,
0.41731897,
-0.3465722799,
-0.0914419815,
-0.0846183151,
-0.1157487258,
-0.2575942576,
-0.1866008043,
-0.3307990134,
0.0978285372,
-0.1791492105,
0.1057736725,
-0.1725385487,
-0.1363509744,
0.003687948,
0.1023929566,
-0.1907086372,
-0.0807413608,
0.0790348276,
0.3291589022,
0.3670134842,
-0.4262425601,
0.2056540847,
0.5175893307,
-0.3233253658,
-0.1374787986,
0.3022693694,
0.1475374252,
-0.1070956588,
0.4146631062,
-0.0502334163,
0.160339728,
0.5822025537,
-0.2052154094,
-0.0157066733,
0.0857766196,
0.3517507613,
-0.1212827563,
-0.0598321632,
0.2842307687,
-0.1013542414,
0.2932529747,
0.2780109048,
0.4115169048,
0.2014755458,
-0.2698428631,
0.284883976,
0.020326104,
-0.155841291,
0.5100287795,
0.0391435772,
1.0379657745,
0.1548093259,
0.2015151232,
0.1303995401,
0.1648876816,
0.4351429343,
0.1203606427,
0.0437972136,
-0.2623808384,
-0.3548328876,
-0.0181119367,
-0.281270802,
-0.0030668015,
-0.14458552,
-0.4378489256,
0.3533286452,
-0.3426380157,
0.0178090464,
-0.0403170921,
-0.2422981858,
-0.2960458994,
-0.1201756001,
0.1541962922,
-0.0456660911,
0.1980380565,
0.6490921974,
-0.2589024305,
-0.1568102688,
-0.3486699462,
-0.412861973,
-0.2985281348,
0.04459887,
-0.071156539,
-0.0084000751,
0.5609118342,
-0.0884364247,
-0.0502098016,
0.2643803954,
0.4029280245,
0.1458510906,
-0.4677395225,
0.1415472776,
-0.0590565614,
0.1639374495,
0.1763229072,
0.0302408338,
0.1974543482,
0.0159925893,
-0.081160821,
0.0317446142,
-0.2748937309,
0.161243692,
0.0895965099,
-0.1340848505,
-0.2504485846,
-0.0825364292,
0.0357696861,
0.133822307,
-0.0820523277,
-0.2147383243,
0.0269698612,
0.2467702925,
-0.2292547226,
0.0371177606,
0.1516254544,
-0.2628642321,
-0.0579268113,
0.4928116202,
-0.3625274599,
-0.0921286345,
0.3880809844,
-0.0145977885,
0.0954931676,
-0.0928231031,
0.1040696502,
0.4269270897,
-0.976048708,
0.136995554,
-0.3507162333,
-0.186985895,
0.0663213283,
0.2257802784,
0.622593224,
-0.1898292601,
-0.1334913373,
-0.1952227056,
-0.3736695051,
0.3917947412,
-0.1390204728,
0.1900529712,
-0.5208063126,
-0.0400967188,
-0.0909902379,
0.2423180789,
-0.1821932644,
0.2231492996,
-0.2955843806,
-0.0602547005,
-0.1686765105,
-0.0956607386,
0.3259903491,
-0.195773989,
-0.1448483318,
-0.0011048159,
-0.3109072447,
-0.0424052104,
-0.3904666305,
0.1731739342,
-0.0693072453,
0.0203191414,
-0.1117979512,
0.4331067502,
0.0426070653,
-0.225369662,
0.0421761349,
0.0808040947,
0.0257406011,
0.296879828,
0.1680522859,
0.0712302178,
-0.2817440629,
0.0882969201,
0.1818268448,
-0.1884129345,
0.0025737584,
0.1377197355,
-0.3446249366,
-0.0645118356,
0.0776907206,
0.2567935288,
-0.1901199967,
0.0443111286,
0.1996379793,
-0.0354053266,
-0.1188759208,
-0.0554231629,
0.3305152059,
0.1367084086,
-0.1593909264,
0.0950240195,
-0.088604942,
0.0568299219,
-0.080278188,
0.1048178524,
0.0006987806,
0.3218780458,
0.0853812397,
0.3245590329,
0.1636345983,
-0.162279889,
0.3574337661,
0.1928838044,
0.1937337518,
-0.1373788565,
0.3820771277,
0.0233726744,
-0.0269826576,
0.4478302598,
0.451541394,
-0.0703796148,
0.279891938,
-0.1402744353,
0.0753537863,
0.2244551331,
-0.2527286112,
0.3090748787,
-0.0772219598,
-0.1782060713,
-0.1137572452,
0.0177016854,
0.0819643214,
0.0149580864,
-0.2574415505,
0.5996485949,
-0.3609506786,
-0.1359236985,
-0.3058481812,
0.0015194165,
0.0109869502,
-0.2694581449,
-0.010166876,
-0.1457813382,
-0.219404161,
-0.2321439087,
0.2223369926,
0.0849190876,
0.0927909315,
0.0869168788,
0.0281846747,
-0.1800114214,
-0.2946360111,
0.3112849593,
0.229245469,
-0.2502248883,
0.1192081198,
0.2937878072,
-0.3188092411,
0.0803373754,
0.6347103715,
0.2948590815,
-0.0884076953,
-0.0761148036,
0.1265599728,
0.1849352121,
-0.0109212957,
-0.0974580497,
0.2389880568,
0.1733165383,
-0.3825126588,
-0.0470268987,
0.0349704996,
-0.1180282086,
0.3365170956,
-0.2178982049,
0.1758396626,
-0.2750439644,
0.5571631789,
-0.2242749184,
0.0447617024,
0.1200988889,
-0.1157296747,
-0.604125917,
0.2573817074,
-0.1305815428,
-0.0253199674,
0.1239656359,
-0.0469639972,
-0.0171992593,
0.0293214694,
0.2097139508,
0.4323585629,
0.2327560335,
-0.270385623,
-0.2339159846,
-0.5324940085,
0.2602302432,
0.1255424172,
0.3936371803,
0.2322480083,
0.2296848148,
-0.0150971226,
-0.0309844427,
0.1089069694,
-0.0064852312,
-0.1906320155,
0.0576797798,
-0.4498602748,
0.0906725973,
-0.0608776249,
-0.0421850309,
0.0336841568,
0.0912771374,
-0.0197498966,
-0.0872879773,
-0.0868773162,
0.1656350791,
-0.2633647919,
-0.0535271466,
0.1447156668,
0.3367939591,
0.1615676582,
0.070088163,
0.0182316154,
-0.1633972824,
-0.2470242381,
-0.191880852,
0.0415444598,
-0.2905659676,
-0.3666418195,
0.4213337302,
-0.292668432,
-0.3824875951,
-0.3238880634,
0.1042758077,
0.1808905751,
-0.104052566,
-0.3937338591,
0.3160434663,
-0.2724987864,
0.0433946289,
0.2825976014,
0.3950484395,
-0.1257578135,
0.3722695112,
-0.2413479984,
-0.0346612297,
0.6833049655,
-0.5341157913,
0.1998462379,
-0.3363789022,
0.3415118456,
0.2691202164,
-0.3390434086,
-0.3327307105,
-0.1527007669,
0.1526992172,
-0.0082968203,
-0.1418137103,
0.0664727539,
-0.1462515146,
-0.0786667317,
-0.0801922083,
0.418396771,
0.1686917394,
-0.092417188,
-0.0271846876,
-0.1010505259
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | To give more context, we are just using the metrics for the `comput_metric` function and nothing else. Is there something else we can use that just applies the function to the full arrays of predictions and labels? Because that's all we need, all the gathering has already been done because the datasets Metric multiprocessing relies on file storage and thus does not work in a multi-node distributed setup (whereas the Trainer does).
Otherwise, we'll have to switch to something else to compute the metrics :-( | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 85 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
To give more context, we are just using the metrics for the `comput_metric` function and nothing else. Is there something else we can use that just applies the function to the full arrays of predictions and labels? Because that's all we need, all the gathering has already been done because the datasets Metric multiprocessing relies on file storage and thus does not work in a multi-node distributed setup (whereas the Trainer does).
Otherwise, we'll have to switch to something else to compute the metrics :-( | [
0.0360684246,
0.0379046649,
0.0943587273,
0.1253584921,
0.1760238707,
0.0507226288,
0.181993857,
0.2530151904,
0.202546075,
0.2519216537,
0.0820865035,
0.1800006926,
-0.3751122653,
0.0665903836,
0.1416406035,
-0.0772905946,
0.0167337395,
0.0033796467,
-0.3133097887,
-0.1381403208,
-0.2079619914,
0.4481795132,
-0.1228017658,
0.0743642151,
-0.5695955753,
0.1498473287,
-0.1701160669,
0.0805466473,
-0.1255109608,
-0.5812904239,
0.3919233382,
0.1458095014,
0.0017235428,
0.4191239476,
-0.0001262029,
-0.1343240142,
0.1936454922,
-0.2333319336,
-0.1047418714,
-0.222505942,
0.126871556,
-0.3101152182,
0.5224649906,
-0.1460824311,
-0.0495463721,
-0.0838490203,
0.0543666258,
-0.421436727,
0.4781624675,
0.1183949113,
0.0403271392,
0.3758565187,
0.0349310786,
-0.2492557317,
-0.0176794194,
-0.0774545819,
0.0102451369,
0.3299075067,
-0.0517304204,
-0.0297791902,
-0.2794109881,
0.243684113,
0.2794157267,
-0.0088511799,
0.3128017187,
-0.0297789015,
-0.0145227127,
-0.1737944037,
-0.1312470287,
0.2608568668,
0.1949299872,
-0.2089758813,
-0.3556878567,
-0.0132942051,
-0.0411428697,
-0.5314989686,
0.1682944298,
-0.107542336,
-0.033615835,
0.0474925637,
-0.1316339672,
-0.1672553569,
-0.0719780326,
0.0874652267,
-0.17601794,
0.0398991182,
-0.1421070695,
0.0187714472,
0.0959141999,
0.2716262639,
-0.2097333074,
0.4107066989,
0.1014253795,
0.1463738233,
-0.3606152833,
0.0825560167,
0.1721965075,
-0.0282379761,
0.0502633601,
0.395039022,
0.0560897365,
0.1017730087,
0.3755907714,
0.0602637455,
-0.0459785163,
0.4311771989,
0.2344249785,
0.1357564926,
0.1160663292,
0.2224401534,
-0.136713773,
-0.2810333967,
0.1833155304,
-0.1449776739,
0.3832839131,
0.224560529,
-0.1369666159,
-0.3769401908,
0.0637288913,
0.2464853525,
-0.0436972678,
-0.2041596472,
0.196501717,
0.2750817537,
0.1341590285,
0.1387402415,
0.1868601143,
0.3029894233,
-0.1722473502,
-0.1656983197,
-0.3989242315,
-0.0746450275,
-0.2094955295,
0.3997784853,
0.1894151568,
-0.0411997996,
0.6616479158,
-0.3590565324,
0.5086172819,
-0.2140401602,
0.1602291912,
0.1458079815,
0.0974275544,
0.2898179889,
-0.0214463472,
0.099196814,
0.3203162551,
0.0583865419,
-0.1933313608,
-0.3334258199,
-0.2335585207,
-0.5762784481,
0.1160745397,
-0.0552394837,
-0.5121428967,
0.2240831852,
0.1648789048,
0.3089998066,
-0.153039977,
-0.1348861456,
0.13651748,
-0.0361999124,
-0.1778377295,
-0.0181774497,
0.6248615384,
0.5545915365,
-0.3065956831,
-0.4664388895,
0.1092805117,
-0.010735509,
-0.1018489003,
0.1589731425,
-0.0603356548,
0.3344111443,
-0.1820726097,
0.0870404541,
0.4544461071,
-0.408008635,
-0.4083964527,
-0.2533524036,
-0.2293062061,
-0.1433101594,
0.0884947181,
-0.0339344516,
0.1266493648,
0.2276445329,
0.075058654,
-0.2100951672,
0.1428103298,
-0.3406259716,
-0.2292550802,
-0.256660372,
-0.0860025883,
0.0165199935,
0.4695231915,
-0.0004946962,
-0.0502111837,
0.0669919178,
0.0552923977,
0.027339166,
0.2316124141,
0.2274736911,
0.1849207729,
-0.0041502286,
0.1867883056,
-0.0504859015,
-0.2215744555,
0.3026395738,
-0.2930178046,
0.0360223018,
0.1714073122,
-0.0174643546,
-0.1159895957,
-0.0948146284,
-0.0791351199,
-0.3731916845,
-0.0357884616,
-0.0878941417,
0.2329109758,
0.1419443339,
-0.22978127,
0.1230703667,
-0.0462219939,
0.2593325078,
-0.3241707981,
0.1597562581,
-0.060949944,
-0.3210315406,
-0.0040194765,
0.2121326029,
-0.1006767154,
0.0424052253,
-0.1941399574,
0.369818747,
-0.1435663104,
0.1763958782,
0.3263868093,
0.1948135942,
-0.094641991,
-0.4680906534,
-0.186837852,
0.02265843,
-0.0780186504,
-0.0425729305,
-0.1915084273,
0.4211588502,
0.1166607514,
0.1690179855,
-0.2548600435,
0.3526451588,
-0.1358471364,
-0.0465469174,
-0.5137107372,
-0.1119696945,
0.3400013447,
-0.422224015,
0.3011054397,
-0.183504954,
-0.0250695571,
-0.054298345,
0.2423658073,
0.1083919331,
0.1255672872,
0.1039019078,
-0.2200977057,
-0.0708635226,
-0.0215850547,
-0.1932944804,
0.4366635978,
0.1965452135,
0.2774855196,
0.0860511735,
0.0942834318,
-0.1230534837,
0.07706514,
0.1215848178,
-0.0986285359,
0.0737261772,
0.1942547858,
0.1487439871,
-0.3226960301,
-0.0074758008,
-0.3011006713,
0.0195340943,
-0.3337207735,
0.0596702173,
-0.2368711978,
0.3630673289,
0.1683838665,
-0.002191402,
-0.2333551049,
-0.270606488,
0.2153630555,
-0.1962650567,
-0.0221890844,
0.3150246143,
-0.2984646559,
0.6049126387,
0.0615460426,
-0.0226249062,
-0.2279763222,
-0.1495146006,
-0.0368127301,
-0.1780065894,
-0.1940801591,
0.096849367,
0.0374842212,
-0.1499426365,
0.142657578,
-0.1025382727,
-0.3601032197,
0.0933406055,
0.0195363611,
0.6119940281,
0.3463464379,
-0.0454068184,
-0.3527433872,
0.0991577804,
0.4500067532,
-0.4079984426,
-0.2767027318,
0.1133298874,
0.031444665,
0.1227022931,
-0.2230904251,
-0.2890613675,
0.0112665016,
-0.2815250754,
0.5991315842,
0.041066356,
-0.0020703636,
-0.2288424224,
-0.023793336,
0.177327767,
-0.3290058672,
0.2469101846,
-0.2935866416,
-0.7149358392,
0.4781394005,
-0.0908920169,
-0.2723421156,
-0.1058706641,
0.0675317347,
0.4347687364,
-0.0653754398,
-0.6561147571,
-0.4437874556,
0.0903980359,
0.0918386281,
-0.0425772257,
0.1582766473,
0.5028114915,
-0.064025484,
0.0077044666,
-0.1060189232,
-0.2926114202,
0.5380766392,
-0.1749774367,
0.1555450708,
-0.2270887494,
0.2927684784,
0.2596622705,
0.9461429119,
0.5229728222,
0.1494289488,
0.2371039987,
0.1094769835,
0.4445637763,
-0.0699855536,
-0.1495136619,
0.3707893491,
0.1129484922,
-0.0162511095,
0.2179200053,
0.0888905674,
0.1427088082,
-0.1463133395,
0.0634610653,
-0.2214709967,
-0.330262661,
-0.0235036276,
-0.139238745,
0.1961390376,
0.0105118826,
0.1161013469,
-0.2785870731,
-0.0160789303,
0.3919545412,
0.2068580389,
0.182901144,
0.2792763412,
-0.2079497278,
-0.1412259787,
-0.5621311665,
0.2452994287,
-0.3804304898,
0.045309756,
-0.1028642282,
-0.2098943442,
0.1184138283,
0.0432681702,
0.4675578475,
0.0269830171,
-0.249911353,
0.0441781282,
-0.1463385522,
-0.3195357621,
-0.1433724165,
-0.0926066414,
-0.0362705812,
0.2236896902,
0.363689065,
-0.3087711334,
-0.2344550192,
-0.0264692903,
-0.0768966079,
-0.2622458339,
-0.230419457,
-0.3326174915,
0.0181398615,
-0.1006649137,
0.2380574495,
-0.2081857324,
0.0290682353,
0.0581596494,
0.0281244814,
-0.1481355578,
-0.0308293365,
0.0667480826,
0.1422158778,
0.2505323291,
-0.2525005937,
0.1854339242,
0.5488846302,
-0.0489840433,
0.0482749045,
0.3769325316,
-0.0075409585,
-0.0765793398,
0.2972931862,
-0.1136458814,
0.4294686019,
0.5456273556,
-0.2386381477,
-0.0288395807,
0.0150821917,
0.3863062859,
-0.1887577623,
0.0419143476,
0.3080279231,
-0.0701057911,
0.3514485955,
0.2013500035,
0.2649248838,
0.3189355731,
-0.1359725893,
0.3833769262,
0.145275414,
-0.1684725583,
0.2444077134,
0.1406890005,
0.9382525682,
0.0755253807,
0.1895242184,
0.1168168038,
0.1613007486,
0.556883812,
0.0864991993,
0.0911396593,
-0.0336713567,
-0.337688148,
-0.0298737884,
-0.1386369169,
-0.0565792769,
-0.2417142987,
-0.1862827092,
0.4872968495,
-0.2020579129,
-0.1370247602,
-0.1831812561,
-0.1983543634,
-0.0345439352,
-0.2534299791,
0.1010972857,
-0.0277632214,
-0.0001072902,
0.4575576484,
-0.1977294385,
-0.1139497608,
-0.2750445306,
-0.3137762845,
-0.2320908904,
-0.0686381906,
-0.3299613297,
-0.0637360364,
0.3890515268,
-0.0742628947,
-0.0691311285,
0.2970837057,
0.4629920721,
0.083980076,
-0.345638752,
0.023330424,
0.0002665706,
0.0664637908,
0.0784114078,
0.0073580742,
0.3216219544,
-0.0662751347,
0.0102042556,
0.064730823,
-0.2347321063,
0.0867331922,
0.1223050654,
-0.1585002393,
-0.0257755257,
-0.0164250098,
0.1262873858,
0.0988368019,
0.03055574,
-0.3260997534,
0.0412871949,
0.2831498384,
-0.2210922837,
0.1518939584,
0.1960550994,
-0.2191640586,
-0.044074785,
0.5630208254,
-0.2119456083,
-0.1510917246,
0.4201501012,
-0.015550755,
-0.0242567733,
-0.0876445845,
-0.0401977897,
0.5411642194,
-0.6515188813,
0.0948459506,
-0.0788781792,
-0.0676446855,
0.0241861902,
0.2146997154,
0.4476056993,
-0.1635665298,
-0.0357842967,
-0.1667281836,
-0.4597929716,
0.3591099083,
-0.2916512489,
0.1436112374,
-0.4845038652,
0.2159707695,
-0.1050218493,
0.3118295968,
-0.2253037393,
0.1461494863,
-0.318045795,
-0.1291920841,
-0.0933290496,
0.0110472804,
0.2815726995,
-0.0890990645,
-0.1196709573,
0.0586083718,
-0.3794821501,
-0.0509907976,
-0.2384001762,
0.1548075974,
-0.0630954951,
0.1762827635,
0.077498354,
0.2939431965,
-0.1219547689,
-0.2028156668,
-0.0756870732,
0.1594338417,
0.0939316601,
0.1515529454,
0.0464459322,
0.166139245,
-0.0703116059,
0.0964263007,
0.2369103432,
-0.1221285164,
-0.168543756,
0.0712453872,
-0.2272640765,
-0.1083661988,
0.1468652636,
0.085944131,
-0.2312790751,
0.1614070684,
0.117042385,
0.0157430489,
-0.3260178566,
-0.0447357148,
0.4571231008,
0.1518581063,
-0.1727565974,
0.1071678624,
-0.0483281314,
0.0419552885,
-0.1785605252,
0.220438689,
0.0308612473,
0.3870634139,
0.0750013888,
0.2414397895,
0.2413668483,
-0.0906781554,
0.1284109354,
0.1479792148,
-0.11518538,
-0.1851308197,
0.3271121085,
-0.0866596848,
0.1272319853,
0.2972465754,
0.5802831054,
0.1529764235,
0.1836087257,
0.0263477974,
-0.1465581059,
0.2086514831,
-0.1388531327,
0.2844024003,
-0.205994159,
-0.092662856,
-0.241039902,
0.0577582195,
0.0592739582,
-0.1637320966,
-0.5557777882,
0.8412665725,
-0.290737927,
-0.3601016998,
-0.1830141991,
0.1928287297,
-0.0918846354,
-0.1444140375,
-0.1394795626,
-0.1471372694,
-0.1995601654,
-0.1521382034,
0.2422815412,
-0.0227825083,
0.2497122586,
0.045651339,
0.0298938677,
-0.234526217,
-0.5137044191,
0.2872683108,
0.264559865,
-0.1797735393,
0.1278647035,
0.1644396931,
-0.3647378385,
0.1035096571,
0.7005337477,
0.2391315401,
-0.0577403083,
-0.1001667082,
0.0456193984,
0.2528731227,
-0.0410718173,
-0.0698785484,
0.1960646212,
0.3556522131,
-0.5932099819,
0.1043870598,
-0.0007089935,
-0.0965858102,
0.4026237726,
-0.2877334654,
0.3789044321,
-0.2412755787,
0.4647477865,
-0.1182560623,
0.0442554504,
0.0336354747,
-0.1767842919,
-0.6990656257,
0.3790701032,
-0.0600345731,
0.0051057562,
0.0743904784,
-0.2577639818,
0.0131136887,
-0.0343389288,
0.2901058793,
0.4205662906,
0.0609131604,
-0.2994612753,
-0.2646946609,
-0.5396562815,
0.1927632987,
0.1414522231,
0.3431107998,
-0.0039901845,
0.3530113697,
0.030911494,
0.0716793537,
0.23825939,
-0.0196619108,
-0.0600883365,
-0.0279568918,
-0.6378627419,
0.1246492267,
-0.1350909919,
0.00223062,
0.1786824763,
0.3266238272,
0.0166262556,
-0.2267826498,
-0.0830881968,
0.3012847006,
-0.0886905342,
-0.1383722425,
0.3098233342,
0.4241037965,
0.0609223247,
0.181089744,
0.0859951228,
-0.0158587322,
-0.1867469549,
-0.1026429981,
-0.0907714814,
-0.0847365856,
-0.279447943,
0.4343640208,
-0.4776816368,
-0.3170906305,
-0.3601941764,
0.0708861649,
0.1506474018,
-0.1097971126,
-0.3143194914,
0.356870383,
-0.1146314293,
0.0508090593,
0.3609868884,
0.3676595092,
-0.0981444865,
0.3505343199,
-0.3764443994,
-0.1778482199,
0.7721174955,
-0.4845305681,
0.2451476008,
-0.3426681459,
0.3337122798,
0.6285012364,
-0.2959863842,
-0.5779030919,
-0.0101888627,
0.1667892337,
0.0562025383,
-0.3047437668,
0.1543694586,
-0.4789429009,
0.0177682936,
-0.0675173774,
0.3459009528,
0.1138319522,
-0.0296077803,
-0.0420081466,
-0.1831468642
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | OK, it definitely leads to a race condition in how it's used right now. Here is how you can reproduce it - by injecting a random sleep time different for each process before the locks are acquired.
```
--- a/src/datasets/metric.py
+++ b/src/datasets/metric.py
@@ -348,6 +348,16 @@ class Metric(MetricInfoMixin):
elif self.process_id == 0:
# Let's acquire a lock on each node files to be sure they are finished writing
+
+ import time
+ import random
+ import os
+ pid = os.getpid()
+ random.seed(pid)
+ secs = random.randint(1, 15)
+ time.sleep(secs)
+ print(f"sleeping {secs}")
+
file_paths, filelocks = self._get_all_cache_files()
# Read the predictions and references
@@ -385,7 +395,10 @@ class Metric(MetricInfoMixin):
if predictions is not None:
self.add_batch(predictions=predictions, references=references)
+ print("FINALIZE START")
+
self._finalize()
+ print("FINALIZE END")
self.cache_file_name = None
self.filelock = None
```
then run with 2 procs: `python -m torch.distributed.launch --nproc_per_node=2`
```
export BS=16; rm -r output_dir; PYTHONPATH=src USE_TF=0 CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2 examples/seq2seq/run_seq2seq.py --model_name_or_path t5-small --output_dir output_dir --adam_eps 1e-06 --do_eval --do_train --do_predict --evaluation_strategy=steps --label_smoothing 0.1 --learning_rate 3e-5 --logging_first_step --logging_steps 1000 --max_source_length 128 --max_target_length 128 --num_train_epochs 1 --overwrite_output_dir --per_device_eval_batch_size $BS --per_device_train_batch_size $BS --predict_with_generate --eval_steps 25000 --sortish_sampler --task translation_en_to_ro --val_max_target_length 128 --warmup_steps 500 --max_train_samples 10 --max_val_samples 10 --max_test_samples 10 --dataset_name wmt16 --dataset_config ro-en --source_prefix "translate English to Romanian: "
```
```
***** Running Evaluation *****
Num examples = 10
Batch size = 16
0%| | 0/1 [00:00<?, ?it/s]FINALIZE START
FINALIZE START
sleeping 11
FINALIZE END
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:11<00:00, 11.06s/it]
sleeping 11
Traceback (most recent call last):
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 368, in _finalize
self.data = Dataset(**reader.read_files([{"filename": f} for f in file_paths]))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 236, in read_files
pa_table = self._read_files(files, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 171, in _read_files
pa_table: pa.Table = self._get_dataset_from_filename(f_dict, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 302, in _get_dataset_from_filename
pa_table = ArrowReader.read_table(filename, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 97, in pyarrow.lib.check_status
FileNotFoundError: [Errno 2] Failed to open local file '/home/stas/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow'. Detail: [errno 2] No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "examples/seq2seq/run_seq2seq.py", line 645, in <module>
main()
File "examples/seq2seq/run_seq2seq.py", line 601, in main
metrics = trainer.evaluate(
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer_seq2seq.py", line 74, in evaluate
return super().evaluate(eval_dataset, ignore_keys=ignore_keys, metric_key_prefix=metric_key_prefix)
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1703, in evaluate
output = self.prediction_loop(
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1876, in prediction_loop
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
File "examples/seq2seq/run_seq2seq.py", line 556, in compute_metrics
result = metric.compute(predictions=decoded_preds, references=decoded_labels)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 402, in compute
self._finalize()
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 370, in _finalize
raise ValueError(
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
``` | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 452 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
OK, it definitely leads to a race condition in how it's used right now. Here is how you can reproduce it - by injecting a random sleep time different for each process before the locks are acquired.
```
--- a/src/datasets/metric.py
+++ b/src/datasets/metric.py
@@ -348,6 +348,16 @@ class Metric(MetricInfoMixin):
elif self.process_id == 0:
# Let's acquire a lock on each node files to be sure they are finished writing
+
+ import time
+ import random
+ import os
+ pid = os.getpid()
+ random.seed(pid)
+ secs = random.randint(1, 15)
+ time.sleep(secs)
+ print(f"sleeping {secs}")
+
file_paths, filelocks = self._get_all_cache_files()
# Read the predictions and references
@@ -385,7 +395,10 @@ class Metric(MetricInfoMixin):
if predictions is not None:
self.add_batch(predictions=predictions, references=references)
+ print("FINALIZE START")
+
self._finalize()
+ print("FINALIZE END")
self.cache_file_name = None
self.filelock = None
```
then run with 2 procs: `python -m torch.distributed.launch --nproc_per_node=2`
```
export BS=16; rm -r output_dir; PYTHONPATH=src USE_TF=0 CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2 examples/seq2seq/run_seq2seq.py --model_name_or_path t5-small --output_dir output_dir --adam_eps 1e-06 --do_eval --do_train --do_predict --evaluation_strategy=steps --label_smoothing 0.1 --learning_rate 3e-5 --logging_first_step --logging_steps 1000 --max_source_length 128 --max_target_length 128 --num_train_epochs 1 --overwrite_output_dir --per_device_eval_batch_size $BS --per_device_train_batch_size $BS --predict_with_generate --eval_steps 25000 --sortish_sampler --task translation_en_to_ro --val_max_target_length 128 --warmup_steps 500 --max_train_samples 10 --max_val_samples 10 --max_test_samples 10 --dataset_name wmt16 --dataset_config ro-en --source_prefix "translate English to Romanian: "
```
```
***** Running Evaluation *****
Num examples = 10
Batch size = 16
0%| | 0/1 [00:00<?, ?it/s]FINALIZE START
FINALIZE START
sleeping 11
FINALIZE END
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:11<00:00, 11.06s/it]
sleeping 11
Traceback (most recent call last):
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 368, in _finalize
self.data = Dataset(**reader.read_files([{"filename": f} for f in file_paths]))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 236, in read_files
pa_table = self._read_files(files, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 171, in _read_files
pa_table: pa.Table = self._get_dataset_from_filename(f_dict, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 302, in _get_dataset_from_filename
pa_table = ArrowReader.read_table(filename, in_memory=in_memory)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/arrow_reader.py", line 322, in read_table
stream = stream_from(filename)
File "pyarrow/io.pxi", line 782, in pyarrow.lib.memory_map
File "pyarrow/io.pxi", line 743, in pyarrow.lib.MemoryMappedFile._open
File "pyarrow/error.pxi", line 122, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 97, in pyarrow.lib.check_status
FileNotFoundError: [Errno 2] Failed to open local file '/home/stas/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow'. Detail: [errno 2] No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "examples/seq2seq/run_seq2seq.py", line 645, in <module>
main()
File "examples/seq2seq/run_seq2seq.py", line 601, in main
metrics = trainer.evaluate(
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer_seq2seq.py", line 74, in evaluate
return super().evaluate(eval_dataset, ignore_keys=ignore_keys, metric_key_prefix=metric_key_prefix)
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1703, in evaluate
output = self.prediction_loop(
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1876, in prediction_loop
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
File "examples/seq2seq/run_seq2seq.py", line 556, in compute_metrics
result = metric.compute(predictions=decoded_preds, references=decoded_labels)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 402, in compute
self._finalize()
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 370, in _finalize
raise ValueError(
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
``` | [
-0.0844156146,
-0.0657198578,
0.065217942,
0.2722369432,
-0.0870864317,
-0.0326645225,
0.164352566,
0.2697965205,
0.331193924,
0.0832053944,
0.0982610062,
0.1941851228,
-0.3580646515,
-0.0889754221,
-0.0156052075,
0.0498044901,
0.0008350275,
-0.124026753,
-0.4149838388,
-0.1004129946,
-0.3499765098,
0.5602379441,
-0.1571681798,
0.0035177283,
-0.4620627761,
0.1669690609,
-0.1928112805,
0.2385936975,
-0.1055250242,
-0.5698459148,
0.2600383759,
0.3488969207,
-0.0173992142,
0.5673112273,
-0.0001218659,
-0.1565489918,
0.2898033857,
-0.1259327382,
-0.0928893685,
-0.1496434659,
0.0516424328,
-0.3085161448,
0.4341693521,
-0.1971868277,
-0.0820829421,
-0.0098870955,
0.075830847,
-0.4951310456,
0.428683579,
0.0644461662,
0.089119032,
0.4689292908,
-0.0374805331,
-0.1865044087,
-0.0093002953,
-0.1247959137,
0.0236410201,
0.359459579,
-0.011902906,
0.0050281063,
-0.2551396787,
0.2341876626,
0.2350952178,
-0.068675831,
0.3630652726,
0.0002417788,
-0.0232570358,
-0.1683154404,
-0.1944668889,
0.3529826403,
0.1351853162,
-0.1981301904,
-0.2590883672,
0.0288158208,
-0.1893539578,
-0.4624348879,
0.1429927349,
-0.1314376295,
0.0301053748,
0.0880871043,
0.0458041243,
0.0006787926,
0.0064208806,
0.0480431132,
-0.0750623345,
0.0268833861,
-0.1102876812,
0.0966648906,
-0.0770760924,
0.2674492896,
-0.3081119955,
0.2120121717,
-0.0161632914,
0.1204223186,
-0.4356834888,
0.0894818902,
0.2162853777,
0.0534019843,
0.0726226792,
0.4313858449,
-0.031930197,
0.1022296622,
0.397911787,
0.1139197052,
-0.0671370104,
0.4132389426,
0.2391610742,
-0.0104752257,
0.1874958724,
0.2739359736,
-0.0854607671,
-0.2305917442,
0.2011027783,
-0.2459424734,
0.3607820272,
0.0959144533,
0.1355985403,
-0.3899983168,
0.0805318505,
0.3011380732,
-0.0750804022,
-0.156828478,
0.1580271125,
0.3855431974,
0.0852396041,
-0.0463730693,
0.3243302107,
0.3055721521,
-0.2333539873,
0.0019364804,
-0.429610014,
-0.2353734374,
-0.1194881126,
0.3071154356,
0.059108153,
-0.0130563304,
0.6048252583,
-0.1873205304,
0.3196307123,
-0.0907877833,
0.3079954386,
0.1744778305,
0.0746119618,
0.2866277397,
-0.0819514468,
0.1851961017,
0.3427527845,
0.1244140714,
-0.2457848787,
-0.286018163,
-0.0978622213,
-0.4152007699,
0.1885589808,
0.0005309498,
-0.506839633,
0.2394952327,
0.4041298032,
0.2187769562,
-0.0533615611,
-0.147634849,
0.2692391276,
-0.0410024002,
-0.1580678523,
-0.0170978718,
0.437728107,
0.8433098197,
-0.2321901917,
-0.4020535946,
0.1269693673,
-0.0551430285,
0.0210925099,
0.2261462361,
-0.0151570253,
0.1313891411,
-0.2769355774,
0.1363203973,
0.3145761192,
-0.3304305375,
-0.5092275143,
-0.066776745,
-0.4374721646,
-0.0340690427,
0.1925943643,
0.1099863797,
0.0847022086,
0.1234845966,
0.1166358888,
-0.2643075883,
0.2656230927,
-0.3458690047,
-0.3824670017,
-0.2897080183,
-0.0370656922,
0.0175565928,
0.3021219373,
0.0015977621,
-0.0653042942,
0.0374569632,
0.1378032267,
0.1916086376,
0.1767642498,
0.2403720021,
0.2350226343,
0.0061781034,
0.2478130162,
0.1133009121,
-0.2364608347,
0.3414189517,
-0.382301569,
0.0915213525,
0.1842736751,
0.0344476253,
-0.1589944065,
-0.1162986681,
-0.1790622771,
-0.4125054777,
0.0269137472,
-0.1299059838,
0.179064244,
0.1748146564,
-0.1957101524,
0.1845988035,
-0.031174019,
0.2445204258,
-0.4866955578,
0.0809157342,
-0.0441643372,
-0.3607321978,
-0.0655890852,
0.0390101038,
0.037620198,
0.0411499627,
-0.1389894485,
0.3688037694,
-0.1038149968,
0.2514375448,
0.2193904817,
0.2252394557,
-0.0583162084,
-0.330986917,
-0.1056121737,
0.1239269748,
0.0039938092,
-0.0451715365,
-0.1759317517,
0.3884069324,
0.0900150388,
0.1461755633,
-0.3499465585,
0.218274653,
-0.0972635001,
-0.1663037837,
-0.5556679368,
-0.0380488783,
0.4728532135,
-0.2647351027,
0.2770220637,
-0.1631319076,
0.0137210339,
0.0223105177,
0.2333166599,
0.1545524597,
0.1547575146,
0.0433732234,
-0.2333094478,
-0.0644799396,
0.0418289825,
-0.1615595222,
0.4600602686,
0.1955895126,
0.2431958765,
0.1327682883,
0.0874554589,
-0.2459523678,
0.0757679567,
0.0206987523,
-0.1691019684,
0.1593047231,
0.151480794,
0.1435247958,
-0.3288495243,
-0.1380015314,
-0.2743547261,
0.0329185277,
-0.3341396153,
0.1020790413,
-0.1349845231,
0.4001248777,
0.1546520591,
0.0645308569,
-0.1272901297,
-0.271653831,
0.1869915277,
-0.1014484465,
0.0305641517,
0.280986011,
-0.2944427133,
0.5300468802,
0.1100016162,
-0.1811982989,
-0.356182605,
-0.2111853361,
-0.0076868213,
-0.1622642875,
-0.1794359833,
0.1930832416,
0.0939459801,
-0.1835870147,
-0.0286585502,
-0.0898884907,
-0.3543089032,
0.0646217912,
0.0579993054,
0.68081218,
0.2862642109,
0.0012737848,
-0.22506392,
0.0685055777,
0.4061905146,
-0.4061918855,
-0.31842044,
0.0798089281,
-0.0271716453,
0.1367469728,
-0.3171882033,
-0.2778086066,
0.1093109623,
-0.3724222779,
0.6279049516,
0.0351588428,
-0.028473258,
-0.1151438504,
-0.0065652765,
0.2452505827,
-0.2081531435,
0.2181218266,
-0.3708446324,
-0.6790495515,
0.3761586249,
-0.0701827854,
-0.2920276523,
-0.1314426959,
0.0402267128,
0.3392744958,
-0.0791689083,
-0.5926972628,
-0.5223212838,
-0.0324286744,
0.3241590261,
0.0222351458,
-0.0101590557,
0.5464063883,
-0.0936023518,
-0.0244744681,
-0.0935699865,
-0.1157884151,
0.6498823762,
-0.1600065529,
0.1253627092,
-0.0418157503,
0.5212672949,
0.140297845,
0.9083523154,
0.5984094143,
0.1078084931,
0.2427328676,
0.08595635,
0.3766081631,
-0.1323868334,
-0.1446454823,
0.3193691969,
0.1811808348,
-0.1200466082,
0.1948687434,
0.0696766078,
0.1736307144,
-0.0670456663,
0.0006565228,
-0.2872378826,
-0.3173184395,
-0.0860210955,
-0.0220398363,
0.1946449876,
0.0260201395,
0.0464017689,
-0.1714584529,
0.0210229568,
0.4557760358,
0.4667408764,
0.2445892096,
0.1518065631,
-0.2346042395,
0.1502716243,
-0.684779048,
0.2808540165,
-0.4843687713,
0.0926618204,
-0.2083668113,
-0.2226662785,
0.1251804531,
-0.1038854122,
0.5619889498,
0.1111254469,
-0.0981001854,
0.0389464945,
-0.20032309,
-0.4362591803,
-0.0758719593,
-0.1520858407,
-0.0548395514,
0.3091388345,
0.4101701975,
-0.4372711182,
-0.2195554674,
-0.1674847901,
-0.1153565794,
-0.2638752759,
-0.3710042536,
-0.3840123117,
-0.0693474114,
-0.1728612185,
0.2685465217,
-0.1494054496,
-0.0337940529,
-0.0002120957,
-0.0176423416,
-0.1778663099,
-0.1178580746,
0.0093378499,
0.1683958471,
0.3239504993,
-0.3101278543,
0.2613334358,
0.5088198781,
-0.2863683701,
0.0070695747,
0.3720147908,
0.1125949845,
0.0114652403,
0.3537887037,
-0.0389566533,
0.1451952159,
0.6060221791,
-0.2288957834,
0.1373125613,
0.0775092691,
0.3824470937,
-0.1733540446,
-0.0328599624,
0.3560642302,
-0.0646376237,
0.3657299876,
0.1943792403,
0.2134577334,
0.255145371,
-0.2164087147,
0.3798711598,
0.0024197921,
-0.1254675984,
0.2896784544,
0.0662676245,
0.9412873983,
0.1287698597,
0.1316239983,
0.3074277043,
0.0300390273,
0.3580705822,
0.1575557888,
0.0769380927,
-0.2155002654,
-0.2878908217,
0.0281257927,
-0.1426004469,
0.0565014444,
-0.2786583304,
-0.199224785,
0.3468227386,
-0.3472925127,
-0.1368995607,
-0.2096109092,
-0.100020729,
-0.2952027023,
-0.1441335678,
0.1108286679,
0.0379241221,
0.0561158657,
0.5053175092,
-0.1962721795,
-0.1456766427,
-0.3237814307,
-0.2361003458,
-0.0989166647,
-0.0192419589,
-0.2128226012,
0.0031836554,
0.2255655676,
-0.0905263126,
-0.1694487482,
0.2623282969,
0.4520953596,
0.1580688804,
-0.2806243896,
0.0806228966,
0.1305502951,
0.0593528375,
0.1180898547,
-0.0961389393,
0.2394876778,
-0.1169214696,
-0.0971451551,
0.0897524953,
-0.241124332,
0.0596835837,
0.0625080466,
-0.1247528419,
-0.0613279194,
-0.0006808415,
0.0187082179,
0.1744978875,
-0.0133293942,
-0.2417226881,
0.0776690915,
0.3355977833,
-0.1957994401,
0.0846763104,
0.1114370972,
-0.2621310949,
-0.1653984785,
0.5599820018,
-0.1930413395,
-0.0438049361,
0.4942458868,
0.0620815754,
-0.0214787945,
-0.089595221,
0.066549316,
0.4614338279,
-0.7272006869,
0.1340646744,
-0.0469327793,
-0.0091643482,
0.1100818068,
0.212738499,
0.3792598844,
-0.1826423854,
-0.0886156559,
-0.1702746153,
-0.3497694135,
0.2320083827,
-0.3465448022,
0.1859749854,
-0.4120007455,
0.0863507465,
-0.2553622127,
0.232076019,
-0.2822773159,
0.1403431296,
-0.3109700084,
-0.1574935913,
-0.085847728,
-0.1043036431,
0.3523842692,
-0.0494927578,
-0.069843106,
0.1268847287,
-0.4691964388,
-0.1015589237,
-0.279707849,
0.1388322711,
-0.1304945052,
0.0779668987,
-0.0051859505,
0.2057713568,
-0.0304877535,
-0.0940724015,
0.018717058,
0.1957744062,
0.0266359989,
0.0420338064,
0.2091132402,
0.13990587,
-0.1949334443,
0.1630114019,
0.0852947533,
-0.2907068133,
-0.115603596,
-0.010771364,
-0.2982536852,
-0.072203055,
0.1672503054,
0.1254224479,
-0.1080611199,
0.1442046463,
0.1264521331,
-0.0076737311,
-0.3007793128,
-0.0219412968,
0.4440126419,
0.1493697315,
-0.1383354366,
-0.1004374996,
-0.1468466669,
0.0966836065,
-0.095638141,
0.2848830819,
-0.0055991206,
0.3573444486,
0.142894119,
0.257946074,
0.1733794659,
0.0006107837,
0.0838580057,
0.2874682844,
0.0338029414,
-0.1745363176,
0.2908104062,
-0.2125219852,
0.1121098697,
0.3444499671,
0.6617327929,
0.043684043,
0.199688226,
-0.1315073669,
-0.0667862967,
0.2362423539,
-0.1090041846,
0.3268864453,
-0.2201711684,
-0.1009958982,
-0.1156374067,
0.0700997934,
-0.1066151261,
-0.1336782873,
-0.378492564,
0.6685905457,
-0.3151121736,
-0.2031569183,
-0.184766233,
0.2426725775,
-0.135775283,
-0.2470301986,
-0.1185153872,
-0.1595749408,
-0.0961113274,
-0.1728560179,
0.21040079,
0.0722270831,
0.1171076149,
0.0522644818,
0.0329216905,
-0.253256321,
-0.4568681419,
0.2618017495,
0.2318311632,
-0.2234418392,
0.0250681993,
0.2808708251,
-0.3499059081,
0.1182914972,
0.6035224199,
0.2322998494,
-0.0582313649,
-0.0834592208,
0.1073074937,
0.3267664313,
0.0169564933,
-0.0994198695,
0.1862423718,
0.3138252199,
-0.5439271331,
0.0629941374,
0.0679737106,
-0.1795364171,
0.3942140341,
-0.2318457067,
0.3350019455,
-0.2536055446,
0.4101295769,
-0.2309691459,
0.1492195874,
0.0442675948,
-0.1309384853,
-0.6219237447,
0.2565176785,
-0.0546293631,
-0.0505178571,
0.0629610866,
-0.2929470837,
0.045598615,
-0.124539189,
0.2503246963,
0.3252750635,
0.0252179485,
-0.3522539437,
-0.1959874332,
-0.5977162719,
0.2946240902,
0.145552516,
0.2486121058,
-0.0141238179,
0.3637275398,
-0.0974157453,
0.082724452,
0.1792983115,
-0.0695216879,
-0.1520160437,
0.0536883622,
-0.5792644024,
0.1903704256,
-0.1522357762,
-0.1525034308,
0.1285755336,
0.1937535405,
0.0038232943,
-0.2565795183,
0.0009938851,
0.2496808171,
-0.2043683529,
0.0098008811,
0.2409564257,
0.4613411129,
0.1374107301,
0.0890945345,
0.0119941384,
-0.0801084563,
-0.2506811619,
-0.1601438522,
-0.1050343513,
-0.1749792993,
-0.332882911,
0.4853776693,
-0.3128299117,
-0.5159755349,
-0.2895017862,
0.2952622771,
0.0879575312,
-0.3334015012,
-0.2367613614,
0.3818426728,
-0.1371771097,
0.0356158428,
0.2386958301,
0.3839944601,
0.0910454914,
0.3809010983,
-0.3290144801,
-0.1753948629,
0.7739758492,
-0.5265747309,
0.248127088,
-0.237990573,
0.3535304666,
0.4005732238,
-0.3686011136,
-0.2991079986,
-0.0865053311,
0.13885957,
0.0136337355,
-0.1585732847,
0.1545864046,
-0.3073345423,
-0.0271866471,
-0.0483862497,
0.1458698064,
0.1749289632,
-0.0491092056,
-0.0030157417,
-0.1477307975
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | I tried to adjust `run_seq2seq.py` and trainer to use the suggested dist env:
```
import torch.distributed as dist
metric = load_metric(metric_name, num_process=dist.get_world_size(), process_id=dist.get_rank())
```
and in `trainer.py` added to call just for rank 0:
```
if self.is_world_process_zero() and self.compute_metrics is not None and preds is not None and label_ids is not None:
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
```
and then the process hangs in a deadlock.
Here is the tb:
```
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/filelock.py", line 275 in acquire
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 306 in _check_all_processes_locks
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 501 in _init_writer
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 440 in add_batch
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 397 in compute
File "examples/seq2seq/run_seq2seq.py", line 558 in compute_metrics
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1876 in prediction_loop
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1703 in evaluate
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer_seq2seq.py", line 74 in evaluate
File "examples/seq2seq/run_seq2seq.py", line 603 in main
File "examples/seq2seq/run_seq2seq.py", line 651 in <module>
```
But this sounds right, since in the above diff I set up a distributed metric and only called one process - so it's blocking on waiting for other processes to do the same.
So one working solution is to leave:
```
metric = load_metric(metric_name)
```
alone, and only call `compute_metrics` from rank 0
```
if self.is_world_process_zero() and self.compute_metrics is not None and preds is not None and label_ids is not None:
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
```
so we now no longer use the distributed env as far as `datasets` is concerned, it's just a single process.
Are there any repercussions/side-effects to this proposed change in Trainer? If it always gathers all inputs on rank 0 then this is how it should have been done in first place - i.e. only run for rank 0. It appears that currently it was re-calculating the metrics on all processes on the same data just to throw the results away other than for rank 0. Unless I missed something.
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 302 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
I tried to adjust `run_seq2seq.py` and trainer to use the suggested dist env:
```
import torch.distributed as dist
metric = load_metric(metric_name, num_process=dist.get_world_size(), process_id=dist.get_rank())
```
and in `trainer.py` added to call just for rank 0:
```
if self.is_world_process_zero() and self.compute_metrics is not None and preds is not None and label_ids is not None:
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
```
and then the process hangs in a deadlock.
Here is the tb:
```
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/filelock.py", line 275 in acquire
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 306 in _check_all_processes_locks
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 501 in _init_writer
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 440 in add_batch
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/metric.py", line 397 in compute
File "examples/seq2seq/run_seq2seq.py", line 558 in compute_metrics
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1876 in prediction_loop
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer.py", line 1703 in evaluate
File "/mnt/nvme1/code/huggingface/transformers-mp-pp/src/transformers/trainer_seq2seq.py", line 74 in evaluate
File "examples/seq2seq/run_seq2seq.py", line 603 in main
File "examples/seq2seq/run_seq2seq.py", line 651 in <module>
```
But this sounds right, since in the above diff I set up a distributed metric and only called one process - so it's blocking on waiting for other processes to do the same.
So one working solution is to leave:
```
metric = load_metric(metric_name)
```
alone, and only call `compute_metrics` from rank 0
```
if self.is_world_process_zero() and self.compute_metrics is not None and preds is not None and label_ids is not None:
metrics = self.compute_metrics(EvalPrediction(predictions=preds, label_ids=label_ids))
```
so we now no longer use the distributed env as far as `datasets` is concerned, it's just a single process.
Are there any repercussions/side-effects to this proposed change in Trainer? If it always gathers all inputs on rank 0 then this is how it should have been done in first place - i.e. only run for rank 0. It appears that currently it was re-calculating the metrics on all processes on the same data just to throw the results away other than for rank 0. Unless I missed something.
| [
0.0881400108,
-0.1092865244,
0.1158301383,
0.2856830955,
0.0845733657,
0.0806559548,
0.2349970192,
0.2377156615,
0.2347910702,
0.2356441617,
0.0788626969,
0.16671215,
-0.3739604354,
-0.2242655754,
0.0668831393,
0.0503918529,
-0.0005043149,
-0.1439884901,
-0.2210303247,
-0.1951079369,
-0.2040553391,
0.5319715142,
-0.0888307095,
-0.0502559356,
-0.5913397074,
0.2881837785,
-0.1651012003,
0.1823536456,
-0.068403095,
-0.6171513796,
0.3700888753,
0.1490112692,
0.0829518139,
0.5546141863,
-0.0001318881,
-0.1994534582,
0.2629811168,
-0.221129775,
-0.1534062028,
-0.2687928677,
0.1527825892,
-0.2575651407,
0.5715032816,
-0.1630464196,
-0.141618371,
-0.087945953,
0.090079248,
-0.4483616352,
0.42596668,
0.1269232035,
-0.0067989677,
0.352801621,
-0.0033906698,
-0.2833948135,
0.0977120697,
-0.1208256483,
0.0339140818,
0.3804104328,
-0.0241648629,
0.0620030314,
-0.3279880583,
0.2068390548,
0.2323786914,
-0.0643238425,
0.3330111504,
-0.078031823,
-0.1014685482,
-0.1370965987,
-0.0787228346,
0.3096218109,
0.2080828398,
-0.1693375707,
-0.2303851843,
0.060358569,
0.0139123909,
-0.3432676196,
0.1953566372,
-0.0388792045,
-0.0130724907,
0.0416937768,
0.0024043694,
-0.1951524019,
0.0043459684,
0.0744458586,
-0.1773837358,
0.129582718,
-0.0661410987,
0.0334933847,
0.0162887536,
0.2591888607,
-0.1607430577,
0.3018883467,
0.0713186339,
0.1371360719,
-0.3619203866,
0.1703315973,
0.2666285634,
0.0578643717,
-0.0056574419,
0.3195265532,
-0.036185246,
0.0114493277,
0.4184843004,
0.0381957181,
-0.138973549,
0.362706095,
0.2786941528,
0.102943778,
0.1162614524,
0.1878525019,
-0.0855454952,
-0.2297569811,
0.0198365599,
-0.1975077391,
0.3149439096,
0.2442039996,
-0.0334215537,
-0.4205733538,
0.1195659563,
0.2952422798,
-0.129090175,
-0.0933247805,
0.1693427861,
0.3671607673,
0.0882644802,
0.0787011459,
0.3353292048,
0.2471681088,
-0.1991894841,
-0.0173156671,
-0.4135643542,
-0.0986427069,
-0.1242087185,
0.2724473178,
0.2196631134,
-0.162900418,
0.6351845264,
-0.1787698269,
0.469688952,
-0.0300204977,
0.1313091516,
0.18577829,
0.1316873729,
0.2409881949,
-0.0306475274,
0.2101946622,
0.3361895084,
0.3006996512,
-0.192524448,
-0.1458503902,
-0.1428042054,
-0.5472247005,
0.1147268713,
-0.0964162126,
-0.4572976232,
0.2726356387,
0.3576217592,
0.1117124781,
-0.107612893,
-0.1904319823,
0.193259865,
-0.0918898582,
-0.099767752,
-0.0660917014,
0.5843696594,
0.7293351889,
-0.2147580683,
-0.4012363553,
0.0988393202,
0.0325714462,
-0.1022925824,
0.1529448181,
-0.0788162947,
0.1933200806,
-0.3543681204,
-0.0127083659,
0.3378301859,
-0.4275301099,
-0.5466859341,
-0.1503922939,
-0.3531807661,
-0.1286652386,
0.2151798457,
0.0550855212,
0.1297098845,
0.1625278443,
0.1186122298,
-0.1818707436,
0.1395098716,
-0.3199317455,
-0.2878008187,
-0.1782020628,
0.0546490327,
0.0048479661,
0.3287586272,
-0.0217357203,
-0.0645740777,
0.1194041818,
0.1203887314,
0.1731574237,
0.1427401006,
0.3467542529,
0.157051757,
0.0381506011,
0.2471779734,
0.1363053918,
-0.3056510687,
0.3053779006,
-0.3536010385,
0.0743501559,
0.0327295549,
0.0219307989,
-0.1138970703,
-0.2049700022,
-0.1658810973,
-0.4732402265,
-0.1058142558,
-0.0443965681,
0.1530438662,
0.1016914248,
-0.1681245863,
0.1105814129,
-0.0655053854,
0.2719030976,
-0.4879295826,
0.0612643361,
-0.063874796,
-0.3398907483,
-0.0635266528,
0.161618039,
0.0212044641,
0.0049573332,
-0.1978688091,
0.3824714422,
-0.1466091424,
0.2167607695,
0.2523963749,
0.1297356188,
-0.0266657323,
-0.5362621546,
-0.0249712914,
0.1701556146,
-0.0606264435,
0.0106333047,
-0.2244263589,
0.3717919886,
0.1169161424,
0.1528424025,
-0.3292895854,
0.2574855089,
-0.1934430003,
-0.0311922133,
-0.4751602411,
-0.0524114296,
0.4612508416,
-0.1249244809,
0.3616276681,
-0.0769222528,
0.0478944331,
-0.0536690243,
0.1575224102,
0.1602759063,
0.0998463184,
0.0825532675,
-0.1949331909,
-0.0521238483,
0.084589161,
-0.2958512604,
0.4181762636,
0.1859459281,
0.2945017815,
0.0671905354,
0.1383650899,
-0.1661347598,
0.0406897515,
-0.0327740349,
-0.1795003116,
0.2156119049,
0.27682212,
0.1916084588,
-0.2365641296,
-0.0302205049,
-0.2584657669,
0.0539821535,
-0.3396646082,
0.0545163751,
-0.211402446,
0.526972115,
0.1609437764,
-0.0017839037,
-0.3060410619,
-0.258770138,
0.1985840946,
-0.1509753466,
0.1073674858,
0.2856146097,
-0.1824020445,
0.366370976,
0.1217018738,
-0.1664311737,
-0.2195232213,
-0.2145324647,
-0.0526190512,
-0.2615050673,
-0.2276444882,
-0.0176192597,
0.1051155627,
-0.1812950671,
-0.0146129951,
-0.0838484466,
-0.323687017,
0.0789871961,
0.0525831766,
0.5157327056,
0.3757802546,
0.0781343579,
-0.3018793464,
0.1728727669,
0.4036899209,
-0.4335684478,
-0.2911329269,
0.0549032316,
0.0282632858,
0.2603509724,
-0.2404194623,
-0.3637055755,
0.0195960533,
-0.3122050762,
0.4976969361,
0.0147306025,
-0.0670965761,
-0.2401186526,
0.0159150362,
0.2207765877,
-0.1401596367,
0.3164167404,
-0.2179737985,
-0.7374212742,
0.4974017441,
-0.0844748169,
-0.2427445203,
-0.0719335899,
0.0304229781,
0.3310317397,
0.0592323765,
-0.6748090386,
-0.377584666,
0.044279933,
0.2911630273,
-0.1235278994,
0.1387414932,
0.5976378322,
-0.0984464437,
0.090632394,
-0.0848379433,
-0.164619416,
0.5146777034,
-0.2370684743,
0.2622247338,
-0.0529934093,
0.3953694999,
0.2051373124,
1.0660387278,
0.4365730286,
0.1375797391,
0.240404129,
0.0790984184,
0.3862742186,
-0.1463311613,
-0.1985453665,
0.292594105,
0.2944891453,
-0.0125391781,
0.1614470035,
0.1040411294,
0.2343328297,
-0.1193147078,
-0.058051303,
-0.2382651269,
-0.2966467738,
0.0124175977,
-0.2467527837,
0.2885682881,
-0.0768275559,
0.1307215691,
-0.2336004972,
-0.0309703052,
0.4748421013,
0.3607443571,
0.3400710523,
0.2659282088,
0.013888739,
0.0185917616,
-0.5698674917,
0.2131744623,
-0.3507133126,
0.0432459526,
-0.1114052087,
-0.2091225982,
0.1995762289,
-0.0166742075,
0.6270532012,
0.2594264448,
-0.132365644,
-0.0344227627,
-0.3146464825,
-0.4235264063,
-0.0951625109,
0.0202624723,
-0.0807247162,
0.2619808316,
0.2443539798,
-0.4367319047,
-0.2846646607,
-0.0777888,
-0.0575864501,
-0.2513467073,
-0.3316428661,
-0.2910421193,
0.0722150803,
-0.2103279829,
0.1533391029,
-0.1221066117,
0.0151935145,
-0.0006408431,
0.061013937,
-0.1972710341,
-0.0585678294,
0.0051172562,
0.0079775453,
0.3492989242,
-0.3426241577,
0.184699133,
0.5938440561,
0.0462311432,
0.0694634542,
0.3731180429,
0.0241028666,
-0.0076119862,
0.3295190334,
-0.1096443683,
0.1984229833,
0.6668336391,
-0.2204792649,
0.0092966873,
-0.1022757441,
0.2657597959,
-0.0840353221,
-0.0240894295,
0.2751722336,
-0.1857891381,
0.3366447985,
0.1338973045,
0.1865536273,
0.3631667197,
-0.0754490048,
0.39289096,
0.0175169427,
-0.1314400136,
0.1697275043,
0.0900112689,
1.0596704483,
0.1733088046,
0.2880571485,
0.0718041211,
0.1134336367,
0.2868269384,
0.0770511925,
0.0670693815,
-0.233656317,
-0.2930971682,
-0.0237972811,
-0.2060995698,
-0.026865799,
-0.2503249943,
-0.0878031328,
0.4228725731,
-0.3943592608,
-0.0798704177,
-0.2098237425,
-0.1599925905,
-0.2157907188,
-0.1572387666,
0.1809965372,
-0.0825240165,
0.0263456218,
0.5747038126,
-0.2157612294,
-0.1310920864,
-0.3793376684,
-0.3467716277,
-0.1881608665,
-0.068369478,
-0.3320902586,
0.0041433387,
0.3321133852,
-0.3493465185,
-0.1050177664,
0.4051668346,
0.4119636118,
0.1500433385,
-0.3158902526,
0.013393404,
0.1522024125,
0.0163303986,
0.0662484244,
0.003769435,
0.2145135254,
-0.0231952481,
-0.0367355645,
0.1018835306,
-0.2719019055,
0.0721197948,
0.0464075059,
-0.1299453229,
0.1035781279,
0.0079813302,
-0.0395115279,
0.0416477397,
0.0088517815,
-0.2246103138,
0.0092117889,
0.3180174232,
-0.3020228148,
-0.1260163486,
0.1327931881,
-0.2384745777,
-0.0902608037,
0.7388561368,
-0.1895924658,
-0.0833806098,
0.508665204,
0.1132061556,
-0.0950124413,
-0.0279415324,
0.0534407049,
0.5871478319,
-0.7964067459,
0.0375652499,
-0.0280947722,
-0.0733697414,
0.0531304665,
0.1573780477,
0.4190034866,
-0.1953763813,
-0.0511976741,
-0.1429458708,
-0.5170992613,
0.1747762561,
-0.4015873969,
0.2136110067,
-0.4579902291,
0.1037901789,
-0.12055213,
0.1368552446,
-0.1713809967,
0.2145245969,
-0.1843111813,
-0.0760751367,
-0.0796446949,
-0.0408803225,
0.2494171709,
-0.083016254,
-0.1323852986,
0.1427086443,
-0.2992050052,
-0.0110411644,
-0.2597683966,
0.2003600597,
-0.1354681551,
0.1764081419,
-0.0286936536,
0.122787267,
-0.0617899895,
-0.1407510936,
0.0586235672,
0.2382397354,
0.1416049898,
0.1680942625,
0.1831713915,
0.205873996,
-0.1489624679,
0.1856941581,
0.1167300791,
-0.1866597235,
-0.122310482,
-0.0412710011,
-0.3114249408,
-0.030534856,
0.074110806,
0.1024202332,
-0.1661491096,
0.2045536488,
0.0971293151,
0.0333640352,
-0.2555372715,
0.0163988248,
0.3647413254,
0.0902379304,
-0.2043004483,
0.0552672669,
-0.00693487,
0.0217386447,
-0.1006235033,
0.28915295,
0.0024502054,
0.3737571239,
-0.0071059167,
0.233938843,
0.155573532,
-0.0368139297,
0.0814674497,
0.2663899064,
-0.095786266,
-0.1646012068,
0.349430114,
-0.0394396074,
0.1696409136,
0.2915657759,
0.5672032237,
0.1148114428,
0.1399039775,
-0.0718223974,
-0.0939023793,
0.342705965,
-0.1072870493,
0.2498674542,
-0.1057757288,
-0.2143327892,
-0.2015595287,
-0.0100699514,
0.1364572048,
-0.241556257,
-0.4650447071,
0.7891877294,
-0.2494341135,
-0.2857101262,
-0.1221793741,
0.2552674115,
-0.1764249057,
-0.1262226254,
-0.0582016706,
-0.1612548232,
-0.1629290134,
-0.082010664,
0.2177404612,
-0.0160788856,
0.1115117967,
0.2160970271,
-0.0037843771,
-0.3101451397,
-0.4137941301,
0.2377392948,
0.2219976485,
-0.1553964019,
0.0707510561,
0.1623303592,
-0.2955524325,
0.1464755535,
0.5265615582,
0.1610318124,
0.0471725091,
-0.1688612401,
0.0138867386,
0.3292218447,
-0.0358104035,
-0.0913411528,
0.1298605651,
0.2126794308,
-0.480481714,
-0.0339949578,
-0.0184380412,
-0.1049447507,
0.3532825708,
-0.2245303243,
0.248627305,
-0.142225638,
0.5203976035,
-0.1859230399,
0.0720938891,
0.0460007451,
-0.1357638687,
-0.6475101709,
0.2183960378,
-0.1087092161,
0.0747859329,
0.0200169012,
-0.2638198733,
-0.0101508405,
-0.1398231387,
0.1454685926,
0.4622711539,
0.0498206019,
-0.4014340341,
-0.2744063735,
-0.5686839819,
0.3184740245,
0.25549227,
0.2237141728,
-0.0390213616,
0.2768023312,
-0.0909412503,
0.0218149424,
0.1413942873,
-0.0164729208,
0.1134105548,
0.1315145642,
-0.5518436432,
0.1067016721,
-0.1301147044,
-0.08022324,
0.067461133,
0.330352962,
-0.1294990033,
-0.1872293055,
-0.0767546147,
0.1827470064,
-0.2880126834,
-0.0719877109,
0.3570815325,
0.3660237491,
-0.006994471,
0.1416058838,
0.0799732953,
-0.1299113929,
-0.200525701,
-0.0048075467,
-0.0236917771,
-0.0690523833,
-0.2420406044,
0.4297696948,
-0.4423103034,
-0.5072240829,
-0.4338132441,
0.2354564965,
0.0585269928,
-0.1644812971,
-0.389999032,
0.3686196208,
-0.2135484219,
0.1273378432,
0.2562367618,
0.3571560383,
0.0298297964,
0.3313846588,
-0.4535014629,
-0.1204371601,
0.7448205352,
-0.5261365771,
0.3190011978,
-0.2616109252,
0.3419785202,
0.4516818225,
-0.3614069819,
-0.5016878843,
0.007907629,
0.1330848336,
-0.0542375334,
-0.161600545,
0.0992288142,
-0.3082866967,
0.101312384,
-0.0707925931,
0.2413410097,
0.1822641641,
0.054124251,
-0.0666987896,
-0.1139494926
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | But no, since
`
metric = load_metric(metric_name)
`
is called for each process, the race condition is still there. So still getting:
```
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
```
i.e. the only way to fix this is to `load_metric` only for rank 0, but this requires huge changes in the code and all end users' code.
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 76 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
But no, since
`
metric = load_metric(metric_name)
`
is called for each process, the race condition is still there. So still getting:
```
ValueError: Error in finalize: another metric instance is already using the local cache file. Please specify an experiment_id to avoid colision between distributed metric instances.
```
i.e. the only way to fix this is to `load_metric` only for rank 0, but this requires huge changes in the code and all end users' code.
| [
-0.0743826032,
0.0189331323,
0.0986257344,
0.2385592163,
0.0015653335,
0.0210895538,
0.1744563729,
0.2997214496,
0.3662587404,
0.1215545982,
0.027227968,
0.1309886575,
-0.3144583404,
-0.0137842074,
-0.0090085324,
-0.035420619,
-0.0174290612,
-0.0118402429,
-0.2157495469,
-0.1407074928,
-0.3549724519,
0.490944773,
-0.0940420479,
-0.0196763873,
-0.4025663733,
0.1862024814,
-0.1208146662,
0.1642287225,
-0.0712106228,
-0.6548618674,
0.3760331571,
0.2119781524,
0.0550229028,
0.5668934584,
-0.000122839,
-0.164762184,
0.222458005,
-0.1654274464,
-0.0895346999,
-0.2089779079,
0.0419956595,
-0.3018235564,
0.4746408165,
-0.1672161222,
-0.0354675166,
-0.0663869604,
0.083405219,
-0.4613867402,
0.4682589173,
-0.0254342668,
0.0765320957,
0.3800906241,
-0.1043442711,
-0.2316597998,
-0.0495067835,
-0.161381647,
0.024174653,
0.4295062125,
-0.0905898958,
0.0036927164,
-0.3325565457,
0.263350904,
0.1676268131,
0.0209395476,
0.245375976,
-0.0559756681,
0.0649292022,
-0.1188938841,
-0.1050409079,
0.3209834099,
0.1520300359,
-0.1883412749,
-0.2298357785,
0.0978396088,
-0.1813545525,
-0.5551990271,
0.1535266191,
-0.0972425938,
0.0039796531,
0.0392486043,
-0.0380286947,
-0.0952161998,
-0.056838423,
0.0566300303,
-0.1502935588,
0.0717537031,
-0.1363276988,
0.0569492392,
-0.0620993488,
0.2655830085,
-0.3586176634,
0.348726809,
-0.0165709201,
0.1322409809,
-0.4776334465,
0.1030917317,
0.2333645672,
0.019275479,
0.0638082474,
0.3710051775,
0.0047829263,
0.1104966477,
0.441955626,
0.0989987254,
-0.064853631,
0.4883593321,
0.2653138936,
-0.0972630531,
0.1724758148,
0.1843728423,
-0.1093546078,
-0.3044107258,
0.1303546727,
-0.2305695415,
0.3590705097,
0.1428925395,
0.0501231551,
-0.430065155,
0.0268942788,
0.3125572801,
-0.0175366178,
-0.1046421379,
0.1151105762,
0.3204917908,
0.0802920759,
0.0376733728,
0.2430396676,
0.3135303259,
-0.2338243723,
0.0523921475,
-0.4096280038,
-0.2092064917,
-0.0599775128,
0.3604542315,
0.0806868821,
-0.0867772996,
0.6622604132,
-0.245279029,
0.4363714457,
-0.0098046809,
0.2046341002,
0.1801849604,
-0.0522347577,
0.3353192806,
0.0205428153,
0.2025182545,
0.4229383469,
0.11806795,
-0.2378180176,
-0.2222478092,
-0.1965864003,
-0.4890337884,
0.1327873915,
-0.0115741696,
-0.4982540011,
0.2183235139,
0.2810570002,
0.2368833721,
-0.0750577897,
-0.1865118146,
0.1740933359,
0.0632326901,
-0.1740946472,
-0.077520743,
0.5013917685,
0.8000811934,
-0.3106372952,
-0.4090103805,
0.0621439964,
0.0198930427,
0.0597446598,
0.1834789515,
-0.0264084395,
0.112223044,
-0.239866823,
0.0430616289,
0.4500412941,
-0.3818280697,
-0.4404714704,
-0.0102442987,
-0.4845240712,
-0.0927991122,
0.1529060602,
0.0130152889,
0.1167050302,
0.1743206233,
0.0596253052,
-0.2304719239,
0.1583906561,
-0.2232082486,
-0.4099034071,
-0.3159251511,
0.0827810317,
0.0175186023,
0.3511084318,
0.0312785432,
-0.0216587745,
0.0502977781,
0.0938850641,
0.2171547115,
0.1900274158,
0.2495133728,
0.2989780605,
0.0074489918,
0.2794368565,
0.062663883,
-0.1863883585,
0.3330546021,
-0.3847154975,
0.1210736334,
0.155823037,
0.0101862177,
-0.2554593682,
-0.2336652279,
-0.2074739933,
-0.381709218,
0.0072999112,
-0.1816834211,
0.2119551003,
0.207486093,
-0.2354961783,
0.1523827612,
0.022761941,
0.2163388282,
-0.4231482744,
0.0148603991,
-0.0297127143,
-0.3313261271,
-0.020222351,
0.060649775,
0.0434454605,
0.0521163344,
-0.2072705626,
0.4505543709,
-0.2041246146,
0.2718483806,
0.3489959538,
0.194997251,
-0.0731451511,
-0.3825896382,
-0.0086811539,
0.1035970151,
-0.0373465531,
0.0034667477,
-0.2115323991,
0.365671277,
0.074035272,
0.1468584388,
-0.3511493206,
0.2701358199,
-0.0933459699,
-0.1494022012,
-0.5120645165,
-0.0564064942,
0.3746652603,
-0.2972074151,
0.3142362833,
-0.134836033,
0.0955447555,
-0.0313754342,
0.2216871381,
0.2303462029,
0.1324337572,
0.0526793636,
-0.2068915814,
-0.0816361606,
0.0431519747,
-0.2377363741,
0.3779175282,
0.1831711233,
0.226216808,
0.080045715,
0.1042995602,
-0.1794142872,
0.0618313439,
0.0174042396,
-0.0939751863,
0.1558280587,
0.2129645944,
0.1486933231,
-0.330837369,
-0.0466649719,
-0.2257913053,
0.0777692422,
-0.3252049685,
0.0463926867,
-0.1676026285,
0.3622935712,
0.1042044237,
0.0760790855,
-0.3205082119,
-0.2006611824,
0.2137134224,
-0.1206794977,
0.0558300763,
0.3729780614,
-0.3125833869,
0.5460642576,
0.1275178492,
-0.1835205704,
-0.2804267704,
-0.2134809494,
-0.0769839287,
-0.1774237752,
-0.1685278714,
0.1996259391,
0.1353265643,
-0.1900616586,
0.0364903957,
0.0679251924,
-0.2923833728,
0.0072686984,
0.0544487312,
0.6009895802,
0.219851166,
0.1127612144,
-0.3618911505,
0.066127792,
0.4660862386,
-0.46539554,
-0.2988181412,
0.1108196825,
-0.0056105889,
0.2263548374,
-0.2566636205,
-0.3030110002,
0.1503474861,
-0.355573833,
0.6306393147,
0.0361633599,
0.0040374268,
-0.1828643531,
-0.0212516636,
0.1606978476,
-0.1455214024,
0.1835407615,
-0.3220936954,
-0.7370197177,
0.3779355586,
-0.0818957537,
-0.2262861282,
-0.0287558958,
0.1227796078,
0.3893446922,
-0.0262096561,
-0.5944402218,
-0.481017381,
0.0726500005,
0.2060050368,
0.0201440733,
0.0549133718,
0.5922463536,
-0.0786325186,
-0.0039966106,
-0.0683100894,
-0.1145826578,
0.5944654942,
-0.1417895108,
0.144472912,
-0.09602274,
0.4891732931,
0.1768222302,
0.902428031,
0.5595241785,
0.3208564818,
0.2859216034,
0.143448472,
0.329437077,
-0.0935018957,
-0.1837988496,
0.3400886059,
0.2125223875,
-0.1073292196,
0.2406833023,
0.1164699793,
0.1155292243,
-0.1355738342,
-0.0306925066,
-0.2832496166,
-0.2555083632,
-0.1000089943,
-0.1111309305,
0.311330229,
-0.0138609521,
-0.0408910662,
-0.2783924937,
0.0155792534,
0.4612797201,
0.426333189,
0.2830251455,
0.2465913892,
-0.204309985,
0.11604736,
-0.6026703119,
0.2750053406,
-0.4224063754,
0.0926006883,
-0.1565951109,
-0.2295764983,
0.0888545141,
-0.0187627096,
0.577018559,
0.1072899327,
-0.1236121207,
0.0174960345,
-0.1422738135,
-0.478292942,
-0.0497221164,
-0.0212903693,
-0.019378081,
0.2601453662,
0.3804853261,
-0.4534819722,
-0.2462221086,
-0.0953103006,
-0.0955962837,
-0.3262086213,
-0.385243237,
-0.4633518457,
-0.0032969266,
-0.1292665899,
0.2367419451,
-0.2321141064,
0.064247787,
-0.026982462,
0.0604758561,
-0.1549198627,
-0.0765355229,
0.0366819128,
0.1297766566,
0.3405135572,
-0.241149798,
0.2002916783,
0.5243806839,
-0.2138915658,
0.0506992266,
0.3819578886,
0.0767323598,
-0.0481712855,
0.3114060163,
-0.005778484,
0.2684096098,
0.5744405389,
-0.2220167667,
0.066104278,
-0.0402306281,
0.3440590203,
-0.1733617485,
-0.081148982,
0.2343105078,
-0.1451412886,
0.3937271535,
0.1732896119,
0.2727485597,
0.2667099535,
-0.1511025131,
0.3070550561,
-0.0249730162,
-0.0635653809,
0.3386532366,
0.0756163597,
0.9486910701,
0.1284905076,
0.0594738834,
0.1395249218,
0.136572957,
0.3250001669,
0.0548073575,
0.1148585528,
-0.2098522186,
-0.2724110484,
0.0157612637,
-0.1055022031,
0.0776121616,
-0.2158323973,
-0.1212311685,
0.4221634567,
-0.3239583075,
-0.1714653969,
-0.2416452765,
-0.1891415119,
-0.2325045019,
-0.2224490047,
0.0643668696,
0.0192139968,
0.0016663424,
0.5368849039,
-0.2295256853,
-0.1801929176,
-0.2658553123,
-0.2863273025,
-0.0787060931,
-0.1080936939,
-0.2214949578,
-0.0420685001,
0.1932157874,
-0.1637633294,
-0.2138139904,
0.2345356792,
0.4904162884,
0.1107981652,
-0.3359956741,
-0.038644243,
0.1185731515,
0.0472977012,
0.1092758551,
-0.0165891722,
0.181206733,
-0.0596784726,
-0.0700881183,
0.0230901949,
-0.2034308016,
0.0294569284,
0.1439749599,
-0.1831576973,
-0.0728731006,
0.0270325132,
0.0658260882,
0.1261665821,
0.1493878812,
-0.2243225276,
0.0851811543,
0.2755257487,
-0.2201798409,
0.0531318039,
0.1045536324,
-0.2454174757,
-0.1072543338,
0.6161943078,
-0.126684159,
-0.082866475,
0.3956397176,
-0.0293206275,
-0.0281582251,
-0.0929279476,
0.0045109168,
0.4716777503,
-0.7318609953,
0.0216790587,
-0.0043270923,
-0.0188714564,
0.0596462712,
0.2084544599,
0.4314271808,
-0.1141752601,
-0.0196106136,
-0.1784307063,
-0.4242087603,
0.292336762,
-0.2917726338,
0.1231568456,
-0.4502271414,
0.1341912448,
-0.1835298538,
0.2939425111,
-0.2639541328,
0.1568068117,
-0.2775065601,
-0.1364923418,
-0.166944325,
-0.0555276424,
0.3321179748,
-0.0281807818,
-0.091153264,
0.1156946123,
-0.3989807963,
-0.0762548,
-0.2312701941,
0.1596202999,
-0.1257577389,
0.0551052913,
0.0076610483,
0.1867501736,
-0.073407732,
-0.1189984679,
0.0783571154,
0.2514325082,
0.0211225394,
0.0762162954,
0.2156772912,
0.1833855659,
-0.126737237,
0.15074642,
0.1054923758,
-0.1982359886,
-0.1485911161,
-0.0037622312,
-0.3089409471,
-0.037823841,
0.2045277059,
0.0469589904,
-0.1688234061,
0.1653286219,
0.0948098376,
-0.0197558459,
-0.3375716209,
0.0840762705,
0.433913976,
0.108473748,
-0.1139040515,
0.0171609223,
-0.0484704785,
0.0941807479,
-0.1545310318,
0.2720602751,
0.0388730094,
0.3606872559,
0.056130141,
0.294547677,
0.2639492154,
-0.0546166524,
0.19453381,
0.2769584656,
0.0772709399,
-0.169298768,
0.3492947519,
-0.2080004215,
0.1232769415,
0.2905895412,
0.6146331429,
0.1501110643,
0.1940247267,
-0.1213557422,
-0.0478097796,
0.1758173406,
-0.1244241446,
0.33462888,
-0.3251514435,
-0.1684167683,
-0.2493239045,
0.0356426425,
-0.0446148627,
-0.1554942727,
-0.4124147594,
0.6928101778,
-0.3596712649,
-0.3196226954,
-0.2396949083,
0.2137311697,
-0.162191391,
-0.1622504741,
-0.1247980371,
-0.1749600917,
-0.1221319437,
-0.1295804381,
0.2556283772,
-0.0027532242,
0.1339075863,
0.0584407523,
-0.0129344799,
-0.1976456642,
-0.5636007786,
0.3360031247,
0.2369901836,
-0.1671221554,
0.0380648896,
0.2539330125,
-0.3203011751,
0.050287243,
0.6612639427,
0.1400389671,
0.020517543,
-0.1053515673,
-0.0068345889,
0.3291794062,
0.0089211017,
-0.0917340219,
0.1179403216,
0.2123983204,
-0.4631313384,
0.0149257407,
0.0473550335,
-0.1401911676,
0.4126121402,
-0.2860367298,
0.3127579987,
-0.252771914,
0.4388634264,
-0.2093799412,
0.138630867,
0.0330515988,
-0.2090285718,
-0.6382531524,
0.2721566856,
-0.1211094707,
0.005566258,
0.0730725974,
-0.3161963224,
0.0315935574,
-0.0653401166,
0.1874741912,
0.3362926841,
-0.0472612604,
-0.3999146819,
-0.3157868981,
-0.628883481,
0.3125655651,
0.1188522428,
0.3920563757,
0.0013995096,
0.3996368051,
-0.139310956,
0.107726261,
0.080646798,
0.0115245357,
-0.1084975973,
0.0430238284,
-0.5370870829,
0.2043403387,
-0.1205613613,
-0.0920398235,
0.0809259564,
0.2269992977,
0.019759899,
-0.1926594377,
-0.0032362863,
0.2365330458,
-0.1842131019,
-0.0328186974,
0.2254682332,
0.3922173083,
0.0917388424,
-0.002601929,
0.0817169249,
-0.0382621177,
-0.2293901145,
-0.1227650791,
-0.1318765432,
-0.0989051461,
-0.3089472055,
0.435929358,
-0.3517576158,
-0.478405565,
-0.3654702902,
0.2429837883,
0.0612316132,
-0.2057874799,
-0.2005642354,
0.2870569229,
-0.0907279179,
0.0183590166,
0.2171035856,
0.31077829,
0.0868571028,
0.3682160974,
-0.3748183846,
-0.1778042316,
0.7422965765,
-0.512637198,
0.1951850057,
-0.2612097263,
0.4384706616,
0.4767012596,
-0.3749292493,
-0.3257743716,
-0.0008095503,
0.1407495737,
0.0945584476,
-0.13392739,
0.1337972134,
-0.4526486695,
-0.040645849,
-0.0503085777,
0.1235827357,
0.1436062604,
0.0090166181,
0.0242891237,
-0.1226509362
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | OK, here is a workaround that works. The onus here is absolutely on the user:
```
diff --git a/examples/seq2seq/run_seq2seq.py b/examples/seq2seq/run_seq2seq.py
index 2a060dac5..c82fd83ea 100755
--- a/examples/seq2seq/run_seq2seq.py
+++ b/examples/seq2seq/run_seq2seq.py
@@ -520,7 +520,11 @@ def main():
# Metric
metric_name = "rouge" if data_args.task.startswith("summarization") else "sacrebleu"
- metric = load_metric(metric_name)
+ import torch.distributed as dist
+ if dist.is_initialized():
+ metric = load_metric(metric_name, num_process=dist.get_world_size(), process_id=dist.get_rank())
+ else:
+ metric = load_metric(metric_name)
def postprocess_text(preds, labels):
preds = [pred.strip() for pred in preds]
@@ -548,12 +552,17 @@ def main():
# Some simple post-processing
decoded_preds, decoded_labels = postprocess_text(decoded_preds, decoded_labels)
+ kwargs = dict(predictions=decoded_preds, references=decoded_labels)
+ if metric_name == "rouge":
+ kwargs.update(use_stemmer=True)
+ result = metric.compute(**kwargs) # must call for all processes
+ if result is None: # only process with rank-0 will return metrics, others None
+ return {}
+
if metric_name == "rouge":
- result = metric.compute(predictions=decoded_preds, references=decoded_labels, use_stemmer=True)
# Extract a few results from ROUGE
result = {key: value.mid.fmeasure * 100 for key, value in result.items()}
else:
- result = metric.compute(predictions=decoded_preds, references=decoded_labels)
result = {"bleu": result["score"]}
prediction_lens = [np.count_nonzero(pred != tokenizer.pad_token_id) for pred in preds]
```
This is not user-friendly to say the least. And it's still wasteful as we don't need other processes to do anything.
But it solves the current race condition.
Clearly this calls for a design discussion as it's the responsibility of the Trainer to handle this and not user's. Perhaps in the `transformers` land? | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 233 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
OK, here is a workaround that works. The onus here is absolutely on the user:
```
diff --git a/examples/seq2seq/run_seq2seq.py b/examples/seq2seq/run_seq2seq.py
index 2a060dac5..c82fd83ea 100755
--- a/examples/seq2seq/run_seq2seq.py
+++ b/examples/seq2seq/run_seq2seq.py
@@ -520,7 +520,11 @@ def main():
# Metric
metric_name = "rouge" if data_args.task.startswith("summarization") else "sacrebleu"
- metric = load_metric(metric_name)
+ import torch.distributed as dist
+ if dist.is_initialized():
+ metric = load_metric(metric_name, num_process=dist.get_world_size(), process_id=dist.get_rank())
+ else:
+ metric = load_metric(metric_name)
def postprocess_text(preds, labels):
preds = [pred.strip() for pred in preds]
@@ -548,12 +552,17 @@ def main():
# Some simple post-processing
decoded_preds, decoded_labels = postprocess_text(decoded_preds, decoded_labels)
+ kwargs = dict(predictions=decoded_preds, references=decoded_labels)
+ if metric_name == "rouge":
+ kwargs.update(use_stemmer=True)
+ result = metric.compute(**kwargs) # must call for all processes
+ if result is None: # only process with rank-0 will return metrics, others None
+ return {}
+
if metric_name == "rouge":
- result = metric.compute(predictions=decoded_preds, references=decoded_labels, use_stemmer=True)
# Extract a few results from ROUGE
result = {key: value.mid.fmeasure * 100 for key, value in result.items()}
else:
- result = metric.compute(predictions=decoded_preds, references=decoded_labels)
result = {"bleu": result["score"]}
prediction_lens = [np.count_nonzero(pred != tokenizer.pad_token_id) for pred in preds]
```
This is not user-friendly to say the least. And it's still wasteful as we don't need other processes to do anything.
But it solves the current race condition.
Clearly this calls for a design discussion as it's the responsibility of the Trainer to handle this and not user's. Perhaps in the `transformers` land? | [
-0.0640708059,
-0.0787642151,
0.0965641737,
0.2534789443,
-0.0056878757,
-0.0212866068,
0.3123218119,
0.2638015151,
0.211024791,
0.2581806481,
0.0451732948,
0.3624681532,
-0.3526071012,
-0.1786395013,
0.0343576595,
-0.0347165614,
-0.044143606,
-0.133979395,
-0.3583229482,
-0.1239944994,
-0.1907992214,
0.4987674952,
-0.034417145,
-0.0297838412,
-0.3931261003,
0.2241770774,
-0.1640092731,
0.1788710654,
0.0643504187,
-0.5618359447,
0.3226024806,
0.1360855997,
0.0452937782,
0.5764808059,
-0.0001236324,
-0.1678352058,
0.1872862726,
-0.2450377047,
-0.1874427199,
-0.2717459798,
0.2271200418,
-0.3114355505,
0.4968201518,
-0.2364773154,
-0.1577859968,
0.0378682539,
-0.0054119481,
-0.3434018493,
0.4236992002,
0.0571899451,
0.0859313756,
0.3330073655,
-0.0323253721,
-0.1901232451,
0.1212333888,
-0.065386191,
-0.0368982442,
0.3913046718,
-0.0438002646,
0.0700681433,
-0.2391370535,
0.2996978462,
0.1755199432,
-0.0402261801,
0.314748466,
-0.0494450293,
-0.2011662275,
-0.077526316,
-0.1052093729,
0.3080211282,
0.2359145135,
-0.2234062403,
-0.2375838608,
0.0531368405,
-0.112919867,
-0.3445172906,
0.1445092708,
-0.0051627643,
-0.0187257826,
0.0235213861,
0.0591397621,
-0.1390452683,
-0.0627025515,
0.0998461619,
-0.0452241376,
0.0139156897,
-0.1335919201,
0.1054867655,
-0.001754249,
0.1352347732,
-0.3275977075,
0.306905061,
0.0391777754,
0.1289854944,
-0.3677839339,
0.1243338957,
0.2337083369,
0.0210838765,
0.0918097198,
0.4560727179,
0.0586395673,
0.0463435128,
0.4054612815,
0.1031669304,
-0.1804311574,
0.4910767376,
0.2882263064,
0.1206454486,
0.1309709251,
0.2508909106,
-0.0607224554,
-0.2949396372,
0.134270668,
-0.258698523,
0.2855694592,
0.2531370521,
0.0631743819,
-0.4151261747,
0.0464722216,
0.1691827327,
-0.0770845413,
-0.1774356365,
0.1314894855,
0.4097198546,
0.148839891,
0.0084703788,
0.3483400345,
0.3792476058,
-0.1620043516,
0.0054707341,
-0.3775832057,
-0.1619018167,
-0.0832369775,
0.3270740509,
0.1530536562,
-0.1119456291,
0.6152178645,
-0.2334931195,
0.4344538748,
-0.0676149502,
0.2086048722,
0.2173033059,
0.0877948701,
0.1757377088,
-0.0002521351,
0.1726159304,
0.3499712646,
0.3042705953,
-0.2953210771,
-0.1659080535,
-0.1075634956,
-0.5697388649,
0.1044745669,
-0.0256730802,
-0.3686075211,
0.2524024844,
0.3646578789,
0.1765206456,
-0.0878769383,
-0.1766685843,
0.2183121443,
0.0424640812,
-0.1899273098,
-0.1322889775,
0.5401468277,
0.7955896854,
-0.2234707475,
-0.3928964734,
0.0861313716,
-0.0302689038,
-0.0688860267,
0.1711061001,
-0.1304828376,
0.1734696627,
-0.3124318719,
0.0132101476,
0.4526493847,
-0.4541596472,
-0.5230976939,
-0.0551195703,
-0.3473706841,
-0.102244094,
0.1810639501,
0.0192641951,
0.1447807997,
0.1427843273,
0.1297153533,
-0.1455386728,
0.205215916,
-0.2678944468,
-0.3874982595,
-0.2713219821,
0.0696764067,
-0.0526350625,
0.2598971426,
0.0193605963,
0.0312664658,
-0.0005676113,
0.1223189011,
0.1125746071,
0.1530450583,
0.3094592094,
0.1673347354,
0.1094968095,
0.2271715552,
0.0797591507,
-0.2535490096,
0.2806078792,
-0.2641233504,
0.0964868963,
0.0547574386,
-0.0591279082,
-0.2017360032,
-0.1917342097,
-0.1189764068,
-0.4788510501,
-0.0093120895,
0.0025926027,
0.1778894961,
0.0980061069,
-0.0833217949,
0.0272524804,
0.002278395,
0.2093141675,
-0.4971339107,
0.0486034676,
-0.1358311325,
-0.3562151492,
-0.0538284965,
0.1624725759,
0.0824346617,
0.027518224,
-0.2172624618,
0.4142549336,
-0.2254364341,
0.2963486314,
0.2505710125,
0.0866476595,
-0.1558144987,
-0.443695724,
0.0706840605,
0.1534368992,
-0.024340516,
0.0595747679,
-0.2230672389,
0.4005186558,
0.0576050095,
0.0881644785,
-0.3915157616,
0.3206072748,
-0.0622931458,
-0.1256735027,
-0.5161962509,
-0.0653317198,
0.4644514024,
-0.2338355035,
0.3068930507,
-0.1485267431,
-0.1147535294,
0.0304504279,
0.0552105233,
0.1526012421,
0.2161267698,
0.0492044128,
-0.2106386274,
0.0539290309,
0.0803137794,
-0.2700405121,
0.4026063979,
0.2301344424,
0.2446662933,
0.0396013409,
0.2069371641,
-0.164956376,
0.0762509704,
-0.0633523613,
-0.1690063626,
0.1789560914,
0.2994231284,
0.1694059968,
-0.2687004209,
-0.0116006956,
-0.3858013153,
-0.0177338012,
-0.3161156178,
0.1085887775,
-0.1061811671,
0.3600186706,
0.0628940612,
0.0278818943,
-0.249700278,
-0.2115066648,
0.2172032893,
-0.128764838,
-0.0082143694,
0.3403941989,
-0.1438347101,
0.4247653484,
0.1005563885,
-0.2894511521,
-0.2572138906,
-0.1998005807,
-0.0760466307,
-0.1793351769,
-0.2656764984,
0.0457002409,
0.2180572748,
-0.1610603631,
0.0553275272,
-0.1083599627,
-0.2996193767,
0.0607740022,
0.0864418074,
0.5465069413,
0.2948967516,
0.0005499199,
-0.1870798916,
0.0301564112,
0.470880717,
-0.4639460742,
-0.2469664365,
0.0697956607,
0.0687670112,
0.3005340993,
-0.3028268218,
-0.4108240604,
0.0369482264,
-0.4303351939,
0.6223109961,
0.0505603254,
0.0192913804,
-0.2766923308,
0.0470694341,
0.1442532241,
-0.1902219653,
0.3671060801,
-0.3098249733,
-0.6642647386,
0.4911518991,
-0.1532043666,
-0.2969998717,
-0.0493564345,
0.0273514688,
0.4338548779,
0.0113271475,
-0.6472849846,
-0.4276646376,
0.0734204203,
0.3311346769,
-0.0820797831,
0.1234305575,
0.5292956829,
-0.1036518663,
0.0432881936,
-0.0859870389,
-0.1932817698,
0.4761618972,
-0.2479907572,
0.2904286385,
-0.0502366498,
0.4710015357,
0.2042385787,
1.0362509489,
0.4318808913,
0.1980847567,
0.240849942,
0.0621297285,
0.3678742647,
-0.1329008341,
-0.2530630827,
0.3594073653,
0.2157839239,
-0.0651660264,
0.1750036478,
0.1401123405,
0.1193042323,
-0.117536746,
-0.1354065835,
-0.3230397105,
-0.3765504956,
0.0584392995,
-0.3119447231,
0.2293656766,
-0.1024175584,
0.1097868755,
-0.2834073305,
-0.0170254968,
0.4524648488,
0.3811733425,
0.2687613368,
0.2686799765,
-0.1061640084,
0.1557905972,
-0.6143526435,
0.3195684552,
-0.4150726199,
0.0204219557,
-0.1964595914,
-0.2585927248,
0.1327748001,
0.0286135189,
0.5701926351,
0.1704187095,
-0.0772310272,
-0.0223532654,
-0.2900950313,
-0.572894454,
-0.1306662261,
0.0335993394,
0.0039162971,
0.3430422544,
0.2983968854,
-0.4971291423,
-0.2174239308,
-0.1325907111,
-0.0143523812,
-0.2948709726,
-0.2986009121,
-0.3033481836,
0.0376957506,
-0.3638681769,
0.1257953346,
-0.1481665671,
0.0019769408,
0.0124047622,
0.0938546956,
-0.17304039,
-0.0333018973,
-0.0419654138,
0.0819389895,
0.4452997148,
-0.368072331,
0.2138538659,
0.4355838001,
-0.0814056993,
0.0465964824,
0.4527381361,
0.0202475041,
-0.0322043598,
0.2494161129,
-0.1702365875,
0.1760366559,
0.5683682561,
-0.2360540628,
-0.0065519474,
-0.1683753729,
0.3352352679,
-0.1813475043,
-0.0947216749,
0.2148929238,
-0.139939189,
0.4028424919,
0.079777956,
0.1491570771,
0.2951566875,
-0.1458461136,
0.3089531958,
-0.0896984041,
-0.1584065855,
0.3474823833,
0.0283564255,
1.0431013107,
0.1858013123,
0.1939705014,
0.0941930413,
0.0594874769,
0.3286899924,
0.0322480127,
0.042359408,
-0.2455369681,
-0.1855898201,
0.0552306548,
-0.1338484883,
-0.0892542377,
-0.2128922343,
-0.2421699166,
0.4602894485,
-0.3985317945,
-0.1187767237,
-0.2751256227,
-0.2161842585,
-0.3653043807,
-0.1540290415,
0.1623812169,
0.0106244534,
0.0389810987,
0.5765026808,
-0.2063028514,
-0.1709362268,
-0.3550461531,
-0.4208045006,
-0.1701825261,
-0.0027210712,
-0.2345432341,
0.012643408,
0.2596084177,
-0.3003922999,
-0.1275203973,
0.3978983164,
0.3753566742,
0.1119140387,
-0.2954158187,
-0.0671834871,
0.0727766305,
0.078079313,
-0.0095543563,
-0.0105061866,
0.2022235841,
-0.0534296632,
-0.0567145944,
0.0370839462,
-0.1553320885,
0.0644344911,
0.0799801946,
-0.1045579612,
-0.0424972512,
-0.1324030459,
0.0671942085,
0.1465182006,
0.0625549257,
-0.2441195846,
0.0631633699,
0.3170137107,
-0.3118458986,
-0.016404409,
0.1300029308,
-0.1914989352,
-0.0526278131,
0.6006051898,
-0.2180669308,
-0.0574903078,
0.4439828992,
0.0863897949,
-0.0900117829,
-0.089664802,
-0.0095597357,
0.4701998234,
-0.8980281949,
0.0632024705,
-0.0824751854,
-0.0375497341,
0.0195323788,
0.2173469663,
0.375993073,
-0.0564710684,
-0.0513752252,
-0.1332805753,
-0.4421543479,
0.2491101623,
-0.2822006643,
0.2810610235,
-0.3498201966,
0.075284481,
-0.1540608704,
0.2654626369,
-0.264413327,
0.1684682369,
-0.1766667962,
-0.0392873362,
-0.0815754309,
0.0030676338,
0.3155995607,
-0.0181359574,
-0.0913405642,
0.1194009781,
-0.3476512432,
-0.0852037519,
-0.2056667656,
0.1578652263,
-0.1684210002,
0.1116118953,
-0.126875937,
0.1770161986,
0.0223466866,
-0.1894458234,
0.1027836949,
0.2703882456,
0.0493440777,
0.2141862959,
0.1147246659,
0.238900423,
-0.0491266176,
0.206923008,
0.0891240686,
-0.0931502879,
-0.1271400452,
0.0269982442,
-0.3199726343,
-0.0794886947,
0.1768059433,
0.0992085412,
-0.0745761618,
0.2261698693,
0.0143870255,
0.0950904191,
-0.3218650818,
-0.0831817091,
0.3944787979,
0.02523873,
-0.2240369469,
0.0014872029,
-0.0390669703,
0.0907803401,
-0.0961546302,
0.2872499526,
-0.0495017096,
0.2823805809,
0.099116303,
0.2836003006,
0.2086954713,
-0.0566606447,
0.0748733506,
0.1841031909,
-0.0485474914,
-0.1143176034,
0.3875469267,
-0.0666379184,
0.0531388372,
0.2888019681,
0.5470500588,
0.1314863563,
0.2106274068,
-0.126486823,
-0.062875703,
0.358679682,
-0.0406753942,
0.2327471673,
-0.068258509,
-0.1420317143,
-0.1085530818,
-0.1151635125,
0.1268029362,
-0.2630986273,
-0.3883002698,
0.6905949116,
-0.1910867095,
-0.2771377563,
-0.0456671268,
0.2752060592,
-0.1335233003,
-0.1646231413,
-0.1777416468,
-0.1472782046,
-0.1358103305,
-0.1351953447,
0.2524802089,
-0.0429371111,
0.1753875017,
0.1447415203,
0.0329224654,
-0.2331382781,
-0.3482343256,
0.2942011952,
0.1979301423,
-0.1560438573,
-0.0297635086,
0.214916423,
-0.2448701411,
0.1404076815,
0.5754083395,
0.2674565315,
-0.04272753,
-0.0580172017,
0.0885278583,
0.2895347774,
0.0024860501,
-0.1617272347,
0.0645732284,
0.1674534976,
-0.4141650796,
0.0965312719,
0.0538550057,
-0.1895057857,
0.4136650264,
-0.2827781439,
0.2279516608,
-0.1521186531,
0.4434227347,
-0.2251651436,
0.1118195429,
-0.0354709737,
-0.1718775481,
-0.6200802922,
0.2504788339,
-0.1109719053,
0.0327467881,
0.09214136,
-0.2865739167,
0.0349233523,
-0.0817041621,
0.1180626675,
0.3765928149,
0.0629151016,
-0.3114417195,
-0.2652309537,
-0.5255100131,
0.3029747605,
0.2050183415,
0.2848010957,
-0.0820882618,
0.3575699627,
-0.0763021111,
0.1108621135,
0.2054356337,
-0.0856381804,
0.0611874685,
0.1663654745,
-0.5453774929,
0.0388832353,
-0.1388043463,
-0.0357952155,
0.1995834261,
0.2482262254,
-0.2193724215,
-0.1935060918,
0.0216776058,
0.165612191,
-0.3292374611,
-0.1445854306,
0.2926347852,
0.4417030811,
0.0879335105,
0.0814845413,
0.0020326823,
-0.0957134068,
-0.190130353,
-0.1178383529,
-0.0816706866,
-0.1169572994,
-0.2638622224,
0.3734413981,
-0.3695465326,
-0.4859574437,
-0.4547898173,
0.2592762709,
0.0549928397,
-0.1221695915,
-0.2049323916,
0.334738493,
-0.1531442553,
0.0585969798,
0.1520990729,
0.3737303317,
0.0079820454,
0.4060660601,
-0.4049214125,
-0.1603149772,
0.7884657383,
-0.4327736497,
0.2788589895,
-0.2969712615,
0.3566490412,
0.4510629475,
-0.3289366961,
-0.485373646,
0.0553711355,
0.1302498281,
0.0357777104,
-0.0032265261,
0.0996123701,
-0.3638702333,
0.0264713019,
-0.0803088546,
0.182939589,
0.1846150756,
0.068811655,
-0.0152423009,
-0.1749408096
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | I don't see how this could be the responsibility of `Trainer`, who hasn't the faintest idea of what a `datasets.Metric` is. The trainer takes a function `compute_metrics` that goes from predictions + labels to metric results, there is nothing there. That computation is done on all processes
The fact a `datasets.Metric` object cannot be used as a simple compute function in a multi-process environment is, in my opinion, a bug in `datasets`. Especially since, as I mentioned before, the multiprocessing part of `datasets.Metric` has a deep flaw since it can't work in a multinode environment. So you actually need to do the job of gather predictions and labels yourself.
The changes you are proposing Stas are making the code less readable and also concatenate all the predictions and labels `number_of_processes` times I believe, which is not going to make the metric computation any faster.
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 144 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
I don't see how this could be the responsibility of `Trainer`, who hasn't the faintest idea of what a `datasets.Metric` is. The trainer takes a function `compute_metrics` that goes from predictions + labels to metric results, there is nothing there. That computation is done on all processes
The fact a `datasets.Metric` object cannot be used as a simple compute function in a multi-process environment is, in my opinion, a bug in `datasets`. Especially since, as I mentioned before, the multiprocessing part of `datasets.Metric` has a deep flaw since it can't work in a multinode environment. So you actually need to do the job of gather predictions and labels yourself.
The changes you are proposing Stas are making the code less readable and also concatenate all the predictions and labels `number_of_processes` times I believe, which is not going to make the metric computation any faster.
| [
-0.0986767784,
0.0874646604,
0.075472258,
0.1241439506,
0.1474274099,
-0.0701842457,
0.3525938392,
0.2356339693,
0.161123991,
0.2098138779,
0.0004213601,
0.2140559852,
-0.3455355167,
0.1935885549,
0.0740701631,
-0.0832459629,
-0.007562492,
-0.1163942963,
-0.2869693637,
0.0001238659,
-0.2612046003,
0.4958013892,
-0.1959232092,
0.1165348142,
-0.493932426,
0.0160646569,
-0.1405583322,
0.0436574817,
-0.1571672857,
-0.6827940941,
0.1779745519,
0.2850542367,
0.059001185,
0.6136053205,
-0.0001197262,
-0.1275454909,
0.1432346851,
-0.0766493604,
-0.1142280698,
-0.1018686444,
0.0062035397,
-0.3750229776,
0.5041076541,
-0.2027846128,
-0.0246386044,
-0.1871018261,
-0.0009074416,
-0.3866352439,
0.4851270318,
0.1960165799,
0.1148970574,
0.3724751174,
-0.0633553118,
-0.2893127501,
-0.0304378457,
-0.1797463745,
-0.0004410148,
0.3063090444,
-0.1180351153,
0.1041779891,
-0.2638344467,
0.2987608314,
0.1747556031,
-0.1579255015,
0.3985741138,
0.010855969,
0.0237382185,
-0.07508488,
-0.0263445433,
0.2280960679,
0.3640274107,
-0.2081996948,
-0.4027555585,
0.0213230401,
-0.1136108786,
-0.4449650645,
0.0867720842,
-0.1544953585,
-0.1124912873,
0.0173529554,
-0.05120093,
-0.0918698087,
-0.120504722,
-0.1065762639,
-0.1400082558,
0.0415050015,
-0.1515364945,
0.1079931483,
0.0986211002,
0.2940389812,
-0.2250002921,
0.4625897408,
0.1078084931,
0.0874909982,
-0.5861873031,
0.0515594482,
0.168926239,
-0.1378931999,
0.0261832848,
0.347635448,
0.139315784,
0.1647455096,
0.4749913216,
0.0469469205,
-0.0859801173,
0.4959519506,
0.1158405244,
-0.0413892232,
0.216643393,
0.2328998148,
-0.1324018389,
-0.3073425591,
0.15093714,
-0.289447099,
0.407508105,
0.1650965661,
-0.0240272693,
-0.3971652389,
0.0111142099,
0.2321660817,
-0.1615180671,
-0.3076351583,
0.1635744721,
0.3015663028,
-0.0188990533,
0.1546540707,
0.2649489641,
0.2677023113,
-0.2153809667,
0.0336759761,
-0.3627003729,
-0.1388626695,
-0.2221658081,
0.5311320424,
-0.0336126685,
-0.1031346917,
0.4908145666,
-0.2240335941,
0.3613179624,
-0.1143145189,
0.1868818402,
0.1512248814,
0.0729254112,
0.1448817402,
-0.0949599147,
0.0191627592,
0.19931449,
-0.0007020235,
-0.2345670164,
-0.3136923611,
-0.1904586405,
-0.5587619543,
0.0460786447,
0.0113417571,
-0.4121508598,
0.1687012911,
0.1926003247,
0.4390545189,
-0.0838324726,
-0.0827375501,
0.1543188095,
-0.0658366084,
-0.2503938675,
-0.053392753,
0.5387085676,
0.657261014,
-0.270811677,
-0.3530544639,
0.1573342085,
-0.0764823556,
0.0305981021,
0.1866124421,
-0.0675417706,
0.3315971196,
-0.1351322681,
0.0338988751,
0.3187025785,
-0.3188117743,
-0.2820296288,
-0.1126638576,
-0.2808200121,
-0.0668340549,
0.182088986,
-0.0332551375,
0.2036234736,
0.1777173579,
0.0793949589,
-0.2812539935,
0.1992521733,
-0.2796541452,
-0.3488394618,
-0.2571010292,
-0.0160806123,
0.0350260735,
0.414249301,
0.144786939,
-0.0155539997,
0.2191109359,
0.0989591181,
0.0124073429,
0.1865233779,
0.1369633526,
0.2173770219,
0.0301680584,
0.2122125924,
-0.0668708831,
-0.259488225,
0.181195125,
-0.2249534875,
0.264439404,
0.1812697649,
-0.1109847575,
-0.131376043,
-0.0319110118,
-0.0450461693,
-0.3800313473,
0.0381469429,
0.0193697214,
0.2461986989,
0.0636056215,
-0.1863691956,
0.2601461112,
-0.0229203627,
0.1970863342,
-0.3077029884,
0.0054114908,
-0.0493267179,
-0.2693817019,
-0.0472346134,
0.2843676805,
0.0135068251,
0.0620762333,
-0.156311065,
0.4471605718,
-0.0971782207,
0.1208168417,
0.2460491806,
0.2522501945,
-0.1001491919,
-0.2472560704,
0.0162441786,
-0.0177386664,
0.0309688337,
-0.0384247452,
-0.1867368519,
0.4048729837,
-0.0063224519,
0.1802697927,
-0.1444070041,
0.2286261022,
-0.1040406823,
-0.014684543,
-0.6104431748,
-0.0441223159,
0.1817785501,
-0.3588337898,
0.2854741216,
-0.0732165799,
-0.1879293919,
0.0873120129,
0.2204127908,
0.0784532204,
-0.0388827883,
0.1840817034,
-0.1221863031,
-0.1293782443,
-0.0313925035,
-0.0752812773,
0.437256515,
0.1981725842,
0.2658681273,
0.1368623525,
0.1789863557,
-0.2482043803,
0.1564267874,
0.1467079371,
-0.1983528435,
0.0213167407,
0.195225805,
0.1843652874,
-0.228649646,
-0.1335852444,
-0.2828974724,
0.0912202895,
-0.3479498029,
0.1114819571,
-0.1617621332,
0.3325736523,
-0.0256054811,
0.0918037593,
-0.1845477223,
-0.2695585489,
0.2477204353,
-0.1749944389,
-0.0806615651,
0.3019293249,
-0.1997396052,
0.5645968914,
0.0458002687,
-0.0543939099,
-0.1539083421,
-0.1289449185,
0.0248093531,
-0.1263792813,
-0.1107395664,
0.2380954921,
0.1443098634,
-0.1409780681,
0.0804133564,
-0.1126380861,
-0.3777040243,
0.0318095535,
-0.0719138533,
0.6136097908,
0.3686550558,
-0.067183122,
-0.3057754934,
0.0558076724,
0.4334176481,
-0.4616742432,
-0.1821402907,
-0.0158737823,
0.0210346244,
0.1119776219,
-0.2284413278,
-0.4090138376,
-0.0333865248,
-0.3924377561,
0.6301875114,
-0.0302863047,
0.0650706887,
-0.1423582137,
0.0048805065,
0.3127190471,
-0.2132500112,
0.2855618596,
-0.3027741313,
-0.6883828044,
0.3615089059,
-0.1526072323,
-0.2784526348,
-0.0922788158,
0.124667719,
0.4610548615,
-0.1010715067,
-0.6390307546,
-0.4874395132,
-0.0583328456,
0.1837802529,
-0.0854853839,
0.1019590795,
0.5934660435,
0.0402632281,
-0.1249963865,
-0.0401612036,
-0.3275842667,
0.6002622843,
-0.2267072201,
-0.0799339861,
-0.0480873659,
0.3807920218,
0.1421987414,
0.9527531266,
0.5312838554,
0.0835819095,
0.2418908328,
0.060900867,
0.4276570976,
-0.1242133602,
-0.2087627649,
0.3594117761,
0.0553972051,
-0.06963332,
0.2196852714,
0.0753752589,
0.2284233421,
-0.093081072,
0.0747023076,
-0.15958938,
-0.3905132413,
0.0516243726,
-0.0709966198,
0.2339845598,
-0.0320986807,
0.0994043201,
-0.2225833982,
0.0105919242,
0.3239012957,
0.3276199996,
0.2672767937,
0.1773557216,
-0.4027303159,
-0.0141268931,
-0.6917688847,
0.2999631166,
-0.3306574225,
0.0550327152,
-0.1726821661,
-0.3012732863,
0.1569827646,
-0.0032646991,
0.5102087855,
0.1670263559,
-0.3256071508,
0.0979510471,
-0.273212254,
-0.3518058658,
-0.0322372429,
-0.1057445407,
-0.092454195,
0.1303736418,
0.462773174,
-0.3822811842,
-0.4021390975,
-0.0575433299,
-0.0055291466,
-0.2024352849,
-0.2122657448,
-0.2753815055,
-0.1906086653,
-0.1053455025,
0.2593508661,
-0.1286327988,
0.1493753791,
-0.0751163363,
0.0105616972,
-0.110957697,
0.0294923335,
0.0498252697,
0.2006750852,
0.3328440487,
-0.2227476686,
0.3211789727,
0.5052853227,
0.0824891329,
0.1085557863,
0.3950394392,
-0.0712282807,
-0.0762941614,
0.2625707984,
-0.0787666291,
0.3977536857,
0.5973268747,
-0.3462335467,
0.0504732393,
0.0018160455,
0.4982073307,
-0.1901552677,
0.0729649737,
0.290230304,
0.0637293607,
0.3196468651,
0.0237539336,
0.2319909036,
0.3206911087,
-0.2400916368,
0.1433270276,
0.0838351548,
-0.2308397889,
0.2619360089,
0.1453872323,
0.9682909846,
0.0424492434,
0.1907758564,
0.2242966741,
0.0791934505,
0.4160701036,
0.0028479174,
0.0480356254,
-0.2100681961,
-0.3447064757,
0.0477673709,
-0.0816004425,
0.0021898029,
-0.2576920986,
-0.3582125008,
0.3531534076,
-0.1213833094,
-0.0985356197,
-0.2188063562,
-0.0329664946,
-0.0591554716,
-0.3148885369,
0.0249200873,
0.0341159552,
0.0702986717,
0.2626047134,
-0.1506990641,
-0.16716066,
-0.220993042,
-0.2636077106,
-0.1547156721,
-0.2230545431,
-0.2821094394,
-0.1693055928,
0.2403289527,
-0.0131513923,
0.0276421532,
0.2988243699,
0.4105799794,
0.093267642,
-0.3142995238,
-0.1085325778,
-0.1605777144,
0.2130302191,
0.1821777374,
-0.0810612738,
0.2795459032,
-0.1318837106,
-0.0807678774,
-0.0057211686,
-0.2108215988,
0.0140889212,
0.0495156646,
-0.145265922,
-0.0465460718,
-0.1562491953,
0.1377622038,
0.0298854616,
-0.0652446449,
-0.3307301402,
0.0995553657,
0.2345897257,
-0.2563972771,
0.2924686968,
0.1632493734,
-0.269954294,
-0.0011779703,
0.4867060781,
-0.0743928999,
-0.0469134636,
0.534265697,
0.0417938232,
-0.0784230456,
-0.1546751261,
-0.0377025083,
0.3748118579,
-0.6188101172,
0.2553616166,
-0.0921286494,
-0.1034342498,
0.0324720442,
0.2574295998,
0.2556152344,
-0.1610132009,
-0.0511724651,
-0.1950925291,
-0.3997428715,
0.3095514774,
-0.2061040103,
0.2349154055,
-0.4314935207,
0.2607112825,
-0.1639532745,
0.2664732635,
-0.3096977472,
0.071242325,
-0.3173696995,
0.0048192814,
-0.0426046699,
-0.0775406957,
0.3540881872,
0.0196157433,
-0.0791099668,
0.1467720717,
-0.3419510424,
-0.1227808669,
-0.2055004239,
0.1432722062,
-0.103079319,
0.0037044287,
0.1053148508,
0.2841438651,
-0.288969785,
-0.2163694352,
0.1446665823,
0.2201306075,
-0.0803182051,
0.1649866104,
0.0227370299,
0.1731149107,
0.0217939541,
0.1275440454,
0.31939587,
-0.1167133451,
-0.2552145123,
0.1674157679,
-0.3102363348,
-0.1582753956,
0.2583625019,
0.2446475327,
-0.0530163534,
0.2476224899,
-0.0005026059,
0.0390849635,
-0.4149766564,
0.0152035169,
0.5152159929,
0.211312145,
0.0148273706,
-0.0132394284,
0.0306367166,
0.0629687905,
-0.2263823152,
0.2248075157,
-0.0241948012,
0.1465189904,
0.182719022,
0.3415965736,
0.2004178762,
-0.008790195,
0.1131296903,
0.1668821275,
-0.0762683228,
-0.0972013026,
0.392945528,
-0.0642776638,
0.0135831833,
0.3271152973,
0.6098663807,
0.0715300441,
0.1015055552,
0.1042545438,
-0.1152171493,
0.3192862868,
-0.0454615355,
0.323151052,
-0.196245715,
-0.0204825252,
-0.1880551428,
0.1027825475,
0.0078948587,
-0.2648671865,
-0.4510084391,
0.7601668835,
-0.2982217371,
-0.428198278,
-0.1595360935,
0.2162977606,
-0.2203802317,
-0.2502736449,
-0.118072629,
-0.2273586392,
-0.1871370673,
-0.1366978586,
0.1595189124,
0.00551973,
0.3172379434,
-0.0935514122,
0.1166388467,
-0.2562333345,
-0.6462208033,
0.1233225614,
0.2115246505,
-0.1426579207,
0.0580328032,
0.1994388998,
-0.263761133,
0.1976293027,
0.7228574157,
0.2519355416,
0.0261538252,
-0.0786249489,
0.0529669225,
0.0839555264,
0.0876960084,
-0.000155732,
0.2521164715,
0.2659592032,
-0.612806201,
0.1696699709,
-0.0114009529,
-0.1481029689,
0.4118885398,
-0.2929539382,
0.3268873394,
-0.3496898115,
0.4951540828,
-0.0461825542,
0.0140399411,
0.0405662283,
-0.142522037,
-0.5717717409,
0.2246555835,
-0.0200585406,
-0.0718961135,
0.0829667225,
-0.3604299128,
0.0544082671,
-0.0736917704,
0.4507356882,
0.2547644377,
0.1308716089,
-0.3133818507,
-0.0882294849,
-0.6293572187,
0.0641283169,
0.2504541278,
0.3122889102,
-0.0201936513,
0.2240372002,
0.0060549788,
0.1538957357,
0.2459774911,
-0.0620284416,
-0.018819388,
-0.0547910966,
-0.6547974944,
0.2398649752,
-0.1182162389,
-0.0279669352,
0.1832564175,
0.1730830818,
0.081300579,
-0.2520496845,
-0.0084850937,
0.2440537065,
-0.06250678,
-0.0889036655,
0.2176324725,
0.4322578311,
0.1216436774,
0.0933316275,
0.0375455469,
-0.0467436351,
-0.2785556316,
-0.1413879395,
-0.1935821027,
-0.0083427001,
-0.1165120527,
0.436283946,
-0.3418883085,
-0.2784528434,
-0.3898365796,
0.2074489594,
0.0935018808,
-0.0974025726,
-0.1647613347,
0.4062651396,
-0.0399051234,
-0.0485422611,
0.2773153782,
0.3931528926,
-0.0078775231,
0.3691280484,
-0.4815252423,
-0.2000587136,
0.7874459624,
-0.5091013908,
0.1622093916,
-0.2533198893,
0.3058440089,
0.453786701,
-0.3390704393,
-0.616296649,
0.0236151218,
0.1255337,
0.0467105396,
-0.1886402816,
0.2845032811,
-0.4269395471,
-0.1252787858,
-0.0973920226,
0.279409647,
0.0648684204,
0.0003611371,
-0.0046300739,
-0.2103042603
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | Right, to clarify, I meant it'd be good to have it sorted on the library side and not requiring the user to figure it out. This is too complex and error-prone and if not coded correctly the bug will be intermittent which is even worse.
Oh I guess I wasn't clear in my message - in no way I'm proposing that we use this workaround code - I was just showing what I had to do to make it work.
We are on the same page.
> The changes you are proposing Stas are making the code less readable and also concatenate all the predictions and labels number_of_processes times I believe, which is not going to make the metric computation any faster.
And yes, this is another problem that my workaround introduces. Thank you for pointing it out, @sgugger
| the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 139 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
Right, to clarify, I meant it'd be good to have it sorted on the library side and not requiring the user to figure it out. This is too complex and error-prone and if not coded correctly the bug will be intermittent which is even worse.
Oh I guess I wasn't clear in my message - in no way I'm proposing that we use this workaround code - I was just showing what I had to do to make it work.
We are on the same page.
> The changes you are proposing Stas are making the code less readable and also concatenate all the predictions and labels number_of_processes times I believe, which is not going to make the metric computation any faster.
And yes, this is another problem that my workaround introduces. Thank you for pointing it out, @sgugger
| [
0.0017206892,
0.0483335853,
0.0741658509,
0.0518633537,
0.003217876,
-0.0890673697,
0.2733486891,
0.3028241992,
0.2559509873,
0.1578378677,
0.0953510031,
0.1828419566,
-0.3078092039,
0.119295381,
0.0508197881,
-0.0088100657,
-0.0289925374,
-0.1110825315,
-0.2872490883,
-0.0546379462,
-0.2833442688,
0.5747833252,
-0.1937363893,
0.0275072195,
-0.4175571501,
0.1800846905,
-0.0967398658,
0.0959720612,
-0.0746750161,
-0.6333746314,
0.3431345224,
0.1919608563,
0.0437810794,
0.4339023232,
-0.00011918,
-0.1534667462,
0.2158040404,
-0.145678103,
-0.0593994521,
-0.1108041108,
0.072156325,
-0.3260092735,
0.4729875326,
-0.1469023377,
-0.0135608017,
-0.0872121602,
0.006510308,
-0.424197197,
0.3669985533,
0.0343662128,
0.1324309707,
0.2945925295,
-0.0216746181,
-0.2921677828,
0.0711685866,
-0.1481089592,
0.0353576094,
0.3607213795,
-0.1339006126,
0.089666307,
-0.2697997391,
0.2661954463,
0.2188458145,
-0.1134707481,
0.3259406686,
0.0134763755,
-0.0287544839,
-0.0886276215,
-0.1024975106,
0.250677079,
0.2891418338,
-0.1953323483,
-0.2615003586,
0.0619683787,
-0.2091753036,
-0.3798716664,
0.1182643026,
-0.1241421476,
-0.0386009365,
0.0183159336,
0.0663574636,
-0.1046715826,
-0.0890833437,
-0.0081153419,
-0.0986795276,
0.0158468168,
-0.1301835775,
0.0467524268,
0.0187201649,
0.1890257895,
-0.3581732213,
0.4645082057,
0.0015239236,
0.1163825393,
-0.452660799,
0.0794148594,
0.2340448797,
0.0034457538,
0.1275366098,
0.4340142906,
0.0486907214,
0.0966067165,
0.4923918843,
0.0736069679,
-0.1517190337,
0.5450244546,
0.1368218213,
-0.0942572355,
0.2153575569,
0.2788960636,
-0.0592238344,
-0.3349407017,
0.2765875757,
-0.2964961231,
0.35461694,
0.1382150799,
0.0487446114,
-0.5286614299,
0.0650703013,
0.17705369,
-0.1106244326,
-0.218483299,
0.1539461762,
0.3556602001,
0.0642176643,
0.0728149116,
0.32316944,
0.2419791222,
-0.2418752611,
0.0730482489,
-0.3921152055,
-0.0962859988,
-0.1561840773,
0.3900669515,
0.0427020118,
-0.0673552603,
0.5413690209,
-0.2072094977,
0.3952540159,
-0.0542631596,
0.2074732184,
0.2733633816,
0.0856188834,
0.2450957447,
-0.0971160457,
0.1106677502,
0.2532867789,
0.0348399356,
-0.2827270627,
-0.3384742737,
-0.1631171554,
-0.5935839415,
0.0358846784,
0.027331084,
-0.4786864519,
0.1723163873,
0.3216296732,
0.3057265282,
-0.1627597511,
-0.1374481767,
0.127164796,
0.0738490447,
-0.1078671813,
-0.0605451912,
0.5470595956,
0.6318545938,
-0.3070906103,
-0.3632519245,
0.0832100883,
-0.0528884605,
0.0849455595,
0.151437223,
0.01826616,
0.1231784672,
-0.2710851431,
0.0660809278,
0.3751963079,
-0.2643638253,
-0.4211468995,
-0.0955257565,
-0.3790400922,
-0.1714235544,
0.148389101,
0.0196403861,
0.2099821568,
0.1557334363,
0.0462166816,
-0.2413897812,
0.2183353603,
-0.2098483145,
-0.4395236373,
-0.3486974835,
0.0186816659,
-0.0433584191,
0.2190105319,
0.0419359691,
0.0363062881,
0.170687288,
0.1256621182,
0.0660398677,
0.1798495054,
0.1231300905,
0.3237702549,
-0.035529945,
0.2978158295,
0.0444419831,
-0.2541935742,
0.1762871593,
-0.3174498677,
0.2136254162,
0.1256954074,
-0.1005329937,
-0.2092877924,
-0.1439250708,
-0.0070458222,
-0.4147659838,
0.0263379365,
-0.0470580906,
0.2701240778,
0.1364540309,
-0.1191124022,
0.1284607947,
0.1072650999,
0.2114131898,
-0.4049315453,
0.0130386651,
-0.0703824759,
-0.3191619813,
0.0298930183,
0.180013448,
0.0329500213,
0.1019104719,
-0.1698642671,
0.3908298016,
-0.1540257037,
0.2272181064,
0.3048533797,
0.2851655483,
-0.1629730761,
-0.3259814382,
0.02203984,
0.1105298996,
-0.0162725672,
0.0648802072,
-0.2022492588,
0.4049360156,
0.0456024297,
0.1391701102,
-0.2539229393,
0.298823595,
-0.0855146721,
-0.0431873053,
-0.5963208079,
-0.0990095362,
0.3305994272,
-0.3289659321,
0.2105220854,
-0.1463254541,
-0.0314472392,
0.0823655128,
0.2473681569,
0.1233501285,
0.0772460476,
0.0978805274,
-0.1947137266,
-0.0856296271,
0.0829137191,
-0.1318768114,
0.3044789732,
0.2182018012,
0.1607310921,
0.1080424935,
0.1258929074,
-0.1936113834,
0.1297408044,
0.0885623991,
-0.1993185282,
-0.0519842878,
0.2548467517,
0.1539467722,
-0.2797268629,
-0.0694240481,
-0.4146265686,
0.0284548607,
-0.3391500115,
0.0933109298,
-0.1681763083,
0.2760872841,
0.0427148417,
0.0185732171,
-0.269207567,
-0.2986366749,
0.2447031736,
-0.2117599398,
-0.1059977859,
0.3040294647,
-0.3279459476,
0.6078710556,
0.0554266423,
-0.14635171,
-0.2999765575,
-0.0674374178,
-0.0466690697,
-0.1146393716,
-0.2106846571,
0.2572806478,
0.2162181884,
-0.1139951795,
0.1512452662,
-0.0240459032,
-0.4987681806,
0.0363869555,
0.0080180671,
0.6357991695,
0.3888979256,
0.0157715995,
-0.2617826164,
0.0316114798,
0.3991931677,
-0.4438587427,
-0.2224481106,
0.0916186273,
-0.0147888549,
0.2465288043,
-0.3239532113,
-0.4238109291,
0.1243079156,
-0.4012075961,
0.7106444836,
-0.0169693977,
0.016824415,
-0.2196319848,
-0.0416194946,
0.215982154,
-0.2111343145,
0.3168080151,
-0.3373788595,
-0.6845719218,
0.4224869013,
-0.1047694683,
-0.2446504235,
-0.0255974308,
0.0576747954,
0.3486337364,
-0.1622273028,
-0.5894122124,
-0.4763508141,
-0.0748244151,
0.1999908388,
-0.0194936804,
0.0612851828,
0.5578104258,
-0.0331990644,
-0.0518376753,
-0.0575106852,
-0.301984936,
0.6153208017,
-0.2054653764,
0.0410428904,
-0.0076326756,
0.4433330894,
0.1814218462,
0.8166952133,
0.5588747263,
0.2393332869,
0.265694499,
0.143451184,
0.305605948,
-0.1414479017,
-0.0946221352,
0.3185797334,
0.1920527518,
-0.06735266,
0.1947140545,
0.1703396291,
0.1952798516,
-0.1114071459,
0.0084030256,
-0.2526727617,
-0.3594486713,
-0.0285915099,
-0.0122558018,
0.1676206291,
-0.0411500558,
0.0335735828,
-0.2244511247,
-0.0431749262,
0.4540453851,
0.3833142519,
0.2945558131,
0.2433978319,
-0.1909148395,
0.1245270222,
-0.7305415869,
0.3448733985,
-0.4099702239,
-0.0014104508,
-0.1546726823,
-0.3290134966,
0.119145602,
-0.0597264841,
0.546821475,
0.226477474,
-0.1941045523,
0.0241149291,
-0.2036361098,
-0.4357720912,
-0.0704818964,
-0.0595809631,
-0.1430880725,
0.2688423097,
0.3386007249,
-0.4449520707,
-0.2397599667,
-0.1034353524,
-0.031264741,
-0.2191045433,
-0.3313039839,
-0.2937084436,
-0.0996043757,
-0.2220917493,
0.2061284482,
-0.1314237714,
0.0842838585,
-0.0277879667,
0.0280270949,
-0.0842381269,
-0.1578319818,
0.0256602783,
0.1693505645,
0.3800822794,
-0.2912945151,
0.3020865619,
0.5443146229,
-0.0482206047,
0.0338873863,
0.3722670972,
0.0880229175,
-0.0148379542,
0.2595534921,
-0.0796131268,
0.3189831972,
0.5834524035,
-0.3298600316,
0.1268919557,
-0.0379149765,
0.4513751864,
-0.2243205756,
-0.1023805141,
0.2226597965,
-0.0440631434,
0.3950964808,
0.1574695557,
0.2453728914,
0.2686672211,
-0.239467442,
0.1392656565,
0.0856609195,
-0.1921716183,
0.3677600622,
0.0769230127,
0.9748387337,
0.1425095499,
0.1349178106,
0.1659116447,
0.0843516812,
0.4501752257,
0.1098286211,
0.0794717148,
-0.2773194313,
-0.2535168529,
0.075246267,
-0.120116964,
-0.1260692328,
-0.3618938625,
-0.3251042068,
0.4081984758,
-0.2714817524,
-0.1056302935,
-0.229064554,
-0.1442243159,
-0.2199870199,
-0.2697544992,
0.0447141565,
0.0453383401,
0.090743795,
0.4384635985,
-0.1836981177,
-0.1840984225,
-0.265237391,
-0.3652841449,
-0.1289608181,
-0.0673223808,
-0.1766268462,
-0.1138676554,
0.279524982,
0.0044975877,
-0.0696092322,
0.1947928667,
0.3574462831,
0.1417984366,
-0.3306594789,
-0.0981989801,
0.0538103022,
0.1065852344,
0.1469285637,
0.0098675787,
0.2544252574,
-0.1431810409,
-0.0182387978,
-0.0244802199,
-0.1513185203,
0.0438992977,
0.0058427304,
-0.2144088894,
-0.060685996,
-0.1018693149,
0.0982543901,
0.107100606,
0.0112330765,
-0.2338056266,
0.1105720401,
0.3235705495,
-0.1865552068,
0.1633545458,
0.2400679886,
-0.225358367,
-0.0512892678,
0.4643435478,
-0.1365623474,
-0.062751472,
0.4527919292,
0.0421352461,
-0.1334979236,
-0.0853753835,
-0.0759427547,
0.3869664967,
-0.8485308886,
0.1576581001,
-0.0338790603,
-0.0515601784,
0.0432767309,
0.2604057789,
0.3408830166,
-0.1588789821,
-0.0822827369,
-0.2311334312,
-0.4042379856,
0.3817248642,
-0.305251658,
0.2306502461,
-0.509955585,
0.1902864128,
-0.281249851,
0.1481527835,
-0.3190318346,
0.0882176459,
-0.2205168903,
-0.1557304561,
-0.1763938814,
-0.0018246472,
0.3478406966,
0.0095735341,
-0.0726351067,
0.2380467951,
-0.3705162406,
-0.0926234499,
-0.2068838477,
0.1353428364,
-0.1777230203,
0.0089452751,
0.0565368421,
0.2277381569,
-0.1598328799,
-0.1410601884,
0.0938876122,
0.1851567328,
-0.0096970499,
0.0880966932,
0.1260212064,
0.0756730884,
-0.0135953128,
0.1977690756,
0.2378552854,
-0.1603399962,
-0.2263822258,
0.1338213682,
-0.460291326,
-0.0737577006,
0.3483997285,
0.2329485416,
-0.1067647412,
0.2461017668,
-0.0235690996,
0.0950314105,
-0.4482335448,
0.0451120399,
0.4445842505,
0.0489913225,
-0.0940127298,
-0.0108752809,
-0.0413271151,
0.0979490727,
-0.1402586102,
0.3291448951,
-0.0622718409,
0.3363090158,
0.1206544936,
0.2768872976,
0.2116225213,
0.0139378905,
0.1684414595,
0.2232992053,
-0.054730989,
-0.0914748609,
0.3676176965,
-0.1182740703,
0.0249970034,
0.3289469182,
0.5605506897,
0.0763847232,
0.1496205628,
-0.1066405252,
-0.0672027841,
0.1484332234,
-0.0661067814,
0.3817591071,
-0.246520713,
-0.0910570994,
-0.157240659,
0.1240018606,
0.006164819,
-0.190174073,
-0.3315699995,
0.6917670965,
-0.1093514636,
-0.2469644248,
-0.1880481243,
0.2186512351,
-0.1606261283,
-0.1849392951,
-0.083889544,
-0.2105692327,
-0.2191711813,
-0.1509546787,
0.1695630252,
0.0185524914,
0.1450960636,
-0.0328526869,
0.1160524338,
-0.1858570129,
-0.5609820485,
0.1418211758,
0.2941443026,
-0.1176033318,
0.0181687828,
0.2244017273,
-0.2845524549,
0.1979003549,
0.6598452926,
0.2762863636,
-0.0298536755,
-0.0884216726,
0.0559421629,
0.1904037297,
0.071514152,
-0.074406758,
0.1938425899,
0.2686853707,
-0.6026166081,
0.1062565595,
0.0361007676,
-0.1714392602,
0.4032698274,
-0.3423995376,
0.3624666333,
-0.244478941,
0.4835076928,
-0.1571034193,
0.1179202422,
0.0028643757,
-0.1929028928,
-0.5489094257,
0.1697089374,
-0.1121427119,
-0.0506033786,
0.0383755043,
-0.3354279399,
0.0636243969,
-0.1064645946,
0.2458087802,
0.264192313,
0.1838690192,
-0.2285054624,
-0.2114781141,
-0.5184432268,
0.1700246334,
0.2240595371,
0.4091752172,
0.0069528632,
0.2396066487,
-0.0212804116,
0.1017894149,
0.2442370355,
0.0565918311,
-0.0967075527,
-0.0916238278,
-0.5518026948,
0.196303308,
-0.0378330462,
-0.1260570288,
0.1373005062,
0.1882483959,
-0.0334712304,
-0.2720348239,
0.0144159943,
0.1643153727,
-0.0944959223,
-0.0992466807,
0.1843338311,
0.4404117167,
0.2046840936,
0.0925298631,
-0.0192534104,
-0.0986651182,
-0.2265909016,
-0.1590010524,
-0.1725433618,
-0.1005912572,
-0.251617223,
0.4177373648,
-0.2840109468,
-0.4044751525,
-0.3830953538,
0.265268296,
0.041874215,
-0.0697606578,
-0.2140800059,
0.4274218678,
-0.1089174375,
-0.0424475558,
0.18213287,
0.2948851287,
0.0220187679,
0.394222796,
-0.3259971738,
-0.2536973357,
0.7968421578,
-0.4785506725,
0.2696473002,
-0.1940094531,
0.3755393624,
0.4057741165,
-0.3926133215,
-0.4662137926,
-0.0183524415,
0.0767980441,
0.1142917126,
-0.1075979918,
0.2290247977,
-0.400103271,
-0.0505048521,
-0.0820424855,
0.3052809238,
0.1680761427,
0.0183530748,
-0.0207120627,
-0.1692498922
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | > The fact a datasets.Metric object cannot be used as a simple compute function in a multi-process environment is, in my opinion, a bug in datasets
Yes totally, this use case is supposed to be supported by `datasets`. And in this case there shouldn't be any collision between the metrics. I'm looking into it :)
My guess is that at one point the metric isn't using the right file name. It's supposed to use one with a unique uuid in order to avoid the collisions. | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 85 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
> The fact a datasets.Metric object cannot be used as a simple compute function in a multi-process environment is, in my opinion, a bug in datasets
Yes totally, this use case is supposed to be supported by `datasets`. And in this case there shouldn't be any collision between the metrics. I'm looking into it :)
My guess is that at one point the metric isn't using the right file name. It's supposed to use one with a unique uuid in order to avoid the collisions. | [
-0.1335233152,
0.0371194482,
0.092597425,
0.2165250629,
0.1155615449,
0.0259055793,
0.2892276645,
0.2465748787,
0.2430667728,
0.1485801041,
0.0223328546,
0.1771806926,
-0.2399051338,
0.1134239659,
-0.0412851162,
-0.0090812463,
-0.03215212,
-0.0648059845,
-0.2784070671,
-0.0532976128,
-0.371278137,
0.4944945574,
-0.2136931568,
0.0575830713,
-0.4752723873,
0.0689633042,
-0.198795557,
0.1876265407,
-0.1207084134,
-0.4874231219,
0.3815602064,
0.1813318878,
0.0492467731,
0.6484948397,
-0.0001204536,
-0.1653450131,
0.2496986091,
-0.0945593044,
-0.1752889603,
-0.2627053857,
-0.0832362473,
-0.3161791265,
0.4207962453,
-0.2127365768,
0.1441107839,
-0.0956148505,
0.032539852,
-0.4734923244,
0.3829441071,
0.0713171437,
0.1092246026,
0.4045967758,
-0.0290137678,
-0.3105271459,
-0.0231042057,
-0.0660713539,
0.0089848712,
0.3649273813,
-0.1367456615,
0.07073275,
-0.2014991492,
0.1797534078,
0.1654922515,
0.0047446489,
0.2488135695,
0.0069310218,
-0.0829745084,
-0.1220627576,
0.0122205317,
0.336471498,
0.2785387337,
-0.2549472153,
-0.3165338039,
0.0869157165,
-0.2146291137,
-0.448187232,
0.1836498678,
-0.1002552509,
0.06786713,
0.0524203703,
-0.056367524,
-0.1756003201,
-0.051679574,
0.0565950014,
-0.159650892,
-0.1350139827,
-0.2554568648,
0.1798074692,
0.0026668597,
0.1714439392,
-0.3266617358,
0.2590617836,
0.0486182198,
0.1027815789,
-0.5279274583,
0.2013274133,
0.173094362,
-0.0289441571,
0.1072826982,
0.4302811027,
-0.0310635846,
0.1589916945,
0.4029906392,
0.194481492,
-0.0551078469,
0.4064207971,
0.1782256365,
0.0014899746,
0.2801540792,
0.1749694645,
-0.223209098,
-0.3422786593,
0.150229454,
-0.3470233679,
0.3246975243,
0.0795395225,
0.179981932,
-0.3473768234,
-0.0161985159,
0.181060493,
-0.1118374914,
-0.1612891853,
0.0996881723,
0.2822613716,
0.186663121,
-0.0516056269,
0.2157450169,
0.4055100381,
-0.1794900298,
0.0225591585,
-0.4205357432,
-0.1213242561,
-0.1182070002,
0.4025936127,
0.0458071679,
-0.176245138,
0.5597070456,
-0.1764396727,
0.3955850303,
-0.149099052,
0.3421093822,
0.1321392208,
0.0209953431,
0.3017513454,
0.0507129356,
0.2459190935,
0.2705453038,
-0.0035894737,
-0.102076225,
-0.3254127502,
-0.1897499859,
-0.5088409781,
0.091111064,
-0.0049068416,
-0.4515232444,
0.2914335728,
0.2894198895,
0.2298970222,
-0.0728458166,
-0.2428897172,
0.0437056199,
0.039170824,
-0.2443352938,
-0.1741628647,
0.5107159615,
0.6710847616,
-0.3306004405,
-0.346932888,
-0.0772821754,
-0.0398801193,
-0.0461151488,
0.2499621063,
-0.0406616107,
0.2462096959,
-0.2826662362,
0.133400321,
0.276478678,
-0.3176389635,
-0.4375064969,
-0.012118794,
-0.3603681028,
-0.1362520307,
0.0509688705,
-0.0041755512,
0.1422250271,
0.1615345627,
0.0578496158,
-0.1849364191,
0.1697217971,
-0.3091490567,
-0.3021736145,
-0.2162057608,
0.012287315,
-0.0623265356,
0.3494413793,
0.0934795514,
0.0866660327,
0.0533546917,
0.0463508032,
0.0644140393,
0.2311800718,
0.2167261392,
0.2976896167,
-0.0360913686,
0.3019511998,
-0.0816925094,
-0.2597102225,
0.3623228073,
-0.1630402505,
0.0962200463,
0.1023660004,
0.026928775,
-0.1756435037,
-0.0892354995,
-0.1704856753,
-0.4365582466,
0.0410100073,
-0.0181953274,
0.2263961583,
0.1631460339,
-0.1785072684,
0.2397714853,
0.1039651334,
0.1861557364,
-0.4449471235,
0.0735413954,
-0.0304206368,
-0.2937314808,
-0.0068504736,
0.1123706698,
0.0569214113,
0.0660692826,
-0.1975280643,
0.4633433521,
-0.0460407175,
0.2614346445,
0.4305958748,
0.1458352357,
-0.1093314961,
-0.3408474922,
-0.0262468737,
-0.0013611242,
0.0535814166,
-0.0572732985,
-0.2175342888,
0.3621325493,
0.0230479743,
0.1479342133,
-0.2466673255,
0.3745266795,
-0.0810132027,
-0.056190446,
-0.5400526524,
-0.0346519202,
0.3048756123,
-0.2932921946,
0.3406457007,
-0.149043113,
-0.0125550926,
-0.0973651186,
0.2400291711,
0.14147982,
0.1339054406,
0.031243708,
-0.3177609146,
-0.0283259526,
0.0000276733,
-0.0193966553,
0.530139327,
0.1484724134,
0.2678799331,
0.1348846704,
0.0870614797,
-0.1628808677,
0.0027409866,
0.0954414979,
0.005738318,
0.0730256587,
0.2060081065,
0.1222143099,
-0.2972864807,
-0.0545393676,
-0.2019448876,
0.0336648971,
-0.3863703012,
0.1017450392,
-0.2474814653,
0.4286674857,
0.1342417598,
0.1097207069,
-0.2200014144,
-0.2977715433,
0.1170778722,
-0.1271333098,
-0.0012807176,
0.312443763,
-0.1825162917,
0.5101053715,
0.1200418696,
-0.1082283184,
-0.2282789946,
-0.1558557898,
-0.0329199657,
-0.1611796468,
-0.1028815508,
0.2642017901,
0.131570518,
-0.1125557572,
0.1509696543,
-0.2350513637,
-0.427885294,
0.063287504,
0.0025525047,
0.7433800697,
0.4111454487,
-0.0036351457,
-0.2799823284,
0.1605405211,
0.4311719537,
-0.4666102529,
-0.2351150364,
0.1392443031,
-0.0111168623,
0.0810802206,
-0.2477898598,
-0.3338744938,
0.0676991642,
-0.3375909328,
0.6766374707,
-0.0679345354,
-0.0718750209,
-0.229868114,
-0.0328819677,
0.259098798,
-0.2438748032,
0.2232502997,
-0.3715210557,
-0.6017814875,
0.3536576033,
-0.1585830152,
-0.3658099771,
-0.0293396339,
0.0894152597,
0.4120745063,
0.0269323215,
-0.5248667002,
-0.4424890578,
-0.0557458773,
0.2325975597,
0.0255856235,
0.0923559666,
0.4470264018,
-0.1033084095,
0.013612695,
-0.2129085362,
-0.3032341301,
0.5885706544,
-0.1253103763,
0.0313166007,
-0.024689056,
0.4544473588,
0.1785294712,
0.8193630576,
0.573984921,
0.2213239521,
0.2435995936,
0.0272401702,
0.4192949235,
0.010075707,
-0.240745455,
0.3913174272,
0.0835296959,
-0.1487622857,
0.1663929373,
0.1160772592,
0.1952005327,
-0.1400610209,
0.0190409794,
-0.3352612555,
-0.2709818482,
-0.1398508102,
-0.0637857467,
0.2285215259,
0.1551154256,
0.0282766372,
-0.253960371,
0.092576772,
0.3488355875,
0.3272806108,
0.1989055872,
0.1866466403,
-0.2482558489,
0.0456896871,
-0.5986584425,
0.3302704692,
-0.4330098331,
0.1360399574,
-0.1070089713,
-0.1760909557,
0.120939292,
-0.0360519141,
0.6881154776,
-0.0618485138,
-0.2268879414,
-0.0087778345,
-0.2107593119,
-0.5308991671,
0.0313062593,
-0.1429648995,
0.0433674529,
0.2762987614,
0.4072863162,
-0.4166150689,
-0.1702868342,
-0.1740849316,
0.0234402381,
-0.2182139307,
-0.3281385601,
-0.3009988368,
-0.0614097193,
-0.2629103065,
0.057481125,
-0.2308471501,
0.0075306296,
-0.0831389874,
0.0304852948,
-0.2600258887,
0.0171638057,
0.0119881593,
0.1035653502,
0.3259924948,
-0.3720790744,
0.2957859635,
0.561434269,
-0.0697113574,
0.1372143328,
0.4020889401,
0.0194237344,
-0.1499095261,
0.3707309961,
-0.0855317414,
0.2840798795,
0.453029871,
-0.3448778987,
-0.0066800229,
-0.01851926,
0.410857141,
-0.1840827018,
-0.0980244055,
0.3913735449,
0.0029594749,
0.2447970808,
0.1449106187,
0.2343635559,
0.3837969899,
-0.1564328671,
0.2894921899,
0.0861305892,
-0.1000659615,
0.2511704564,
0.0149510428,
0.8027768135,
0.0547762811,
0.0158251524,
0.1936587542,
0.0883166343,
0.5205698609,
0.1113067567,
-0.0199036039,
-0.2456326932,
-0.3328892589,
0.0005449951,
-0.1819362044,
0.0331190601,
-0.2277821898,
-0.253883332,
0.3600266278,
-0.1687325835,
-0.1300897598,
-0.1787784696,
-0.049532,
-0.238525331,
-0.1991710812,
0.1351705492,
0.0546531305,
0.1097515821,
0.4981808662,
-0.170625329,
-0.1447527707,
-0.2722233534,
-0.3178370595,
-0.2482748032,
-0.0900001153,
-0.1977527738,
-0.0111125782,
0.2499722987,
-0.0639047176,
-0.0252632983,
0.2677812278,
0.5074549913,
0.0472655185,
-0.3525553048,
0.0034764241,
0.0130432919,
0.1437118053,
0.116122663,
-0.1211204678,
0.2373103499,
-0.0689930692,
-0.0926970243,
-0.0657149553,
-0.1648803055,
0.1208206043,
0.1018047333,
-0.1147971302,
-0.2134728283,
-0.0577050969,
0.0344792753,
0.0207517389,
0.0414713919,
-0.373603344,
0.0758249387,
0.2722466886,
-0.1465555876,
0.17229864,
0.1113969386,
-0.2608965933,
-0.0141504025,
0.5147312284,
-0.2373295426,
-0.030342415,
0.410490483,
0.1308342516,
-0.0283202529,
-0.1000232771,
0.0894251987,
0.4136511087,
-0.710258007,
0.1728536487,
-0.0983268321,
0.1189660132,
0.0743288994,
0.1602977067,
0.2923332453,
-0.223300904,
-0.0529581159,
-0.3134145141,
-0.3251957595,
0.2963857651,
-0.3277480602,
0.1088097692,
-0.3996323943,
0.1098277718,
-0.1114411429,
0.2352117896,
-0.2688810825,
0.2411135733,
-0.2456685603,
-0.1734367758,
0.0200255327,
0.0298221931,
0.3511941731,
-0.059526518,
-0.0846800432,
0.2033029646,
-0.357853353,
-0.0760752037,
-0.2357935905,
0.1430658549,
-0.0966828912,
0.0890797526,
0.0271422081,
0.1730924249,
-0.1648470759,
-0.1966083646,
0.0914930701,
0.2636886239,
-0.0502029322,
0.0855075717,
0.1101142988,
0.1596431434,
-0.0471668132,
0.1796096861,
0.1540254503,
-0.1877578199,
-0.1319302171,
0.1650541574,
-0.2592051327,
-0.0534540489,
0.2167652845,
0.0128263682,
-0.129029423,
0.1538249403,
0.0696357489,
-0.0815031081,
-0.3034445643,
-0.0409246832,
0.586925149,
0.229036361,
-0.1482739896,
0.0973779038,
-0.0870539322,
0.0747682378,
-0.1308404058,
0.2859938145,
0.0804743469,
0.2450645864,
0.1103253514,
0.194886446,
0.147214666,
0.0447324365,
-0.033263661,
0.1988372803,
-0.0040262081,
-0.2192634046,
0.3405558467,
-0.1342507601,
0.0446110219,
0.395119369,
0.5971513391,
0.1730560213,
0.1452833265,
0.1526365876,
-0.0955006853,
0.2259517461,
-0.1051745266,
0.317173481,
-0.0828348473,
-0.1924596131,
-0.0912107229,
0.0132809132,
-0.0581707656,
-0.2416954488,
-0.4781819284,
0.7504735589,
-0.3104274869,
-0.2048028857,
-0.2497940809,
0.2282944024,
-0.1533307433,
-0.1966205388,
-0.1678115129,
-0.1948922873,
-0.1005985141,
-0.1622462124,
0.190017134,
0.0440238081,
0.2180061638,
-0.0209908001,
0.1075952202,
-0.3651961684,
-0.5040787458,
0.2400594503,
0.0598788112,
-0.1469106972,
-0.0378541611,
0.242474705,
-0.3101006448,
0.1404375434,
0.6258373857,
0.3339038789,
0.0355101489,
-0.1565676332,
0.095711939,
0.2010085732,
0.00261911,
-0.1123623997,
0.1926766634,
0.1783116013,
-0.4704276025,
0.2172094285,
0.0601484701,
-0.1306461394,
0.2637720108,
-0.2489309013,
0.3911227286,
-0.3739432991,
0.4142304063,
-0.3403812051,
0.1366836578,
0.0856259763,
-0.1665982604,
-0.5777475238,
0.2351000011,
0.0391672999,
-0.0788601488,
0.1047068089,
-0.3096340597,
0.0495044366,
-0.1320972592,
0.2266010493,
0.3706276119,
0.1260541677,
-0.301186353,
-0.249292329,
-0.7142770886,
0.3186247051,
0.1262674332,
0.3032761216,
-0.0899853483,
0.3800607324,
-0.0494270138,
0.0609033853,
0.2452539206,
0.0120625384,
-0.0164226294,
0.0251141246,
-0.5366045833,
0.054859098,
-0.2019568831,
-0.1890057921,
0.1295390427,
0.2296339869,
0.0115955099,
-0.2051223069,
-0.0056057051,
0.0943916589,
-0.0876889899,
-0.106506303,
0.1045286208,
0.4748960733,
0.2246722579,
0.1173731536,
-0.0440186709,
-0.0084057897,
-0.2767063081,
-0.109207429,
-0.0843694955,
-0.0301335752,
-0.3517560959,
0.4323346615,
-0.3502947688,
-0.3416683078,
-0.3375523686,
0.1549045146,
0.1130458489,
-0.1122547537,
-0.2595621347,
0.4067007601,
-0.1336850673,
0.0444316491,
0.2433537245,
0.377145499,
-0.0471836068,
0.4603344202,
-0.2675012648,
-0.1776427031,
0.7530431747,
-0.4552133977,
0.132674396,
-0.2201702893,
0.3438362181,
0.3547530174,
-0.4648525119,
-0.3702157736,
0.0004915744,
0.1647121757,
0.0972571075,
-0.2037417889,
0.1568378359,
-0.3975246847,
-0.0358707607,
-0.0410868637,
0.2923148572,
0.1856916845,
-0.0826073438,
0.1278859079,
-0.031436149
] |
https://github.com/huggingface/datasets/issues/1942 | [experiment] missing default_experiment-1-0.arrow | I just opened #1966 to fix this :)
@stas00 if have a chance feel free to try it ! | the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you. | 19 | [experiment] missing default_experiment-1-0.arrow
the original report was pretty bad and incomplete - my apologies!
Please see the complete version here: https://github.com/huggingface/datasets/issues/1942#issuecomment-786336481
------------
As mentioned here https://github.com/huggingface/datasets/issues/1939 metrics don't get cached, looking at my local `~/.cache/huggingface/metrics` - there are many `*.arrow.lock` files but zero metrics files.
w/o the network I get:
```
FileNotFoundError: [Errno 2] No such file or directory: '~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow
```
there is just `~/.cache/huggingface/metrics/sacrebleu/default/default_experiment-1-0.arrow.lock`
I did run the same `run_seq2seq.py` script on the instance with network and it worked just fine, but only the lock file was left behind.
this is with master.
Thank you.
I just opened #1966 to fix this :)
@stas00 if have a chance feel free to try it ! | [
0.0502676368,
-0.0307856351,
0.0585256964,
0.1668777764,
0.0344584174,
0.0303149968,
0.1830065846,
0.293956995,
0.2358810604,
0.1017370373,
0.1023577824,
0.1887699515,
-0.321457386,
0.0011921301,
0.0528261848,
-0.0287235174,
-0.050264366,
-0.0096708201,
-0.358310163,
-0.1254033446,
-0.2325506806,
0.5447105169,
-0.1571904868,
-0.0045520701,
-0.508487165,
0.2233061194,
-0.1877895743,
0.1402112991,
-0.0968824551,
-0.553837955,
0.4047329426,
0.1505976021,
0.0910342783,
0.4610120654,
-0.0001223977,
-0.1811484545,
0.1949374974,
-0.1929724514,
-0.1423967779,
-0.106889084,
-0.006171003,
-0.2863014042,
0.5193980336,
-0.1492535621,
-0.06222165,
-0.0420733131,
0.0282989834,
-0.2860836983,
0.5111196041,
0.0969570354,
0.1117405593,
0.382009685,
0.0355152488,
-0.3040630221,
0.0467134416,
-0.0588731207,
-0.0233320668,
0.3352956772,
-0.1401586235,
0.0310186744,
-0.1860098839,
0.279789269,
0.2617022991,
-0.0712250471,
0.2279257029,
0.0110843405,
-0.1231671497,
-0.1699433327,
0.0016833721,
0.2876954377,
0.2455522418,
-0.1002673358,
-0.2289028913,
0.135090664,
-0.1488544345,
-0.2994298637,
0.151556015,
-0.1007178426,
0.0996645764,
0.113329038,
0.0295816548,
-0.1865184605,
-0.1075744405,
0.0627110153,
-0.1474568099,
-0.0327550992,
-0.1966676116,
0.0187705979,
-0.0397601537,
0.0962036103,
-0.3239342272,
0.4986530542,
-0.0832983702,
0.1876076907,
-0.3713002503,
0.0709056258,
0.2739275694,
0.0947959498,
0.0921185613,
0.3463566899,
-0.0080722701,
0.079876855,
0.4184753299,
0.081495434,
-0.0967770591,
0.4281896353,
0.2606996298,
-0.0058042742,
0.1606585085,
0.2437302023,
-0.2130865753,
-0.2938712835,
0.1996714324,
-0.2233796418,
0.3632268012,
0.0774253458,
0.0587492883,
-0.4959129095,
0.1117426977,
0.2486954629,
-0.1679580808,
-0.207285583,
0.1042116657,
0.3869488239,
0.1424579769,
-0.0502152294,
0.2787383199,
0.3803990781,
-0.1684308052,
-0.0360854194,
-0.4457575679,
-0.1796837747,
-0.0967058688,
0.3202261329,
0.1431294978,
-0.0712345243,
0.6667428017,
-0.1645594537,
0.3357943892,
-0.1272890121,
0.250703007,
0.2788631022,
0.0503678471,
0.2963224351,
0.0175311938,
0.1571663022,
0.3135043979,
0.1565277874,
-0.1912124753,
-0.3347263932,
-0.1577663869,
-0.5147595406,
0.0790074244,
0.0095946034,
-0.3871157467,
0.1222409979,
0.4855632186,
0.2077404261,
-0.1191088334,
-0.1844087243,
0.1640954316,
0.0471464172,
-0.0238040015,
-0.0801770687,
0.5420817137,
0.6158709526,
-0.3337561786,
-0.3810563385,
0.0857882649,
-0.0635163635,
-0.0396001413,
0.2201703638,
-0.0056289807,
0.139670819,
-0.2328994274,
0.1088680923,
0.3703118861,
-0.2679547071,
-0.5212198496,
-0.0552310757,
-0.3564216197,
-0.1850545406,
0.0904624164,
0.0134213418,
0.0928365663,
0.2254571021,
0.0927810296,
-0.2245486528,
0.2309309393,
-0.3117345572,
-0.3569084704,
-0.2340022326,
0.0308412686,
-0.0627723634,
0.3183654249,
0.0701390132,
0.0489343964,
0.0875611901,
0.1221266836,
0.0490468107,
0.137917757,
0.2284752727,
0.405058831,
0.0881535113,
0.2707898021,
0.1175324693,
-0.1957638711,
0.2127176225,
-0.3725411892,
0.147275269,
0.1539964974,
-0.0014505386,
-0.2130687684,
-0.1569034755,
-0.0808191523,
-0.497097373,
0.0302162617,
-0.1321008503,
0.254652828,
0.1813735366,
-0.1608059853,
0.0999847725,
0.0706337616,
0.2504445016,
-0.4489240646,
0.0808875188,
-0.1008736789,
-0.3557091653,
0.010535039,
0.12481758,
0.0659185499,
0.0568635501,
-0.2100695372,
0.3609478176,
-0.196022898,
0.2700867057,
0.3957690597,
0.1259014308,
-0.17008394,
-0.4662602544,
0.0135637131,
0.0963738412,
-0.1102106944,
0.034308359,
-0.229419142,
0.3718854189,
0.1129395515,
0.0852605551,
-0.3511000872,
0.3622634411,
-0.041050788,
-0.1646937579,
-0.4694804549,
-0.1569864452,
0.3380330801,
-0.2630415559,
0.1648105383,
-0.1352575421,
-0.0066631958,
0.0267158262,
0.1505911648,
0.1934216321,
0.1921197474,
0.0640549958,
-0.2622755766,
-0.1365640461,
0.0338334143,
-0.1024968922,
0.367092371,
0.195018515,
0.2114623189,
0.0737836063,
0.1446099281,
-0.1691471934,
0.0960003212,
-0.0084994808,
-0.1350521892,
0.0892327577,
0.2438503206,
0.189613089,
-0.3362029195,
-0.0146995857,
-0.3934497535,
0.0733063072,
-0.3259212971,
-0.0099368766,
-0.163627252,
0.3344519436,
0.183931008,
0.0512284189,
-0.2567329407,
-0.2762012482,
0.2306937277,
-0.2379373461,
-0.0242274292,
0.2272980511,
-0.3410355747,
0.5317053795,
0.134836182,
-0.0654155388,
-0.2855355144,
-0.1330368817,
-0.0507568754,
-0.1544406116,
-0.2762284279,
0.2058389187,
0.1834728867,
-0.1706163585,
0.0736598074,
-0.0385265388,
-0.4664660096,
0.0471298844,
-0.0269776471,
0.6298751831,
0.329867512,
0.0796628147,
-0.2673367858,
0.0408606865,
0.4414917529,
-0.4318048358,
-0.2710913718,
0.0875407755,
0.0048037581,
0.1961856484,
-0.2463934124,
-0.3488063812,
0.0998701602,
-0.331379205,
0.6622591019,
0.0467878431,
-0.0492480695,
-0.206979081,
-0.0708792359,
0.2037697732,
-0.294033438,
0.3168636262,
-0.3551405668,
-0.6845228672,
0.4514543414,
-0.152281031,
-0.279117465,
0.0029129162,
0.0694524795,
0.4111828804,
-0.1043940485,
-0.5750929713,
-0.4180472493,
-0.0496869832,
0.2071719319,
-0.0043237191,
0.0652836785,
0.4947386682,
-0.1225789338,
-0.0312671587,
-0.0818336532,
-0.2440458238,
0.5067591667,
-0.0428347588,
0.1758794039,
-0.124929443,
0.44918257,
0.1300964206,
0.8017881513,
0.5838682652,
0.2834764123,
0.2731365263,
0.061596036,
0.3118467331,
-0.1129288375,
-0.2101449668,
0.3428442776,
0.2255456746,
-0.0879664272,
0.2291420102,
0.1611827761,
0.1728361845,
-0.112584956,
-0.0583816916,
-0.3652353287,
-0.3408975005,
-0.0870220885,
-0.0281338487,
0.2828760743,
-0.0352772586,
0.0498063266,
-0.2052810043,
0.0224271864,
0.4554182887,
0.3214439154,
0.1747523695,
0.201236248,
-0.1860411018,
0.0822222382,
-0.6240666509,
0.2883467674,
-0.5249202847,
0.1255549043,
-0.2078336477,
-0.3026986122,
0.1475294232,
0.028411366,
0.5838213563,
0.0967597663,
-0.1193182841,
-0.0351821594,
-0.2145914137,
-0.4352737069,
-0.0572161488,
-0.1044352055,
-0.0560940877,
0.362449795,
0.3101023436,
-0.3499494791,
-0.1848711818,
0.0001568124,
-0.1191486716,
-0.2800284624,
-0.3077386916,
-0.2760515213,
0.0416325927,
-0.1664546728,
0.1114424989,
-0.2791820765,
-0.0038867183,
0.0408115759,
0.0349707156,
-0.1482826024,
-0.1122354865,
0.0149490684,
0.1882762909,
0.370434165,
-0.2815082669,
0.2173378319,
0.5262380242,
-0.1605110466,
0.030057786,
0.3995454907,
0.0934049636,
0.0072202124,
0.2162244767,
-0.126062572,
0.312265873,
0.479080826,
-0.3064969778,
0.0584019236,
-0.0449392498,
0.4254007638,
-0.1694804281,
-0.1294704527,
0.3659029007,
-0.1286428273,
0.406768173,
0.2168536782,
0.2520106435,
0.1982664317,
-0.1372208893,
0.19229877,
0.1228666008,
-0.1298316717,
0.4037844837,
0.0725144297,
0.906662643,
0.2284973264,
0.1569652706,
0.1142865568,
0.1287258416,
0.4183023572,
0.1660072207,
0.0168893561,
-0.2391741723,
-0.2163566798,
0.0218368918,
-0.1294558793,
-0.0922044665,
-0.3341310322,
-0.3491735458,
0.3339422047,
-0.416236341,
-0.1714327335,
-0.2300451994,
-0.2589848638,
-0.2749918103,
-0.1125247031,
0.0832924917,
0.0393722728,
0.1273488104,
0.5895766616,
-0.1984668076,
-0.1865691543,
-0.308735311,
-0.3782607317,
-0.1994605064,
-0.0075616091,
-0.1780513227,
-0.0446497463,
0.3563722074,
-0.1140277833,
-0.1081578434,
0.2747522295,
0.424838841,
0.1410204917,
-0.4095394015,
-0.0447907224,
0.0970979333,
-0.0064654201,
0.0882154554,
0.0607837029,
0.2227433324,
-0.0427296832,
-0.0432672948,
0.0264808815,
-0.1708240658,
0.1513323933,
-0.0115303397,
-0.1838585436,
-0.2010068446,
-0.1706020236,
-0.0206781253,
0.1199316084,
0.0131870434,
-0.2438447028,
0.0945434198,
0.3195438981,
-0.2366113216,
0.1028823853,
0.2528242469,
-0.223063767,
-0.0671369731,
0.578999579,
-0.3150368035,
-0.0664430708,
0.4722224772,
0.0892630294,
-0.0668759272,
-0.150577575,
0.0398671031,
0.4476165175,
-0.7930828333,
0.0682799444,
-0.0052739903,
0.0292085558,
0.1369701326,
0.1824986339,
0.3927781284,
-0.2080863714,
-0.0718287006,
-0.1729664207,
-0.4252093136,
0.2752407789,
-0.3139513433,
0.1846447736,
-0.5226022005,
0.0904924273,
-0.1134835631,
0.324596256,
-0.2781523168,
0.1821646392,
-0.2456913292,
-0.1479724646,
-0.1570419967,
-0.056665156,
0.3577842712,
-0.003350541,
-0.0518476926,
0.1187390313,
-0.3425807357,
-0.0974131748,
-0.2557144761,
0.1442090422,
-0.1051665917,
0.1344411671,
-0.0343742855,
0.2559404373,
-0.0785719082,
-0.0954964906,
0.0442784801,
0.1784218699,
0.0884612352,
0.2135746628,
0.1516380757,
0.0300448574,
-0.0655760467,
0.1676619202,
0.1536466777,
-0.2112170458,
-0.0985517725,
0.0039767297,
-0.3737254143,
-0.080363974,
0.2566914856,
0.1499453634,
-0.2190530896,
0.1602819264,
0.031648308,
0.0055188476,
-0.3093928099,
-0.0150977112,
0.4742202759,
0.1363733262,
-0.1317465454,
0.0929517895,
-0.0725810528,
0.1205276921,
-0.1349429935,
0.3159371614,
0.0219195895,
0.3718647361,
-0.0062720254,
0.2041679323,
0.0908579975,
-0.0251448825,
0.0929936692,
0.215562582,
-0.039973747,
-0.1805282086,
0.3490176201,
-0.0985131115,
0.0237436146,
0.2782060504,
0.5448619127,
0.1011171341,
0.1475643814,
-0.053107921,
-0.0803423524,
0.1710300297,
-0.1602042615,
0.3541840017,
-0.1122850031,
-0.1044133008,
-0.1984597296,
0.0329713598,
0.0545219891,
-0.1761419475,
-0.4291501939,
0.8031214476,
-0.1808703989,
-0.2441589534,
-0.1937090158,
0.2028092593,
-0.1682159901,
-0.1527667344,
-0.1624448895,
-0.2118791938,
-0.1720937192,
-0.1443361342,
0.2267824709,
0.0363527313,
0.0690791458,
0.1380732208,
0.0689272061,
-0.3447803855,
-0.4604191184,
0.2070624232,
0.1764552146,
-0.1624523699,
-0.0082210107,
0.1841731966,
-0.3044585288,
0.1713965237,
0.6305413842,
0.2583574951,
0.0060696024,
-0.1108447909,
0.111041531,
0.2097602785,
-0.0499811843,
-0.1143165827,
0.1608442366,
0.2634719908,
-0.5008488894,
0.036257863,
0.0499520525,
-0.1437648237,
0.3807952702,
-0.3335393071,
0.3961321115,
-0.1687196046,
0.4106987119,
-0.2003556341,
0.1380611509,
-0.0227758326,
-0.2151287347,
-0.6650733948,
0.217035979,
-0.1144521013,
-0.0244506337,
0.0889874995,
-0.2698172033,
0.0400800258,
-0.1257956475,
0.2337580025,
0.414493233,
0.1051799506,
-0.2935045063,
-0.3102203012,
-0.508348763,
0.2386899889,
0.2074898779,
0.3287261724,
-0.0308864899,
0.397841543,
-0.0995379016,
0.0373932272,
0.2820659876,
-0.0539018288,
-0.076370053,
-0.0365376435,
-0.532299757,
0.2037282586,
-0.0945926011,
-0.0557685047,
0.1298421919,
0.3552445769,
-0.0351367258,
-0.2536828816,
-0.0055234581,
0.1682239175,
-0.2035001218,
-0.1541195959,
0.1702325493,
0.4273455143,
0.1514545381,
0.0878896713,
-0.0181999505,
-0.0674418211,
-0.23749125,
-0.1905188262,
-0.0524032861,
-0.1506196707,
-0.2223513871,
0.4425064325,
-0.3113907576,
-0.4462625682,
-0.3600743115,
0.2234850228,
0.0713067204,
-0.1215925217,
-0.2035282552,
0.3574289978,
-0.0377723873,
-0.0298091769,
0.278734535,
0.3011472225,
-0.0210173167,
0.3851297796,
-0.3222492337,
-0.2338647246,
0.7657583952,
-0.477670908,
0.2892949283,
-0.2233699858,
0.3952085674,
0.3707592487,
-0.4478724003,
-0.4201608896,
-0.0033489019,
0.1377255619,
0.0659001321,
-0.0941235423,
0.1670298129,
-0.2764945626,
0.0034389347,
-0.0241550058,
0.3220016956,
0.2488746345,
0.0636227503,
-0.035592325,
-0.1095626056
] |
https://github.com/huggingface/datasets/issues/1941 | Loading of FAISS index fails for index_name = 'exact' | Works great 👍 I just put a minor comment on the commit, I think you meant to pass the `train_size` from the one obtained from the config.
Thanks for a quick response! | Hi,
It looks like loading of FAISS index now fails when using index_name = 'exact'.
For example, from the RAG [model card](https://huggingface.co/facebook/rag-token-nq?fbclid=IwAR3bTfhls5U_t9DqsX2Vzb7NhtRHxJxfQ-uwFT7VuCPMZUM2AdAlKF_qkI8#usage).
Running `transformers==4.3.2` and datasets installed from source on latest `master` branch.
```bash
(venv) sergey_mkrtchyan datasets (master) $ python
Python 3.8.6 (v3.8.6:db455296be, Sep 23 2020, 13:31:39)
[Clang 6.0 (clang-600.0.57)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration
>>> tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-nq")
>>> retriever = RagRetriever.from_pretrained("facebook/rag-token-nq", index_name="exact", use_dummy_dataset=True)
Using custom data configuration dummy.psgs_w100.nq.no_index-dummy=True,with_index=False
Reusing dataset wiki_dpr (/Users/sergey_mkrtchyan/.cache/huggingface/datasets/wiki_dpr/dummy.psgs_w100.nq.no_index-dummy=True,with_index=False/0.0.0/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb)
Using custom data configuration dummy.psgs_w100.nq.exact-50b6cda57ff32ab4
Reusing dataset wiki_dpr (/Users/sergey_mkrtchyan/.cache/huggingface/datasets/wiki_dpr/dummy.psgs_w100.nq.exact-50b6cda57ff32ab4/0.0.0/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb)
0%| | 0/10 [00:00<?, ?it/s]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 425, in from_pretrained
return cls(
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 387, in __init__
self.init_retrieval()
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 458, in init_retrieval
self.index.init_index()
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 284, in init_index
self.dataset = load_dataset(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/load.py", line 750, in load_dataset
ds = builder_instance.as_dataset(split=split, ignore_verifications=ignore_verifications, in_memory=keep_in_memory)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/builder.py", line 734, in as_dataset
datasets = utils.map_nested(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/utils/py_utils.py", line 195, in map_nested
return function(data_struct)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/builder.py", line 769, in _build_single_dataset
post_processed = self._post_process(ds, resources_paths)
File "/Users/sergey_mkrtchyan/.cache/huggingface/modules/datasets_modules/datasets/wiki_dpr/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb/wiki_dpr.py", line 205, in _post_process
dataset.add_faiss_index("embeddings", custom_index=index)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/arrow_dataset.py", line 2516, in add_faiss_index
super().add_faiss_index(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/search.py", line 416, in add_faiss_index
faiss_index.add_vectors(self, column=column, train_size=train_size, faiss_verbose=faiss_verbose)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/search.py", line 281, in add_vectors
self.faiss_index.add(vecs)
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/faiss/__init__.py", line 104, in replacement_add
self.add_c(n, swig_ptr(x))
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/faiss/swigfaiss.py", line 3263, in add
return _swigfaiss.IndexHNSW_add(self, n, x)
RuntimeError: Error in virtual void faiss::IndexHNSW::add(faiss::Index::idx_t, const float *) at /Users/runner/work/faiss-wheels/faiss-wheels/faiss/faiss/IndexHNSW.cpp:356: Error: 'is_trained' failed
>>>
```
The issue seems to be related to the scalar quantization in faiss added in this commit: 8c5220307c33f00e01c3bf7b8. Reverting it fixes the issue.
| 32 | Loading of FAISS index fails for index_name = 'exact'
Hi,
It looks like loading of FAISS index now fails when using index_name = 'exact'.
For example, from the RAG [model card](https://huggingface.co/facebook/rag-token-nq?fbclid=IwAR3bTfhls5U_t9DqsX2Vzb7NhtRHxJxfQ-uwFT7VuCPMZUM2AdAlKF_qkI8#usage).
Running `transformers==4.3.2` and datasets installed from source on latest `master` branch.
```bash
(venv) sergey_mkrtchyan datasets (master) $ python
Python 3.8.6 (v3.8.6:db455296be, Sep 23 2020, 13:31:39)
[Clang 6.0 (clang-600.0.57)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration
>>> tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-nq")
>>> retriever = RagRetriever.from_pretrained("facebook/rag-token-nq", index_name="exact", use_dummy_dataset=True)
Using custom data configuration dummy.psgs_w100.nq.no_index-dummy=True,with_index=False
Reusing dataset wiki_dpr (/Users/sergey_mkrtchyan/.cache/huggingface/datasets/wiki_dpr/dummy.psgs_w100.nq.no_index-dummy=True,with_index=False/0.0.0/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb)
Using custom data configuration dummy.psgs_w100.nq.exact-50b6cda57ff32ab4
Reusing dataset wiki_dpr (/Users/sergey_mkrtchyan/.cache/huggingface/datasets/wiki_dpr/dummy.psgs_w100.nq.exact-50b6cda57ff32ab4/0.0.0/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb)
0%| | 0/10 [00:00<?, ?it/s]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 425, in from_pretrained
return cls(
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 387, in __init__
self.init_retrieval()
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 458, in init_retrieval
self.index.init_index()
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/transformers/models/rag/retrieval_rag.py", line 284, in init_index
self.dataset = load_dataset(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/load.py", line 750, in load_dataset
ds = builder_instance.as_dataset(split=split, ignore_verifications=ignore_verifications, in_memory=keep_in_memory)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/builder.py", line 734, in as_dataset
datasets = utils.map_nested(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/utils/py_utils.py", line 195, in map_nested
return function(data_struct)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/builder.py", line 769, in _build_single_dataset
post_processed = self._post_process(ds, resources_paths)
File "/Users/sergey_mkrtchyan/.cache/huggingface/modules/datasets_modules/datasets/wiki_dpr/8a97e0f4fa5bc46e179474db6a61b09d5d2419d2911835bd3f91d110c936d8bb/wiki_dpr.py", line 205, in _post_process
dataset.add_faiss_index("embeddings", custom_index=index)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/arrow_dataset.py", line 2516, in add_faiss_index
super().add_faiss_index(
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/search.py", line 416, in add_faiss_index
faiss_index.add_vectors(self, column=column, train_size=train_size, faiss_verbose=faiss_verbose)
File "/Users/sergey_mkrtchyan/workspace/huggingface/datasets/src/datasets/search.py", line 281, in add_vectors
self.faiss_index.add(vecs)
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/faiss/__init__.py", line 104, in replacement_add
self.add_c(n, swig_ptr(x))
File "/Users/sergey_mkrtchyan/workspace/cformers/venv/lib/python3.8/site-packages/faiss/swigfaiss.py", line 3263, in add
return _swigfaiss.IndexHNSW_add(self, n, x)
RuntimeError: Error in virtual void faiss::IndexHNSW::add(faiss::Index::idx_t, const float *) at /Users/runner/work/faiss-wheels/faiss-wheels/faiss/faiss/IndexHNSW.cpp:356: Error: 'is_trained' failed
>>>
```
The issue seems to be related to the scalar quantization in faiss added in this commit: 8c5220307c33f00e01c3bf7b8. Reverting it fixes the issue.
Works great 👍 I just put a minor comment on the commit, I think you meant to pass the `train_size` from the one obtained from the config.
Thanks for a quick response! | [
-0.0889263824,
-0.1103832573,
0.0109984204,
0.0540015623,
0.4101673961,
-0.0569178164,
0.296813041,
0.2508975267,
0.2696115077,
0.199755013,
-0.2217197269,
0.1398443133,
0.1363164186,
-0.1376476288,
-0.1762494147,
0.2843900323,
0.2129580677,
0.1230241731,
-0.0870198235,
-0.102934137,
-0.2702937722,
0.2029289901,
-0.2149683088,
0.2133807838,
-0.3408920765,
0.4708951414,
-0.1025646925,
0.1511470675,
-0.1406459063,
-0.5798462033,
0.3232377172,
-0.0542234071,
0.2287581563,
0.2792838216,
-0.0001177264,
0.1951244175,
0.5912725329,
-0.063896738,
-0.3142659068,
0.0912115052,
0.1551291049,
0.2524675727,
0.1884163916,
-0.1187292188,
-0.2891575694,
-0.2747848928,
0.1698506922,
-0.0614709556,
0.1880252659,
0.2096890509,
0.1545541883,
0.1129424348,
0.3915560246,
-0.2391125262,
0.3550430536,
-0.4309916496,
-0.2338061333,
0.4001961946,
0.0305758175,
-0.0565609112,
-0.0222201422,
0.0928999335,
0.1818180084,
-0.0270643961,
-0.0330317281,
-0.1413351595,
0.3804670274,
0.0840320364,
0.2681782842,
0.1212426424,
0.2935321629,
-0.0311048701,
-0.4443511069,
0.0353070498,
0.1401629448,
0.0986521319,
0.2189644575,
-0.405819118,
0.0952556431,
0.2862178981,
0.0806032121,
0.0194828026,
0.0560734272,
0.2368136346,
-0.0019366667,
0.3903173506,
-0.1297438741,
0.0979247168,
0.0764912963,
-0.2108368576,
-0.2144451588,
0.0950193182,
-0.2728639543,
-0.0225363076,
-0.6444888711,
0.1160784662,
0.0512807965,
-0.2159345597,
0.0352994576,
0.0424009189,
-0.2709307969,
0.2906610966,
0.1008520946,
0.0380019397,
-0.0305453539,
0.0233892426,
-0.0013716631,
0.006833382,
0.1125111654,
0.0044881888,
-0.1529048532,
-0.0514263362,
-0.2755157053,
-0.0540645793,
-0.6907650232,
-0.3722564876,
-0.01457192,
-0.2811404467,
-0.407347858,
-0.0867936909,
-0.2471618205,
0.1289560199,
0.2523565292,
0.7027876377,
0.0485682562,
-0.0792193115,
0.0272351801,
0.164396584,
-0.1352498531,
0.1639369726,
-0.2256950438,
-0.3065268099,
-0.2085853219,
-0.022372691,
0.3425043225,
-0.2723258734,
0.2192731202,
0.0867487192,
-0.0316481665,
0.062571153,
-0.2385466993,
-0.0671571791,
0.1852019727,
0.3195191622,
-0.2793675065,
0.1558533907,
0.2347677201,
0.0704577863,
0.0292038769,
-0.0564666688,
-0.1806446761,
-0.3224923015,
0.1147128344,
0.172033444,
-0.0485992581,
-0.1230745092,
0.3416301906,
0.0278547518,
-0.16015324,
0.1821376532,
-0.0173215382,
-0.2101819515,
-0.2364780456,
0.0118759125,
0.4402869046,
0.3116512895,
-0.1029744223,
-0.2342443168,
-0.0569790825,
0.0164470524,
-0.0192495771,
0.1451807618,
-0.1825053096,
0.1064563245,
-0.3530341983,
0.0480398238,
0.2476073802,
-0.2162939906,
-0.0942615047,
0.0434445739,
0.0055865273,
0.0486932322,
0.002649419,
-0.2383571863,
-0.079919979,
0.1082668751,
0.1997910142,
0.1565125585,
0.2076742351,
-0.1675228477,
-0.2941003144,
-0.4546545148,
0.0983942226,
0.320599854,
0.0272102505,
0.0472555608,
0.0218185652,
-0.1154952943,
0.3780494034,
-0.1594728529,
-0.051047042,
0.110924378,
0.3147562146,
0.2252461165,
0.5496631265,
-0.1197108477,
-0.028309565,
0.4137915671,
-0.1743459553,
0.0491158627,
0.0225036722,
0.081197843,
-0.3368808627,
0.0625073388,
-0.0826764703,
-0.0540841036,
0.0509438589,
0.1031427383,
0.1682229638,
-0.2200391591,
-0.2104472518,
0.242129162,
-0.3746234775,
0.2127413452,
-0.2988702357,
0.1358974278,
-0.1310337633,
-0.304317683,
-0.0259594098,
0.1280352026,
0.3351846933,
-0.2089738846,
-0.1438957453,
0.3537580967,
0.127534166,
-0.536349833,
0.3886003196,
-0.0711897612,
-0.2205823958,
-0.258574307,
-0.0819247738,
0.4026680291,
0.0992993265,
0.0287670344,
-0.1217484623,
0.4608261585,
-0.0165670626,
0.2404153794,
-0.0125435442,
-0.1394444406,
0.3590927124,
-0.1018888876,
0.0247067362,
-0.2612523437,
0.1761954874,
0.1161092371,
0.1975639164,
-0.2353007793,
0.0291064978,
-0.1079018191,
-0.1206867099,
0.1251835227,
-0.0890992284,
0.179740876,
-0.1933807135,
-0.0180684514,
-0.1605872363,
-0.4791120589,
0.2939758599,
0.2113706619,
-0.2033742219,
-0.2445665747,
-0.0433395728,
-0.2511077523,
0.4474210143,
0.0702860504,
-0.1768760532,
0.1726855785,
0.1754901856,
-0.2408616841,
-0.451646179,
-0.1435260326,
0.1325662136,
-0.0127123846,
-0.3138778806,
0.1683809459,
-0.0572852045,
0.0419996455,
-0.0414761901,
-0.2650161386,
-0.3554299176,
-0.19075647,
0.1494047195,
0.0260539167,
-0.0674644858,
0.4098314345,
-0.1587002873,
-0.0234235972,
0.1372430772,
-0.2612145841,
-0.3281145096,
-0.1754925251,
-0.3357376754,
0.0243653432,
-0.1582144797,
-0.1349012256,
-0.0027413582,
0.0413389578,
-0.1111521423,
-0.4048223794,
-0.2892027795,
0.2777304053,
-0.0531001426,
0.1911065578,
0.0515837409,
-0.1222085059,
-0.0527465343,
-0.0224400237,
0.3487251699,
-0.2121925652,
0.0221100971,
0.1077087224,
-0.0921970457,
-0.0403692275,
-0.1590489149,
-0.0769053474,
0.0546552613,
-0.2552224994,
0.2541983724,
-0.098238185,
0.0092389407,
0.1670092046,
0.0950178578,
0.1037696302,
-0.2616461813,
0.0031436011,
-0.3037978709,
-0.0857034698,
0.3209960759,
0.0239907876,
-0.2724446356,
-0.0610824339,
0.0665889457,
-0.2728339732,
0.2826733291,
-0.0803061202,
-0.2875350714,
-0.0009768493,
0.2117815763,
0.1499057114,
0.187191233,
-0.071641326,
-0.2176821232,
-0.1821731329,
-0.0779757798,
-0.0572836027,
0.355032742,
-0.1895642132,
0.228737995,
0.0150353648,
-0.0614270642,
-0.3030913472,
0.6028996706,
0.0793464631,
-0.2462950796,
0.4174587429,
-0.0564925373,
0.2219807506,
-0.1502773017,
-0.1705418229,
0.2374232262,
0.0112561062,
0.2298956811,
0.2118574381,
-0.0392454974,
0.1808157712,
0.0655930415,
0.2212320119,
-0.0633067638,
-0.28131634,
0.1477482915,
0.211394608,
0.1398190856,
0.0216099471,
0.2803670764,
-0.1312763393,
-0.0448245853,
0.4365181923,
0.4124362469,
-0.093806237,
-0.058458373,
-0.0734543577,
-0.094199881,
-0.2126456499,
0.346675396,
-0.0420308299,
0.6930007935,
-0.010785833,
0.009976007,
0.3116781414,
-0.0148643088,
1.0188853741,
0.0143524911,
0.1796170771,
0.4301277995,
0.0960996896,
-0.5665027499,
-0.1661123782,
-0.1035227329,
-0.1321615875,
0.6401349306,
0.5198690295,
-0.3144191504,
-0.3671022356,
0.1703525186,
-0.0147355832,
-0.1291888356,
-0.4554666579,
-0.4305739403,
0.0800217092,
-0.1321149617,
0.0471692495,
-0.0632530302,
0.2533339858,
0.0612500012,
0.3186871409,
-0.5520003438,
-0.1595084816,
-0.016210746,
0.1304313391,
0.2088962793,
-0.0035232231,
0.4091516137,
0.4179936647,
0.274035573,
-0.0023830179,
0.5052621961,
-0.1862425208,
-0.3868593276,
0.3231744468,
0.1895129681,
-0.1025227606,
0.4031800032,
-0.124461323,
0.0519953966,
-0.190173015,
0.1065804735,
-0.133567512,
0.1749093384,
0.0733134449,
0.0210778378,
-0.0387687013,
-0.3867219687,
0.6415423155,
-0.1238804087,
-0.0281987339,
0.1370671988,
0.5084989071,
-0.2649711072,
0.8774347901,
-0.1680253595,
0.7281205654,
-0.1413463801,
-0.0527822599,
0.2257828712,
-0.1320694536,
0.2003127337,
-0.4170895815,
0.1278451085,
-0.5279293656,
0.0115581453,
0.1474643201,
-0.022724852,
0.0728059784,
0.2921898067,
0.013981171,
0.2591738701,
0.0157226175,
0.2496102452,
0.0998790488,
0.4061025381,
0.1601370722,
-0.2743450403,
-0.2626030147,
0.1274070144,
0.2499640286,
0.3748721778,
0.0817506313,
-0.1256571412,
-0.158352077,
0.0780417845,
-0.1681939662,
0.4189149737,
-0.2086646259,
0.2526069582,
0.4359017015,
-0.0881339312,
0.0852473602,
-0.3859871626,
0.6924884915,
-0.0135186343,
-0.2834367752,
0.474570781,
0.0331095457,
-0.1972589791,
-0.1850912273,
-0.1159420758,
0.0572613925,
0.1282583028,
-0.1469918638,
-0.1105545387,
-0.0610213988,
-0.1273231208,
-0.1444335431,
-0.0581155941,
0.0174230076,
-0.3384709954,
0.0029890127,
-0.1440847665,
0.2218134701,
-0.2125498652,
0.1255456954,
-0.0039673802,
-0.1156828701,
0.0639204979,
-0.0838063136,
-0.3261517584,
-0.0504811071,
0.4979912043,
-0.3048431873,
0.0334971771,
0.4874073267,
-0.0755842477,
-0.2245248258,
-0.0711385459,
-0.5798051357,
0.4502626061,
-0.4499553442,
-0.0604626536,
0.2341179252,
-0.3475359976,
0.2754639387,
-0.0626305714,
0.1548950672,
-0.3116753101,
-0.187979728,
-0.1831901371,
-0.0942264721,
0.0886205733,
0.0754399747,
0.2557686567,
-0.0081492886,
0.1618413627,
-0.2867749333,
0.0140154678,
-0.2770028114,
0.1075770184,
-0.0721934512,
0.2139977068,
-0.2397188395,
-0.483212024,
0.2989868522,
-0.0265959166,
0.1560491025,
0.0845015496,
-0.1904973686,
-0.1558910161,
-0.2289706618,
0.1160233542,
0.0690035075,
-0.1123571098,
-0.4639087915,
-0.0516463444,
0.0010402389,
-0.1568175852,
0.2237975299,
0.2394794375,
-0.0870601237,
0.021417357,
0.1709784716,
-0.1172272936,
-0.0764699057,
0.2197504938,
-0.0321822725,
0.3317804635,
0.3021885455,
0.2104404569,
-0.6520295143,
0.0660036504,
-0.119557634,
0.1622675359,
0.0261327829,
-0.0447575301,
0.1173152477,
-0.268381238,
-0.4649254084,
0.0705434754,
0.1218458265,
0.5309673548,
-0.1870385259,
-0.40476349,
0.166649729,
0.1591293961,
-0.2034613788,
-0.1616841853,
0.0730251968,
-0.1118106991,
0.1455478817,
0.0349663906,
0.4118669629,
-0.3259599805,
-0.1759109497,
0.1035070792,
0.2216588557,
0.315219909,
-0.3731231093,
-0.0197301097,
0.2062294781,
0.0610961467,
0.5360978842,
0.0945842862,
0.247702539,
0.6870852113,
-0.0722829029,
0.1744622737,
0.0550851971,
0.2124180198,
-0.0244811364,
-0.2739833295,
0.2910193205,
0.3236260414,
-0.0894855112,
0.0141979484,
-0.0831026882,
0.3698722124,
0.0705714747,
-0.163081184,
-0.1223595291,
-0.0045949277,
-0.0648391992,
-0.4671500921,
0.0585068837,
-0.0793078244,
-0.2404512465,
0.1137237474,
-0.2591263056,
-0.0938859284,
-0.0230940431,
0.1891687065,
-0.0880203322,
-0.3320337534,
-0.5754649043,
0.2790210843,
0.3300843537,
-0.2184932828,
0.1983409822,
0.2257538587,
-0.2282556444,
0.3201930225,
0.0007119831,
0.4408043027,
0.0708033815,
-0.0725467652,
-0.053132914,
0.0030458756,
0.010416165,
-0.0953287333,
0.0662745237,
0.0054898374,
-0.2781653404,
0.2291246951,
0.086631164,
-0.2112829685,
-0.6455814242,
-0.0891910717,
0.4423403144,
-0.0844528824,
0.1159544662,
0.1088769734,
0.2360855639,
-0.145878911,
0.0194317475,
-0.4001851082,
0.3257297277,
0.3039853275,
-0.0512607023,
0.0937301069,
0.268938005,
0.0631137639,
-0.0552853942,
0.2625034451,
0.2940627038,
0.0623607188,
-0.4068159461,
-0.1730079055,
-0.2244133651,
0.2328392565,
-0.1030966043,
0.3240922689,
0.4435864985,
0.373513937,
-0.0393701643,
-0.1675023437,
0.141190514,
-0.0199436974,
0.2401416749,
0.3795912266,
-0.2141231447,
0.3025227785,
-0.2922197282,
0.049027577,
0.0471593663,
-0.3806376159,
0.0363537297,
-0.0276331212,
-0.0515836887,
-0.0018897261,
-0.2796109021,
0.2360594571,
-0.3397529423,
0.3566702604,
-0.0204697456,
0.0925085694,
-0.0411742292,
-0.2193615139,
-0.3757415712,
-0.3629161417,
-0.1448071748,
0.0370060503,
-0.2165197134,
-0.0487193875,
-0.0600932091,
-0.2090695351,
-0.2312407792,
0.3326087594,
0.0224139169,
-0.1425607055,
-0.1646386385,
0.1618475169,
-0.1754698157,
0.2252356708,
-0.1536871195,
0.4404304028,
0.1206705794,
0.3056816757,
-0.2830801606,
-0.1984689832,
0.7958440781,
-0.3033993244,
-0.2339418381,
-0.2163998485,
0.3711205721,
0.183614403,
0.2305793315,
-0.6851294041,
0.2322596759,
0.2757993937,
0.14512299,
-0.2586945593,
-0.1075072438,
0.2922549248,
0.0834403336,
-0.3089337349,
-0.0486343279,
0.4374200702,
0.0018164814,
-0.0098789185,
-0.071688883
] |
https://github.com/huggingface/datasets/issues/1940 | Side effect when filtering data due to `does_function_return_dict` call in `Dataset.map()` | Thanks for the report !
Currently we don't have a way to let the user easily disable this behavior.
However I agree that we should support stateful processing functions, ideally by removing `does_function_return_dict`.
We needed this function in order to know whether the `map` functions needs to write data or not. if `does_function_return_dict` returns False then we don't write anything.
Instead of checking the output of the processing function outside of the for loop that iterates through the dataset to process it, we can check the output of the first processed example and at that point decide if we need to write data or not.
Therefore it's definitely possible to fix this unwanted behavior, any contribution going into this direction is welcome :) | Hi there!
In my codebase I have a function to filter rows in a dataset, selecting only a certain number of examples per class. The function passes a extra argument to maintain a counter of the number of dataset rows/examples already selected per each class, which are the ones I want to keep in the end:
```python
def fill_train_examples_per_class(example, per_class_limit: int, counter: collections.Counter):
label = int(example['label'])
current_counter = counter.get(label, 0)
if current_counter < per_class_limit:
counter[label] = current_counter + 1
return True
return False
```
At some point I invoke it through the `Dataset.filter()` method in the `arrow_dataset.py` module like this:
```python
...
kwargs = {"per_class_limit": train_examples_per_class_limit, "counter": Counter()}
datasets['train'] = datasets['train'].filter(fill_train_examples_per_class, num_proc=1, fn_kwargs=kwargs)
...
```
The problem is that, passing a stateful container (the counter,) provokes a side effect in the new filtered dataset obtained. This is due to the fact that at some point in `filter()`, the `map()`'s function `does_function_return_dict` is invoked in line [1290](https://github.com/huggingface/datasets/blob/96578adface7e4bc1f3e8bafbac920d72ca1ca60/src/datasets/arrow_dataset.py#L1290).
When this occurs, the state of the counter is initially modified by the effects of the function call on the 1 or 2 rows selected in lines 1288 and 1289 of the same file (which are marked as `test_inputs` & `test_indices` respectively in lines 1288 and 1289. This happens out of the control of the user (which for example can't reset the state of the counter before continuing the execution,) provoking in the end an undesired side effect in the results obtained.
In my case, the resulting dataset -despite of the counter results are ok- lacks an instance of the classes 0 and 1 (which happen to be the classes of the first two examples of my dataset.) The rest of the classes I have in my dataset, contain the right number of examples as they were not affected by the effects of `does_function_return_dict` call.
I've debugged my code extensively and made a workaround myself hardcoding the necessary stuff (basically putting `update_data=True` in line 1290,) and then I obtain the results I expected without the side effect.
Is there a way to avoid that call to `does_function_return_dict` in map()'s line 1290 ? (e.g. extracting the required information that `does_function_return_dict` returns without making the testing calls to the user function on dataset rows 0 & 1)
Thanks in advance,
Francisco Perez-Sorrosal
| 123 | Side effect when filtering data due to `does_function_return_dict` call in `Dataset.map()`
Hi there!
In my codebase I have a function to filter rows in a dataset, selecting only a certain number of examples per class. The function passes a extra argument to maintain a counter of the number of dataset rows/examples already selected per each class, which are the ones I want to keep in the end:
```python
def fill_train_examples_per_class(example, per_class_limit: int, counter: collections.Counter):
label = int(example['label'])
current_counter = counter.get(label, 0)
if current_counter < per_class_limit:
counter[label] = current_counter + 1
return True
return False
```
At some point I invoke it through the `Dataset.filter()` method in the `arrow_dataset.py` module like this:
```python
...
kwargs = {"per_class_limit": train_examples_per_class_limit, "counter": Counter()}
datasets['train'] = datasets['train'].filter(fill_train_examples_per_class, num_proc=1, fn_kwargs=kwargs)
...
```
The problem is that, passing a stateful container (the counter,) provokes a side effect in the new filtered dataset obtained. This is due to the fact that at some point in `filter()`, the `map()`'s function `does_function_return_dict` is invoked in line [1290](https://github.com/huggingface/datasets/blob/96578adface7e4bc1f3e8bafbac920d72ca1ca60/src/datasets/arrow_dataset.py#L1290).
When this occurs, the state of the counter is initially modified by the effects of the function call on the 1 or 2 rows selected in lines 1288 and 1289 of the same file (which are marked as `test_inputs` & `test_indices` respectively in lines 1288 and 1289. This happens out of the control of the user (which for example can't reset the state of the counter before continuing the execution,) provoking in the end an undesired side effect in the results obtained.
In my case, the resulting dataset -despite of the counter results are ok- lacks an instance of the classes 0 and 1 (which happen to be the classes of the first two examples of my dataset.) The rest of the classes I have in my dataset, contain the right number of examples as they were not affected by the effects of `does_function_return_dict` call.
I've debugged my code extensively and made a workaround myself hardcoding the necessary stuff (basically putting `update_data=True` in line 1290,) and then I obtain the results I expected without the side effect.
Is there a way to avoid that call to `does_function_return_dict` in map()'s line 1290 ? (e.g. extracting the required information that `does_function_return_dict` returns without making the testing calls to the user function on dataset rows 0 & 1)
Thanks in advance,
Francisco Perez-Sorrosal
Thanks for the report !
Currently we don't have a way to let the user easily disable this behavior.
However I agree that we should support stateful processing functions, ideally by removing `does_function_return_dict`.
We needed this function in order to know whether the `map` functions needs to write data or not. if `does_function_return_dict` returns False then we don't write anything.
Instead of checking the output of the processing function outside of the for loop that iterates through the dataset to process it, we can check the output of the first processed example and at that point decide if we need to write data or not.
Therefore it's definitely possible to fix this unwanted behavior, any contribution going into this direction is welcome :) | [
-0.4214116037,
0.0289576501,
-0.1575037539,
0.0184576046,
-0.0844258517,
-0.3201850653,
0.2051184177,
0.194657743,
0.2660330534,
0.1206626147,
0.2073836327,
0.5199098587,
-0.0788979828,
0.0967266038,
0.030145973,
0.2262496054,
0.1668144166,
-0.0325007401,
-0.1896088719,
-0.1237708181,
-0.207030341,
-0.0553833172,
-0.2351947725,
0.0294675622,
0.1121999025,
-0.2066931576,
0.2833378911,
0.0864986181,
0.1280571073,
-0.2339867651,
0.3797760904,
0.2778169513,
-0.3832843006,
0.4839722812,
-0.0001190907,
0.0343094468,
0.1299142838,
-0.0802712664,
-0.2356486022,
-0.4382510185,
-0.2481365204,
0.007895343,
0.1348448098,
-0.1925273389,
0.2354867011,
-0.1303694695,
-0.4192909896,
-0.3500103354,
0.5882075429,
0.2834123373,
0.1079882756,
0.1053067073,
-0.3432337046,
0.1123892516,
0.4360823333,
0.315669775,
0.1410354078,
0.3769834936,
0.2226105928,
-0.4172934294,
-0.0897968486,
0.6554314494,
-0.4193486571,
0.1199868023,
-0.1365879029,
0.1172131747,
0.1939311326,
0.0126647223,
0.3233864009,
0.1725642085,
0.1055689752,
-0.3248953223,
-0.1459571123,
-0.4208645821,
-0.1007800177,
0.0005768612,
0.075097464,
-0.2886886597,
-0.2021785975,
-0.1206651032,
-0.1208330542,
-0.012386933,
-0.0200271532,
0.1681484282,
-0.2169066221,
0.1855427474,
0.0811172798,
0.2631781697,
0.0817807913,
-0.0514533408,
0.209052816,
-0.3612850606,
0.206083566,
0.2147340178,
0.1196143404,
0.0732958764,
0.3768187165,
-0.478985697,
0.124123469,
-0.1366597414,
0.1146183014,
0.1747402549,
-0.2579140961,
0.2301837653,
0.4253012538,
0.0850004107,
-0.1979261339,
0.6640270948,
0.1729906499,
-0.1068491936,
-0.0037355423,
0.0037668357,
0.2818770707,
0.0269302987,
0.2999613881,
-0.1395049095,
0.1103356034,
-0.1578752697,
-0.2358426601,
0.4175381064,
-0.359054774,
0.2780915201,
0.0907846764,
0.0966717601,
0.3004825711,
0.4036151469,
0.168298319,
-0.1376377791,
0.0033887923,
0.5379951596,
-0.1066150218,
-0.3244780004,
-0.2710617483,
-0.0770937949,
-0.0119903386,
0.0786592662,
0.016985476,
0.110454157,
-0.0109128989,
-0.0541715771,
0.0757819563,
-0.2363330126,
0.7913027406,
0.4118510783,
-0.3480502665,
0.173743248,
0.2374269664,
-0.2243676186,
-0.126040861,
0.2483869046,
-0.0546915568,
-0.1973662674,
0.185219124,
0.1147135124,
0.005986467,
0.3745833039,
-0.0160885677,
0.1991492808,
0.4185601473,
-0.4719982147,
0.3922097087,
-0.5384452343,
-0.3335291743,
-0.2172564864,
0.1258206367,
0.5259780884,
-0.8197143078,
-0.1214057505,
-0.2428822368,
-0.1178345308,
0.0761058033,
0.174031198,
-0.2996687293,
-0.0011382774,
-0.2255411148,
0.1341027468,
0.3071457744,
-0.4831513166,
-0.7054834962,
0.0982056111,
-0.2767548859,
0.49920398,
-0.1460387707,
0.0213065185,
0.3568472564,
-0.0081650391,
0.326733619,
0.1796563417,
-0.0669032261,
0.1762678921,
0.0203268155,
0.1293672025,
0.3243872225,
-0.0954311341,
-0.0437129661,
-0.077088058,
-0.0749708265,
-0.4875873029,
0.2810600102,
0.0821086913,
0.192464292,
-0.3172641695,
0.4052565098,
0.1898687482,
0.007957194,
-0.2883606553,
-0.2915592194,
0.0308365151,
0.0936339423,
-0.2490016669,
-0.2050818503,
-0.1570571661,
-0.1389681399,
0.3914355934,
-0.0558800288,
0.1572160125,
0.1144860089,
-0.0418151319,
0.005538702,
0.0628045946,
0.0852906704,
0.0804480985,
-0.4057525992,
-0.0257784463,
0.0938626081,
0.2385647297,
0.3390493095,
-0.2808124423,
-0.2351311892,
0.1347767115,
0.0226554573,
-0.0478545092,
-0.2866384685,
0.2192193866,
0.5091301203,
-0.0345372222,
-0.2714931369,
0.0800037533,
-0.102727592,
0.4487372041,
-0.0084449509,
0.305834502,
0.1903858483,
-0.2705368102,
-0.099763751,
0.4954223037,
0.0765306503,
0.4916888773,
-0.1645123065,
0.136452511,
0.2814357877,
0.2763489783,
-0.3366052508,
-0.5444406867,
-0.2632300258,
-0.080550395,
0.1870287955,
-0.1061947644,
-0.1559697688,
-0.0380204841,
0.084356457,
-0.0368898027,
0.1383832395,
0.103854686,
-0.0823240355,
-0.0003455728,
0.4746934175,
0.4552513659,
0.5759059787,
0.0555184372,
-0.1408444047,
0.0276645217,
-0.0354821943,
-0.0613750145,
0.1117444932,
0.2467333078,
0.1628266275,
0.3262066841,
0.3910290897,
-0.1081223786,
-0.2600621581,
-0.1121190637,
-0.1188486665,
-0.1379217058,
-0.2318262756,
0.3826010823,
-0.1168807223,
-0.2817380428,
-0.2118336111,
-0.1495020688,
0.0519159958,
-0.1294425726,
-0.0069108023,
0.4019419551,
-0.3745885491,
0.0674527287,
-0.0789708942,
0.3998450637,
0.1533495933,
-0.4363386929,
0.1451999247,
-0.3576054275,
-0.149545908,
-0.0052298419,
0.0020298697,
0.0059774518,
0.4475277662,
0.2823251486,
-0.2104866058,
0.0630711094,
-0.109297432,
0.1017172411,
-0.007605562,
-0.0080981683,
0.33187747,
0.0187595692,
-0.2166742682,
0.0701481476,
0.0484458618,
0.0242533237,
-0.1498596966,
-0.1106221974,
0.0191022828,
0.2351118028,
-0.2889122665,
0.1924328059,
-0.1723961532,
-0.1195877343,
0.1122068018,
-0.1309980303,
0.2291353345,
0.1243815422,
0.2222428918,
0.0680093616,
-0.0073055103,
-0.2129321992,
-0.4279070497,
-0.0134777874,
-0.1785435528,
-0.1449948251,
0.0711149871,
-0.2463676631,
-0.2123690099,
0.1431716532,
0.5977714062,
-0.5670166612,
-0.19801265,
0.1690310836,
0.2804597616,
-0.1854308099,
0.0043557752,
0.3448855281,
0.1838812381,
-0.0802196711,
-0.3217478693,
-0.0644872338,
0.197025016,
0.3125905693,
0.0231231004,
0.0797689408,
0.1033606753,
0.0153852403,
0.6843208075,
0.2735558152,
-0.1441138089,
-0.0708044469,
0.2465561777,
0.2138898075,
-0.3259015679,
-0.0774518549,
-0.3477695584,
-0.233711198,
-0.1915653944,
-0.1238082349,
0.0483661518,
-0.2100283504,
0.1475331932,
-0.1117217317,
-0.2634608746,
-0.4573830366,
0.250960052,
-0.2607239783,
0.3891144097,
0.0045571849,
-0.0390345603,
-0.395788312,
0.0963751823,
-0.0857026428,
-0.2023215741,
0.3929490149,
-0.2158953547,
-0.2131983936,
-0.0444582589,
0.08769086,
0.2771397233,
0.3724202812,
0.2909670472,
-0.0069644675,
-0.1946189702,
-0.0075317472,
0.1920889169,
0.9212180376,
-0.1532569975,
0.1145692766,
0.3606547713,
-0.1634964943,
-0.4281954467,
-0.0288702436,
-0.381183207,
0.3040295541,
0.1288408935,
0.136218518,
-0.1125413626,
0.081065461,
-0.175234288,
0.1800233871,
-0.230507046,
-0.0291539952,
-0.0458448976,
0.0430696458,
-0.2418697178,
0.414095968,
-0.0426528901,
-0.3411509991,
-0.0270897001,
-0.3010747433,
0.1418936253,
-0.2788734436,
-0.1683322489,
0.0251772963,
0.3258305192,
0.0281417966,
-0.033845108,
0.4214905798,
-0.0445802361,
0.3595963717,
0.2097104788,
-0.1842393428,
-0.099346675,
-0.3385730684,
0.2808580995,
0.0884959996,
0.6351360679,
-0.066212669,
0.2925786674,
0.0539203659,
0.0340459421,
0.0468429737,
0.052845791,
0.1087766364,
0.3228061795,
0.0372579694,
-0.5029550791,
0.2584249079,
0.1158449352,
-0.4988928437,
0.5093587041,
0.0707159117,
-0.4502278268,
0.5003752112,
0.0967449695,
0.8437862396,
0.3119133711,
0.1658477485,
0.2260126173,
-0.1717869639,
0.0884204432,
0.2525969744,
-0.0557096936,
0.1190142706,
0.1527563334,
-0.035819836,
-0.290184468,
-0.0203855671,
-0.1074929386,
0.0465304554,
0.0796531141,
-0.5716612339,
0.3465051055,
-0.1252949089,
-0.2016382217,
0.2813825607,
-0.3633782566,
-0.0759186223,
0.1183448136,
0.1558743715,
0.0072762333,
-0.1174486354,
0.1039034873,
0.0316311792,
-0.1071538329,
-0.1751723886,
-0.2920618057,
-0.419413209,
0.4033287168,
0.0758900642,
-0.4115694761,
-0.0596608594,
-0.0202500429,
-0.4361622036,
0.2700867057,
0.0537091419,
-0.0371725298,
0.4989472926,
0.5238935351,
0.1007666588,
-0.2694999576,
0.4783496559,
0.1840207875,
-0.2807707191,
0.0061872965,
-0.0833481327,
0.0214817226,
-0.1123805791,
0.0961130187,
0.3758648932,
-0.2928253114,
0.0643376485,
0.0271968823,
-0.0029286891,
-0.1025000364,
0.0415785983,
-0.0710351095,
-0.0433691181,
0.2548505664,
0.0600101091,
-0.0355919898,
-0.0283804256,
0.2195602953,
0.1335603595,
0.2223019004,
0.3728732169,
-0.1903458238,
-0.1306004822,
-0.1904114038,
-0.2018485069,
0.1169960648,
0.0854392052,
0.4244583249,
0.011542201,
-0.2601807117,
-0.3701818883,
0.1487353444,
-0.0050655785,
0.2873673737,
-0.5665979981,
-0.3418367207,
-0.2227963507,
0.1422398984,
0.1213003919,
0.1461367905,
-0.2059535384,
0.203411445,
-0.3286548853,
0.198247999,
-0.2763056159,
-0.0888120979,
0.076014176,
0.3025521636,
0.072930336,
0.2834868133,
-0.259880662,
-0.1926092654,
0.0222351886,
0.4459892213,
-0.2452252507,
-0.1933342218,
-0.0131519176,
0.1412336975,
-0.0306064133,
-0.3593328893,
0.0489732698,
-0.3224497437,
0.1845188737,
-0.2329693735,
0.4509952366,
0.114058882,
-0.0886012018,
-0.266124934,
-0.0722014308,
0.1520369649,
0.3301410973,
0.4199129045,
-0.282502979,
-0.2660821974,
-0.0245818235,
0.1808161736,
0.0654083788,
-0.0692420825,
-0.0678169355,
0.1880919933,
0.1825222373,
0.1244097948,
0.1734901667,
-0.0926332325,
-0.2610257268,
0.2713045478,
0.3550501764,
0.1851457357,
-0.3361179531,
-0.1769834757,
0.2839799821,
0.1789635718,
-0.2347436547,
-0.2329944223,
-0.0508181192,
-0.1223192215,
0.1850057989,
0.3649446964,
-0.023766689,
-0.2316483557,
-0.0074907802,
0.1311028898,
-0.0525209345,
-0.045991607,
0.1973969042,
0.3307219744,
-0.0605215579,
0.0740697235,
0.203822121,
0.3356961608,
0.1484808624,
0.4495687187,
0.3902043402,
0.4044878781,
0.2652181089,
0.1011538208,
0.0109557416,
-0.3769710064,
0.1383007467,
0.2418542653,
-0.6026107669,
0.2768877745,
0.2579663396,
-0.1434608698,
0.0388518237,
-0.1774979234,
-0.1597566009,
0.0448752828,
-0.2500065863,
-0.1211037934,
-0.0018083751,
-0.1462198794,
0.2350992262,
-0.114502579,
-0.0338258632,
-0.1258026063,
0.4603756964,
-0.0914186537,
-0.009352494,
-0.1481796503,
-0.3311114311,
-0.217255801,
0.2395835668,
-0.1705410182,
-0.2877802253,
0.1675364673,
0.2746527791,
-0.1172757894,
-0.1353557259,
0.2621349096,
0.2212352604,
-0.2120893449,
-0.0261261687,
-0.0590667054,
-0.0073863678,
-0.1609936804,
0.135212481,
0.0356557667,
-0.1427990794,
0.0698154569,
0.1292562783,
-0.1309076399,
0.0188168678,
0.385050714,
0.2576454282,
-0.3966677189,
0.2288760841,
-0.0653827712,
0.0065331161,
-0.3422940969,
-0.0566751845,
0.0717607439,
-0.0921386406,
0.3632926941,
-0.1265349239,
0.3559890687,
0.0683032274,
0.0675616339,
-0.169596523,
0.2754690349,
-0.0937736928,
0.1342550218,
-0.176861167,
-0.1477606893,
-0.3707115948,
-0.0324153304,
-0.2415804565,
-0.0101430565,
0.0788471401,
-0.0222219639,
0.1696500182,
0.385727644,
-0.073442474,
-0.0960478336,
-0.2717598975,
-0.1786079556,
-0.0953546464,
-0.2409930527,
-0.5097749233,
-0.2365780026,
0.3167961538,
-0.393630147,
0.0446067378,
-0.2630875707,
0.1287750453,
0.1057693958,
0.2196850181,
-0.1327523887,
0.214607805,
0.3450618386,
0.4004011154,
0.0616492219,
0.0566400886,
0.1534950584,
0.0584255978,
-0.0649761856,
-0.3119412065,
0.1824388504,
-0.2225930393,
0.3479801416,
-0.3402343094,
-0.2343779802,
0.1301046014,
-0.2715991735,
0.2384702563,
-0.1920745075,
-0.2710632682,
0.0018823668,
-0.2772067785,
0.0974602401,
-0.2875548303,
0.1892950684,
-0.0015225001,
0.1094882637,
0.0225035846,
-0.2070254236,
0.6020872593,
-0.3814716637,
-0.3790596128,
-0.2499881387,
0.2925051749,
-0.1566463709,
-0.3191486597,
-0.1879809797,
-0.1680212021,
0.3404831886,
-0.2098422945,
-0.254611969,
-0.1149112433,
-0.0779229105,
-0.0458498634,
-0.2578430772,
0.060182631,
-0.0492762662,
-0.148709625,
0.304651618,
-0.2471381426
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | Thanks for reporting and for all the details and suggestions.
I'm totally in favor of having a HF_DATASETS_OFFLINE env variable to disable manually all the connection checks, remove retries etc.
Moreover you may know that the use case that you are mentioning is already supported from `datasets` 1.3.0, i.e. you already can:
- first load datasets and metrics from an instance with internet connection
- then be able to reload datasets and metrics from another instance without connection (as long as the filesystem is shared)
This is already implemented, but currently it only works if the requests return a `ConnectionError` (or any error actually). Not sure why it would hang instead of returning an error.
Maybe this is just a issue with the timeout value being not set or too high ?
Is there a way I can have access to one of the instances on which there's this issue (we can discuss this offline) ?
| This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 156 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
Thanks for reporting and for all the details and suggestions.
I'm totally in favor of having a HF_DATASETS_OFFLINE env variable to disable manually all the connection checks, remove retries etc.
Moreover you may know that the use case that you are mentioning is already supported from `datasets` 1.3.0, i.e. you already can:
- first load datasets and metrics from an instance with internet connection
- then be able to reload datasets and metrics from another instance without connection (as long as the filesystem is shared)
This is already implemented, but currently it only works if the requests return a `ConnectionError` (or any error actually). Not sure why it would hang instead of returning an error.
Maybe this is just a issue with the timeout value being not set or too high ?
Is there a way I can have access to one of the instances on which there's this issue (we can discuss this offline) ?
| [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | I'm on master, so using all the available bells and whistles already.
If you look at the common issues - it for example tries to look up files if they appear in `_PACKAGED_DATASETS_MODULES` which it shouldn't do.
--------------
Yes, there is a nuance to it. As I mentioned it's firewalled - that is it has a network but making any calls outside - it just hangs in:
```
sin_addr=inet_addr("xx.xx.xx.xx")}, [28->16]) = 0
close(5) = 0
socket(AF_INET, SOCK_STREAM|SOCK_CLOEXEC, IPPROTO_TCP) = 5
connect(5, {sa_family=AF_INET, sin_port=htons(3128), sin_addr=inet_addr("yy.yy.yy.yy")}, 16^C) = ? ERESTARTSYS (To be restarted if SA_RESTART is set)
```
until it times out.
That's why we need to be able to tell the software that there is no network to rely on even if there is one (good for testing too).
So what I'm thinking is that this is a simple matter of pre-ambling any network call wrappers with:
```
if HF_DATASETS_OFFLINE:
assert "Attempting to make a network call under Offline mode"
```
and then fixing up if there is anything else to fix to make it work.
--------------
Otherwise I think the only other problem I encountered is that we need to find a way to pre-cache metrics, for some reason it's not caching it and wanting to fetch it from online.
Which is extra strange since it already has those files in the `datasets` repo itself that is on the filesystem.
The workaround I had to do is to copy `rouge/rouge.py` (with the parent folder) from the datasets repo to the current dir - and then it proceeded. | This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 257 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
I'm on master, so using all the available bells and whistles already.
If you look at the common issues - it for example tries to look up files if they appear in `_PACKAGED_DATASETS_MODULES` which it shouldn't do.
--------------
Yes, there is a nuance to it. As I mentioned it's firewalled - that is it has a network but making any calls outside - it just hangs in:
```
sin_addr=inet_addr("xx.xx.xx.xx")}, [28->16]) = 0
close(5) = 0
socket(AF_INET, SOCK_STREAM|SOCK_CLOEXEC, IPPROTO_TCP) = 5
connect(5, {sa_family=AF_INET, sin_port=htons(3128), sin_addr=inet_addr("yy.yy.yy.yy")}, 16^C) = ? ERESTARTSYS (To be restarted if SA_RESTART is set)
```
until it times out.
That's why we need to be able to tell the software that there is no network to rely on even if there is one (good for testing too).
So what I'm thinking is that this is a simple matter of pre-ambling any network call wrappers with:
```
if HF_DATASETS_OFFLINE:
assert "Attempting to make a network call under Offline mode"
```
and then fixing up if there is anything else to fix to make it work.
--------------
Otherwise I think the only other problem I encountered is that we need to find a way to pre-cache metrics, for some reason it's not caching it and wanting to fetch it from online.
Which is extra strange since it already has those files in the `datasets` repo itself that is on the filesystem.
The workaround I had to do is to copy `rouge/rouge.py` (with the parent folder) from the datasets repo to the current dir - and then it proceeded. | [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | Ok understand better the hanging issue.
I guess catching connection errors is not enough, we should also avoid all the hangings.
Currently the offline mode tests are only done by simulating an instant connection fail that returns an error, let's have another connection mock that hangs instead.
I'll also take a look at why you had to do this for `rouge`.
| This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 61 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
Ok understand better the hanging issue.
I guess catching connection errors is not enough, we should also avoid all the hangings.
Currently the offline mode tests are only done by simulating an instant connection fail that returns an error, let's have another connection mock that hangs instead.
I'll also take a look at why you had to do this for `rouge`.
| [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | FWIW, I think instant failure on the behalf of a network call is the simplest solution to correctly represent the environment and having the caller to sort it out is the next thing to do, since here it is the case of having no functional network, it's just that the software doesn't know this is the case, because there is some network. So we just need to help it to bail out instantly rather than hang waiting for it to time out. And afterwards everything else you said. | This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 88 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
FWIW, I think instant failure on the behalf of a network call is the simplest solution to correctly represent the environment and having the caller to sort it out is the next thing to do, since here it is the case of having no functional network, it's just that the software doesn't know this is the case, because there is some network. So we just need to help it to bail out instantly rather than hang waiting for it to time out. And afterwards everything else you said. | [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | Update on this:
I managed to create a mock environment for tests that makes the connections hang until timeout.
I managed to reproduce the issue you're having in this environment.
I'll update the offline test cases to also test the robustness to connection hangings, and make sure we set proper timeouts where it's needed in the code. This should cover the _automatic_ section you mentioned. | This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 65 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
Update on this:
I managed to create a mock environment for tests that makes the connections hang until timeout.
I managed to reproduce the issue you're having in this environment.
I'll update the offline test cases to also test the robustness to connection hangings, and make sure we set proper timeouts where it's needed in the code. This should cover the _automatic_ section you mentioned. | [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1939 | [firewalled env] OFFLINE mode | I lost access to the firewalled setup, but I emulated it with:
```
sudo ufw enable
sudo ufw default deny outgoing
```
(thanks @mfuntowicz)
I was able to test `HF_DATASETS_OFFLINE=1` and it worked great - i.e. didn't try to reach out with it and used the cached files instead.
Thank you! | This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks. | 51 | [firewalled env] OFFLINE mode
This issue comes from a need to be able to run `datasets` in a firewalled env, which currently makes the software hang until it times out, as it's unable to complete the network calls.
I propose the following approach to solving this problem, using the example of `run_seq2seq.py` as a sample program. There are 2 possible ways to going about it.
## 1. Manual
manually prepare data and metrics files, that is transfer to the firewalled instance the dataset and the metrics and run:
```
DATASETS_OFFLINE=1 run_seq2seq.py --train_file xyz.csv --validation_file xyz.csv ...
```
`datasets` must not make any network calls and if there is a logic to do that and something is missing it should assert that this or that action requires network and therefore it can't proceed.
## 2. Automatic
In some clouds one can prepare a datastorage ahead of time with a normal networked environment but which doesn't have gpus and then one switches to the gpu instance which is firewalled, but it can access all the cached data. This is the ideal situation, since in this scenario we don't have to do anything manually, but simply run the same application twice:
1. on the non-firewalled instance:
```
run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
which should download and cached everything.
2. and then immediately after on the firewalled instance, which shares the same filesystem
```
DATASETS_OFFLINE=1 run_seq2seq.py --dataset_name wmt16 --dataset_config ro-en ...
```
and the metrics and datasets should be cached by the invocation number 1 and any network calls be skipped and if the logic is missing data it should assert and not try to fetch any data from online.
## Common Issues
1. for example currently `datasets` tries to look up online datasets if the files contain json or csv, despite the paths already provided
```
if dataset and path in _PACKAGED_DATASETS_MODULES:
```
2. it has an issue with metrics. e.g. I had to manually copy `rouge/rouge.py` from the `datasets` repo to the current dir - or it was hanging.
I had to comment out `head_hf_s3(...)` calls to make things work. So all those `try: head_hf_s3(...)` shouldn't be tried with `DATASETS_OFFLINE=1`
Here is the corresponding issue for `transformers`: https://github.com/huggingface/transformers/issues/10379
Thanks.
I lost access to the firewalled setup, but I emulated it with:
```
sudo ufw enable
sudo ufw default deny outgoing
```
(thanks @mfuntowicz)
I was able to test `HF_DATASETS_OFFLINE=1` and it worked great - i.e. didn't try to reach out with it and used the cached files instead.
Thank you! | [
-0.4748801291,
0.1685330868,
-0.0625535473,
0.1050562933,
0.0443116091,
-0.2284920067,
0.4647973478,
0.071679756,
0.2545911968,
0.04853075,
0.0184396878,
0.0636150688,
0.0845068842,
0.1993041933,
0.0180914924,
0.1079673618,
0.0199452154,
-0.1096388996,
-0.2528401017,
0.0637163669,
-0.2635913491,
-0.0536728948,
-0.2544520199,
-0.2575805783,
0.0783677995,
-0.5081003904,
-0.2168779224,
-0.0105695147,
0.0096479878,
-0.2275566161,
0.3089391589,
0.2612625062,
0.251886487,
0.1530118287,
-0.0001084152,
-0.0537568554,
-0.1393842846,
-0.0361675024,
-0.2728874385,
0.0223879069,
-0.3179338574,
-0.3802424073,
0.1650777757,
-0.4613102078,
-0.1713552326,
-0.1747253239,
0.1464557499,
-0.6594021916,
0.515466094,
-0.0763157308,
0.0945165455,
0.0003515705,
-0.3278794885,
-0.0994149595,
-0.3226186633,
-0.1980502754,
-0.1814403832,
0.1710342467,
0.088378109,
-0.2174991071,
-0.0455865376,
0.395288229,
-0.3122881055,
0.1565714777,
0.3260604143,
0.0182257667,
-0.3381130397,
-0.118822284,
-0.1281383932,
0.3376221955,
0.1812190861,
-0.2216328532,
-0.426237762,
-0.4555668235,
-0.0790791214,
-0.3703311682,
0.3025759161,
0.088930808,
-0.116917491,
0.2809564769,
-0.3825721741,
-0.2138665318,
-0.1247217283,
-0.0012390222,
-0.0308276489,
0.3738209605,
0.0736685842,
0.1736759841,
-0.0692528859,
0.1060306951,
0.0134039372,
-0.3340771496,
0.2569809854,
0.0270411409,
-0.5665540695,
0.0265183151,
0.1224477589,
-0.0067838691,
0.0168157369,
0.4544950426,
-0.1779251993,
-0.0429257229,
0.4983155429,
0.1950367391,
0.0398953408,
0.2975893021,
0.1519621909,
-0.1790608764,
0.3776466846,
0.1602639109,
-0.0568714999,
-0.1345084757,
-0.1136466265,
-0.07069242,
0.1811161041,
0.0640621036,
0.119768396,
-0.1377027631,
-0.1166460812,
0.0469651818,
-0.0323215947,
0.2401375175,
0.3313706517,
0.309374094,
-0.0555983633,
-0.2358307391,
0.2333135009,
0.1667182595,
-0.1382888854,
-0.0603913069,
0.0725907683,
-0.098499611,
-0.1313182265,
0.0826691091,
0.3286069036,
0.2563714981,
0.2747863531,
-0.1755118072,
0.1528573781,
0.0238282606,
0.5519109368,
-0.2458669841,
0.1138634235,
0.3021979332,
0.1545312554,
-0.2306012064,
0.0980399549,
0.281317085,
-0.037536338,
0.0506099835,
0.079942748,
-0.3858197927,
0.404483676,
0.1439911276,
-0.1975331008,
-0.0476034991,
0.178219676,
0.074253343,
-0.0721476227,
-0.4519343376,
0.1831779927,
0.0667775869,
-0.1193483621,
-0.0543197542,
0.4079636931,
0.4936105013,
-0.4202960134,
0.022811912,
-0.2217298448,
0.0191004276,
-0.1197147146,
0.1713059992,
-0.2794616818,
0.1356070042,
-0.1801661998,
-0.2079317123,
0.2443794906,
-0.3084554672,
-0.3225480616,
0.4538068473,
-0.1916148663,
0.2314611077,
0.3152421117,
0.1994497329,
0.2296577394,
-0.0338120274,
0.1222660691,
0.3941686749,
-0.2495331913,
-0.0643457472,
-0.149642542,
-0.1908912212,
0.2760186195,
0.1059188545,
0.2374100387,
-0.1076447964,
-0.0861400664,
-0.2940387726,
-0.035080459,
-0.08320795,
0.2003033161,
-0.1944312155,
0.0856610611,
0.0736428201,
0.0493784621,
0.1411922574,
-0.4604979753,
0.3037664592,
-0.2521933317,
0.0747481287,
-0.3391274214,
-0.3241584301,
0.2313054055,
-0.0713331476,
-0.1031432971,
-0.3441323638,
0.0740437731,
0.0512405112,
-0.1017183289,
-0.0374328606,
-0.0173989236,
0.0242225416,
-0.087022543,
0.113507621,
0.1581825614,
-0.0672888011,
0.0744351968,
0.0992200524,
-0.2595793009,
-0.1118376255,
-0.0186912902,
-0.1419193447,
-0.0033472832,
0.5071068406,
0.0670178533,
0.4341065884,
0.1801056862,
0.4344967604,
0.100204207,
0.3320182562,
0.2241401225,
0.0292813145,
0.0667586625,
-0.0281433016,
-0.5865287781,
0.8847306967,
-0.077290535,
0.4949630499,
-0.0253684744,
-0.0896621868,
0.2581320405,
-0.0700585991,
-0.5748964548,
-0.0896325558,
-0.0292922631,
0.2472512424,
0.1246779561,
-0.0695688277,
-0.3645501137,
0.1309467256,
0.0615998954,
0.211324811,
0.4444037676,
-0.2755912542,
0.2542672753,
-0.0863385648,
0.4632288814,
0.2097622603,
0.2965559661,
0.2528911233,
0.0414114445,
0.1776763499,
-0.0744992942,
-0.2386738658,
0.364202559,
0.1546071321,
0.3154805303,
0.1600137651,
0.0825261101,
-0.1450217962,
-0.165667057,
-0.2701162696,
0.1441671997,
-0.2285314351,
-0.0524325408,
0.3188202679,
-0.3138053417,
-0.5460375547,
-0.0330593437,
0.2567464411,
-0.1731722355,
-0.1265536994,
0.2325973511,
0.1215339601,
-0.3687726855,
0.3979505002,
-0.2202256024,
0.4702409208,
-0.0649727508,
-0.4042861164,
0.2372109741,
-0.373627305,
0.097589165,
0.0727020875,
0.2432581633,
-0.0904773995,
0.682811141,
0.2652266026,
-0.0664931536,
-0.172498852,
-0.2048209906,
0.1804468334,
0.0453943498,
0.5188972354,
0.1410696507,
0.097885415,
-0.1078776419,
0.0392980427,
0.2668644786,
-0.131492883,
0.2002530694,
-0.1014679074,
0.005211316,
0.1161195189,
-0.2464787513,
-0.5637351274,
-0.2847305834,
-0.2684565187,
0.1167268232,
-0.222254619,
0.2011252642,
-0.023112908,
0.1001847535,
-0.0503061526,
0.1435731947,
0.0995233655,
-0.032247372,
-0.568808496,
0.1139891297,
-0.1637168229,
0.099777177,
0.0580961704,
-0.0110152997,
0.3906190395,
0.586969316,
-0.4705567062,
-0.2902962863,
0.3895429969,
0.2864701152,
0.0516614839,
-0.3373516202,
0.615039289,
0.0365746282,
-0.0880158767,
-0.0884277746,
0.3548744321,
0.4860229492,
-0.1864489913,
-0.0539081357,
0.1791346073,
0.7724381685,
0.202255547,
0.2877364755,
-0.1666051149,
0.4881625772,
0.2260148227,
0.1102788225,
0.3049393594,
-0.1625729799,
0.0179587156,
0.20340693,
-0.0815214664,
-0.5336043239,
0.0716209561,
0.0470551699,
-0.0527304187,
-0.3611356914,
-0.2582451701,
-0.1032527089,
-0.3916655183,
0.2961249352,
-0.1127898693,
0.4995017648,
0.1178962737,
0.3723381162,
-0.2182423174,
0.1661283672,
-0.1418525428,
0.15018408,
0.5369583964,
-0.012886012,
-0.1799635887,
0.0348872244,
-0.2227435559,
0.2784568369,
0.009933643,
0.3417261839,
-0.0363918915,
0.0174976513,
0.0015467331,
-0.0763284862,
0.446313113,
-0.1146828458,
0.0632685423,
0.0081594363,
-0.2919883728,
-0.2112849802,
-0.1305244565,
-0.0428580381,
-0.0211980492,
0.0649971366,
0.195253104,
-0.2032026052,
-0.0390481874,
-0.318778336,
0.1351860017,
-0.3197741807,
-0.2367113531,
0.0234855488,
0.1267213225,
-0.6222807765,
0.4204728901,
-0.157197088,
-0.1905482858,
-0.2480875403,
-0.0747949705,
-0.1783073843,
0.481212467,
-0.1459497213,
0.0565130711,
0.1470394433,
-0.5123633146,
0.1858814359,
0.1123725399,
0.1500663459,
0.3838641942,
0.0461330973,
0.1363832057,
0.2845409214,
0.0637651905,
0.0664847121,
0.2051707804,
0.4261293113,
-0.1089275777,
-0.0030096434,
0.1616743207,
0.3030037582,
-0.3588069975,
-0.3979019821,
0.1224461868,
0.1558634341,
0.0335064009,
-0.4153930843,
0.0671156943,
0.0460529923,
-0.2596212626,
0.044562757,
0.1521451771,
-0.3728852868,
0.2617143989,
0.263071537,
0.9972519875,
0.2139619142,
0.2179921567,
0.4170044661,
0.2532131374,
0.4726444781,
-0.6332616806,
0.2983070612,
0.1272925735,
-0.1670440733,
-0.149963215,
-0.0255903974,
0.2321767211,
0.0998707563,
-0.2685554922,
0.5433548689,
-0.0701307058,
0.1014338732,
-0.243312344,
-0.0296448022,
-0.2376531065,
-0.4036458731,
-0.2663610578,
0.1260659546,
0.2334098965,
-0.0104810726,
-0.3751176596,
-0.0689792261,
0.1925515234,
0.1631510258,
0.1454557627,
0.0721767545,
0.0428949669,
0.2766603529,
-0.4192085266,
0.0559383333,
-0.2763842344,
0.1090795621,
0.0180984363,
0.2108996809,
-0.2886297405,
0.125995338,
-0.0419276915,
0.1140724346,
-0.153917253,
-0.2668181956,
0.3729190528,
-0.1662878692,
-0.2487161458,
-0.2205847204,
-0.3582705259,
-0.0771526098,
0.2514948845,
-0.3500297964,
0.0939493701,
-0.0691656023,
0.0619472973,
0.1937756836,
-0.1183085889,
-0.2158385813,
0.0828799605,
0.1098429561,
-0.0477759503,
0.2642478347,
0.08337529,
-0.0296080038,
-0.1844732761,
0.522472918,
-0.012487933,
0.2983218729,
0.329159528,
0.1709975004,
-0.242272824,
-0.3916771412,
-0.1179048344,
-0.0944123715,
0.1713486016,
0.0689706951,
0.0006623305,
-0.0254699886,
0.0580741838,
-0.2008395642,
0.1766275615,
-0.0651127994,
-0.1013034582,
-0.4263277054,
-0.0957949758,
0.3383996487,
0.2018853724,
0.0942364261,
0.1214300394,
0.1285647154,
0.0334424824,
0.0116607901,
-0.3435806334,
-0.1933898032,
-0.266688168,
0.2216061503,
-0.0113146817,
-0.0477260873,
0.1774717271,
-0.0150594003,
0.0563594922,
-0.0292198826,
-0.2538237274,
-0.1619532406,
-0.1136551946,
0.2258028537,
-0.1483043283,
-0.3157638609,
0.2160576284,
-0.4238920808,
-0.3664361238,
-0.3153071702,
0.0918551236,
-0.0170712993,
-0.1677894443,
-0.4335836172,
-0.1071412936,
0.3017936051,
-0.2243884802,
0.0949886143,
-0.0977729261,
-0.2725068927,
-0.3345751166,
0.0538964123,
-0.0734190121,
-0.1096149385,
-0.0395690314,
0.1027320474,
0.2106037289,
0.2055788338,
0.4639980197,
0.0892976075,
-0.1204226092,
0.1604590267,
0.6960183978,
0.1433716565,
0.1893890947,
0.1112303659,
0.0307054818,
0.2147426158,
-0.4058088958,
0.1708360463,
-0.2121230215,
-0.1652801484,
0.2011715323,
0.2182682455,
0.4211680293,
0.1228165627,
0.2115367055,
-0.0768138021,
0.3960928023,
-0.363971293,
0.2452290356,
-0.0232141688,
-0.0485097393,
0.1692207754,
-0.0433971547,
0.1081220359,
0.3069613576,
0.0682092458,
-0.269303441,
0.1668251157,
0.4557293355,
0.2022745609,
-0.0908759683,
-0.4782034755,
-0.1009482741,
0.0956824794,
-0.0603445023,
0.0244359002,
-0.4648662806,
-0.0542510822,
-0.3395823836,
-0.1913395822,
-0.2145866901,
0.602753222,
-0.3315453827,
-0.1462775171,
-0.4204240441,
-0.4396710396,
0.011094898,
0.0038174987,
0.1870265901,
-0.1384874731,
-0.1736918092,
0.0796871409,
0.2774658501,
-0.271273762,
0.0969134346,
-0.0292455368,
0.143199265,
-0.3236311376,
0.1587521136,
0.5755218863,
0.0011658836,
0.0627888441,
0.2858875096,
0.2266619503,
0.0022770073,
-0.1239222288,
0.1276148111,
0.1615412831,
0.0533407629,
-0.3756060004,
-0.0559494495,
-0.0167080238,
0.2881964147,
0.3220772743,
0.1017412543,
-0.0451429486,
0.5039827228,
-0.0574037172,
-0.3350549638,
-0.2178630233,
-0.0598232523,
0.4229665101,
0.1130176336,
-0.0961264372,
-0.0181787759,
-0.1123340428,
-0.0917094126,
0.2774384022,
-0.161667645,
0.0388009399,
0.0190164745,
0.0622580275,
-0.191754669,
0.5251252651,
-0.0828251615,
0.1001401544,
-0.3585447073,
-0.0513545498,
-0.60318923,
-0.2333036661,
-0.0481281169,
0.0508252084,
-0.1225792915,
0.1338447332,
0.2245395035,
0.1548327506,
-0.250117898,
0.0248773731,
0.1090886071,
0.0784568191,
-0.3683552444,
0.0319491178,
0.0217684433,
-0.0349100716,
0.1462562978,
-0.2422400266,
0.2108057737,
-0.1415776759,
0.067228213,
0.3073818088,
0.0241462439,
0.0134564787,
0.1558466107,
0.7136696577,
0.0963874534,
0.3541610241,
0.0330818519,
-0.0392262638,
-0.1355392188,
0.0061805099,
-0.1747148484,
0.0417086259,
0.0420962125,
0.2525863349,
-0.1677736938,
-0.1356782168,
-0.173233822,
0.0797531828,
-0.0598759465,
-0.0068256855,
-0.1402484477,
-0.1277012825,
-0.1939339638,
0.1754539609,
-0.0835496038,
0.2089726925,
0.12800093,
0.2270999402,
-0.2550810575,
-0.1245741323,
0.4326337874,
-0.0996425375,
0.1882971078,
-0.3496018052,
0.2039178908,
0.2293762267,
0.170838967,
-0.0759380311,
-0.0538150966,
0.1604666412,
-0.030330386,
0.1078234315,
-0.0209006071,
0.053911522,
-0.1801204085,
-0.1716645062,
0.2066053748,
-0.3249641359,
-0.3378338218,
-0.2800191045,
-0.24342224
] |
https://github.com/huggingface/datasets/issues/1924 | Anonymous Dataset Addition (i.e Anonymous PR?) | Hi !
I guess you can add a dataset without the fields that must be kept anonymous, and then update those when the anonymity period is over.
You can also make the PR from an anonymous org.
Pinging @yjernite just to make sure it's ok | Hello,
Thanks a lot for your librairy.
We plan to submit a paper on OpenReview using the Anonymous setting. Is it possible to add a new dataset without breaking the anonimity, with a link to the paper ?
Cheers
@eusip | 45 | Anonymous Dataset Addition (i.e Anonymous PR?)
Hello,
Thanks a lot for your librairy.
We plan to submit a paper on OpenReview using the Anonymous setting. Is it possible to add a new dataset without breaking the anonimity, with a link to the paper ?
Cheers
@eusip
Hi !
I guess you can add a dataset without the fields that must be kept anonymous, and then update those when the anonymity period is over.
You can also make the PR from an anonymous org.
Pinging @yjernite just to make sure it's ok | [
-0.2978557944,
0.5536888838,
-0.0238175578,
-0.1903211474,
-0.0758947879,
-0.1079587638,
0.5019997358,
0.0080970339,
0.0642963499,
0.0534692258,
0.0712104514,
0.1038332283,
0.0041556433,
0.1634862572,
-0.0382043794,
0.0795786455,
0.0454187915,
-0.0966099054,
0.0962190628,
-0.2013755441,
-0.475312233,
-0.2437649816,
0.0791261643,
-0.5910321474,
0.1504401565,
-0.2484027445,
-0.2008708715,
-0.2064756304,
-0.0254559405,
-0.4402409792,
-0.2508313358,
0.2532380223,
0.0878260136,
0.1687666327,
-0.0001136982,
0.0365014672,
0.2275119722,
0.0229098797,
-0.2287046015,
-0.1465034038,
-0.4085339308,
-0.124110803,
-0.0218315255,
-0.1218807697,
0.0511341169,
-0.3378620148,
0.1586950868,
-0.4428369403,
-0.292034179,
0.0187054612,
0.1875342578,
0.2154201418,
0.1018248349,
-0.1916252822,
0.1721317172,
0.3090620041,
-0.0694507435,
-0.0765852034,
-0.2525061369,
0.23809129,
0.1200366914,
0.1911776364,
0.2108568102,
-0.3427753448,
0.2185996771,
0.2559672594,
0.1358453929,
-0.0866366178,
0.0983041078,
-0.0139205325,
0.5642252564,
0.1111583859,
-0.2762332261,
0.0066861808,
0.1424971819,
-0.2253249884,
0.1219129488,
0.360276252,
-0.1878941953,
0.0065117683,
-0.3873699009,
-0.4332364202,
-0.1487915516,
0.2721282542,
0.3960340023,
-0.082706973,
0.2204150856,
0.2198987901,
-0.1080730855,
-0.0076146713,
-0.0657847226,
-0.1311267018,
-0.1911905259,
0.0271341726,
0.2725529671,
-0.0197922066,
0.0595822632,
0.0854260921,
-0.0407376662,
0.1603175104,
0.0219724104,
-0.2648990154,
-0.2661925256,
-0.1018881723,
0.1051793545,
-0.2567985356,
-0.032910008,
0.2582084239,
0.1456508785,
-0.1770041883,
0.5126863122,
-0.1028511226,
0.2065977156,
0.2185720503,
-0.1635547578,
-0.4257533848,
0.0924937725,
-0.2446935177,
-0.0193251781,
0.1227923781,
-0.13861534,
0.1385224164,
-0.3871080875,
0.1158608198,
0.1389349997,
-0.3016420603,
0.2873122096,
-0.0588759556,
0.0300294012,
-0.1079090312,
0.0076900981,
-0.1463463306,
-0.1829976737,
0.02203146,
0.1334697008,
0.1433649659,
0.0655954108,
0.0237890892,
0.32793203,
0.2830860317,
0.3779454827,
0.0594573654,
0.158207804,
0.0805629641,
0.0714399815,
-0.2646239996,
-0.0486892574,
-0.5071716905,
0.0606982261,
0.2869587541,
0.4154133797,
0.3485382199,
-0.2451437265,
0.1902704239,
-0.1741664708,
-0.2699772716,
0.2741082907,
0.2919698954,
-0.0598985702,
-0.1097662598,
0.3346721232,
0.2572620511,
0.037102215,
-0.0198332183,
-0.0654721707,
0.3630783558,
-0.1588662267,
-0.2072475553,
0.1318630576,
-0.1387540549,
-0.2335203439,
-0.003538996,
-0.2095681578,
-0.199700743,
0.0260432139,
-0.2894258499,
-0.0541627854,
-0.2797903419,
-0.3725436628,
-0.3235584199,
-0.255299747,
-0.2175354809,
0.4006324112,
0.579221487,
0.2217823267,
-0.1478642225,
-0.2846816182,
-0.0699518919,
-0.1696241349,
-0.3042343259,
-0.4500306845,
-0.1257380843,
0.132505998,
0.0718635023,
-0.2021481097,
0.3229564726,
0.3146027327,
-0.158500284,
0.2660982609,
-0.2799394727,
0.1645662487,
0.1382401288,
0.5464157462,
-0.0631279871,
0.1361646205,
0.2277242392,
-0.1811690032,
-0.2020768523,
-0.3287793994,
0.4175467491,
0.1670810282,
-0.5736305714,
0.2955127656,
-0.2195339799,
0.0891281217,
-0.177730009,
0.0052080341,
0.2768188715,
-0.2847760618,
-0.2003248036,
-0.0757935122,
0.1646344364,
0.0238511711,
0.1015505344,
-0.4024478197,
-0.2272194922,
0.1277789921,
0.051037129,
-0.2499551773,
-0.0462140627,
-0.1039532572,
-0.1574062258,
0.1313717663,
0.1555236578,
-0.1213009506,
-0.0245547704,
0.7049095631,
0.4160329103,
-0.0812955201,
-0.2752478719,
0.1455886811,
0.1336465627,
-0.1570244581,
0.2209076881,
-0.662438333,
0.5039728284,
-0.0585589372,
-0.0115830824,
-0.1644262224,
0.2266272604,
0.4233987629,
-0.2268243432,
-0.1009387299,
-0.2740346789,
0.2443836331,
0.5010877848,
0.2135252804,
-0.1423729509,
-0.4439186752,
0.2676128745,
-0.4807434976,
-0.0149002224,
-0.0389984101,
-0.0193485823,
-0.1518962532,
0.2312800139,
-0.0101372693,
0.1605867147,
0.1751213819,
0.3742697537,
-0.2108211517,
0.0669728443,
0.2131952345,
-0.1076785251,
0.2763510346,
-0.0201306231,
-0.3504347801,
-0.15473786,
0.4559476972,
0.2296286374,
0.0018171966,
0.0319887549,
0.2925583124,
0.1714341938,
-0.0211069584,
-0.0420378447,
0.2286534607,
-0.1715064347,
0.133668676,
0.2258003801,
-0.2456110418,
-0.0075086392,
0.3010828793,
0.0826035142,
-0.1294430047,
0.2552321255,
-0.2911657095,
0.812264204,
-0.3899852633,
0.1093773991,
0.0818706155,
-0.2424033582,
0.0765797049,
0.1667840332,
0.1588558257,
-0.3950016201,
0.5738779306,
0.1488124728,
0.0059595853,
-0.4898708463,
-0.4616577625,
-0.0331034996,
0.1855695993,
0.0171720367,
-0.1285820603,
-0.2001143545,
-0.2259711325,
-0.0200330764,
-0.0314346515,
-0.2088885009,
-0.2280632257,
-0.1323081255,
-0.185437724,
-0.0510283858,
-0.1308765858,
-0.2060874403,
-0.0663329363,
0.2258480191,
0.4678016603,
0.4428368509,
-0.0063270256,
0.2578885555,
-0.1487092078,
-0.2320686281,
-0.1050002351,
-0.3201143742,
-0.2184939831,
-0.4413895309,
0.2819328606,
-0.3696287572,
-0.247701332,
0.2761836052,
-0.1013433039,
-0.041812975,
0.365200758,
-0.5292717218,
-0.0816141292,
0.144163385,
0.3721830845,
0.167689845,
-0.1993679255,
0.503202498,
0.192081511,
-0.2018285096,
-0.1152357236,
-0.1001848504,
0.2119533271,
-0.0725309476,
0.5440943837,
-0.3213919699,
-0.0509776548,
-0.0702492967,
0.6683696508,
0.4097921848,
-0.0773720592,
-0.0310786813,
-0.1300681084,
0.5164962411,
0.0133105442,
-0.072720781,
-0.0798473135,
0.1519531757,
-0.0881562531,
0.0063353963,
-0.0149986353,
-0.4001383781,
-0.0334775187,
-0.0946289673,
-0.1591326296,
-0.2538814545,
0.0483877584,
-0.0147430962,
0.6296268106,
0.1038354039,
-0.1671636254,
-0.5689088702,
-0.6112786531,
-0.0166112855,
0.2107751667,
0.097506538,
-0.3711679578,
-0.1986087263,
0.403889358,
-0.3924842477,
0.4433426261,
-0.2406225502,
0.0674040616,
-0.0974950641,
0.0603987724,
0.0572578758,
0.2134021819,
0.4243108928,
-0.4223253727,
-0.3225738406,
0.1250232458,
0.3005472124,
0.1358095109,
0.2328310013,
-0.2882538438,
0.3327240944,
0.1789937764,
0.0821347982,
-0.1581817716,
0.2696544528,
0.5318700075,
-0.1897791177,
-0.3482927978,
-0.1841911823,
-0.2772928178,
0.0357521027,
-0.09448006,
-0.0110327937,
-0.3482286334,
-0.0295590777,
-0.0094440207,
0.2832340598,
-0.1524952352,
0.4053570032,
0.227245152,
-0.3571995199,
-0.0322473608,
0.5289491415,
0.2385686934,
0.2302051932,
0.1427941024,
0.4895367622,
0.4533395469,
-0.1047908962,
-0.4532611668,
-0.0820490494,
-0.46717453,
0.2918985188,
0.4869163632,
0.3948284984,
0.4081509709,
0.1663913578,
0.0477410816,
-0.0323489718,
0.0627355129,
-0.08194758,
-0.1082178503,
-0.1452765167,
-0.4273102283,
0.1304294765,
-0.0571514368,
-0.1948763132,
0.3306123018,
-0.0339325368,
-0.1191854626,
0.5634449124,
0.1458995938,
1.102309823,
0.1004172564,
0.2189496458,
-0.2815108895,
-0.1477195472,
0.4086611271,
-0.0295333341,
0.19761765,
0.0389006063,
-0.0740404502,
0.0169317424,
-0.0460317321,
0.1066624299,
-0.0323698297,
-0.0779179335,
-0.1003016084,
-0.1667450815,
0.0400969945,
-0.1215350404,
0.4067530334,
0.1403792053,
-0.1238912046,
0.2923856676,
0.1070314348,
0.1914249808,
-0.3018865883,
-0.0045079477,
-0.2803260684,
-0.1367900819,
-0.1383552849,
-0.0493658185,
-0.0047561303,
0.2829568088,
-0.2809302211,
0.0539185703,
-0.3358975053,
0.3144949079,
0.2217281163,
0.4640052617,
0.0496844277,
-0.2218733728,
0.0630489737,
0.0547410212,
0.0449578948,
0.2886953056,
-0.0992083028,
-0.1164653897,
-0.006829679,
-0.2520045638,
0.4243527651,
0.0123269185,
-0.4694528878,
-0.3177476823,
-0.2418299019,
0.333938241,
-0.3270103335,
0.5733968019,
0.2793071568,
-0.3338205814,
-0.1607429385,
0.1331073493,
0.207985118,
0.0421005823,
0.1667077094,
0.2086301744,
-0.0422760174,
-0.1200370938,
0.2904540598,
0.2368240803,
-0.2219739556,
0.3192645907,
-0.0267110914,
-0.1747163534,
-0.1491133571,
0.1113801003,
0.0908228755,
-0.5180431604,
-0.1389475763,
0.2150217891,
0.5172665119,
-0.1865575314,
0.0901886448,
-0.0313063264,
0.2822613418,
-0.3256953955,
-0.1173382103,
-0.3173074722,
0.1764897704,
-0.0433793627,
0.0434846096,
0.132183224,
0.1180728227,
-0.0843106508,
0.1214713603,
-0.3499597013,
-0.1642445922,
0.0788094923,
0.1941696852,
-0.0780225247,
0.1549398154,
0.2916840911,
-0.3930840492,
0.0790205821,
-0.334344238,
0.1350066364,
-0.1498426497,
-0.062908724,
0.1342588961,
0.0922142714,
-0.3548137844,
-0.2271990776,
-0.2885836959,
0.0102205351,
0.155758068,
0.0311390683,
0.0854367465,
-0.1410773396,
-0.5126223564,
0.03281416,
0.1637097448,
-0.0851256698,
-0.2230620384,
-0.0045079254,
0.2441368103,
-0.0929356068,
0.1326642781,
-0.2574228048,
0.0306834728,
0.3204218149,
0.1555725783,
0.3057706654,
0.1402734816,
0.1473254859,
-0.1621404588,
0.2386341393,
0.0103779808,
0.1027918756,
-0.0022573713,
-0.3916285932,
0.1462595165,
0.2996272743,
0.205855459,
-0.3969429731,
-0.0135457739,
-0.0898564458,
0.348826617,
0.1341759861,
0.3458151519,
0.2564820051,
-0.130851686,
-0.0300773568,
0.0337994844,
-0.0981148034,
-0.3168903589,
0.0707858205,
0.0121033713,
0.0776243508,
0.074657917,
0.3548710942,
0.527747035,
-0.1667528301,
-0.0421075374,
0.1757557392,
0.1529542059,
0.0668763071,
-0.0905279145,
-0.0371427722,
0.0062319562,
0.2874759734,
0.0308411419,
0.3539131284,
0.4143078327,
0.2471044958,
0.0791501552,
0.1839891523,
0.3479205966,
-0.0506792478,
0.2434343696,
0.25242576,
0.1177732795,
-0.4141277671,
-0.0449197963,
0.0372038782,
0.0642394274,
0.1411057413,
-0.2051167935,
0.0313309133,
0.1824623644,
-0.1503525376,
-0.2200533748,
0.1741537452,
0.178161025,
0.3247282505,
-0.2266157269,
0.0150914695,
0.1836251616,
-0.0367107503,
0.373590529,
0.7002022266,
-0.0724854469,
-0.0966845304,
0.0389507748,
0.4729219675,
0.0671326667,
-0.0448374711,
0.0145803243,
0.1043499932,
0.3690758348,
-0.0846205652,
-0.1591273546,
0.0622223467,
-0.0456526317,
0.3700080812,
0.1289580464,
-0.0270937849,
-0.2512516081,
0.2030462921,
0.1691949069,
-0.1808615923,
-0.1311333179,
-0.1924047619,
-0.0200506821,
-0.0679726452,
0.5362576842,
-0.087204054,
-0.1096590832,
-0.0797272772,
0.0757282227,
-0.0774374753,
0.0412249826,
0.06765607,
0.1005144417,
-0.2797423601,
-0.2162722647,
-0.0529519618,
0.4314089715,
-0.2296756208,
-0.1711851954,
-0.2231499702,
0.0299841985,
0.4677250087,
0.0866040885,
0.2552156448,
-0.0955869406,
-0.3125904202,
0.4435619414,
-0.1287368983,
-0.0321026295,
-0.2192428112,
0.3003249764,
0.07728073,
0.1784944534,
0.2000688314,
0.2640064657,
0.0090310574,
-0.1725067496,
-0.0694137514,
-0.0369102247,
0.3054789603,
0.2970539927,
-0.1720964313,
0.2769837976,
-0.1313275993,
-0.2637746334,
0.1774758697,
-0.0812478513,
-0.4348012209,
0.1676121801,
0.2094634473,
-0.2985936701,
-0.3462677002,
-0.1960133016,
0.253947556,
0.0089699142,
-0.3364591002,
-0.0469115227,
-0.2473500669,
0.0109994933,
-0.0972688124,
-0.0149677843,
-0.0300884265,
-0.0220290292,
0.0159425959,
-0.0463808402,
-0.2196148932,
-0.0853784084,
0.2916967869,
-0.5352241397,
0.2132211626,
0.005520273,
0.3361282349,
-0.3151446879,
-0.3718152642,
-0.4345382452,
0.0649128705,
0.2189655304,
-0.0057225209,
0.0519206971,
-0.2206097394,
-0.0004887246,
0.0172971487,
-0.1764286608,
0.4233128726,
-0.103806749,
-0.0114995837,
-0.1624679863,
-0.1279364526
] |
https://github.com/huggingface/datasets/issues/1924 | Anonymous Dataset Addition (i.e Anonymous PR?) | Hello,
I would prefer to do the reverse: adding a link to an anonymous paper without the people names/institution in the PR. Would it be conceivable ?
Cheers
| Hello,
Thanks a lot for your librairy.
We plan to submit a paper on OpenReview using the Anonymous setting. Is it possible to add a new dataset without breaking the anonimity, with a link to the paper ?
Cheers
@eusip | 28 | Anonymous Dataset Addition (i.e Anonymous PR?)
Hello,
Thanks a lot for your librairy.
We plan to submit a paper on OpenReview using the Anonymous setting. Is it possible to add a new dataset without breaking the anonimity, with a link to the paper ?
Cheers
@eusip
Hello,
I would prefer to do the reverse: adding a link to an anonymous paper without the people names/institution in the PR. Would it be conceivable ?
Cheers
| [
-0.2011667788,
0.4423331022,
0.0201971494,
-0.0765715465,
-0.0947151184,
-0.0944124982,
0.4387367964,
-0.0027227094,
0.0362778082,
0.0934400558,
-0.0507460125,
0.0669828951,
0.0037659407,
0.0874346569,
0.137011528,
0.1177665442,
0.1308903992,
-0.0895970836,
0.2099278867,
-0.2495175302,
-0.5651834011,
-0.1786225289,
0.0916348547,
-0.5738096833,
0.2315873206,
-0.2072768062,
-0.2125303745,
-0.166072309,
0.0858120397,
-0.3669402301,
-0.2833179235,
0.1583455652,
-0.0273780786,
0.2683498859,
-0.0001262668,
0.1422165483,
0.2734459043,
0.0246181414,
-0.2487216145,
-0.1675762087,
-0.5025140047,
-0.0314129256,
-0.0990060493,
-0.1659387797,
0.0814034343,
-0.3552339077,
0.1146846861,
-0.3773091137,
-0.3385234177,
0.0273259878,
0.0901735276,
0.1165460348,
0.0729364008,
-0.0961632282,
0.1395260692,
0.4059197605,
-0.0186561197,
0.0075167082,
-0.2865668833,
0.2626429796,
0.219978109,
0.0724458098,
0.1970393062,
-0.3708331585,
0.2289114892,
0.1912171245,
0.0853179693,
-0.0382952318,
0.0687063411,
0.1095336527,
0.6216088533,
0.0948710591,
-0.2760728896,
0.0327649713,
0.1474634111,
-0.085282892,
0.0675752908,
0.3558513522,
-0.1859708875,
0.0162691865,
-0.4244500399,
-0.5069785714,
-0.1157959327,
0.2613271177,
0.3833505213,
-0.1147989631,
0.3029622138,
0.2377170324,
-0.1090037823,
0.0225539915,
-0.1252333075,
-0.1423777193,
-0.2535926998,
0.1280825883,
0.3269245625,
-0.050358396,
0.0798264593,
0.0344047546,
-0.0289013721,
0.0962053537,
0.0563226379,
-0.2872753143,
-0.2855662405,
-0.0859596431,
0.0643909201,
-0.316010505,
-0.0880989209,
0.2874580622,
0.0498651043,
-0.1583643258,
0.5381064415,
-0.0784930065,
0.1278111935,
0.2147776037,
-0.2309503853,
-0.4649636745,
0.0294684842,
-0.2643435597,
-0.0542346574,
0.0519964136,
-0.253179431,
0.1072177887,
-0.4038321674,
0.1170247421,
0.0738791972,
-0.346156925,
0.3234758377,
-0.0996326208,
-0.0508285798,
0.0475883745,
0.0394909643,
-0.0566449463,
-0.2179218233,
-0.079860121,
0.127335459,
0.2161311656,
0.0571416989,
0.0441981405,
0.2492774725,
0.2355626523,
0.2972958684,
-0.0089327805,
0.2083232701,
0.0413715392,
0.1397214234,
-0.2625544369,
-0.0491478704,
-0.6527944803,
0.0888419151,
0.3641911149,
0.2272873074,
0.2947357893,
-0.2152951807,
0.0771596506,
-0.1591787189,
-0.2361368984,
0.1870181859,
0.1990447044,
-0.1162703782,
-0.1540000141,
0.2643205523,
0.3141644597,
-0.0525800809,
-0.0466040559,
0.017729193,
0.4519747496,
-0.2121917009,
-0.2798966169,
0.2454179078,
-0.1773502678,
-0.1309851408,
0.127519533,
-0.1967163384,
-0.124519214,
-0.0778546855,
-0.3614667356,
0.0002922267,
-0.2300684452,
-0.3864836991,
-0.2217601091,
-0.3067473173,
-0.1487670243,
0.3727559447,
0.648047924,
0.1632295251,
-0.0901930481,
-0.1348421872,
-0.0588172227,
-0.0446886569,
-0.2744155824,
-0.4228569269,
-0.1009270698,
0.1049103886,
-0.0332544595,
-0.3364387751,
0.460716635,
0.3744396865,
-0.1109759063,
0.37315768,
-0.2698196769,
0.1481584013,
0.2017127872,
0.5984318852,
-0.0929670781,
0.149168089,
0.3131767511,
-0.0584282465,
-0.2318140268,
-0.3104813695,
0.349332571,
0.1211913377,
-0.5198779106,
0.2872048318,
-0.1309356391,
0.1329521537,
-0.0196611788,
-0.1120870113,
0.1630741507,
-0.33604002,
-0.102652885,
-0.1313272864,
0.2828585207,
0.0319539607,
0.1305519938,
-0.3676550984,
-0.1331423372,
0.2234247476,
0.1517512053,
-0.2283906341,
0.0358832739,
-0.0845750868,
-0.1171073169,
0.1633032262,
0.1684955955,
-0.1520996988,
-0.0657867789,
0.7105240822,
0.4017373621,
-0.0259385966,
-0.3726827502,
0.1247756854,
0.1605212688,
-0.0716077164,
0.0807437599,
-0.6590613723,
0.4140939713,
-0.0702204555,
0.0705800056,
-0.1083903536,
0.1661628187,
0.2895026803,
-0.2371482551,
0.0395897925,
-0.1682228297,
0.3143057525,
0.5559506416,
0.3530570865,
-0.153545782,
-0.4435144067,
0.105155535,
-0.6972008348,
-0.0495924242,
-0.0260427669,
-0.002795334,
-0.2676231265,
0.2278952003,
-0.040353585,
0.2359809577,
0.2688289285,
0.2979460061,
-0.1101178676,
0.1627375633,
0.2760839164,
-0.0626735017,
0.2161821276,
-0.1061522141,
-0.3277378678,
-0.1313814074,
0.4049284458,
0.2382097393,
0.0116228312,
0.0085199103,
0.3042289615,
0.1753626019,
-0.1000527889,
-0.0469265319,
0.2736751735,
-0.1836228371,
0.132582888,
0.2519229949,
-0.2054319084,
-0.0217876732,
0.2373285592,
0.1182723567,
-0.0469167121,
0.1121252701,
-0.3568474054,
0.7788481712,
-0.4367330372,
-0.0640920624,
0.0964713693,
-0.1648803651,
0.1710065454,
0.1025387123,
0.0354841836,
-0.1941317022,
0.4717493057,
0.1212737784,
-0.009404473,
-0.4361032248,
-0.3261879385,
-0.0210465081,
0.2483043969,
-0.0025580395,
-0.0462467447,
-0.1519554704,
-0.1663589478,
-0.0428269655,
-0.0476743095,
-0.0424825735,
-0.2063823938,
-0.1596757919,
-0.231605947,
-0.1420591772,
-0.0228530951,
-0.0611934215,
-0.1075329706,
0.3066418171,
0.4745169282,
0.3733966947,
0.0308913589,
0.1602348089,
-0.1567804664,
-0.2718473971,
-0.0966224521,
-0.3991083205,
-0.0479863733,
-0.4468702078,
0.2575536668,
-0.3110265136,
-0.2784335613,
0.3430236578,
-0.1163652986,
-0.017909199,
0.4083649218,
-0.5690828562,
-0.0278034061,
0.1275938302,
0.3076960742,
0.1703377813,
-0.1437243074,
0.3931420743,
0.2275319993,
-0.082277514,
-0.0902018845,
-0.1061114967,
0.204621464,
-0.1738819331,
0.6187599301,
-0.1944048405,
-0.0548234917,
-0.0545139797,
0.6473226547,
0.3969497383,
-0.1100331172,
-0.0457416177,
-0.2120824009,
0.5558550954,
0.0750209913,
-0.0713286325,
-0.128147319,
0.2026109397,
-0.1250329763,
-0.074445188,
0.0493145213,
-0.483283788,
-0.1288199127,
-0.119301118,
-0.1127938554,
-0.1850577891,
-0.0140775265,
-0.0746961683,
0.566765964,
0.0115627497,
-0.2575371265,
-0.5660694242,
-0.6033062935,
0.0378772095,
0.3398483396,
0.1747937351,
-0.3895837069,
-0.3328223825,
0.4341337383,
-0.4290967286,
0.3453397155,
-0.2231279612,
0.0909884125,
-0.1166273803,
0.0546609014,
0.1256179065,
0.1978277415,
0.4831281006,
-0.4202387631,
-0.2678388357,
0.0679990053,
0.3786559105,
0.0961248428,
0.2741272748,
-0.2173536718,
0.3001218438,
0.2447848767,
-0.0258017294,
-0.1339325011,
0.2969378233,
0.5221359134,
-0.2260957807,
-0.3237432837,
-0.2412979156,
-0.2852935195,
0.0302048549,
0.0038252994,
-0.0078398548,
-0.3472955227,
-0.0410927646,
0.092320621,
0.1792479157,
-0.1938169599,
0.3915729523,
0.343472898,
-0.3964322209,
-0.0315436088,
0.5583866835,
0.2310681343,
0.2091258913,
0.3514365256,
0.4544341564,
0.5043780804,
-0.1293909252,
-0.5598924756,
-0.1983911246,
-0.4881313741,
0.4070115685,
0.4231128395,
0.3770062923,
0.3981899619,
0.1953666508,
-0.0679778457,
0.063485235,
0.140625447,
-0.1637904793,
-0.047614079,
-0.353659302,
-0.3614880443,
0.1960398406,
-0.0669638664,
-0.2215824723,
0.3081966043,
0.0110014603,
-0.1212780401,
0.4656130672,
0.0476929471,
1.238451004,
0.1484941691,
0.2753530145,
-0.2489495128,
-0.2222339511,
0.3537950516,
0.0325671136,
0.1756274551,
-0.0171393082,
-0.0671030805,
-0.0676111132,
-0.1809383333,
0.0076241898,
-0.0023862659,
0.0968144834,
-0.1408823729,
-0.244505167,
0.0033183843,
-0.0653604865,
0.3175384402,
0.2996057272,
-0.0296008345,
0.4043976068,
-0.0052496009,
0.2315816283,
-0.2421018332,
-0.0196748599,
-0.2431137711,
-0.2138371766,
-0.1342339069,
-0.0893024504,
0.0681935698,
0.1340528578,
-0.31129691,
0.1171625555,
-0.447260499,
0.3500978053,
0.3386387825,
0.4484480619,
0.1044708043,
-0.1836316139,
0.0932415649,
0.0463463441,
0.1008625478,
0.3841865063,
-0.0712477267,
-0.2078347951,
0.0892880559,
-0.3228289783,
0.4911770821,
-0.0261773802,
-0.5840346813,
-0.2656639516,
-0.2430944443,
0.2465313673,
-0.3386535645,
0.4108411372,
0.3529491425,
-0.2228948921,
-0.0658856332,
0.0265276879,
0.236474961,
0.0466573909,
0.1349187791,
0.181542173,
-0.0505536273,
-0.0214173011,
0.4060077667,
0.2141212821,
-0.3812198639,
0.3060284257,
-0.1139823198,
-0.186562866,
-0.0384656675,
0.0719916672,
0.174349308,
-0.4641197324,
-0.2478680313,
0.189019233,
0.6258869767,
-0.2097506821,
0.0078873057,
-0.0960949212,
0.3676455021,
-0.3700948954,
-0.1063321009,
-0.2826553881,
0.2064352483,
-0.115742445,
0.0402161404,
0.0490543097,
0.0395376608,
-0.043816939,
0.0087192561,
-0.2136474848,
-0.1027080491,
0.1055814847,
0.1609469354,
0.0182481483,
0.3611711562,
0.3024439514,
-0.414524138,
-0.0036732741,
-0.4310127497,
0.1613377631,
-0.0555241071,
-0.0313929506,
0.1859136969,
0.1197884977,
-0.3572767675,
-0.2396792769,
-0.357696712,
0.0524084456,
0.1926746666,
0.0640170351,
0.122562021,
-0.1955525577,
-0.5717163682,
0.1367691755,
0.1894460022,
-0.2239124626,
-0.1017909795,
0.1240300834,
0.3057493269,
-0.1502552032,
0.0802022293,
-0.2521093488,
0.0395031422,
0.3099377751,
0.1254523695,
0.3709876537,
0.231478706,
0.1872219592,
-0.1610581726,
0.2410394549,
-0.0610753447,
0.0096614733,
-0.0496631339,
-0.4228900969,
0.1595315635,
0.3035747707,
0.1041654944,
-0.3014523983,
-0.0901675075,
-0.0021966174,
0.4176015556,
0.1107092798,
0.2904235721,
0.2070519477,
-0.1417417377,
-0.0988253057,
-0.0486247987,
-0.0972355902,
-0.3975602388,
-0.0232786294,
0.0920950621,
0.0603302717,
0.1328024119,
0.228908211,
0.5250608921,
-0.2251093984,
0.0746503323,
0.1632906795,
0.1939116269,
0.2640427053,
-0.0256817862,
-0.0109263211,
0.1031353474,
0.2265006006,
0.0186881125,
0.3572963178,
0.3870841265,
0.2988615632,
-0.0848571509,
0.2265390158,
0.4148774147,
0.0197619125,
0.2069387883,
0.2649191916,
0.0552268475,
-0.4714699984,
-0.0823116302,
0.0401577652,
0.0133799016,
0.1034477204,
-0.2998309135,
-0.0817497522,
0.1620904505,
-0.010219913,
-0.1705801934,
0.1244458705,
0.2051192224,
0.2919786274,
-0.2277559787,
-0.0460024998,
0.06939888,
0.1477602571,
0.4056126177,
0.7547986507,
-0.0657601953,
-0.0475361943,
0.2017755061,
0.4158106148,
0.0178020298,
-0.0005627722,
0.0580631942,
0.0738009438,
0.3980825543,
-0.1837504208,
-0.1494769901,
-0.0211103987,
0.0649082512,
0.4278397262,
0.134667322,
0.0423229858,
-0.2120673507,
0.2639536262,
-0.0169522557,
-0.1686989069,
-0.1762657613,
-0.1584698558,
0.0956152529,
0.0273570344,
0.5971824527,
-0.1300954819,
-0.13465029,
0.0912897885,
-0.0000076741,
-0.0564623624,
-0.0904305577,
0.1892033219,
0.1403956562,
-0.2802186608,
-0.2236588001,
-0.0279043987,
0.3802797198,
-0.1861333251,
-0.2230953574,
-0.3326605558,
-0.061839357,
0.3705801964,
0.0530228987,
0.0935086682,
0.0273808688,
-0.3798121214,
0.4845601916,
-0.111927487,
-0.1011022776,
-0.2941743135,
0.3170862198,
-0.0849199295,
0.1796785295,
0.152898103,
0.3086315691,
-0.0950192735,
-0.1213015318,
-0.0762389749,
-0.1598171294,
0.3058773279,
0.2785099149,
-0.19577308,
0.3706884384,
-0.1379020512,
-0.2090938985,
0.290596962,
-0.0943000019,
-0.3691658378,
0.2377328277,
0.1939806044,
-0.3000845313,
-0.297524035,
-0.0845024139,
0.3556576669,
-0.0637445599,
-0.2942963243,
-0.0366500169,
-0.3430094421,
-0.0321010835,
-0.1096607223,
-0.0095350966,
0.0639313459,
-0.0277892686,
0.0190484133,
-0.0419584885,
-0.156174615,
0.0116827935,
0.2327779531,
-0.5458317995,
0.3114070296,
-0.0218560249,
0.4344708323,
-0.4064273238,
-0.3519240022,
-0.3572444618,
-0.0659035593,
0.2870923579,
-0.0447316207,
0.0349984765,
-0.2243575305,
-0.0411721319,
0.0790597647,
-0.1507352889,
0.3140511811,
-0.0783636868,
-0.0839312896,
-0.1645743251,
-0.1246091798
] |
https://github.com/huggingface/datasets/issues/1922 | How to update the "wino_bias" dataset | Hi @JieyuZhao !
You can edit the dataset card of wino_bias to update the URL via a Pull Request. This would be really appreciated :)
The dataset card is the README.md file you can find at https://github.com/huggingface/datasets/tree/master/datasets/wino_bias
Also the homepage url is also mentioned in the wino_bias.py so feel free to update it there as well.
You can create a Pull Request directly from the github interface by editing the files you want and submit a PR, or from a local clone of the repository.
Thanks for noticing ! | Hi all,
Thanks for the efforts to collect all the datasets! But I think there is a problem with the wino_bias dataset. The current link is not correct. How can I update that?
Thanks! | 89 | How to update the "wino_bias" dataset
Hi all,
Thanks for the efforts to collect all the datasets! But I think there is a problem with the wino_bias dataset. The current link is not correct. How can I update that?
Thanks!
Hi @JieyuZhao !
You can edit the dataset card of wino_bias to update the URL via a Pull Request. This would be really appreciated :)
The dataset card is the README.md file you can find at https://github.com/huggingface/datasets/tree/master/datasets/wino_bias
Also the homepage url is also mentioned in the wino_bias.py so feel free to update it there as well.
You can create a Pull Request directly from the github interface by editing the files you want and submit a PR, or from a local clone of the repository.
Thanks for noticing ! | [
-0.3595491052,
0.1663746387,
-0.0910966992,
0.1053339988,
0.0010983972,
0.1791954935,
-0.0393258519,
0.0637115687,
0.0622650683,
-0.1287822425,
-0.2621756792,
-0.0001705661,
0.2578837574,
0.159308821,
0.2776814699,
-0.3877265751,
0.1113261282,
0.1100791767,
0.1167293936,
-0.2110314965,
-0.3091870546,
-0.082206808,
-0.2789872289,
0.0970850438,
-0.0021145977,
-0.0778444484,
-0.5115852952,
0.27758044,
-0.2706758976,
-0.4728705883,
0.3751762807,
0.4130044878,
-0.0921291411,
0.3960894048,
-0.0001064684,
-0.2544564307,
0.4334570169,
0.1170404255,
-0.1512612402,
-0.0557769239,
-0.1133985296,
-0.1855168641,
-0.3816219568,
-0.1367889941,
0.0211596265,
-0.1777677089,
0.13675116,
0.1966269016,
0.1025710553,
0.4926283956,
0.2142111212,
0.2089529186,
0.3577847481,
-0.1554494053,
0.3749153316,
0.2227900326,
0.256405443,
0.3468583822,
0.0206911471,
-0.1691048741,
0.1622936726,
0.2506521642,
-0.1467976868,
-0.1894209087,
-0.0663959384,
-0.0445827357,
0.0470029004,
-0.1570635289,
0.127535969,
0.1428318471,
0.1548642218,
-0.0296165496,
-0.358740449,
-0.0778490752,
0.1910635978,
-0.7335625291,
0.1653981954,
0.0577255264,
0.1552005112,
-0.0117981965,
0.1111091748,
-0.5181042552,
-0.1214333102,
0.3081213832,
0.1125403121,
0.101073876,
-0.0518351868,
0.1025831997,
0.1344676614,
0.0219501108,
0.3273353279,
0.0410104804,
-0.2244394124,
-0.0568372719,
-0.2383378893,
-0.1983615011,
-0.1284374744,
0.462798357,
0.1581487209,
-0.0870216638,
-0.2016758025,
-0.0480875149,
-0.2221377045,
-0.1606108993,
0.3178703189,
-0.026414454,
-0.1392502487,
-0.1350998282,
0.1666117311,
0.198597461,
-0.0380893536,
0.0863722861,
-0.0215295739,
-0.3497814536,
-0.1778567135,
-0.0127986129,
0.2721113861,
-0.0768699422,
-0.3750707209,
-0.0947863609,
0.0259465352,
-0.1254573762,
0.0305153541,
0.0824878216,
-0.1555778384,
0.219055593,
-0.0637160167,
0.1419501603,
-0.2471201718,
-0.4623444676,
-0.2801394761,
0.0670007914,
-0.2803138494,
-0.0033135936,
0.1079284251,
-0.1923787743,
0.4917117655,
-0.0540024415,
-0.1442542523,
-0.0563038364,
-0.2679827511,
0.0572680049,
-0.0964816064,
0.0256994739,
-0.0513382778,
0.0214460641,
0.0096339704,
0.2765557468,
-0.0625868067,
0.1380600035,
-0.382538408,
-0.2012335062,
-0.0192093197,
0.0240554102,
-0.0979100764,
-0.3637700975,
-0.3967582881,
0.2190154195,
-0.1084559858,
-0.0248066187,
0.0903780907,
-0.0839592665,
-0.3254859447,
-0.1047410518,
0.3892813623,
0.4097489715,
0.0005987063,
0.2222696245,
-0.0285614543,
0.0435795188,
0.0198258944,
0.1556759477,
-0.1559699178,
0.0572351925,
-0.0255528539,
-0.2854389548,
0.2380370796,
-0.464219749,
-0.4162274897,
-0.264711678,
-0.2668763995,
-0.3149488866,
-0.2621735632,
0.1892431676,
-0.2789890468,
0.0009048581,
-0.1267003119,
0.4232994914,
-0.1128282547,
-0.0350896865,
-0.2599847317,
-0.1326120794,
-0.1460778266,
0.1583600938,
0.0443663001,
0.1021289974,
-0.0254889652,
-0.0300820656,
0.4650369287,
0.105911985,
-0.0444544926,
0.0974042267,
0.4143878818,
0.167291224,
0.0513468087,
-0.0249311775,
-0.1223680377,
-0.1004391164,
0.0400926769,
-0.0212420542,
0.325835228,
-0.0977121741,
-0.4292384386,
0.0725059509,
-0.0063625239,
-0.1980825216,
0.1160107106,
0.0325334109,
0.1374223679,
0.1323656738,
-0.1745866239,
-0.0510635301,
-0.1280090213,
0.2269713581,
0.2016354054,
0.3748256266,
-0.1444935501,
-0.0283052959,
0.4443068504,
0.2948619723,
0.161873579,
-0.0843275785,
-0.1209078729,
0.5419948101,
-0.1632205546,
0.1217005998,
0.4820056856,
0.2558828592,
0.2863235176,
-0.5517088175,
0.1105967164,
0.1785250902,
-0.1446703672,
0.3147726357,
-0.1308722347,
0.1476727873,
-0.2084830701,
0.0985419899,
-0.3138096333,
0.4175607264,
0.2844437957,
-0.0370112993,
0.1749882996,
0.2364445031,
-0.0953708738,
-0.0353617705,
0.0069963709,
-0.2239558548,
-0.3241181374,
0.3490576744,
0.1075028628,
-0.0751613155,
-0.1951500773,
0.0845119208,
-0.4713924825,
-0.1856357455,
-0.1845973581,
0.1250655651,
0.2852446139,
0.0596685931,
0.0369321927,
0.1783206016,
0.0880989954,
-0.1388008893,
0.2750366628,
0.2534804344,
-0.1154990569,
-0.1864896566,
0.1920571625,
0.1174202487,
-0.2700319588,
0.2500794828,
0.1069786176,
0.3870289922,
-0.1194835007,
-0.0423878469,
0.1071498692,
-0.2513253093,
-0.1118522659,
-0.2089512795,
-0.3875475526,
-0.2502337098,
0.1265204847,
0.2735847235,
-0.0377792791,
0.3236306906,
-0.0978945568,
0.4073317051,
0.2222722769,
0.2817662656,
-0.0579493456,
-0.192430079,
-0.0709383637,
-0.0261851195,
0.1622070372,
0.007729318,
0.1705098003,
-0.3521749973,
0.3545432091,
-0.5018277764,
-0.2015231252,
-0.0719842017,
0.0605513863,
0.4237367213,
0.1837676018,
0.0672683865,
-0.0619833246,
-0.0454571992,
0.2018214166,
-0.5429756045,
0.1377968192,
-0.1011146009,
0.0560784116,
0.1222833842,
0.2462978661,
-0.2002688348,
0.0757782087,
-0.1129234433,
-0.0110853016,
0.0062893108,
0.2626054287,
0.0833494216,
0.0803582668,
-0.1830536276,
-0.1386596262,
-0.1123514026,
-0.2118853033,
-0.5318338275,
0.2945449352,
-0.359564513,
-0.4036682546,
0.0698619634,
-0.0449922383,
0.1299846172,
-0.2164372355,
-0.455337137,
-0.1812924147,
-0.1102126539,
-0.0260296315,
-0.017629765,
0.2905606031,
0.2023492903,
-0.250267148,
-0.2868279219,
-0.061381951,
-0.2773210406,
-0.0220622458,
0.1599816382,
0.4155291021,
0.2630162239,
-0.1384329051,
-0.0856815651,
0.7499743104,
0.1584287137,
-0.3041236699,
0.2625769377,
-0.1049518138,
0.3394190073,
0.2625928223,
-0.4821256101,
0.0951005891,
0.0758166984,
0.3618255258,
0.4159469604,
0.21476686,
-0.0936457142,
-0.1254830211,
-0.4458153248,
-0.3002959192,
-0.3494910002,
-0.0799217224,
-0.2334311306,
0.2280254662,
0.3162689805,
0.0644520372,
-0.3139526844,
-0.3706804514,
0.4793215692,
0.2478989959,
0.0487986468,
-0.0248305276,
-0.0488770343,
-0.1790973246,
-0.2587457597,
0.100715667,
-0.3959198296,
0.3303527832,
0.1559309363,
0.2400743514,
0.1629255265,
-0.0136304125,
0.5150209069,
-0.3189904988,
0.2145065963,
0.0065915212,
0.4062414169,
0.0342264995,
0.0536104888,
-0.1371552944,
0.4261341095,
0.3031488657,
0.0954751149,
-0.0536198989,
-0.3796595037,
0.5107845068,
0.0088636391,
-0.3696417212,
-0.1735326052,
0.0432640463,
-0.1732383817,
0.1945694834,
0.1986499131,
0.003381595,
0.2432299852,
-0.173266232,
0.3855415583,
-0.1760900468,
0.1895329058,
0.1059627831,
-0.1900243163,
0.1902853847,
0.7589068413,
-0.0688140541,
0.2169294655,
0.3593293726,
0.0102547985,
0.2289910316,
-0.0221395083,
-0.0018982328,
-0.3797620535,
-0.0291330703,
0.3100038767,
0.3471753001,
0.1730179489,
-0.143930763,
-0.0786146894,
-0.076165691,
-0.266587913,
0.256415993,
0.4641890824,
0.0551709607,
0.0940660089,
-0.451341629,
0.3257634938,
-0.0963439941,
-0.2576680779,
0.3563810587,
0.677089572,
-0.0954729468,
0.1394957155,
0.4004176557,
0.9049201012,
0.08276847,
0.1737267226,
0.0031516049,
-0.3393121064,
0.5974231362,
0.2014865577,
0.2052601129,
-0.2889651656,
-0.1158254445,
0.0493203253,
-0.0234537199,
-0.1391365677,
-0.1694777608,
-0.0496769845,
0.1778846532,
0.1009410173,
-0.0762550458,
0.2097302228,
0.0811128914,
0.3292877376,
0.0470665731,
0.0342060812,
0.1318622231,
-0.0655900389,
0.3627303541,
-0.0238695741,
-0.0672174618,
0.1665794104,
-0.0531841666,
-0.4540297687,
-0.1060018763,
-0.3741121292,
-0.1598392427,
-0.2068897486,
-0.2787749767,
0.089974463,
0.2630539238,
0.6369228959,
0.1170048639,
-0.2022815198,
0.2968301177,
0.209212631,
-0.3399342895,
-0.0385755748,
-0.0325205997,
0.0729541853,
-0.0469718352,
-0.4824801385,
0.0205872282,
0.0569696277,
-0.3097610176,
-0.2386479974,
-0.3444825411,
-0.2155518234,
-0.0426006243,
-0.0406220183,
0.1557124704,
0.2891095877,
-0.2364448756,
0.1308779866,
0.2382086515,
-0.0215799846,
0.0094801523,
0.1763737202,
-0.1295788884,
-0.0420934185,
0.3489995599,
-0.212296471,
-0.2513976395,
0.4640516639,
0.0533142984,
0.0571855381,
-0.2307001352,
-0.0966277421,
0.0322855152,
-0.1737258732,
-0.2553487122,
-0.2653813958,
0.2728779614,
-0.0655220151,
0.0965376645,
0.1904216707,
0.1249187216,
0.1873115003,
-0.3786252141,
-0.1960244179,
-0.0279772617,
0.0701157823,
0.0200257115,
-0.1266752183,
-0.3186695576,
0.3089509904,
0.023886947,
-0.3352240026,
0.2886680961,
0.0549336821,
-0.090195179,
0.0819556937,
0.1310852021,
0.0562632605,
-0.1661138535,
0.0168973133,
-0.6879586577,
-0.1011052728,
-0.1165051684,
-0.2289209068,
0.180902347,
0.0273050219,
-0.1927421093,
-0.1123914868,
-0.3219530582,
-0.0460023582,
0.107562989,
-0.2701085806,
-0.0007016435,
-0.0720351338,
0.1723214388,
-0.2173938006,
0.3391173184,
0.037873283,
0.1605909765,
0.0547614656,
0.3583118618,
0.1378595233,
0.2793385088,
0.1034465134,
-0.3125388026,
0.120746851,
0.1390588135,
-0.2034009248,
0.1799467504,
0.0968891382,
-0.1967435777,
0.0019370168,
0.5520458221,
0.3395652175,
0.5244820118,
0.0260486156,
0.1232300997,
0.3496302068,
0.2413619012,
-0.3485978246,
0.2696169317,
0.3537513018,
0.1770883799,
-0.0470032357,
-0.0910043567,
0.4353305399,
-0.0034166723,
-0.2699351311,
0.220396623,
0.4291741848,
-0.1460222751,
0.076917395,
0.020686999,
0.0055980086,
0.0343428031,
0.2224044055,
0.5141208768,
0.0484518856,
0.2780652046,
0.0842241794,
0.1572731435,
0.0436355807,
-0.0590578653,
-0.1049679145,
0.1966391653,
0.0795729533,
0.1122420356,
0.0878397971,
-0.0910462886,
-0.1542136073,
0.7406965494,
-0.1039515287,
-0.2170756161,
-0.2388748825,
0.1248540431,
0.179828167,
-0.346788466,
-0.671862483,
-0.1101689115,
-0.074315533,
-0.1182857603,
0.1658358276,
-0.1971793622,
0.3857592642,
-0.1451541483,
-0.1421557665,
-0.4889585376,
0.0479590222,
0.2954271138,
0.2731300294,
0.0242914185,
-0.0154145937,
0.2184960693,
0.0382711925,
0.2569573224,
0.6370871067,
0.0527469441,
-0.0318295136,
-0.3823572993,
0.0200689249,
-0.0201666504,
-0.1323196739,
0.1026280224,
-0.0901867375,
-0.0984988064,
0.0167686529,
0.1813510954,
0.1151837558,
-0.04494619,
0.2071424723,
0.1966821998,
0.2826370299,
-0.0761732459,
0.1413919926,
-0.019568095,
-0.162072137,
-0.1833897531,
-0.2084058821,
-0.0501036793,
0.0226606615,
0.213099435,
-0.2613894343,
-0.0142905004,
-0.2763114572,
0.0691143349,
0.0182072446,
0.4819695055,
0.4473840892,
0.2496089339,
-0.2478420883,
-0.2606365681,
-0.6751897335,
0.070212014,
0.0466932952,
-0.2329176366,
0.0193470344,
-0.0835054144,
-0.1857015193,
0.1238350272,
0.2564578056,
-0.117608279,
0.0093840584,
0.3501384854,
-0.3854681849,
-0.0368202031,
-0.1467264146,
0.4041582644,
0.1732928753,
0.198158294,
0.1174556836,
-0.0660072193,
-0.0030950457,
-0.003050616,
0.102330029,
-0.10300567,
0.1763141453,
0.1233725101,
-0.0009334981,
0.0599504523,
0.0501686782,
-0.0178510919,
-0.6541122198,
-0.2283634245,
-0.466280967,
0.0908215046,
-0.1706440896,
0.0184336752,
-0.2503195405,
-0.5449123979,
-0.0179039575,
0.1748427302,
0.0182508305,
-0.427233845,
-0.3343033195,
0.2882150114,
0.1454262584,
-0.0655942783,
0.2387571633,
-0.0146010928,
-0.0937179178,
0.1880163252,
-0.2161308825,
-0.2531245649,
0.4511773884,
-0.7091345787,
-0.0033164769,
-0.2580905557,
0.4379966855,
0.4351279736,
0.1337573528,
-0.1770674139,
0.1712414622,
0.1312956512,
0.2485462725,
-0.3428422213,
0.1815025061,
-0.2439761311,
-0.0710948482,
0.123007454,
0.6898621917,
0.1534519643,
-0.4521770477,
-0.1873075664,
0.0191909745
] |
https://github.com/huggingface/datasets/issues/1919 | Failure to save with save_to_disk | Hi thanks for reporting and for proposing a fix :)
I just merged a fix, feel free to try it from the master branch ! | When I try to save a dataset locally using the `save_to_disk` method I get the error:
```bash
FileNotFoundError: [Errno 2] No such file or directory: '/content/squad/train/squad-train.arrow'
```
To replicate:
1. Install `datasets` from master
2. Run this code:
```python
from datasets import load_dataset
squad = load_dataset("squad") # or any other dataset
squad.save_to_disk("squad") # error here
```
The problem is that the method is not creating a directory with the name `dataset_path` for saving the dataset in (i.e. it's not creating the *train* and *validation* directories in this case). After creating the directory the problem resolves.
I'll open a PR soon doing that and linking this issue.
| 25 | Failure to save with save_to_disk
When I try to save a dataset locally using the `save_to_disk` method I get the error:
```bash
FileNotFoundError: [Errno 2] No such file or directory: '/content/squad/train/squad-train.arrow'
```
To replicate:
1. Install `datasets` from master
2. Run this code:
```python
from datasets import load_dataset
squad = load_dataset("squad") # or any other dataset
squad.save_to_disk("squad") # error here
```
The problem is that the method is not creating a directory with the name `dataset_path` for saving the dataset in (i.e. it's not creating the *train* and *validation* directories in this case). After creating the directory the problem resolves.
I'll open a PR soon doing that and linking this issue.
Hi thanks for reporting and for proposing a fix :)
I just merged a fix, feel free to try it from the master branch ! | [
-0.0436870642,
0.1429485828,
0.0570115224,
0.1895615757,
0.5194174647,
0.2885245681,
0.1922132075,
0.2388130724,
-0.1313642859,
0.1859050393,
0.141578123,
0.3680595458,
-0.3196107149,
-0.2508215606,
0.0698748976,
0.0474061258,
0.2884761095,
0.0203923956,
-0.0373369195,
-0.0828114748,
-0.1238712445,
0.2648705542,
0.0996690542,
0.0421989411,
-0.5081452727,
-0.0061077178,
-0.1618734896,
0.3145305514,
0.0010721609,
-0.2673491836,
0.1534150541,
-0.3283669353,
0.2090513259,
0.5840063095,
-0.0001187165,
0.0369130224,
-0.038238287,
-0.1583705544,
-0.4442633092,
-0.4077844024,
-0.2219836712,
-0.1730086356,
0.1148862317,
-0.4092870951,
0.070168063,
-0.050697796,
0.2495626062,
-0.1575471312,
0.387182951,
0.2822136283,
0.1549114287,
-0.038205117,
0.1740318835,
-0.2288419604,
-0.1663123071,
0.6161907911,
-0.3467488289,
0.2955403626,
-0.0347838178,
-0.1686116159,
0.3699816763,
0.0192848966,
-0.0258869752,
-0.1304123998,
0.2818757594,
0.1215029731,
0.0884763449,
-0.330726862,
-0.0136279296,
0.0122263245,
0.6147156954,
-0.6311990023,
-0.5074195862,
0.2244338244,
0.0965239108,
0.0193154216,
0.1897400916,
0.249519214,
-0.2263126224,
0.1998586357,
-0.1867232323,
-0.3792485297,
-0.3171934485,
0.2024368048,
0.2337696254,
-0.1623827815,
-0.0914280862,
-0.0537082963,
0.1051191241,
-0.0304780714,
-0.0507368743,
-0.2843352556,
0.08436656,
0.1800297797,
-0.1861408204,
-0.1386528164,
-0.1079161167,
-0.0201206282,
-0.1373758465,
0.120416455,
0.1969715953,
-0.2985675037,
0.0002791397,
0.3721285462,
0.1506848931,
0.0598140843,
0.2832765281,
0.3988431692,
0.3854604661,
0.090272814,
0.0267786607,
-0.0459407493,
-0.0410896577,
-0.1973221004,
0.5454723239,
-0.009781234,
0.3190213144,
-0.1826925576,
-0.2289984375,
0.4254074097,
-0.1825689077,
0.1510067433,
0.0385161228,
0.2409754843,
0.1118041053,
-0.0846430659,
-0.0101933181,
0.4569027424,
0.0640194267,
0.3838764429,
-0.1783627421,
0.1913471967,
-0.0088192895,
0.2253189683,
0.2253584117,
-0.1830711663,
0.184874773,
0.0567195863,
-0.0188270733,
-0.016478397,
-0.0026840717,
-0.2802462876,
0.0282900892,
0.3156392276,
0.0725493208,
0.1089624614,
0.1198394448,
-0.4458801746,
-0.1219691336,
0.2643063962,
-0.0774843991,
-0.4749076068,
-0.2428726405,
0.0788343474,
0.0020113587,
-0.0123703387,
-0.2620708942,
-0.1679549813,
0.2051439881,
-0.1154923886,
-0.0640136525,
0.0571933724,
-0.0576199964,
-0.4168857932,
0.4127008617,
0.36203444,
-0.5066762567,
0.2732729912,
0.0345087163,
-0.1230076402,
0.1597744077,
0.558768332,
-0.0151763707,
0.3559603989,
-0.2971853912,
0.3118731678,
0.5128964186,
-0.4045401514,
-0.2994145155,
0.1477933377,
-0.1420669854,
-0.344843179,
0.0593090765,
0.1312282979,
-0.130954802,
-0.1517181247,
0.037294291,
0.0817089081,
-0.1437608749,
0.0410484821,
-0.4923841357,
0.1228280365,
-0.0063292519,
0.0801652223,
-0.1582184732,
-0.0098181963,
0.2346525043,
0.0741201788,
0.3167812824,
-0.2465213835,
0.1298096329,
0.4385488331,
0.7910141349,
-0.1569510847,
-0.1734232157,
-0.0515779108,
-0.3099276125,
0.1068462431,
0.2395673394,
-0.0529064946,
-0.1526202857,
-0.095318377,
-0.1524759382,
0.0408611149,
-0.4206056893,
0.0266654342,
0.0151802897,
-0.0559187829,
0.1668252498,
0.1911594719,
-0.3386027813,
0.1531254798,
-0.3734539747,
0.0766412169,
0.0540728606,
0.4180465043,
-0.340133369,
-0.3832111955,
-0.2234161347,
0.0831990093,
0.026224032,
-0.0200844109,
-0.1568187177,
0.4635841846,
0.0581359453,
0.1698821187,
0.1769727468,
0.0594830811,
0.1188472435,
-0.1710502356,
-0.0211700089,
0.3229103684,
0.3793407381,
0.0707868636,
-0.6439213753,
-0.0796292275,
-0.0559943169,
0.0869724825,
0.1420306563,
0.0986384749,
0.1159186587,
0.0188671798,
0.0087372437,
-0.0919282958,
0.1190078408,
0.0041112937,
0.1784569472,
-0.2022560686,
-0.3054652214,
-0.053745959,
0.0045894878,
0.0100241676,
0.4323052168,
0.1164100915,
0.1207398027,
-0.0221864469,
-0.0730504766,
0.469450146,
0.539755702,
-0.0449206717,
-0.0467780903,
0.0906324759,
-0.2236267328,
-0.1323502958,
0.1095442921,
-0.1445534825,
0.2041362673,
0.578312993,
0.0744420886,
0.139869988,
-0.2375941277,
-0.0113389567,
-0.0262816921,
0.2607399523,
-0.2932001948,
-0.1521641016,
-0.2273587584,
0.1628910303,
-0.2301878631,
0.0590161905,
-0.0771371871,
-0.2169805169,
0.0288821682,
0.3544382751,
-0.0217196718,
-0.069999069,
0.0079927668,
0.0586091429,
-0.1687672287,
-0.3593933284,
-0.0122329444,
-0.0431216992,
-0.0647559464,
-0.0405166633,
0.2432489395,
-0.0931788832,
0.4074337184,
-0.1362607181,
0.0064518303,
-0.6529040933,
0.041798681,
-0.0146365911,
0.0060269348,
0.3284972608,
0.0468836017,
0.3231472969,
-0.2639583945,
-0.2300034165,
0.3766773939,
0.0304340571,
-0.1758867353,
0.1453617513,
0.1346268356,
-0.310898602,
-0.1181408912,
-0.3630983531,
-0.3233221471,
-0.2766489983,
0.1131915003,
-0.2195639759,
0.1284824908,
-0.0175775718,
0.2915683091,
0.2227850705,
0.1160495728,
-0.0855982453,
-0.0912681893,
-0.3929235935,
0.6367708445,
-0.3100696504,
-0.4007938504,
0.1726481467,
0.1183609068,
0.0293653831,
-0.0310871843,
-0.3465647101,
-0.3220783174,
0.0218670983,
0.1416010559,
-0.1745147407,
0.2809892595,
0.3479141593,
-0.0128051229,
0.02595146,
0.0195511132,
-0.3645201325,
0.1986775994,
0.5634405017,
0.5095307827,
-0.274174422,
0.4664590657,
0.0063155219,
0.5923214555,
0.1623980701,
0.2559559941,
0.4696495235,
0.0460420996,
0.3521331251,
-0.2076815665,
-0.3169224262,
0.1021851897,
-0.1217057109,
-0.5222662091,
-0.171311304,
0.1042050272,
0.040589489,
-0.0762660429,
-0.0276086703,
-0.1950878501,
-0.1804352701,
-0.0646871775,
0.1493257582,
0.2803892493,
-0.1482632309,
0.1099717915,
0.2512289286,
-0.0083693415,
0.0718715191,
0.6023398042,
0.3001722395,
0.0348980092,
-0.5934092999,
-0.3321790397,
-0.3693141043,
0.0797098354,
0.0514099784,
0.3805754185,
-0.2038464546,
-0.0741525069,
-0.0068886988,
0.0717541203,
0.4177289009,
-0.2680717111,
-0.2005079538,
0.0655287281,
0.0476388112,
-0.444698602,
-0.1667145789,
0.3305512071,
0.349088937,
-0.1856459677,
0.4512583613,
-0.0317449495,
-0.1143832058,
-0.0933342129,
0.2862457335,
-0.1979822218,
-0.2021403015,
-0.1551750451,
-0.2627655864,
-0.353759706,
-0.113311097,
-0.002702415,
0.0357786641,
0.3868083954,
0.0661629513,
0.0273660403,
0.0027788579,
0.1347557902,
-0.173217088,
0.4466360211,
-0.2300790846,
0.1627963781,
-0.0883872956,
0.0243432634,
0.3703675568,
0.3753552735,
-0.2585773766,
-0.3253381252,
0.0951187909,
-0.2750014365,
0.4798432887,
0.1903962642,
-0.1140700355,
-0.4817686975,
-0.0603417158,
-0.0389530733,
-0.228341639,
-0.2157604098,
0.2810062468,
-0.1916753948,
-0.1766626686,
-0.001936242,
0.1306397319,
0.2120263577,
0.2215736657,
0.1068675444,
0.0158492252,
-0.0663998425,
0.33100757,
0.1926986724,
0.8347560167,
-0.021093078,
0.3929270804,
0.1046826541,
-0.2786883712,
0.1136673465,
-0.1644261926,
0.174298197,
-0.5152050257,
0.0956637338,
-0.0492585413,
-0.0825592428,
0.2677946091,
0.256013304,
-0.2164880931,
0.395359695,
-0.3980212808,
0.1224346757,
-0.187114507,
0.0445940197,
-0.294190228,
-0.2515768111,
-0.2279292792,
0.1340619475,
0.0702631548,
0.1735771149,
-0.039029263,
-0.0485875532,
-0.3068463206,
-0.0457440801,
-0.2921006083,
0.2171728611,
-0.2839551568,
0.4295625091,
-0.1510761976,
-0.5352058411,
-0.0205071531,
0.7050193548,
0.455648303,
0.0605425909,
-0.2067080736,
-0.0671711117,
-0.15872401,
0.1630186886,
-0.0978154615,
-0.1728441268,
0.3130176365,
-0.0761650652,
-0.0021596998,
-0.0566106662,
-0.1997147053,
-0.2288072705,
-0.1366556734,
-0.0496363938,
0.1111470535,
-0.1250344366,
-0.5298015475,
0.0124669485,
-0.1250457019,
-0.1918316036,
0.0934786573,
0.1643522382,
-0.117880933,
0.374989152,
-0.3534037471,
0.0724099576,
0.1048420668,
0.4314040542,
-0.3103459179,
0.0340767205,
0.4716997147,
-0.1298159361,
0.0600450262,
-0.3047809601,
0.4256793857,
0.0011188164,
-0.1358043402,
0.144616276,
-0.2254028618,
-0.012412861,
-0.0816781521,
-0.0176234022,
0.3689906597,
-0.255494833,
-0.0857539251,
-0.2979806066,
-0.5189902186,
-0.0992816687,
0.4103381038,
0.1895849109,
-0.0605893992,
-0.124457784,
0.1943559945,
-0.275088191,
-0.2191738188,
-0.0253717657,
-0.3023045361,
0.0205047913,
0.0584354289,
0.0555096269,
0.3383272588,
-0.3429076076,
0.0912238434,
0.016808629,
-0.1370320171,
-0.16761145,
-0.038636826,
0.1491855532,
0.1423886865,
0.1044515669,
0.0018481016,
-0.3341523409,
-0.0731880516,
-0.0511821173,
-0.274012506,
0.477329731,
0.0286542866,
0.0739973113,
-0.0087247733,
0.0509583689,
0.1287954301,
-0.1969233602,
0.0527655929,
0.2355608642,
0.0527852029,
-0.1015890315,
0.1415499896,
-0.2127240151,
-0.324586153,
-0.0788837373,
-0.1340296268,
0.169349879,
-0.0163829997,
-0.3777160645,
-0.1813932061,
0.0101543367,
0.4741106331,
0.3265454471,
-0.311891526,
0.2228897512,
0.3820927143,
0.1247720718,
0.0235109925,
-0.2684885561,
0.2837251723,
-0.1158550158,
-0.2796922922,
0.0808056518,
-0.0752430484,
0.0504001379,
-0.4147300124,
0.0888022855,
0.3177280128,
-0.3490394354,
0.1107111126,
0.1824828386,
-0.0235075355,
0.3092254996,
-0.1106090695,
-0.1716780663,
0.1599219888,
0.2186937332,
-0.1734765768,
0.2385257781,
0.301281184,
0.149350211,
0.2953529656,
-0.0135152563,
-0.0069186855,
-0.4171994627,
-0.0982949436,
-0.0134434663,
-0.0001410581,
0.3464177251,
-0.4165669978,
-0.2585894465,
-0.171969682,
0.0722390115,
-0.266956687,
0.0741029829,
-0.0763529465,
-0.1190555319,
0.2702915072,
-0.0594436862,
-0.1385220587,
-0.1295193732,
-0.0523250178,
0.1955132633,
0.1951882541,
-0.0229410678,
0.1453986019,
0.2591454387,
0.0271720774,
-0.196170181,
-0.0074243871,
0.0663922951,
-0.3078883588,
-0.39982149,
0.3634174764,
0.4397543371,
0.553585887,
-0.0218441579,
0.2691556513,
-0.2226391733,
0.0166127831,
0.2642593086,
0.0386485569,
-0.1029786021,
-0.0857439041,
0.422095269,
0.0959093273,
0.0597939678,
0.2503316998,
-0.0150831193,
0.1046725512,
-0.0081346892,
0.2377558351,
-0.0240384303,
-0.2090129405,
-0.0002202904,
0.0854361057,
-0.2198100388,
-0.0536928847,
0.5168392658,
0.0682666153,
0.1285762787,
-0.1078775525,
0.0383698568,
-0.0365348086,
0.3760025501,
0.2762748301,
-0.2577252984,
-0.5194138288,
0.0330716968,
-0.620077908,
-0.1018304676,
0.0556328446,
0.2281338274,
-0.2808685601,
0.2638736367,
-0.1216256842,
0.2111624479,
0.1909234822,
0.1290504187,
-0.1083806008,
0.5609633327,
-0.1707861722,
-0.0435361266,
0.0092587005,
0.0681425035,
0.1391111165,
-0.3969308734,
0.1743019968,
0.1783903539,
-0.0282298513,
0.0432104096,
-0.0229311213,
-0.2206308693,
0.3385378718,
0.4254107475,
-0.0666353628,
0.1587460637,
0.0038033426,
-0.3034678698,
0.1541689187,
0.0924080461,
-0.0381419994,
0.1844717711,
-0.0478864796,
0.5128149986,
-0.3749786913,
-0.0932085514,
-0.4210632443,
-0.0058485344,
-0.0919634476,
-0.0159094092,
-0.0792491212,
0.0060057491,
0.1355395317,
0.1587671936,
0.2504426539,
-0.0072605144,
-0.199528113,
0.3668052554,
-0.3987041116,
-0.2346241176,
0.576382637,
0.0544466488,
-0.0320315994,
-0.0823911503,
0.2976130545,
0.2576942146,
0.1391413659,
-0.3864375055,
-0.2610428035,
0.2535919845,
-0.0958448797,
-0.2886846066,
0.0853897929,
-0.1614725292,
0.1608715355,
0.1543490738,
0.3828742206,
-0.0747899562,
-0.0117753521,
0.151154086,
-0.334856987
] |
https://github.com/huggingface/datasets/issues/1915 | Unable to download `wiki_dpr` | Thanks for reporting ! This is a bug. For now feel free to set `ignore_verifications=False` in `load_dataset`.
I'm working on a fix | I am trying to download the `wiki_dpr` dataset. Specifically, I want to download `psgs_w100.multiset.no_index` with no embeddings/no index. In order to do so, I ran:
`curr_dataset = load_dataset("wiki_dpr", embeddings_name="multiset", index_name="no_index")`
However, I got the following error:
`datasets.utils.info_utils.UnexpectedDownloadedFile: {'embeddings_index'}`
I tried adding in flags `with_embeddings=False` and `with_index=False`:
`curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False, embeddings_name="multiset", index_name="no_index")`
But I got the following error:
`raise ExpectedMoreDownloadedFiles(str(set(expected_checksums) - set(recorded_checksums)))
datasets.utils.info_utils.ExpectedMoreDownloadedFiles: {‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_5’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_15’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_30’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_36’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_18’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_41’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_13’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_48’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_10’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_23’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_14’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_34’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_43’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_40’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_47’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_3’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_24’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_7’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_33’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_46’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_42’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_27’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_29’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_26’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_22’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_4’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_20’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_39’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_6’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_16’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_8’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_35’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_49’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_17’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_25’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_0’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_38’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_12’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_44’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_1’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_32’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_19’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_31’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_37’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_9’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_11’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_21’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_28’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_45’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_2’}`
Is there anything else I need to set to download the dataset?
**UPDATE**: just running `curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False)` gives me the same error.
| 22 | Unable to download `wiki_dpr`
I am trying to download the `wiki_dpr` dataset. Specifically, I want to download `psgs_w100.multiset.no_index` with no embeddings/no index. In order to do so, I ran:
`curr_dataset = load_dataset("wiki_dpr", embeddings_name="multiset", index_name="no_index")`
However, I got the following error:
`datasets.utils.info_utils.UnexpectedDownloadedFile: {'embeddings_index'}`
I tried adding in flags `with_embeddings=False` and `with_index=False`:
`curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False, embeddings_name="multiset", index_name="no_index")`
But I got the following error:
`raise ExpectedMoreDownloadedFiles(str(set(expected_checksums) - set(recorded_checksums)))
datasets.utils.info_utils.ExpectedMoreDownloadedFiles: {‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_5’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_15’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_30’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_36’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_18’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_41’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_13’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_48’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_10’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_23’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_14’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_34’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_43’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_40’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_47’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_3’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_24’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_7’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_33’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_46’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_42’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_27’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_29’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_26’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_22’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_4’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_20’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_39’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_6’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_16’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_8’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_35’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_49’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_17’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_25’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_0’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_38’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_12’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_44’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_1’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_32’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_19’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_31’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_37’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_9’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_11’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_21’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_28’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_45’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_2’}`
Is there anything else I need to set to download the dataset?
**UPDATE**: just running `curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False)` gives me the same error.
Thanks for reporting ! This is a bug. For now feel free to set `ignore_verifications=False` in `load_dataset`.
I'm working on a fix | [
-0.3591079414,
-0.3630982935,
-0.0504426472,
0.3148261905,
0.3717033267,
0.4270487726,
0.3198951483,
0.0625340343,
0.2385216802,
0.1804990768,
0.1647018045,
-0.0206089392,
0.0954112411,
-0.0210909098,
-0.1673092246,
-0.0334007256,
0.1001308262,
0.1070558727,
0.3710586429,
-0.0622111484,
-0.3295958042,
0.2548210621,
-0.3576647043,
-0.2318953425,
-0.1536285579,
-0.1435201168,
-0.0590854362,
-0.0766760409,
-0.4447589815,
-0.4724355936,
0.5257073045,
0.2139313072,
0.4566455781,
0.2094179988,
-0.0001233339,
-0.0081855133,
0.2570952177,
-0.2760653794,
-0.226743862,
-0.1994098425,
-0.5099616051,
-0.2134324908,
0.1228157058,
-0.2729626596,
0.2952132821,
-0.2128576636,
-0.0749511495,
-0.092896156,
0.0745316744,
0.1510063261,
0.1449033916,
-0.0625399053,
0.1508242041,
-0.2355566323,
0.0945448875,
-0.1769308448,
-0.1318903863,
0.1710798144,
-0.0272566825,
0.1341178566,
0.269108206,
0.0560871139,
-0.2518815994,
0.227773726,
0.1170532182,
0.0164905638,
0.2614759803,
-0.3041638136,
0.4016200304,
0.5526137948,
0.9237311482,
0.1302186549,
-0.2386935353,
0.0382867754,
0.2114564329,
-0.375774622,
0.1635374427,
0.2258235663,
-0.2530865669,
0.2006118447,
-0.2965895236,
-0.0683534294,
-0.0212515667,
0.3237016201,
-0.0989622921,
0.3529933095,
0.2022208273,
0.1463738531,
0.1856655478,
-0.203629002,
0.0743473396,
-0.1115097106,
-0.2810246944,
0.0874458253,
-0.2282610536,
0.0073460862,
-0.1403036565,
0.0571534336,
0.2315741777,
0.3335158825,
0.0137926154,
0.1705389917,
0.4314413369,
0.10148298,
0.2653805017,
0.2508426309,
-0.1195530742,
0.2461989522,
0.3145948946,
0.547288537,
0.2549657822,
-0.0376865119,
-0.0095996,
-0.0076848306,
-0.0618709475,
-0.110142082,
-0.0867578536,
-0.3650256097,
-0.2998701334,
0.1207371354,
-0.2721976936,
-0.185774222,
-0.2455057651,
0.39892748,
0.0512361489,
0.4941690564,
0.1345020682,
0.1692252308,
-0.0389334336,
0.0455489345,
-0.096598044,
-0.0991688371,
-0.1295962036,
0.0827542767,
0.3832921386,
-0.3321831822,
0.3954803646,
0.035917908,
0.0670720339,
0.08984752,
-0.0672727302,
-0.0200422853,
0.2051385939,
0.3828621209,
0.0285898,
0.2937914729,
0.1074254736,
-0.2261043191,
-0.024616383,
0.277497381,
-0.1818650514,
-0.433575809,
-0.0587908402,
0.1249007732,
-0.2760755122,
-0.0717420876,
-0.3263550103,
0.1071711332,
0.0178033188,
-0.1447196901,
0.1583663076,
0.0138513502,
-0.2004510313,
-0.3192542791,
0.3162531257,
0.5299693346,
-0.5378867388,
-0.0565232821,
-0.0876787901,
-0.2763262391,
0.1355167627,
0.0153359957,
-0.2887879312,
-0.0896553546,
-0.3782500029,
0.1199999154,
0.130803287,
-0.4048188627,
-0.5335977077,
0.4260770977,
0.016574502,
0.1425586641,
0.3088408709,
0.0687620193,
0.2286394984,
0.0907654539,
-0.1096129864,
0.3609176874,
0.1335098743,
-0.2713666856,
-0.0036886036,
-0.3318435848,
0.380472064,
0.064660266,
-0.0475531593,
-0.0033563338,
0.1183110625,
0.1810514927,
0.4451653361,
0.0042786193,
0.0312241688,
0.2706511915,
0.1180465519,
0.0901453122,
0.13392739,
0.1229389757,
-0.5137768388,
0.2290320694,
0.1504508853,
0.0457542911,
-0.3320195377,
-0.0949367508,
-0.3201953173,
-0.087365672,
-0.3744760454,
-0.2137420475,
-0.0082012005,
0.1704930663,
-0.0325005651,
-0.1602597535,
-0.0984632373,
0.0415373668,
-0.0527912676,
0.0277334712,
-0.1782588065,
0.146972999,
-0.137459591,
0.0143732429,
0.1123540476,
-0.0683239102,
0.3189949989,
-0.050568413,
0.0679064021,
0.3047975302,
0.2043270767,
-0.0166686773,
-0.0066900766,
-0.0056712329,
0.2587312758,
-0.5035477877,
-0.0205894373,
0.3803390861,
0.1576838195,
-0.0605142713,
-0.2632325292,
0.2244814038,
-0.0290870685,
0.3685745597,
-0.2061022669,
0.0416201353,
0.3276458085,
-0.1760828942,
0.228708595,
-0.3677020669,
0.2900663614,
0.5791851878,
-0.2493052185,
-0.1534954458,
-0.0482501239,
0.0391809978,
0.1322935224,
-0.208743751,
-0.0871027857,
0.0236713905,
0.0552036837,
-0.048158396,
0.1448452771,
0.0600990802,
0.3681732714,
0.2472549081,
0.0936578065,
-0.1286987364,
0.1636076868,
-0.3218725324,
0.2134934366,
-0.0119465441,
-0.1579785943,
0.5060125589,
-0.082175523,
-0.0530169122,
-0.2347955704,
-0.033047948,
0.2623146176,
0.2831532657,
-0.3046385944,
-0.1779408455,
-0.6171993613,
0.0053768903,
0.0949760377,
-0.1705057621,
-0.203748554,
-0.2974212766,
0.1051950753,
-0.0748012289,
0.124833107,
0.1406848431,
-0.2971716523,
-0.1109370887,
-0.1286423355,
-0.2025417238,
-0.0407080203,
-0.0536280684,
-0.3318493366,
-0.0249450151,
0.2382587045,
0.3014805615,
0.2516849339,
0.0587189272,
-0.3458539546,
-0.3976095021,
-0.0931583941,
0.0518293791,
-0.1394763589,
0.0614192933,
0.3366080225,
0.4813328087,
-0.052812878,
-0.3433488011,
0.0972989798,
-0.0614316538,
-0.0334287211,
-0.0621373057,
-0.0878676474,
-0.0453505032,
-0.1144122258,
-0.6167298555,
-0.133022368,
-0.313521862,
-0.0966890827,
0.0850570723,
-0.0683745742,
-0.0727717727,
0.1985065937,
-0.1263793409,
0.2616435885,
-0.0850996524,
-0.1912422478,
-0.1607694626,
0.5849160552,
-0.1287933737,
-0.3757937253,
0.5222774148,
0.0111421421,
-0.0262259077,
0.3314106464,
-0.5823149681,
0.1069681868,
0.1025797948,
0.2082472742,
0.0025483835,
0.231055811,
0.3058458567,
0.0286072977,
-0.0586488955,
-0.0875663236,
-0.093997106,
-0.0564807728,
-0.2831984162,
0.5092186332,
-0.0105563672,
0.2368226349,
-0.1851424873,
0.6094354391,
0.3000097275,
0.1178232208,
0.3601542711,
-0.035562247,
0.1675812006,
0.1629152298,
-0.1403384805,
0.0368614867,
-0.1783125848,
0.3249067962,
0.2848486304,
-0.1618456542,
-0.33210361,
-0.3920762241,
0.1011416763,
-0.0891899988,
-0.0727448836,
-0.0491276383,
0.3427486122,
0.1343056709,
-0.1356297135,
0.3029321432,
0.0821177959,
-0.3334835768,
0.2686664164,
0.2534909546,
-0.1429877579,
0.1484564394,
0.1589110494,
-0.0379804298,
-0.3448108435,
0.3444148898,
0.1638857573,
0.3087618649,
0.0045744032,
0.1655190736,
0.1895902455,
-0.0242544524,
0.5605655909,
-0.4350597858,
0.3597973287,
0.1688044518,
0.0517792106,
-0.2673885524,
-0.1922028363,
-0.2611429691,
-0.276124835,
-0.1781180203,
0.2215981781,
-0.6027705669,
0.0161780082,
0.1669196934,
0.161229074,
-0.1238547564,
-0.359395504,
-0.3704820871,
-0.4479211867,
-0.4778937697,
-0.1964291334,
0.0075106695,
0.3619407713,
-0.2136184573,
-0.0925563499,
-0.0091365296,
0.1171009243,
-0.001581084,
-0.0716376677,
0.2431966662,
0.4718741179,
-0.0696358308,
0.2932137847,
0.1787507832,
0.5283370614,
0.6097087264,
-0.0627237409,
-0.1071097106,
0.0484472141,
-0.2018222213,
0.0192039013,
0.2239027917,
-0.0920594856,
0.1804145575,
0.1474186182,
0.0573407821,
-0.0472669601,
-0.2881350815,
0.3782991767,
-0.255780071,
-0.3657978773,
-0.4157607555,
0.466252774,
0.04135032,
0.1092936993,
0.1936930716,
0.2587583363,
-0.1668643057,
0.0226926319,
0.3577559292,
1.0175698996,
0.1065523326,
-0.0188474897,
0.2060740292,
0.014623858,
0.2137940526,
-0.2915169597,
0.2040701807,
-0.0980691835,
-0.4103772044,
-0.2600044012,
0.0024991035,
0.1532847583,
0.3295855224,
-0.4880577624,
0.3298540115,
-0.2777177989,
0.1029104441,
-0.0117820948,
0.2915962934,
-0.220928371,
-0.1896231622,
-0.2440837324,
0.0250633508,
0.1062117964,
0.1561801732,
-0.1630842984,
-0.0633790493,
0.0575041473,
-0.3713616133,
-0.4828014374,
0.058330439,
-0.2181004137,
0.4266484678,
-0.3948786557,
-0.2592288852,
0.2925424278,
0.4851697087,
0.1716977954,
-0.0103368107,
-0.3101625443,
0.2248513997,
-0.1228220761,
-0.4960903227,
-0.2240564525,
0.0299590044,
0.0838921517,
-0.0808787942,
-0.5226160884,
0.1020057946,
-0.037109822,
-0.0976727679,
0.0695047081,
-0.0406801514,
-0.148755461,
-0.2177063078,
-0.0691506788,
-0.0016988479,
0.1558316946,
-0.0840197876,
0.036230851,
-0.0019893423,
-0.1585761607,
0.1211211979,
0.1971922666,
-0.1881342381,
-0.166673243,
0.6800677776,
0.021305263,
-0.2687132359,
0.6212596893,
0.2700417042,
-0.3788089454,
-0.2264260501,
-0.0876421928,
0.1271990985,
-0.2000418007,
0.0818252265,
0.3339782655,
0.1755504608,
0.125036478,
0.4918980002,
0.2114367187,
0.0397879034,
-0.0355798341,
-0.5814547539,
-0.0940761864,
0.0889580622,
-0.080797337,
0.1880506128,
-0.469819963,
-0.3154391944,
0.2788206339,
0.0934569985,
-0.1850285977,
0.0440965667,
-0.4358551502,
-0.080707252,
0.1609925032,
0.023494957,
0.1360042542,
0.1081748679,
-0.0183175094,
0.0517950989,
-0.0362447537,
-0.0480694138,
-0.0843036026,
0.2136484981,
0.0365363732,
-0.1229961067,
0.0750055611,
-0.4660218358,
-0.3347611129,
0.0175155215,
0.2713813186,
0.3464637399,
0.0112376744,
0.0264790431,
0.3368586898,
0.0637378842,
-0.5087183118,
0.0850757807,
0.1238570586,
0.3832838833,
0.3522446454,
-0.0603413992,
0.0815019161,
0.0552271903,
-0.0290250219,
0.0364019498,
0.1451058388,
-0.0185585245,
0.4161346853,
-0.2846485972,
-0.0397412106,
0.3598662615,
0.0802498609,
0.4701504707,
-0.1421667784,
0.1650277972,
0.1924475133,
0.0728163719,
-0.4741269946,
-0.2790477276,
0.322547853,
0.088057518,
-0.1475119889,
0.0822518319,
0.197687462,
-0.5292576551,
-0.198570475,
0.0116610527,
0.2377843559,
-0.2428720593,
0.1522593796,
0.5658451319,
-0.0603008866,
0.0266814306,
-0.1705100983,
0.1049015224,
0.3029582798,
-0.0226077046,
-0.3302283585,
0.323898077,
-0.051720798,
0.2576179802,
-0.2167515457,
-0.581649363,
0.1976688504,
0.2486559749,
0.109131664,
-0.06238731,
0.1428007483,
0.28139171,
0.1161990166,
0.1848137528,
-0.5019871593,
0.1146366671,
-0.0847101361,
-0.0868119448,
-0.1550210714,
-0.1631115526,
-0.3408387601,
0.3697254658,
0.2506074011,
-0.3877317607,
0.1391343772,
0.2234000564,
-0.3519897461,
-0.3351709247,
-0.0046413606,
0.0475754514,
0.1984974742,
-0.2501713037,
-0.0406994969,
0.4264983833,
-0.0355310887,
0.5507057309,
0.115885891,
0.2798181176,
0.1993607283,
-0.0190245025,
0.0986390859,
0.1593253464,
0.0265294835,
0.0264791325,
0.3988072276,
0.184543103,
0.1278179884,
0.1725085229,
-0.0023973547,
-0.1505350322,
-0.3022057712,
0.1991574913,
0.1582883149,
0.1563820541,
-0.0906075686,
0.0002231766,
0.119393006,
-0.1225256249,
-0.0520094819,
-0.2637984753,
0.2559805214,
0.2613161802,
0.096773684,
0.099728629,
0.1218368262,
-0.0077643916,
-0.2113357931,
0.489398241,
0.3439131677,
-0.0329996459,
-0.3170143664,
-0.1657277048,
-0.8635806441,
0.1909137666,
-0.0018570956,
-0.198791638,
0.2767448723,
0.1448273957,
0.0629289076,
0.1501016766,
0.0517332554,
0.0493550561,
0.355802536,
0.5278561711,
-0.0388546437,
-0.3366439342,
0.057761386,
0.0212413315,
-0.2305416763,
-0.2928698659,
0.0717011541,
-0.0456008241,
-0.1349541247,
-0.1366090178,
-0.0527432151,
-0.1673204005,
0.2791331112,
0.2438107878,
0.0376396701,
0.6168966293,
-0.0421946384,
0.03400518,
-0.216089204,
0.0339433402,
-0.1781508923,
0.3692154288,
0.1981419027,
0.0402885601,
-0.2612345219,
0.1893489659,
-0.255158186,
0.1336953044,
0.1048163399,
0.3364699483,
-0.4344372451,
-0.0075038709,
-0.0616459623,
0.0245792456,
0.1752415001,
-0.1018627882,
-0.1526125968,
-0.0320740268,
-0.1232205704,
-0.3546846509,
0.6694047451,
-0.4451084733,
-0.1035619229,
-0.0715577826,
0.1026606113,
-0.2589514852,
0.1466145664,
-0.782880187,
0.2994580865,
0.2722882032,
-0.0191891473,
-0.5205339789,
0.0897957683,
-0.1968085617,
0.0039333105,
-0.1593838036,
0.1601533145,
0.1927387714,
-0.2259235382,
-0.1817130744,
-0.2133520842
] |
https://github.com/huggingface/datasets/issues/1915 | Unable to download `wiki_dpr` | I just merged a fix :)
We'll do a patch release soon. In the meantime feel free to try it from the master branch
Thanks again for reporting ! | I am trying to download the `wiki_dpr` dataset. Specifically, I want to download `psgs_w100.multiset.no_index` with no embeddings/no index. In order to do so, I ran:
`curr_dataset = load_dataset("wiki_dpr", embeddings_name="multiset", index_name="no_index")`
However, I got the following error:
`datasets.utils.info_utils.UnexpectedDownloadedFile: {'embeddings_index'}`
I tried adding in flags `with_embeddings=False` and `with_index=False`:
`curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False, embeddings_name="multiset", index_name="no_index")`
But I got the following error:
`raise ExpectedMoreDownloadedFiles(str(set(expected_checksums) - set(recorded_checksums)))
datasets.utils.info_utils.ExpectedMoreDownloadedFiles: {‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_5’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_15’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_30’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_36’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_18’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_41’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_13’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_48’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_10’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_23’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_14’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_34’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_43’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_40’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_47’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_3’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_24’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_7’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_33’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_46’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_42’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_27’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_29’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_26’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_22’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_4’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_20’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_39’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_6’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_16’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_8’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_35’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_49’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_17’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_25’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_0’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_38’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_12’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_44’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_1’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_32’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_19’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_31’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_37’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_9’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_11’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_21’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_28’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_45’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_2’}`
Is there anything else I need to set to download the dataset?
**UPDATE**: just running `curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False)` gives me the same error.
| 29 | Unable to download `wiki_dpr`
I am trying to download the `wiki_dpr` dataset. Specifically, I want to download `psgs_w100.multiset.no_index` with no embeddings/no index. In order to do so, I ran:
`curr_dataset = load_dataset("wiki_dpr", embeddings_name="multiset", index_name="no_index")`
However, I got the following error:
`datasets.utils.info_utils.UnexpectedDownloadedFile: {'embeddings_index'}`
I tried adding in flags `with_embeddings=False` and `with_index=False`:
`curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False, embeddings_name="multiset", index_name="no_index")`
But I got the following error:
`raise ExpectedMoreDownloadedFiles(str(set(expected_checksums) - set(recorded_checksums)))
datasets.utils.info_utils.ExpectedMoreDownloadedFiles: {‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_5’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_15’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_30’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_36’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_18’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_41’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_13’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_48’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_10’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_23’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_14’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_34’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_43’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_40’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_47’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_3’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_24’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_7’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_33’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_46’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_42’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_27’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_29’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_26’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_22’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_4’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_20’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_39’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_6’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_16’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_8’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_35’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_49’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_17’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_25’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_0’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_38’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_12’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_44’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_1’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_32’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_19’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_31’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_37’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_9’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_11’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_21’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_28’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_45’, ‘https://dl.fbaipublicfiles.com/rag/rag_multiset_embeddings/wiki_passages_2’}`
Is there anything else I need to set to download the dataset?
**UPDATE**: just running `curr_dataset = load_dataset("wiki_dpr", with_embeddings=False, with_index=False)` gives me the same error.
I just merged a fix :)
We'll do a patch release soon. In the meantime feel free to try it from the master branch
Thanks again for reporting ! | [
-0.3591079414,
-0.3630982935,
-0.0504426472,
0.3148261905,
0.3717033267,
0.4270487726,
0.3198951483,
0.0625340343,
0.2385216802,
0.1804990768,
0.1647018045,
-0.0206089392,
0.0954112411,
-0.0210909098,
-0.1673092246,
-0.0334007256,
0.1001308262,
0.1070558727,
0.3710586429,
-0.0622111484,
-0.3295958042,
0.2548210621,
-0.3576647043,
-0.2318953425,
-0.1536285579,
-0.1435201168,
-0.0590854362,
-0.0766760409,
-0.4447589815,
-0.4724355936,
0.5257073045,
0.2139313072,
0.4566455781,
0.2094179988,
-0.0001233339,
-0.0081855133,
0.2570952177,
-0.2760653794,
-0.226743862,
-0.1994098425,
-0.5099616051,
-0.2134324908,
0.1228157058,
-0.2729626596,
0.2952132821,
-0.2128576636,
-0.0749511495,
-0.092896156,
0.0745316744,
0.1510063261,
0.1449033916,
-0.0625399053,
0.1508242041,
-0.2355566323,
0.0945448875,
-0.1769308448,
-0.1318903863,
0.1710798144,
-0.0272566825,
0.1341178566,
0.269108206,
0.0560871139,
-0.2518815994,
0.227773726,
0.1170532182,
0.0164905638,
0.2614759803,
-0.3041638136,
0.4016200304,
0.5526137948,
0.9237311482,
0.1302186549,
-0.2386935353,
0.0382867754,
0.2114564329,
-0.375774622,
0.1635374427,
0.2258235663,
-0.2530865669,
0.2006118447,
-0.2965895236,
-0.0683534294,
-0.0212515667,
0.3237016201,
-0.0989622921,
0.3529933095,
0.2022208273,
0.1463738531,
0.1856655478,
-0.203629002,
0.0743473396,
-0.1115097106,
-0.2810246944,
0.0874458253,
-0.2282610536,
0.0073460862,
-0.1403036565,
0.0571534336,
0.2315741777,
0.3335158825,
0.0137926154,
0.1705389917,
0.4314413369,
0.10148298,
0.2653805017,
0.2508426309,
-0.1195530742,
0.2461989522,
0.3145948946,
0.547288537,
0.2549657822,
-0.0376865119,
-0.0095996,
-0.0076848306,
-0.0618709475,
-0.110142082,
-0.0867578536,
-0.3650256097,
-0.2998701334,
0.1207371354,
-0.2721976936,
-0.185774222,
-0.2455057651,
0.39892748,
0.0512361489,
0.4941690564,
0.1345020682,
0.1692252308,
-0.0389334336,
0.0455489345,
-0.096598044,
-0.0991688371,
-0.1295962036,
0.0827542767,
0.3832921386,
-0.3321831822,
0.3954803646,
0.035917908,
0.0670720339,
0.08984752,
-0.0672727302,
-0.0200422853,
0.2051385939,
0.3828621209,
0.0285898,
0.2937914729,
0.1074254736,
-0.2261043191,
-0.024616383,
0.277497381,
-0.1818650514,
-0.433575809,
-0.0587908402,
0.1249007732,
-0.2760755122,
-0.0717420876,
-0.3263550103,
0.1071711332,
0.0178033188,
-0.1447196901,
0.1583663076,
0.0138513502,
-0.2004510313,
-0.3192542791,
0.3162531257,
0.5299693346,
-0.5378867388,
-0.0565232821,
-0.0876787901,
-0.2763262391,
0.1355167627,
0.0153359957,
-0.2887879312,
-0.0896553546,
-0.3782500029,
0.1199999154,
0.130803287,
-0.4048188627,
-0.5335977077,
0.4260770977,
0.016574502,
0.1425586641,
0.3088408709,
0.0687620193,
0.2286394984,
0.0907654539,
-0.1096129864,
0.3609176874,
0.1335098743,
-0.2713666856,
-0.0036886036,
-0.3318435848,
0.380472064,
0.064660266,
-0.0475531593,
-0.0033563338,
0.1183110625,
0.1810514927,
0.4451653361,
0.0042786193,
0.0312241688,
0.2706511915,
0.1180465519,
0.0901453122,
0.13392739,
0.1229389757,
-0.5137768388,
0.2290320694,
0.1504508853,
0.0457542911,
-0.3320195377,
-0.0949367508,
-0.3201953173,
-0.087365672,
-0.3744760454,
-0.2137420475,
-0.0082012005,
0.1704930663,
-0.0325005651,
-0.1602597535,
-0.0984632373,
0.0415373668,
-0.0527912676,
0.0277334712,
-0.1782588065,
0.146972999,
-0.137459591,
0.0143732429,
0.1123540476,
-0.0683239102,
0.3189949989,
-0.050568413,
0.0679064021,
0.3047975302,
0.2043270767,
-0.0166686773,
-0.0066900766,
-0.0056712329,
0.2587312758,
-0.5035477877,
-0.0205894373,
0.3803390861,
0.1576838195,
-0.0605142713,
-0.2632325292,
0.2244814038,
-0.0290870685,
0.3685745597,
-0.2061022669,
0.0416201353,
0.3276458085,
-0.1760828942,
0.228708595,
-0.3677020669,
0.2900663614,
0.5791851878,
-0.2493052185,
-0.1534954458,
-0.0482501239,
0.0391809978,
0.1322935224,
-0.208743751,
-0.0871027857,
0.0236713905,
0.0552036837,
-0.048158396,
0.1448452771,
0.0600990802,
0.3681732714,
0.2472549081,
0.0936578065,
-0.1286987364,
0.1636076868,
-0.3218725324,
0.2134934366,
-0.0119465441,
-0.1579785943,
0.5060125589,
-0.082175523,
-0.0530169122,
-0.2347955704,
-0.033047948,
0.2623146176,
0.2831532657,
-0.3046385944,
-0.1779408455,
-0.6171993613,
0.0053768903,
0.0949760377,
-0.1705057621,
-0.203748554,
-0.2974212766,
0.1051950753,
-0.0748012289,
0.124833107,
0.1406848431,
-0.2971716523,
-0.1109370887,
-0.1286423355,
-0.2025417238,
-0.0407080203,
-0.0536280684,
-0.3318493366,
-0.0249450151,
0.2382587045,
0.3014805615,
0.2516849339,
0.0587189272,
-0.3458539546,
-0.3976095021,
-0.0931583941,
0.0518293791,
-0.1394763589,
0.0614192933,
0.3366080225,
0.4813328087,
-0.052812878,
-0.3433488011,
0.0972989798,
-0.0614316538,
-0.0334287211,
-0.0621373057,
-0.0878676474,
-0.0453505032,
-0.1144122258,
-0.6167298555,
-0.133022368,
-0.313521862,
-0.0966890827,
0.0850570723,
-0.0683745742,
-0.0727717727,
0.1985065937,
-0.1263793409,
0.2616435885,
-0.0850996524,
-0.1912422478,
-0.1607694626,
0.5849160552,
-0.1287933737,
-0.3757937253,
0.5222774148,
0.0111421421,
-0.0262259077,
0.3314106464,
-0.5823149681,
0.1069681868,
0.1025797948,
0.2082472742,
0.0025483835,
0.231055811,
0.3058458567,
0.0286072977,
-0.0586488955,
-0.0875663236,
-0.093997106,
-0.0564807728,
-0.2831984162,
0.5092186332,
-0.0105563672,
0.2368226349,
-0.1851424873,
0.6094354391,
0.3000097275,
0.1178232208,
0.3601542711,
-0.035562247,
0.1675812006,
0.1629152298,
-0.1403384805,
0.0368614867,
-0.1783125848,
0.3249067962,
0.2848486304,
-0.1618456542,
-0.33210361,
-0.3920762241,
0.1011416763,
-0.0891899988,
-0.0727448836,
-0.0491276383,
0.3427486122,
0.1343056709,
-0.1356297135,
0.3029321432,
0.0821177959,
-0.3334835768,
0.2686664164,
0.2534909546,
-0.1429877579,
0.1484564394,
0.1589110494,
-0.0379804298,
-0.3448108435,
0.3444148898,
0.1638857573,
0.3087618649,
0.0045744032,
0.1655190736,
0.1895902455,
-0.0242544524,
0.5605655909,
-0.4350597858,
0.3597973287,
0.1688044518,
0.0517792106,
-0.2673885524,
-0.1922028363,
-0.2611429691,
-0.276124835,
-0.1781180203,
0.2215981781,
-0.6027705669,
0.0161780082,
0.1669196934,
0.161229074,
-0.1238547564,
-0.359395504,
-0.3704820871,
-0.4479211867,
-0.4778937697,
-0.1964291334,
0.0075106695,
0.3619407713,
-0.2136184573,
-0.0925563499,
-0.0091365296,
0.1171009243,
-0.001581084,
-0.0716376677,
0.2431966662,
0.4718741179,
-0.0696358308,
0.2932137847,
0.1787507832,
0.5283370614,
0.6097087264,
-0.0627237409,
-0.1071097106,
0.0484472141,
-0.2018222213,
0.0192039013,
0.2239027917,
-0.0920594856,
0.1804145575,
0.1474186182,
0.0573407821,
-0.0472669601,
-0.2881350815,
0.3782991767,
-0.255780071,
-0.3657978773,
-0.4157607555,
0.466252774,
0.04135032,
0.1092936993,
0.1936930716,
0.2587583363,
-0.1668643057,
0.0226926319,
0.3577559292,
1.0175698996,
0.1065523326,
-0.0188474897,
0.2060740292,
0.014623858,
0.2137940526,
-0.2915169597,
0.2040701807,
-0.0980691835,
-0.4103772044,
-0.2600044012,
0.0024991035,
0.1532847583,
0.3295855224,
-0.4880577624,
0.3298540115,
-0.2777177989,
0.1029104441,
-0.0117820948,
0.2915962934,
-0.220928371,
-0.1896231622,
-0.2440837324,
0.0250633508,
0.1062117964,
0.1561801732,
-0.1630842984,
-0.0633790493,
0.0575041473,
-0.3713616133,
-0.4828014374,
0.058330439,
-0.2181004137,
0.4266484678,
-0.3948786557,
-0.2592288852,
0.2925424278,
0.4851697087,
0.1716977954,
-0.0103368107,
-0.3101625443,
0.2248513997,
-0.1228220761,
-0.4960903227,
-0.2240564525,
0.0299590044,
0.0838921517,
-0.0808787942,
-0.5226160884,
0.1020057946,
-0.037109822,
-0.0976727679,
0.0695047081,
-0.0406801514,
-0.148755461,
-0.2177063078,
-0.0691506788,
-0.0016988479,
0.1558316946,
-0.0840197876,
0.036230851,
-0.0019893423,
-0.1585761607,
0.1211211979,
0.1971922666,
-0.1881342381,
-0.166673243,
0.6800677776,
0.021305263,
-0.2687132359,
0.6212596893,
0.2700417042,
-0.3788089454,
-0.2264260501,
-0.0876421928,
0.1271990985,
-0.2000418007,
0.0818252265,
0.3339782655,
0.1755504608,
0.125036478,
0.4918980002,
0.2114367187,
0.0397879034,
-0.0355798341,
-0.5814547539,
-0.0940761864,
0.0889580622,
-0.080797337,
0.1880506128,
-0.469819963,
-0.3154391944,
0.2788206339,
0.0934569985,
-0.1850285977,
0.0440965667,
-0.4358551502,
-0.080707252,
0.1609925032,
0.023494957,
0.1360042542,
0.1081748679,
-0.0183175094,
0.0517950989,
-0.0362447537,
-0.0480694138,
-0.0843036026,
0.2136484981,
0.0365363732,
-0.1229961067,
0.0750055611,
-0.4660218358,
-0.3347611129,
0.0175155215,
0.2713813186,
0.3464637399,
0.0112376744,
0.0264790431,
0.3368586898,
0.0637378842,
-0.5087183118,
0.0850757807,
0.1238570586,
0.3832838833,
0.3522446454,
-0.0603413992,
0.0815019161,
0.0552271903,
-0.0290250219,
0.0364019498,
0.1451058388,
-0.0185585245,
0.4161346853,
-0.2846485972,
-0.0397412106,
0.3598662615,
0.0802498609,
0.4701504707,
-0.1421667784,
0.1650277972,
0.1924475133,
0.0728163719,
-0.4741269946,
-0.2790477276,
0.322547853,
0.088057518,
-0.1475119889,
0.0822518319,
0.197687462,
-0.5292576551,
-0.198570475,
0.0116610527,
0.2377843559,
-0.2428720593,
0.1522593796,
0.5658451319,
-0.0603008866,
0.0266814306,
-0.1705100983,
0.1049015224,
0.3029582798,
-0.0226077046,
-0.3302283585,
0.323898077,
-0.051720798,
0.2576179802,
-0.2167515457,
-0.581649363,
0.1976688504,
0.2486559749,
0.109131664,
-0.06238731,
0.1428007483,
0.28139171,
0.1161990166,
0.1848137528,
-0.5019871593,
0.1146366671,
-0.0847101361,
-0.0868119448,
-0.1550210714,
-0.1631115526,
-0.3408387601,
0.3697254658,
0.2506074011,
-0.3877317607,
0.1391343772,
0.2234000564,
-0.3519897461,
-0.3351709247,
-0.0046413606,
0.0475754514,
0.1984974742,
-0.2501713037,
-0.0406994969,
0.4264983833,
-0.0355310887,
0.5507057309,
0.115885891,
0.2798181176,
0.1993607283,
-0.0190245025,
0.0986390859,
0.1593253464,
0.0265294835,
0.0264791325,
0.3988072276,
0.184543103,
0.1278179884,
0.1725085229,
-0.0023973547,
-0.1505350322,
-0.3022057712,
0.1991574913,
0.1582883149,
0.1563820541,
-0.0906075686,
0.0002231766,
0.119393006,
-0.1225256249,
-0.0520094819,
-0.2637984753,
0.2559805214,
0.2613161802,
0.096773684,
0.099728629,
0.1218368262,
-0.0077643916,
-0.2113357931,
0.489398241,
0.3439131677,
-0.0329996459,
-0.3170143664,
-0.1657277048,
-0.8635806441,
0.1909137666,
-0.0018570956,
-0.198791638,
0.2767448723,
0.1448273957,
0.0629289076,
0.1501016766,
0.0517332554,
0.0493550561,
0.355802536,
0.5278561711,
-0.0388546437,
-0.3366439342,
0.057761386,
0.0212413315,
-0.2305416763,
-0.2928698659,
0.0717011541,
-0.0456008241,
-0.1349541247,
-0.1366090178,
-0.0527432151,
-0.1673204005,
0.2791331112,
0.2438107878,
0.0376396701,
0.6168966293,
-0.0421946384,
0.03400518,
-0.216089204,
0.0339433402,
-0.1781508923,
0.3692154288,
0.1981419027,
0.0402885601,
-0.2612345219,
0.1893489659,
-0.255158186,
0.1336953044,
0.1048163399,
0.3364699483,
-0.4344372451,
-0.0075038709,
-0.0616459623,
0.0245792456,
0.1752415001,
-0.1018627882,
-0.1526125968,
-0.0320740268,
-0.1232205704,
-0.3546846509,
0.6694047451,
-0.4451084733,
-0.1035619229,
-0.0715577826,
0.1026606113,
-0.2589514852,
0.1466145664,
-0.782880187,
0.2994580865,
0.2722882032,
-0.0191891473,
-0.5205339789,
0.0897957683,
-0.1968085617,
0.0039333105,
-0.1593838036,
0.1601533145,
0.1927387714,
-0.2259235382,
-0.1817130744,
-0.2133520842
] |
https://github.com/huggingface/datasets/issues/1911 | Saving processed dataset running infinitely | am suspicious of this thing? what's the purpose of this? pickling and unplickling
`self = pickle.loads(pickle.dumps(self))`
```
def save_to_disk(self, dataset_path: str, fs=None):
"""
Saves a dataset to a dataset directory, or in a filesystem using either :class:`datasets.filesystem.S3FileSystem` or any implementation of ``fsspec.spec.AbstractFileSystem``.
Args:
dataset_path (``str``): path (e.g. ``dataset/train``) or remote uri (e.g. ``s3://my-bucket/dataset/train``) of the dataset directory where the dataset will be saved to
fs (Optional[:class:`datasets.filesystem.S3FileSystem`,``fsspec.spec.AbstractFileSystem``], `optional`, defaults ``None``): instance of :class:`datasets.filesystem.S3FileSystem` or ``fsspec.spec.AbstractFileSystem`` used to download the files from remote filesystem.
"""
assert (
not self.list_indexes()
), "please remove all the indexes using `dataset.drop_index` before saving a dataset"
self = pickle.loads(pickle.dumps(self))
``` | I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function. | 103 | Saving processed dataset running infinitely
I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function.
am suspicious of this thing? what's the purpose of this? pickling and unplickling
`self = pickle.loads(pickle.dumps(self))`
```
def save_to_disk(self, dataset_path: str, fs=None):
"""
Saves a dataset to a dataset directory, or in a filesystem using either :class:`datasets.filesystem.S3FileSystem` or any implementation of ``fsspec.spec.AbstractFileSystem``.
Args:
dataset_path (``str``): path (e.g. ``dataset/train``) or remote uri (e.g. ``s3://my-bucket/dataset/train``) of the dataset directory where the dataset will be saved to
fs (Optional[:class:`datasets.filesystem.S3FileSystem`,``fsspec.spec.AbstractFileSystem``], `optional`, defaults ``None``): instance of :class:`datasets.filesystem.S3FileSystem` or ``fsspec.spec.AbstractFileSystem`` used to download the files from remote filesystem.
"""
assert (
not self.list_indexes()
), "please remove all the indexes using `dataset.drop_index` before saving a dataset"
self = pickle.loads(pickle.dumps(self))
``` | [
-0.2176188827,
0.1967647076,
-0.1150819436,
0.2636954784,
0.1485696435,
-0.0731193721,
0.1913732737,
0.2333742529,
-0.1932330579,
-0.0838293731,
0.0687723756,
0.3565648496,
-0.1087137684,
0.2635344863,
0.018014688,
0.2948513627,
0.2866329253,
0.1167819798,
0.2218980491,
0.0443395227,
-0.1144506186,
-0.0392292067,
-0.105300732,
0.0729433,
-0.2676576078,
-0.2402476966,
0.0895975754,
-0.0787026137,
-0.3543222547,
-0.2827144861,
-0.0402493514,
0.2020812631,
-0.0207410082,
0.5996129513,
-0.0001320559,
-0.3027281761,
0.2690655291,
-0.0898029208,
-0.5009801984,
0.091837734,
0.1252098083,
-0.027451802,
-0.0236186646,
-0.4111997485,
0.3176500797,
0.0550618097,
0.1007526591,
-0.2602962255,
0.2392186224,
0.0291387588,
0.0483903289,
0.1627708375,
-0.1344564259,
0.4667538404,
0.3288219869,
0.5879322886,
-0.0920099244,
0.0681962669,
0.2598215342,
-0.1367047876,
-0.0111972727,
0.2131494284,
-0.4668719769,
-0.1813861728,
0.3238170147,
-0.0626366287,
-0.164229095,
-0.4855025113,
0.1900699288,
0.1010173783,
0.3124757707,
-0.5746485591,
-0.3016626835,
-0.2907088399,
-0.3085362613,
-0.4616626501,
0.3031603992,
0.2828127444,
-0.1148830503,
0.0892650187,
-0.3573214412,
-0.0885956585,
-0.0714922249,
0.0139243547,
-0.1595257074,
-0.0218514539,
-0.2625795901,
0.2816874087,
0.1571232975,
0.1349898279,
0.2138314098,
-0.5275250077,
0.2432622015,
0.1830548942,
-0.2409774661,
-0.0892679244,
-0.1281012893,
0.2305010259,
0.0684549138,
0.2216308415,
-0.2834173441,
0.0942214429,
0.1737059653,
0.0587531142,
0.3496393561,
-0.1559290588,
0.1240392327,
0.1688315719,
0.1533986628,
-0.3493703306,
-0.1055042893,
0.098983489,
0.1423722208,
0.1292575002,
0.2541599572,
-0.4392640293,
0.0998978466,
-0.1352924556,
-0.1503776461,
0.293504864,
-0.1735001802,
-0.0440575965,
-0.0193764344,
0.2404391468,
-0.1896394044,
-0.1738826632,
-0.2136648297,
-0.122686699,
-0.2828748822,
0.1256302893,
0.0212365687,
-0.0136182187,
-0.076416105,
0.1896993518,
-0.0462354235,
-0.0983828679,
-0.0973694623,
-0.0382033326,
-0.2951149344,
0.0666370317,
0.1504891813,
-0.6224696636,
0.2766811848,
0.5634760857,
0.0076078922,
0.3507174551,
-0.128665626,
-0.1917124391,
-0.1637354493,
0.3982416391,
-0.0726505369,
-0.1950868368,
-0.0195658654,
-0.044941783,
-0.2271587849,
0.0960474759,
-0.4880657196,
0.0129516572,
0.4187316,
-0.3198015094,
0.1130045131,
-0.2247400582,
-0.4087261856,
-0.2898057699,
0.0422478765,
0.312944293,
-0.652502656,
0.1624652743,
-0.2219654918,
0.0037870184,
0.3684931099,
0.2907517552,
-0.0184266437,
0.3519964814,
-0.2924844325,
0.4722722173,
0.4642393589,
0.1513803005,
-0.5824524164,
0.2184978426,
-0.0032660365,
0.0871214345,
-0.167991519,
0.280618161,
0.3480524421,
-0.1936859488,
0.0448205583,
0.1299174726,
0.0902845934,
-0.0739289373,
-0.2451008856,
-0.1558863223,
0.1333709657,
0.1133299246,
-0.1586102545,
-0.0949895009,
0.1699925363,
-0.281157434,
0.3692250848,
0.1862330735,
0.2380424738,
0.226243943,
0.4387607872,
0.3766168058,
-0.0299832914,
-0.2525387704,
0.0410478786,
-0.0314511545,
0.2027604878,
-0.2562843561,
-0.3878393173,
-0.344466269,
-0.0672248453,
0.3747114539,
-0.195119381,
0.0566943213,
-0.1595135331,
0.1894021928,
0.0622432083,
0.2978357077,
-0.176330924,
0.1621254683,
-0.4285734594,
-0.0729778707,
-0.2045768201,
0.1255356669,
0.0948562846,
-0.0802919492,
-0.2075642347,
-0.0644675866,
0.1118550226,
-0.0987706929,
0.0527720265,
0.1788642108,
0.3130967319,
0.0153764486,
-0.0880966783,
0.2765066326,
0.0521607399,
-0.0698792487,
-0.0566344857,
-0.1350645274,
0.437130928,
0.0080774799,
-0.1692183018,
0.1955078542,
-0.0692286342,
0.2699981332,
0.0169119686,
-0.2877047062,
0.1833030581,
0.124882035,
-0.089494586,
0.1032607257,
0.131716311,
0.2835095525,
0.3211946487,
0.2160080969,
-0.1547554582,
0.4590101242,
0.6726173162,
-0.3963782191,
0.0546124242,
0.3500018716,
0.1655369848,
-0.3123157322,
-0.1193109229,
0.1344990432,
0.4746419787,
0.0471060239,
0.1163870096,
-0.0722296759,
0.056672819,
-0.1296166033,
0.3597225547,
0.1670878977,
0.2787836492,
0.2663930655,
0.2064011991,
-0.0926340073,
-0.3180207014,
-0.2046257854,
0.2067175359,
0.3337040842,
-0.2496601641,
0.4450703263,
-0.2661705911,
-0.1963254511,
-0.0248066261,
-0.0397421643,
-0.101184532,
-0.1752339453,
-0.2606776059,
0.2353003919,
-0.0188821889,
-0.0016379021,
0.2302643806,
0.2999732494,
0.1067367643,
-0.3236912191,
-0.2955579162,
-0.0433313586,
-0.2829550505,
-0.0933735073,
0.3848827779,
0.1502902508,
0.4872847497,
-0.035730008,
-0.2362872213,
-0.6306732893,
-0.3318598866,
0.1534072161,
-0.0029993835,
0.553743124,
0.0621285327,
0.231398046,
-0.2958819568,
0.1094464064,
-0.057073392,
-0.0468584597,
-0.2712247968,
-0.1202102453,
0.0726316869,
-0.1153619289,
0.0249980688,
0.0948594585,
-0.4184023142,
0.0370900258,
0.3668836951,
-0.3680100739,
0.3206293285,
-0.2364814728,
-0.1106600538,
0.1673945785,
-0.0467545539,
-0.1276376545,
-0.2547929883,
-0.3890379369,
0.2348026931,
0.1060609296,
-0.2417602688,
-0.0012644231,
0.0302167907,
-0.2923388779,
0.2514455914,
-0.667049706,
0.1034716144,
-0.25750947,
0.1251005232,
0.0880423039,
0.095470801,
0.3562362194,
-0.1353863329,
0.1612422317,
-0.004498221,
-0.4499638677,
-0.0907287002,
-0.060586486,
0.1561435461,
0.2193525732,
0.6643556952,
0.0217190012,
0.6045765877,
0.0560658053,
0.1380068958,
0.2675991058,
-0.0902758911,
0.3034043312,
-0.2566717863,
-0.308006227,
0.1277372539,
-0.1734943092,
-0.5134419799,
0.0484150723,
-0.267170161,
-0.0654827803,
0.2267851979,
-0.1906283051,
-0.2014591545,
-0.1861757785,
0.0661618784,
0.3981273174,
0.1973188221,
0.2608335316,
-0.0373024195,
-0.0074186698,
-0.1218076348,
-0.2753039896,
0.3208585978,
0.3104403019,
-0.0027600192,
-0.3376791179,
-0.1947740614,
-0.4489061236,
0.167063728,
0.081419602,
0.1557715833,
-0.0812252164,
-0.0599583313,
0.3341171741,
-0.3577573895,
1.1544932127,
-0.0068501271,
0.112806201,
0.2271644026,
-0.1863551438,
-0.4312824309,
0.0836013034,
0.0544865131,
0.1243982837,
0.2215549052,
0.2606631517,
-0.2197369784,
-0.1267493665,
0.0469346307,
0.3274631202,
-0.1696224809,
-0.1616918743,
-0.2280382216,
-0.5100157261,
-0.2329774499,
0.2316287458,
0.0015793219,
0.0847226977,
0.0655162334,
-0.0458719917,
-0.2126408517,
-0.0030376613,
0.1310040653,
-0.0917726085,
0.1487096846,
-0.1269325763,
0.1700149179,
-0.3102151453,
0.0544445068,
0.3360916078,
0.1814395487,
0.0254546702,
0.2124774158,
0.4325070679,
-0.1095377281,
0.3074271083,
0.421472162,
-0.0701554641,
0.1852405071,
-0.0510540493,
0.3726034164,
-0.3217010796,
0.0800599158,
0.1405323446,
0.2599297166,
-0.4864938855,
-0.3555611372,
0.2350630313,
0.1999844313,
-0.0872699767,
0.2623471022,
-0.0226396583,
-0.0963537544,
0.2893200815,
0.1828494072,
1.30864048,
-0.1450112313,
0.3485220969,
0.3548603654,
-0.0636640266,
0.0992137492,
-0.4697080851,
-0.0752162784,
-0.150297001,
-0.0368538573,
-0.170392096,
-0.173016265,
0.1589582115,
0.0185333844,
-0.0591553599,
0.1629076302,
-0.1580059975,
-0.0016330145,
0.0588590205,
0.3480600119,
-0.2200701982,
-0.0782796144,
-0.2369823754,
-0.0147664696,
0.1073827371,
-0.0454218797,
-0.0758808851,
0.0863575041,
0.0776352957,
-0.0572424755,
-0.1449042112,
-0.1484404802,
-0.226564616,
0.1288760006,
-0.2935390472,
-0.1118462086,
0.1564642936,
-0.1301864684,
-0.1242422312,
0.2798476815,
0.1314186901,
0.1870655417,
0.0642459691,
0.2229291052,
0.0615285635,
-0.4931619465,
0.2082463056,
-0.0806249902,
-0.0249037743,
-0.0510763228,
-0.1839709282,
-0.3373160958,
-0.0620779991,
0.1841900498,
0.5335727334,
-0.2247229367,
-0.3462597728,
-0.1785235405,
-0.1496078521,
-0.1413446963,
-0.0690938681,
0.1150558814,
0.1077010781,
0.6988788843,
-0.5259567499,
-0.3785373569,
-0.2415051013,
0.2715921402,
0.0896577239,
0.1711592972,
0.3277641237,
-0.0070921853,
-0.3488295078,
-0.1859961748,
0.3130154908,
-0.1552766412,
-0.0183441117,
0.4162687361,
-0.0939853638,
0.1282290816,
-0.1468787938,
0.0273753107,
-0.0787055567,
-0.1212730333,
-0.3803351223,
-0.5441792607,
-0.1124779284,
-0.1328877956,
0.3512420058,
0.0402089916,
0.3490841985,
0.0829243064,
0.0400355384,
-0.0158003718,
-0.1107849628,
0.1925852299,
-0.1334846765,
0.2333538234,
0.076591976,
-0.0061088772,
0.1610465944,
-0.1598842442,
-0.1560812891,
0.3592750132,
-0.1628751457,
-0.0988479108,
-0.1282745004,
0.2784000635,
-0.0087861959,
-0.1810811907,
-0.0937768966,
-0.3695254922,
-0.1319839209,
-0.1144261286,
-0.0155969299,
0.2512644231,
0.0957467556,
-0.5581240654,
0.191504389,
-0.1907063425,
-0.109484911,
0.0942845196,
-0.0859409571,
-0.0040377639,
0.0022503883,
0.2778497636,
0.1359619051,
-0.0499163419,
-0.284709692,
0.0633431673,
0.2024171054,
-0.172654748,
0.2768597305,
-0.2688004971,
-0.1116402075,
0.0199813135,
0.3108311594,
0.6949667931,
-0.1796551198,
-0.4286448061,
0.3348476589,
-0.0296764467,
-0.0848098025,
-0.1932091266,
0.0858063102,
-0.0390491448,
-0.1912899762,
0.0999154821,
-0.0937541276,
0.2890971303,
-0.3029479682,
0.101022169,
0.546503067,
-0.0691457838,
-0.203859508,
0.3043861389,
0.0957866758,
0.3228175342,
0.3069993556,
0.2502882779,
0.0553603433,
0.6206977367,
0.0220015533,
0.3816456795,
0.3313222229,
0.3392210901,
-0.1184369773,
-0.3111567199,
0.1606085598,
0.4570243359,
-0.4098160863,
0.1371448636,
0.080412358,
0.1331513822,
0.0033823252,
-0.3533459306,
-0.0817737728,
0.6286836863,
-0.0968445688,
-0.1043325067,
-0.0912948996,
-0.142957598,
0.3363593221,
0.197630927,
-0.0739197582,
-0.1162398607,
0.036283914,
0.2106340677,
0.0515608788,
-0.1486435235,
-0.0090334732,
-0.1144568324,
0.0577784628,
-0.1716115475,
-0.0355438404,
-0.1341074109,
0.1001670286,
-0.3189407885,
0.0707942024,
0.5186488032,
0.4143278599,
0.0960569978,
0.0883212984,
-0.0924513862,
0.0018731803,
-0.1517197937,
0.3243770003,
0.007516386,
0.2404129356,
0.2899965048,
-0.035259407,
0.0400817245,
-0.2816099524,
0.275796324,
0.3581039608,
-0.2363889515,
0.1167817116,
0.0016279425,
0.0580271184,
-0.1721298397,
0.2312396616,
-0.5059823394,
0.2320761681,
0.7975668311,
-0.2769778371,
0.2970380783,
-0.0013091378,
-0.0100843403,
-0.3129545748,
0.5281898975,
0.0810494572,
0.0797505528,
-0.2492759079,
-0.0456881002,
-0.5394473076,
-0.148283124,
-0.0268195122,
-0.093561247,
0.0284003168,
0.1371171176,
0.1103590876,
0.051377248,
-0.0754792988,
-0.0340373181,
-0.4058741033,
0.2675478458,
-0.0699572563,
0.1924787015,
-0.2088575065,
0.2203046829,
-0.1897623688,
-0.5497689247,
0.5792176127,
-0.1915828139,
-0.1039918214,
-0.0177280772,
0.0784667432,
0.0395193994,
0.0994429737,
0.3121793568,
0.1372772753,
0.130153954,
-0.0450151563,
0.2122038603,
0.1963950843,
0.3255819082,
-0.0211296342,
0.1904251873,
0.0014327485,
0.1711434424,
-0.20031479,
0.3008947074,
-0.1542052925,
0.2673220932,
0.0009627789,
0.0915154815,
0.2174801081,
-0.0998980552,
-0.0176700205,
0.4172852039,
0.0027415194,
0.308870852,
0.1381883621,
0.2813309729,
-0.4911983609,
-0.1034869403,
0.5883194804,
-0.5534769893,
-0.3947318196,
-0.1415629089,
0.1200660914,
0.4194648266,
-0.1703367233,
-0.4737052023,
-0.2043702304,
0.4016352892,
-0.0639478415,
-0.3335828781,
0.3400069773,
-0.3495186865,
0.2096219361,
-0.1204766482,
0.5618266463,
-0.1222990751,
-0.3039700389,
0.0734525472,
-0.3393238485
] |
https://github.com/huggingface/datasets/issues/1911 | Saving processed dataset running infinitely | Tried finding the root cause but was unsuccessful.
I am using lazy tokenization with `dataset.set_transform()`, it works like a charm with almost same performance as pre-compute. | I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function. | 26 | Saving processed dataset running infinitely
I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function.
Tried finding the root cause but was unsuccessful.
I am using lazy tokenization with `dataset.set_transform()`, it works like a charm with almost same performance as pre-compute. | [
-0.3561963439,
0.1683893204,
-0.0956781507,
0.2705769539,
0.2809063792,
-0.0555668175,
0.2274561077,
0.2229019552,
-0.3108582497,
-0.1395980418,
0.1082133725,
0.1759863198,
-0.1438535601,
0.234498173,
0.0023733862,
0.2562199235,
0.3514586687,
0.1079184785,
0.3704063594,
0.1087034121,
-0.0698907673,
0.0259362627,
-0.1417381763,
0.0397689492,
-0.3694447577,
-0.3420149386,
0.1376418173,
-0.2580527067,
-0.3062803447,
-0.2669491172,
-0.0369218662,
0.3850828409,
0.0711700618,
0.5333967209,
-0.0001309843,
-0.3164227903,
0.2313468605,
-0.0396904349,
-0.2950723171,
0.1473657936,
0.1879240721,
-0.0768320337,
-0.0761473849,
-0.3226695359,
0.2280895263,
-0.0560970604,
0.0666026324,
-0.1789616644,
0.3069123328,
-0.0999139696,
0.0244625863,
0.0963330418,
-0.2024441063,
0.3944405615,
0.0781612694,
0.5052488446,
-0.0299686119,
-0.0143227782,
0.241749227,
-0.1317527443,
-0.0599419028,
0.2605648935,
-0.4225962758,
-0.1757381558,
0.2483234704,
-0.0603323728,
-0.1893056929,
-0.4330960512,
0.3536807299,
0.1250942945,
0.2908189893,
-0.5002195239,
-0.2704517245,
-0.2737933099,
-0.2251418829,
-0.4966007471,
0.3358698785,
0.287578702,
-0.0780525133,
0.0703708678,
-0.4501781166,
-0.0643268749,
-0.121984005,
-0.0093970634,
-0.2441366762,
0.0015156306,
-0.2214838117,
0.1797711104,
-0.0281713419,
0.0159975383,
0.2974229157,
-0.4438254833,
0.119784154,
0.0765058994,
-0.2945413589,
-0.0409832932,
-0.3404233456,
0.2206652761,
0.1488776952,
0.1337822974,
-0.2470908016,
0.1759067476,
0.2328188568,
-0.0440793298,
0.1266121268,
0.0395752974,
0.0933193117,
0.1374984384,
0.2706355751,
-0.4037395418,
-0.1725403816,
0.0876228064,
0.0085120834,
0.1798761636,
0.4154669642,
-0.5015189052,
-0.0385922231,
-0.156815052,
-0.1673544198,
0.299908787,
-0.1609705091,
-0.0266306158,
-0.0335096754,
0.3120589554,
-0.2557437122,
-0.1600604355,
-0.1566642225,
-0.0816944316,
-0.1733273268,
-0.0099959597,
0.048499357,
-0.0495842472,
-0.0274325535,
0.2023168206,
0.0920677632,
-0.0578010716,
-0.0601685494,
0.0159268007,
-0.232803449,
-0.0403909795,
0.1218574569,
-0.6531038284,
0.3523283303,
0.3579157889,
0.0929572284,
0.4303993583,
-0.2356290817,
-0.1576523334,
-0.0987184718,
0.3413204551,
0.0289168395,
-0.1152944118,
0.0900055766,
-0.0103596188,
-0.2273552865,
0.0456303805,
-0.3860264719,
0.1007697657,
0.4957955182,
-0.1020925418,
-0.031296391,
-0.2901703119,
-0.4178996384,
-0.1631324738,
0.1399660408,
0.2322126329,
-0.5348829031,
0.1970509887,
-0.1168975383,
0.1143589169,
0.3382967412,
0.4622503817,
0.0029968992,
0.3917591572,
-0.2535491884,
0.3449242115,
0.3348971605,
0.2179573923,
-0.5527258515,
0.1113337874,
0.010500297,
-0.0713314861,
-0.1493802667,
0.2181281447,
0.6505596638,
-0.2163798958,
0.1967262924,
0.1443398297,
0.0347694978,
0.0030679256,
-0.3002890348,
-0.2102754116,
0.0704103783,
0.1161041185,
-0.0734509453,
-0.1079779714,
-0.026213672,
-0.0873258337,
0.2607066035,
0.0873323679,
0.3065141737,
0.3071137369,
0.2978537977,
0.310849756,
0.038542375,
-0.2755224705,
0.09263473,
-0.0664935187,
0.2263990939,
-0.2249598801,
-0.3080346584,
-0.3852898479,
-0.0595656931,
0.3317053318,
-0.1536320895,
0.0673740134,
-0.0793265179,
0.191582486,
-0.0316917971,
0.28129825,
-0.0916689038,
0.3534352183,
-0.2316100895,
0.0081599765,
-0.0361625589,
0.0677645653,
0.1614734679,
-0.2569531798,
0.0036000386,
0.0137709454,
0.1429989785,
-0.0656036586,
0.0499034524,
0.1705204993,
0.0855396539,
0.0511173755,
-0.1799759865,
0.2203267366,
0.0789979175,
-0.1262027174,
0.009175444,
-0.1297917664,
0.2911992669,
-0.0562337935,
-0.1374333352,
0.2095733881,
-0.026498558,
0.2317999899,
0.0407147855,
-0.2505152822,
0.2457155883,
0.0342696533,
-0.011594642,
0.0257079527,
0.2258250415,
0.1869432181,
0.2858631611,
0.1908425242,
-0.1367454529,
0.4230034947,
0.6380717158,
-0.2912368476,
-0.0326800682,
0.2630532682,
0.0070479214,
-0.2236629426,
-0.0833157897,
0.0391301885,
0.2409750819,
0.1381785274,
0.0359576195,
-0.0949810296,
0.0806816369,
-0.112868607,
0.3657851815,
0.1423153877,
0.3048703671,
0.2377584726,
0.2720299959,
-0.0556192622,
-0.3268781304,
-0.115410015,
0.3128302991,
0.3043536246,
-0.1911306381,
0.2493323386,
-0.3187403381,
-0.2655820251,
-0.0087980349,
0.0150733925,
-0.209875688,
-0.1644063294,
-0.2142039984,
0.290217042,
0.0964984745,
0.1142698228,
0.2372080982,
0.1645355821,
0.0896115601,
-0.3337946534,
-0.288640976,
-0.0336717032,
-0.2719532847,
-0.1009411886,
0.344932884,
0.1193484366,
0.5602350831,
-0.0377672277,
-0.201066047,
-0.5184050202,
-0.3823206425,
0.183689788,
0.0954876989,
0.4375213981,
-0.072078079,
0.2666134536,
-0.2688076198,
0.0775143206,
-0.0349434912,
-0.1127792001,
-0.1897430122,
-0.0965391919,
0.0818460584,
-0.1162788272,
0.0815772712,
0.1189093515,
-0.4141136408,
0.081656538,
0.2215961814,
-0.4189991951,
0.2398498803,
-0.4049995542,
-0.1667508334,
-0.0364277288,
-0.030109426,
-0.1684181392,
-0.1815479547,
-0.3694896996,
0.4045924842,
0.1032461375,
-0.2782887518,
0.085425809,
0.0241050422,
-0.1648417562,
0.3543707728,
-0.5036890507,
0.292039901,
-0.2637020051,
0.1855167747,
0.1028891653,
-0.0470127612,
0.4037770629,
-0.0545964018,
0.1448313296,
0.0343244523,
-0.5332233906,
-0.0501895808,
-0.1899868995,
0.144188121,
0.2329203486,
0.7339384556,
0.0965074822,
0.5828557014,
0.0113102719,
0.0956184566,
0.1882005483,
-0.0941794291,
0.1587114334,
-0.188386336,
-0.3327955604,
0.0815476924,
-0.0654176176,
-0.3301578462,
0.1679269373,
-0.1776426286,
-0.1029428691,
0.2416676283,
-0.3118292987,
-0.2376960218,
-0.235197708,
0.0346931741,
0.3311671317,
0.3532546461,
0.3883739114,
0.0201950446,
-0.0997946858,
-0.2928072512,
-0.2255048454,
0.2581327558,
0.2102110386,
-0.0397795588,
-0.2918078303,
-0.0236837976,
-0.5580815077,
0.2260975689,
-0.0011214651,
0.3068580329,
0.016296953,
0.0465433151,
0.3327514231,
-0.3871509433,
1.0841895342,
0.0571328178,
0.1120699793,
0.1409698278,
-0.3501552045,
-0.3101100922,
0.0681097806,
0.0868622363,
0.0650252476,
0.4142942131,
0.3152955174,
-0.2865533829,
-0.2286938131,
0.0804544017,
0.3835462928,
-0.2443113923,
-0.2377671003,
-0.1378659755,
-0.4442846179,
-0.3062328696,
0.238686204,
0.0170173571,
0.1890955716,
0.0787004381,
-0.1602079719,
-0.3057709336,
-0.1577474773,
0.3004530072,
-0.0766149461,
0.2009888142,
-0.2700322866,
0.3185226023,
-0.2576546669,
0.1249515861,
0.3866302669,
-0.1018245518,
0.1028977782,
0.0692249388,
0.3182426393,
-0.1466760784,
0.311928004,
0.3617823422,
-0.1181294695,
0.1253963858,
-0.0126627805,
0.3950272799,
-0.3305842578,
0.1631427109,
0.0222368278,
0.3152641952,
-0.6721394062,
-0.2041842341,
0.0585967824,
0.2648920417,
-0.1384115964,
0.1423315257,
-0.1764370054,
-0.2093855292,
0.3884708881,
0.2383250892,
1.1850357056,
-0.3527076244,
0.2435463518,
0.0962540507,
-0.0012013614,
0.1224104017,
-0.5848093033,
0.0443327539,
-0.1987951994,
-0.2017736882,
-0.1024802849,
-0.1412131637,
0.0470493138,
0.0428305343,
-0.0479662605,
0.1785009205,
-0.0131100714,
0.1646737158,
0.045717217,
0.4429644346,
-0.1965350211,
-0.2629643083,
-0.2347230911,
-0.0292725917,
0.0182910152,
-0.0972595662,
-0.0533504002,
0.0304029137,
0.1121017337,
0.0048080534,
-0.1400815994,
-0.2070684135,
-0.2857976258,
-0.0379704423,
-0.3523787558,
-0.1300926954,
0.1904643923,
-0.1281272173,
0.1564736366,
0.3663038909,
0.0848680958,
0.1745134294,
-0.0607503802,
0.1701620966,
0.0446170196,
-0.457350105,
0.0846156403,
-0.1292465031,
0.0339117646,
-0.1513442099,
-0.0740291476,
-0.2199122012,
-0.1011072323,
-0.0188079625,
0.5681501031,
-0.244674623,
-0.1460636854,
-0.2114470601,
-0.0063431859,
-0.1912202835,
-0.0381627716,
0.1370898187,
0.0802055597,
0.7246431112,
-0.4206167758,
-0.2836775482,
-0.2777165473,
0.3438785374,
0.17981565,
-0.0019148812,
0.3709088862,
0.0575134531,
-0.3377860785,
-0.1966435909,
0.2740467787,
-0.1098371372,
-0.0943763778,
0.3875783682,
-0.0744719803,
0.0293175131,
-0.0829981342,
0.0094575584,
-0.0782549828,
-0.291531831,
-0.2505191267,
-0.525945425,
-0.282207787,
-0.1161349118,
0.3359450996,
-0.0582361259,
0.3865966201,
0.256059587,
0.0006351247,
0.0651099831,
-0.1440391541,
0.0952230468,
-0.0528994873,
0.2378826439,
-0.1445235163,
-0.0852373838,
0.0821095705,
-0.1076262295,
-0.1013876349,
0.4470530748,
-0.0315739848,
-0.0804556087,
-0.1055285782,
0.2624733448,
-0.1276791394,
-0.1190460473,
-0.0676206276,
-0.3544045985,
-0.1504695415,
-0.2015369236,
0.0297060609,
0.1752886474,
0.0456516594,
-0.741725862,
0.2148524076,
-0.06610246,
-0.1919353008,
-0.0027080877,
-0.0215967596,
0.034685865,
-0.0384615809,
0.3330563307,
0.1243119687,
-0.054538168,
-0.2384003997,
-0.0041591153,
0.2349290252,
-0.2059052736,
0.3036857843,
-0.3696600795,
-0.2139865458,
0.2145575881,
0.2094155401,
0.5877885222,
-0.0653167665,
-0.4183363914,
0.380753994,
0.0052930941,
-0.0946512297,
-0.0843187571,
0.1214045584,
-0.1532377601,
-0.2048007548,
0.1553563774,
0.0319791064,
0.157907337,
-0.1818760633,
0.1740531921,
0.7348407507,
-0.2052144706,
-0.1840087175,
0.2037712038,
0.1761213243,
0.3332897425,
0.3962142169,
0.34134835,
0.0233669952,
0.6390283704,
0.1276099086,
0.3665175736,
0.4090359211,
0.5006254911,
-0.1466985792,
-0.3013935983,
-0.0524774045,
0.4722623825,
-0.2299553305,
0.1820887923,
-0.0172430947,
-0.0034151785,
0.2032733411,
-0.4290622473,
-0.1983247399,
0.7001780868,
-0.1493079662,
0.0222712867,
-0.1310428977,
-0.1185642555,
0.2914767265,
0.2307429016,
-0.0646342188,
-0.0569516011,
-0.0748467594,
0.2232404649,
0.0209223479,
-0.1831177771,
-0.0129006142,
-0.1347300708,
0.0248774812,
-0.1974099576,
0.0989501178,
-0.1873246282,
-0.0104270801,
-0.2813014984,
0.1564616114,
0.4901656806,
0.4058196843,
0.103632167,
0.1096381098,
-0.1395998895,
-0.0003487915,
-0.1513185352,
0.3752331138,
0.0393167809,
0.3430397809,
0.1373867244,
-0.0394749157,
0.0164961256,
-0.3794522285,
0.2590900064,
0.2656537294,
-0.0957699418,
0.1269500703,
0.0949983895,
-0.0701044053,
-0.1572814584,
0.067624636,
-0.3923460245,
0.3463863432,
0.7331265211,
-0.3003650904,
0.2092429996,
-0.0336299613,
0.0196809173,
-0.3313454688,
0.403632611,
0.0891081169,
-0.0353798084,
-0.2302407324,
-0.196440801,
-0.5508621335,
-0.1725220978,
-0.0550134107,
0.0583998114,
0.0377015695,
0.1008859202,
0.1468233317,
0.0002136277,
-0.1338798851,
-0.0363392457,
-0.3546833396,
0.2725558579,
-0.1193494648,
0.2490174472,
-0.2044499815,
0.2637100518,
-0.1720294207,
-0.4543081522,
0.6403275728,
-0.2540268004,
-0.0691938549,
-0.059238147,
0.1509839147,
-0.0337427333,
-0.0266966745,
0.2175128162,
0.2582066655,
0.0115284845,
-0.0121625811,
0.3897278905,
0.2276839316,
0.420509994,
-0.0297247265,
0.2628472745,
0.1851284504,
0.2558898926,
-0.1835882962,
0.2455415726,
-0.0848855227,
0.1309694648,
-0.0105225518,
0.0494291857,
0.1391809285,
0.0279732868,
0.0576982349,
0.4323115647,
0.0486310422,
0.351611346,
0.1439344883,
0.2393043041,
-0.6242008209,
-0.1817733049,
0.4588896632,
-0.6276183724,
-0.5190659761,
-0.1725154072,
0.1477305442,
0.5300551057,
-0.1651885659,
-0.3037887216,
-0.0740635693,
0.3971313238,
-0.0567140132,
-0.3319042623,
0.3381712437,
-0.2208521366,
0.1761684716,
-0.1379644722,
0.3688358068,
-0.0206711367,
-0.3688695133,
-0.0293013491,
-0.1754116863
] |
https://github.com/huggingface/datasets/issues/1911 | Saving processed dataset running infinitely | Hi ! This very probably comes from the hack you used.
The pickling line was added an a sanity check because save_to_disk uses the same assumptions as pickling for a dataset object. The main assumption is that memory mapped pyarrow tables must be reloadable from the disk. In your case it's not possible since you altered the pyarrow table.
I would suggest you to rebuild a valid Dataset object from your new pyarrow table. To do so you must first save your new table to a file, and then make a new Dataset object from that arrow file.
You can save the raw arrow table (without all the `datasets.Datasets` metadata) by calling `map` with `cache_file_name="path/to/outut.arrow"` and `function=None`. Having `function=None` makes the `map` write your dataset on disk with no data transformation.
Once you have your new arrow file, load it with `datasets.Dataset.from_file` to have a brand new Dataset object :)
In the future we'll have a better support for the fast filtering method from pyarrow so you don't have to do this very unpractical workaround. Since it breaks somes assumptions regarding the core behavior of Dataset objects, this is very discouraged. | I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function. | 191 | Saving processed dataset running infinitely
I have a text dataset of size 220M.
For pre-processing, I need to tokenize this and filter rows with the large sequence.
My tokenization took roughly 3hrs. I used map() with batch size 1024 and multi-process with 96 processes.
filter() function was way to slow, so I used a hack to use pyarrow filter table function, which is damm fast. Mentioned [here](https://github.com/huggingface/datasets/issues/1796)
```dataset._data = dataset._data.filter(...)```
It took 1 hr for the filter.
Then i use `save_to_disk()` on processed dataset and it is running forever.
I have been waiting since 8 hrs, it has not written a single byte.
Infact it has actually read from disk more than 100GB, screenshot below shows the stats using `iotop`.
Second process is the one.
<img width="1672" alt="Screenshot 2021-02-19 at 6 36 53 PM" src="https://user-images.githubusercontent.com/20911334/108508197-7325d780-72e1-11eb-8369-7c057d137d81.png">
I am not able to figure out, whether this is some issue with dataset library or that it is due to my hack for filter() function.
Hi ! This very probably comes from the hack you used.
The pickling line was added an a sanity check because save_to_disk uses the same assumptions as pickling for a dataset object. The main assumption is that memory mapped pyarrow tables must be reloadable from the disk. In your case it's not possible since you altered the pyarrow table.
I would suggest you to rebuild a valid Dataset object from your new pyarrow table. To do so you must first save your new table to a file, and then make a new Dataset object from that arrow file.
You can save the raw arrow table (without all the `datasets.Datasets` metadata) by calling `map` with `cache_file_name="path/to/outut.arrow"` and `function=None`. Having `function=None` makes the `map` write your dataset on disk with no data transformation.
Once you have your new arrow file, load it with `datasets.Dataset.from_file` to have a brand new Dataset object :)
In the future we'll have a better support for the fast filtering method from pyarrow so you don't have to do this very unpractical workaround. Since it breaks somes assumptions regarding the core behavior of Dataset objects, this is very discouraged. | [
-0.3835662007,
0.2207917422,
-0.0753483921,
0.2524214983,
0.1676334441,
-0.0859725475,
0.099591352,
0.2425321341,
-0.1649822891,
-0.0584909841,
0.0330234617,
0.4020827413,
-0.0848711729,
0.2319577932,
0.017286066,
0.2511723638,
0.1903818548,
0.0988867879,
0.1988590062,
0.1922226995,
-0.0973071456,
-0.0212856438,
-0.1073063761,
0.0238848515,
-0.3752512038,
-0.3421776891,
0.1020547673,
-0.1764685512,
-0.2869735658,
-0.3915144801,
0.0258450136,
0.2738235593,
0.0088606402,
0.4969803393,
-0.0001254971,
-0.2490328401,
0.2621250153,
-0.0777265951,
-0.3850048184,
0.1543707997,
0.1895152479,
-0.0919576958,
-0.0632524043,
-0.3167693913,
0.2532064319,
-0.084028855,
-0.0716400519,
-0.1480088979,
0.3763957918,
-0.0836272389,
0.0935049355,
0.1139937341,
0.0000267997,
0.389632225,
0.1677816063,
0.5001137853,
-0.0288457498,
0.0879573375,
0.3362616599,
-0.0670926124,
-0.1371208578,
0.2783449888,
-0.4867143035,
-0.1411878318,
0.2645100653,
-0.0966664329,
-0.1675315052,
-0.3948977292,
0.3739181757,
0.2074772269,
0.2618885338,
-0.5467395186,
-0.2214096785,
-0.3172277808,
-0.1709031314,
-0.5744588375,
0.3465483487,
0.3022805154,
-0.1129225791,
0.0919189453,
-0.3701658547,
-0.1759994179,
-0.1269788295,
0.0259890668,
-0.1026379168,
-0.0435412265,
-0.2155543268,
0.2533546686,
0.135345608,
-0.0528707244,
0.2719110847,
-0.3670329452,
0.0432741866,
0.1267949045,
-0.3480746448,
-0.0000374019,
-0.4093144238,
0.211443454,
0.1972281188,
0.1091900766,
-0.1748604029,
0.1767201424,
0.2158227861,
-0.02454165,
0.1190610081,
-0.0190226808,
0.0418545343,
0.1248178035,
0.1343435943,
-0.3414921463,
-0.1594240218,
0.0596138686,
0.0896654278,
0.038978491,
0.3420501351,
-0.4655123949,
0.072036393,
-0.1747670472,
-0.093022339,
0.273494333,
-0.1519592702,
-0.0404584557,
-0.0540088862,
0.3045741618,
-0.2286710143,
-0.174548462,
-0.130506143,
-0.0846112221,
-0.1934617907,
-0.0743906498,
0.026220642,
-0.0696874708,
-0.1318625659,
0.2878906727,
0.165646553,
-0.0967977345,
-0.0424889922,
-0.0020869114,
-0.2887095511,
-0.0223021805,
0.0240139812,
-0.5841045976,
0.4081670642,
0.4192660153,
0.0451572388,
0.3685526252,
-0.2013342381,
-0.1447465122,
-0.0956553817,
0.466476053,
-0.068585597,
-0.1822186708,
0.026173288,
0.0107505126,
-0.1690288484,
0.0607695691,
-0.5576754212,
0.0929525122,
0.4866019189,
-0.0743255392,
-0.0206744745,
-0.2441196293,
-0.3163403571,
-0.2649130821,
0.1202724874,
0.2557486892,
-0.6958670616,
0.2083716691,
-0.2188269198,
0.1379348934,
0.3864470124,
0.4180859327,
-0.0557119995,
0.2867589891,
-0.209656924,
0.4172908664,
0.2335224152,
0.280186832,
-0.6572467685,
0.0674072802,
-0.0122791603,
-0.0770692527,
-0.2024061382,
0.2173898369,
0.5708348751,
-0.1448688805,
0.0347845517,
0.1399071366,
0.0250297282,
0.022794202,
-0.3262348175,
-0.2562336624,
0.0538616665,
0.114188239,
-0.1550350487,
-0.2059198618,
0.001463782,
-0.2663286924,
0.3237321377,
0.0752421618,
0.3228730261,
0.3021536767,
0.268881321,
0.2784594893,
0.0033833496,
-0.323317647,
0.000104472,
0.0160899684,
0.1061025411,
-0.2561633885,
-0.4138256311,
-0.3335954249,
-0.0771506652,
0.3796462417,
-0.1549663544,
0.1034819707,
-0.0588507205,
0.0857050046,
0.0420031548,
0.1270332634,
-0.0741483122,
0.2337340266,
-0.2826394141,
-0.0001097657,
-0.1158529595,
0.0829700828,
0.0813679025,
-0.3049839735,
0.0064166784,
-0.0618372224,
0.1211540103,
-0.0209337734,
0.0009329952,
0.2589648962,
0.1091074795,
0.0138093792,
-0.1826982796,
0.1420536488,
0.0493601412,
-0.2159519345,
-0.005300899,
-0.2205975503,
0.3107071817,
-0.0643308312,
-0.128713578,
0.22958529,
-0.0448106527,
0.2971574366,
-0.0409098156,
-0.2173623741,
0.1645671725,
-0.0057498589,
0.0359959602,
0.0714560449,
0.07664074,
0.167894572,
0.2786031961,
0.2042236775,
-0.1401464194,
0.4174444377,
0.6351091266,
-0.2277690023,
-0.0020449888,
0.2752582431,
-0.0467979796,
-0.2161700726,
-0.0883962587,
0.0545100868,
0.2103078216,
0.1702260971,
-0.0626563728,
-0.1257079244,
0.0470383912,
-0.1102022901,
0.3756610751,
0.1766788661,
0.3735616803,
0.1716317236,
0.2496703416,
-0.1166418716,
-0.3490364254,
-0.0377164632,
0.2515731454,
0.3713638783,
-0.2111777067,
0.2102582753,
-0.3709143698,
-0.2967138886,
-0.0247602127,
-0.0927342623,
-0.1991699934,
-0.1360979229,
-0.1633739769,
0.2816186547,
0.0414894745,
0.2130254805,
0.2939043641,
0.2184885889,
0.0265320018,
-0.2598506808,
-0.3517385721,
-0.1167459488,
-0.3186903894,
-0.0677340999,
0.4512524903,
0.1509425491,
0.5623543859,
0.1044082046,
-0.1226968765,
-0.5189983845,
-0.2806612253,
0.1452243328,
0.0051961802,
0.4073173702,
-0.0914311484,
0.3319066167,
-0.3093451262,
0.0313943177,
-0.0224244837,
-0.1245276034,
-0.2509892583,
-0.0310690664,
0.0188747458,
-0.0936467201,
0.121101439,
-0.0205940641,
-0.3942090273,
0.0286189765,
0.2887546718,
-0.3527418375,
0.3154902756,
-0.2839595675,
-0.0240325779,
0.0274490379,
-0.0858806893,
-0.1939318925,
-0.2213402689,
-0.3033427894,
0.3021123707,
0.1005963087,
-0.2718930542,
0.1194723696,
0.0966609195,
-0.0227282904,
0.2998021245,
-0.4964137673,
0.310754776,
-0.3027848899,
0.1872269362,
0.0311479382,
-0.068636708,
0.4508423507,
-0.009657111,
0.0962056369,
0.0648790523,
-0.4937606454,
-0.0692878887,
-0.1915511787,
0.1666079611,
0.1861972958,
0.6179504991,
0.0621534884,
0.67126894,
0.0298435241,
0.0587154105,
0.1744911969,
-0.0372414961,
0.186165154,
-0.2480671257,
-0.3244780302,
0.0117503256,
-0.1864840984,
-0.2262554169,
0.0575555526,
-0.1923344433,
-0.1853729188,
0.2674933672,
-0.1907461733,
-0.3303471804,
-0.2938374281,
0.0812422782,
0.3420725465,
0.3041873276,
0.3475703597,
-0.0553810969,
-0.1528905034,
-0.3027535379,
-0.1691554934,
0.1518637091,
0.2205616683,
-0.0215394441,
-0.37341851,
-0.1219280064,
-0.5277500749,
0.2789166272,
-0.0333706923,
0.2458920181,
0.062991634,
0.0062376857,
0.3658822179,
-0.4122896791,
0.9922735095,
0.0536893085,
0.0677827373,
0.1915002316,
-0.1122046486,
-0.3611142337,
0.0798028558,
0.1436755359,
0.1254971921,
0.2814243436,
0.1921368688,
-0.3499977589,
-0.1675754189,
0.1012261286,
0.4437808096,
-0.2382509559,
-0.1989260912,
-0.1163625568,
-0.4190784395,
-0.3543669283,
0.2736727595,
0.0291243941,
0.1738575548,
0.1009289995,
-0.0984691158,
-0.2437911779,
-0.1644493341,
0.160319373,
-0.0630563349,
0.2521679699,
-0.1471995115,
0.3327231407,
-0.221974045,
0.1688555181,
0.4433542192,
0.0932507515,
0.0593428724,
0.1484787613,
0.2607511878,
-0.0668409988,
0.2663197517,
0.436547935,
-0.0338199809,
0.2442590892,
0.0333703011,
0.3509741127,
-0.2852977216,
0.1120600402,
-0.0206118822,
0.2415351272,
-0.6335097551,
-0.2218268216,
0.1848454624,
0.2303270102,
-0.143434152,
0.3211608827,
-0.0970836207,
-0.1378962547,
0.3642771542,
0.3419321775,
1.3057326078,
-0.4013215899,
0.1721461266,
0.1258505285,
-0.1251711845,
0.0361405872,
-0.3602975011,
0.0009562895,
-0.2630104721,
-0.0570890829,
-0.0198606253,
-0.1382505298,
0.0301307235,
0.1163861528,
-0.087732546,
0.0985184535,
-0.0142496452,
0.0629533231,
0.0384842679,
0.425583303,
-0.198021695,
-0.3105161786,
-0.1420005262,
0.0185981393,
-0.094105728,
-0.1785995364,
-0.0326000229,
0.0235857219,
0.1079504043,
-0.0005844608,
-0.2096890509,
-0.1743987203,
-0.3854607642,
-0.047823377,
-0.2607673407,
-0.0625416115,
0.124436222,
-0.1577664912,
-0.0354484618,
0.3297656476,
0.0586728901,
0.2710846364,
-0.0003178697,
0.1032804251,
0.0483014546,
-0.477030158,
0.1158821657,
-0.1223640367,
-0.0531711504,
-0.0253545009,
-0.1195459962,
-0.3043123186,
-0.1572328359,
0.0488528311,
0.6313863397,
-0.192030549,
-0.1679517031,
-0.1920223534,
-0.1115257442,
-0.2200875878,
-0.0146433348,
0.1289256811,
0.1512727141,
0.7495434284,
-0.3222370148,
-0.2972279787,
-0.2453293651,
0.3345035613,
0.2095810622,
0.0449315608,
0.3827111721,
0.1830311716,
-0.3001342416,
-0.2251049429,
0.2039883435,
-0.0447253659,
-0.2874954939,
0.3544316292,
-0.0558970422,
0.1160111576,
0.0527334511,
0.065396741,
0.0482635498,
-0.1968935281,
-0.2552433908,
-0.4429485202,
-0.3174458146,
-0.0582750514,
0.3190536499,
-0.0096050482,
0.344111681,
0.2328620702,
-0.0540824383,
0.1911132336,
-0.1861877739,
0.1707054377,
-0.0204901211,
0.261490047,
-0.1396443993,
-0.0499601662,
0.1205211878,
-0.1675787568,
-0.0861780196,
0.44233042,
-0.0704083219,
-0.132602185,
-0.054318402,
0.2066000849,
-0.1244501323,
-0.1713792384,
-0.1712510288,
-0.4104072154,
-0.0668743849,
-0.1763727367,
0.032145828,
0.1435270607,
-0.0098114386,
-0.7529995441,
0.2587637007,
-0.1436861604,
-0.1073426083,
0.0183959398,
-0.1062030122,
0.0172515661,
0.0108025223,
0.304266274,
0.1916282177,
-0.1061212346,
-0.2100139707,
-0.0201885961,
0.1744356304,
-0.1991692483,
0.2817910314,
-0.3842070103,
-0.2681799531,
0.1518538892,
0.2423109859,
0.6086739302,
0.0183300525,
-0.3965164125,
0.4168857932,
0.0804785341,
-0.1517600715,
-0.0787134469,
0.1096322834,
-0.1024196446,
-0.1905667186,
0.1284279078,
0.0097172707,
0.211280942,
-0.2009210438,
0.2207413614,
0.6288782954,
-0.2152167708,
-0.2538821697,
0.3068974912,
0.1504749656,
0.2075532824,
0.4781114757,
0.3338555396,
0.1287805438,
0.5598872304,
0.0369234569,
0.4385351539,
0.4149486423,
0.4798613191,
-0.0740119368,
-0.3737940788,
-0.0722604915,
0.4444384575,
-0.1859979928,
0.1629436612,
0.0284209289,
0.1739249825,
0.2753219903,
-0.5637278557,
-0.1189808398,
0.6395542622,
-0.1324682236,
0.0351375602,
-0.2842959762,
-0.12481305,
0.2628365457,
0.2157151103,
-0.0875985622,
-0.1145732924,
0.0797307938,
0.1870060265,
-0.0750070885,
-0.2530845106,
-0.0606536455,
-0.1116713434,
0.0507920794,
-0.1934710741,
0.0569864362,
-0.1657605171,
-0.040058732,
-0.2671602666,
0.1219037324,
0.4641970396,
0.3988073468,
0.0146514066,
0.1256381273,
-0.0709118992,
0.0146943033,
-0.1193475276,
0.3011447191,
0.1262504011,
0.2292539775,
0.2591571212,
0.004254967,
-0.0350129493,
-0.3886090815,
0.1917255968,
0.3899108768,
-0.005960986,
-0.0864132345,
0.10857521,
-0.1012050807,
-0.1906958669,
0.0765013546,
-0.3310647607,
0.3115874529,
0.7496414185,
-0.3366104066,
0.1730407029,
-0.0520020351,
0.0414757431,
-0.2468411475,
0.5603488684,
0.1954561472,
-0.0417719446,
-0.308290422,
-0.1963680983,
-0.5810982585,
-0.1768035442,
-0.1830585599,
0.0588071346,
0.0534623377,
0.1389306784,
0.1888882816,
0.1059294119,
-0.0484947786,
-0.0238037016,
-0.3080117702,
0.3683890104,
-0.182207033,
0.2372139841,
-0.0119722821,
0.2303600013,
-0.0999855846,
-0.4714619815,
0.6159066558,
-0.0906797945,
-0.0516414605,
-0.105369173,
0.0749473572,
0.0398689955,
0.1452368945,
0.3274846971,
0.2118892819,
0.0348480865,
0.0449148938,
0.3216600418,
0.0835496634,
0.3057259023,
-0.0545347482,
0.2176977694,
0.1881707162,
0.3021154106,
-0.0890843719,
0.2686498165,
-0.174184829,
0.1284355372,
-0.0778661221,
0.1572754085,
0.1311725676,
-0.036222972,
0.1362513155,
0.4410624802,
-0.0110283978,
0.3235625029,
0.1653806418,
0.3450426161,
-0.6625385284,
-0.2042626143,
0.5496144295,
-0.4917224646,
-0.440237999,
-0.0959463269,
0.2014236748,
0.4864785373,
-0.1716353744,
-0.3719890118,
-0.12934044,
0.3550191522,
-0.0672079772,
-0.3207605481,
0.2895029783,
-0.2783168256,
0.2445471287,
-0.1638460159,
0.4990862012,
-0.0096896533,
-0.3690646589,
-0.1437181532,
-0.2491854429
] |
https://github.com/huggingface/datasets/issues/1907 | DBPedia14 Dataset Checksum bug? | Hi ! :)
This looks like the same issue as https://github.com/huggingface/datasets/issues/1856
Basically google drive has quota issues that makes it inconvenient for downloading files.
If the quota of a file is exceeded, you have to wait 24h for the quota to reset (which is painful).
The error says that the checksum of the downloaded file doesn't match because google drive returns a text file with the "Quota Exceeded" error instead of the actual data file. | Hi there!!!
I've been using successfully the DBPedia dataset (https://huggingface.co/datasets/dbpedia_14) with my codebase in the last couple of weeks, but in the last couple of days now I get this error:
```
Traceback (most recent call last):
File "./conditional_classification/basic_pipeline.py", line 178, in <module>
main()
File "./conditional_classification/basic_pipeline.py", line 128, in main
corpus.load_data(limit_train_examples_per_class=args.data_args.train_examples_per_class,
File "/home/fp/dev/conditional_classification/conditional_classification/datasets_base.py", line 83, in load_data
datasets = load_dataset(self.name, split=dataset_split)
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/load.py", line 609, in load_dataset
builder_instance.download_and_prepare(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/builder.py", line 526, in download_and_prepare
self._download_and_prepare(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/builder.py", line 586, in _download_and_prepare
verify_checksums(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://drive.google.com/uc?export=download&id=0Bz8a_Dbh9QhbQ2Vic1kxMmZZQ1k']
```
I've seen this has happened before in other datasets as reported in #537.
I've tried clearing my cache and call again `load_dataset` but still is not working. My same codebase is successfully downloading and using other datasets (e.g. AGNews) without any problem, so I guess something has happened specifically to the DBPedia dataset in the last few days.
Can you please check if there's a problem with the checksums?
Or this is related to any other stuff? I've seen that the path in the cache for the dataset is `/home/fp/.cache/huggingface/datasets/d_bpedia14/dbpedia_14/2.0.0/a70413e39e7a716afd0e90c9e53cb053691f56f9ef5fe317bd07f2c368e8e897...` and includes `d_bpedia14` instead maybe of `dbpedia_14`. Was this maybe a bug introduced recently?
Thanks! | 75 | DBPedia14 Dataset Checksum bug?
Hi there!!!
I've been using successfully the DBPedia dataset (https://huggingface.co/datasets/dbpedia_14) with my codebase in the last couple of weeks, but in the last couple of days now I get this error:
```
Traceback (most recent call last):
File "./conditional_classification/basic_pipeline.py", line 178, in <module>
main()
File "./conditional_classification/basic_pipeline.py", line 128, in main
corpus.load_data(limit_train_examples_per_class=args.data_args.train_examples_per_class,
File "/home/fp/dev/conditional_classification/conditional_classification/datasets_base.py", line 83, in load_data
datasets = load_dataset(self.name, split=dataset_split)
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/load.py", line 609, in load_dataset
builder_instance.download_and_prepare(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/builder.py", line 526, in download_and_prepare
self._download_and_prepare(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/builder.py", line 586, in _download_and_prepare
verify_checksums(
File "/home/fp/anaconda3/envs/conditional/lib/python3.8/site-packages/datasets/utils/info_utils.py", line 39, in verify_checksums
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://drive.google.com/uc?export=download&id=0Bz8a_Dbh9QhbQ2Vic1kxMmZZQ1k']
```
I've seen this has happened before in other datasets as reported in #537.
I've tried clearing my cache and call again `load_dataset` but still is not working. My same codebase is successfully downloading and using other datasets (e.g. AGNews) without any problem, so I guess something has happened specifically to the DBPedia dataset in the last few days.
Can you please check if there's a problem with the checksums?
Or this is related to any other stuff? I've seen that the path in the cache for the dataset is `/home/fp/.cache/huggingface/datasets/d_bpedia14/dbpedia_14/2.0.0/a70413e39e7a716afd0e90c9e53cb053691f56f9ef5fe317bd07f2c368e8e897...` and includes `d_bpedia14` instead maybe of `dbpedia_14`. Was this maybe a bug introduced recently?
Thanks!
Hi ! :)
This looks like the same issue as https://github.com/huggingface/datasets/issues/1856
Basically google drive has quota issues that makes it inconvenient for downloading files.
If the quota of a file is exceeded, you have to wait 24h for the quota to reset (which is painful).
The error says that the checksum of the downloaded file doesn't match because google drive returns a text file with the "Quota Exceeded" error instead of the actual data file. | [
-0.2483564168,
0.35080567,
-0.1227643862,
0.262452811,
0.1046654582,
-0.0048443824,
0.3205700517,
0.5194305182,
-0.087741822,
-0.0413589478,
0.0776253417,
-0.1982266605,
0.0031297319,
0.3474114835,
-0.244494468,
-0.009974461,
0.0986697376,
-0.154382214,
-0.0066189021,
0.0506975278,
-0.1762790382,
0.0808000043,
-0.2772865593,
-0.306119591,
-0.002176322,
0.2340682149,
0.0149133429,
-0.1259736866,
-0.001067277,
-0.2701269984,
0.5489286184,
0.1795444638,
0.023815088,
0.4143281877,
-0.0001126272,
-0.0021211058,
0.3832173944,
-0.0588618591,
-0.3282299638,
0.1053186655,
-0.31564188,
-0.1499087512,
-0.0790164471,
-0.08250691,
-0.1530303657,
-0.5050343275,
-0.2010856718,
-0.1114247963,
-0.1040297225,
0.2618599832,
0.217597425,
0.2089744806,
0.0917270109,
-0.0789879858,
-0.1643638462,
0.0381903201,
-0.0346373059,
0.0000729915,
0.1079570055,
-0.1750724018,
0.0917512327,
0.1509218365,
-0.1619755924,
0.0288009867,
0.1156820059,
-0.2075261027,
0.0970108509,
-0.2640064359,
0.2317044735,
0.2405157536,
0.2926249504,
-0.1287696809,
-0.3720597625,
-0.297703743,
-0.0750020593,
-0.2456415594,
0.2302349508,
0.1716050506,
-0.0673036054,
-0.0363246985,
-0.3061077595,
0.2523925602,
0.1045452058,
0.0709877759,
-0.1234945357,
0.149882257,
0.0577275269,
-0.0749312118,
0.106335789,
-0.041352056,
0.6009082794,
-0.1663403213,
-0.2048627436,
0.0807093233,
-0.4355801642,
-0.1634500325,
0.0560004264,
0.6840983629,
0.1832090914,
0.4219740927,
-0.1995081306,
0.1657821089,
-0.1209181175,
-0.0922685415,
0.1504205614,
-0.0343499184,
-0.0519278683,
0.4296019375,
0.2829833031,
0.3001468182,
0.128410846,
0.1946968287,
0.1358948946,
-0.2661389709,
0.2309339345,
0.0963650048,
0.0433473289,
-0.2907248735,
-0.2783107162,
0.2760262787,
-0.1270308793,
-0.138053298,
0.2978389561,
0.2877977192,
-0.4330212176,
0.1959684342,
0.0082534626,
0.0210148692,
0.044229269,
-0.251160562,
-0.1368150711,
-0.1471278071,
-0.0469174385,
0.1370758861,
0.118085824,
0.0314625949,
0.2896966338,
-0.0034998879,
0.1563294232,
-0.1104827672,
0.2000197023,
-0.3637543619,
0.2052348554,
0.517328918,
-0.0956809744,
0.208198607,
0.0894038081,
-0.3298083544,
-0.0150807574,
-0.102418378,
-0.2392274886,
-0.1406961977,
-0.3736909628,
0.2286853492,
-0.3321994841,
0.0105252229,
-0.5877591372,
-0.2102765292,
0.3507716954,
-0.2136425972,
0.075219363,
-0.3401680589,
-0.2169395685,
-0.1177301332,
0.3215621114,
0.3882475197,
-0.2872912288,
0.1926520616,
-0.505841434,
-0.0116741937,
0.1311635077,
0.0566375293,
-0.3111608028,
0.3517678976,
-0.3844764531,
-0.2458634377,
0.0601756126,
-0.5608973503,
-0.3146587312,
0.1328957379,
0.1920130253,
0.3282433748,
-0.1377325207,
-0.2058591694,
0.068196021,
-0.2593607903,
0.1219717339,
0.0948207974,
0.0682111979,
0.158423394,
-0.2493801862,
-0.0130493864,
0.2194643021,
0.1278645396,
0.1831165999,
0.0032977574,
0.0568598509,
0.1512577981,
0.2725537717,
-0.1036928147,
-0.0148236584,
0.1306651235,
0.2006483227,
0.0827575177,
0.0509695038,
-0.0756182671,
-0.3890631795,
0.1855843812,
-0.2650210261,
0.0332781188,
-0.1553547084,
-0.0455866382,
-0.1992483437,
-0.1255377233,
-0.429751724,
-0.3456072807,
0.2111904472,
0.2249654531,
-0.1547850072,
0.153678298,
0.0469173789,
0.1888699979,
-0.386210978,
-0.0471784882,
-0.0983200818,
0.3950824142,
-0.0024917498,
0.0562325343,
0.0013946518,
0.0921591669,
0.1276094466,
-0.0379030034,
-0.0082376543,
0.4010096192,
0.3606114686,
0.0389327034,
0.1908299327,
0.351999402,
0.2566122115,
-0.2802023292,
-0.264305979,
0.3049584329,
0.1222957671,
-0.0713121295,
0.1305746585,
0.3840273023,
-0.2733517289,
0.0554502942,
0.1262115389,
0.0377751291,
0.4955727458,
-0.1079362035,
-0.23691836,
-0.2576626539,
0.1765875667,
0.1049607843,
0.0675177425,
0.1393898129,
-0.0053885952,
0.0798842907,
0.2055532336,
-0.080689393,
-0.1177331209,
-0.0426338091,
0.1219715104,
-0.0356093906,
0.0879072472,
0.751088202,
0.4312457144,
-0.0813289732,
0.0834226161,
0.1749543846,
-0.0722505897,
-0.0217373297,
0.070721209,
0.078365624,
0.1058776677,
0.3351550996,
0.0585348345,
0.0462718569,
-0.1494187862,
-0.2088984102,
0.173096925,
0.3105881512,
-0.3180756569,
0.0239008889,
-0.3514174223,
-0.2420356572,
-0.3531622589,
-0.1688135564,
-0.2286225855,
-0.4381591678,
-0.0157748759,
0.3220817745,
0.0368192419,
0.2074559629,
-0.449282974,
0.0925887004,
0.1287261397,
-0.2775694132,
0.0639830083,
0.1320364177,
-0.0243531596,
0.0451561362,
0.5737255216,
0.2313752919,
0.2702274024,
-0.3618480861,
-0.129732132,
-0.5231909156,
-0.3683935404,
0.0594918802,
0.0910139382,
0.3205639124,
0.1584274471,
0.1651458591,
0.1059408933,
-0.3236397207,
0.0281089395,
0.3177773654,
-0.0200343505,
-0.0149844177,
-0.0636157095,
0.0657199696,
0.0275882632,
0.0795960352,
-0.256221801,
-0.0792466477,
-0.2330018282,
0.1119934246,
0.0559556112,
0.0892031491,
-0.0778989419,
0.1029611677,
-0.0143481679,
0.0297972336,
-0.1901175231,
-0.3761087656,
0.1693157852,
-0.0902888626,
-0.4417481422,
-0.0958379656,
-0.0300431401,
0.3719251752,
-0.0784174576,
-0.3715617061,
0.0062591769,
-0.1335083842,
0.1186245978,
0.2444374114,
-0.384608537,
0.2823685408,
0.0073238648,
0.016855523,
-0.1849657893,
0.0201205984,
-0.0567025132,
-0.2161253691,
0.5088827014,
-0.1245596185,
-0.1940551996,
0.0573148057,
0.3122808635,
0.4735924602,
-0.1680440903,
0.2759517133,
0.0246085171,
0.4678418636,
-0.1390939951,
-0.4603391886,
-0.1930563748,
-0.0950100273,
-0.0335224569,
-0.0393402651,
-0.2474112064,
-0.1724463403,
-0.137442708,
0.0313695222,
-0.3403827548,
-0.2995960414,
-0.1047632098,
-0.3917803466,
0.2362752855,
0.0033103004,
0.091602087,
0.1571808755,
-0.3026347756,
0.1535651684,
0.3441469669,
-0.150244832,
-0.1311244816,
-0.4588466287,
-0.0238260422,
0.2610837221,
0.168689087,
0.0576810129,
0.5540508032,
-0.0254557505,
0.0201966763,
-0.227448076,
-0.0330082029,
0.3528865874,
-0.513967514,
0.3113611341,
-0.013579797,
0.0258567333,
-0.0678248033,
-0.2351484746,
0.0483183116,
-0.1841360331,
0.1932885051,
0.5447886586,
-0.048257716,
-0.126422286,
0.2902932167,
0.3781236112,
-0.1861048341,
-0.0063398741,
-0.2953608632,
-0.1861953735,
-0.0172108002,
0.097582072,
0.1861977726,
0.2329003364,
-0.3318364024,
-0.1802389324,
-0.0489784665,
0.0665419847,
0.022489816,
0.0602440052,
0.2055639923,
0.0275994204,
0.3083540499,
0.1284882426,
-0.138871491,
0.774851501,
0.2862087786,
-0.2472216189,
-0.2631367445,
0.0567612015,
-0.1023990139,
-0.0323166586,
0.0993444324,
-0.1670053452,
-0.0142989494,
0.3959566951,
-0.092776835,
-0.1880432516,
0.0810124874,
0.2037274987,
-0.0244848039,
-0.4803302884,
-0.520072937,
-0.125157401,
0.1447145343,
0.0110958368,
0.4486394525,
0.0692510605,
-0.2791031301,
0.3141691387,
0.0963628218,
0.4851429164,
0.1838278025,
-0.0944213271,
-0.070691973,
-0.0477723442,
-0.0933509693,
-0.1017455161,
-0.0060382467,
-0.3348732293,
-0.6285867691,
-0.1254465431,
-0.1433282048,
0.2475312352,
-0.2586276829,
0.1138378158,
0.365952462,
-0.0879791379,
0.0187847111,
0.024982471,
0.0654745847,
-0.0623461977,
-0.0027056355,
-0.3052215278,
0.135548681,
0.0379382558,
0.1731661111,
-0.1275609434,
0.0060585737,
-0.1111587882,
-0.1593548357,
-0.5275296569,
0.0831995159,
0.0327204801,
0.0402565338,
0.211758852,
0.0777601898,
-0.208278358,
0.5434346199,
0.2335903645,
0.3035466075,
-0.1847198755,
0.1648022234,
0.06131294,
0.2169542909,
0.0758585632,
-0.2602528334,
0.539942801,
-0.1387322694,
-0.2067428976,
-0.0027814228,
-0.1383003294,
-0.320225805,
0.111430347,
0.0230444651,
-0.1734693944,
-0.1759434491,
-0.021959763,
-0.1599994153,
0.1275669634,
-0.2879591584,
0.1467721313,
0.0805837661,
-0.0266170539,
0.1897022873,
0.1364037842,
-0.3712583184,
-0.0933347642,
0.6231889129,
-0.037038736,
-0.1861688793,
0.355568707,
-0.0803419501,
-0.1102820188,
-0.2094679773,
0.1944052428,
0.2133214325,
-0.3284568191,
-0.162713334,
-0.4114900529,
0.2058758885,
0.0871243775,
0.4304662049,
0.2160064578,
0.0801108629,
0.1100862473,
-0.7900108099,
-0.0305310562,
0.0986932516,
-0.1488038152,
0.107732445,
-0.1543703079,
0.1304600537,
0.0958675593,
-0.2670512795,
-0.2456793487,
0.0759126693,
-0.4292775691,
0.0968830734,
-0.3269151449,
0.0146332635,
0.2100940347,
-0.0985255986,
0.1508654356,
0.0151351374,
-0.059697777,
-0.1588755697,
-0.2603532374,
0.1143745631,
0.20526281,
-0.1401932687,
0.1857996583,
-0.1429643333,
-0.0799992979,
-0.1082237065,
-0.0873484313,
0.0758005008,
0.070032157,
0.1423604786,
0.3404821455,
0.0620092489,
-0.0689226761,
-0.2004764378,
-0.0650644377,
-0.0370064303,
0.2050034404,
0.1779526174,
-0.069851011,
-0.0259231851,
0.0382796377,
0.0719283521,
-0.2412769496,
0.1140168607,
0.330791086,
-0.2690096498,
0.0512698144,
0.4562601745,
0.5259580016,
0.490016371,
-0.4191826582,
-0.1510872841,
0.3108332157,
0.1734562218,
-0.334433794,
-0.1327140778,
0.3407928646,
0.19404836,
-0.0793225393,
0.1123547703,
0.3362830281,
-0.3540165126,
0.3828051984,
0.105366312,
0.1520604044,
-0.5242891312,
0.5086354017,
0.4963036776,
-0.3074858785,
0.1020005941,
0.2030337304,
-0.1898430437,
-0.1762793958,
0.55033499,
-0.0935349464,
0.3178267181,
-0.1157833487,
0.0076929941,
-0.0532062314,
-0.5167145133,
0.4916115403,
0.1496074796,
-0.0728207827,
-0.0603761114,
0.0231313668,
0.0525234342,
-0.1056097895,
0.0569681041,
-0.3598282933,
-0.0102922451,
-0.0256155636,
0.0309299119,
0.1982893348,
-0.3819049001,
-0.1348474026,
0.1083564162,
0.1899414212,
0.024669297,
0.1944096386,
0.091519542,
-0.5324067473,
-0.4129909575,
0.1037877053,
-0.0287562255,
0.0480684936,
-0.1690663099,
0.1855584383,
0.2086616606,
-0.0036589857,
-0.0355610028,
0.1527328938,
0.2425486743,
0.4049271941,
0.242044732,
-0.1356984675,
-0.0280001499,
0.079108566,
0.0403712839,
0.4793929756,
0.223982811,
0.1611831486,
0.3548165858,
0.1577808261,
-0.1032980084,
0.1443248093,
0.0364899933,
0.2156118453,
-0.2185447812,
0.5506975651,
0.0730590373,
0.1857667714,
-0.2539485693,
0.1162547097,
-0.0487988964,
-0.0968164653,
0.3136622012,
0.2028796524,
0.1977776587,
-0.0651572645,
0.0681771934,
0.1485725939,
0.4755244851,
0.2595830858,
-0.1090378836,
-0.1856628656,
-0.3761807382,
-0.728256464,
0.1537903696,
-0.0555228144,
0.5085517764,
0.0304752178,
0.1653368175,
-0.1357874274,
0.136076957,
0.3877426386,
-0.4676199555,
0.2673336267,
0.0702679604,
-0.1864009202,
-0.0820701867,
-0.2114672512,
0.0746219903,
-0.0708475634,
-0.2945508659,
0.2579194009,
-0.0986898392,
0.1454252303,
0.1770617068,
-0.5136871338,
0.020894058,
0.0947820544,
0.2509295344,
0.0693166479,
0.7262421846,
0.010047026,
-0.1033883244,
-0.4000114799,
-0.0479570851,
0.0073249266,
0.5848953128,
0.096426636,
0.0247241855,
0.1275391877,
-0.0883764625,
-0.3506603241,
0.4802133739,
0.0932057351,
-0.3719084263,
-0.4586521387,
-0.0097273774,
0.1669661403,
-0.0739348531,
-0.0917092189,
-0.154352203,
-0.1256268024,
0.4537493587,
-0.0129288882,
-0.234623313,
0.3987482786,
-0.291744113,
0.0939346775,
-0.1376037151,
0.2120365798,
-0.1579275429,
-0.220851481,
-0.6547482014,
0.0706729591,
0.4977369905,
0.0051652826,
-0.0674567968,
0.0658463389,
-0.177076295,
0.3712736964,
-0.0562771633,
0.4282297492,
-0.0078678653,
-0.2420660406,
0.2888297439,
0.0225235447
] |
https://github.com/huggingface/datasets/issues/1906 | Feature Request: Support for Pandas `Categorical` | We already have a ClassLabel type that does this kind of mapping between the label ids (integers) and actual label values (strings).
I wonder if actually we should use the DictionaryType from Arrow and the Categorical type from pandas for the `datasets` ClassLabel feature type.
Currently ClassLabel corresponds to `pa.int64()` in pyarrow and `dtype('int64')` in pandas (so the label names are lost during conversions).
What do you think ? | ```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints. | 69 | Feature Request: Support for Pandas `Categorical`
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints.
We already have a ClassLabel type that does this kind of mapping between the label ids (integers) and actual label values (strings).
I wonder if actually we should use the DictionaryType from Arrow and the Categorical type from pandas for the `datasets` ClassLabel feature type.
Currently ClassLabel corresponds to `pa.int64()` in pyarrow and `dtype('int64')` in pandas (so the label names are lost during conversions).
What do you think ? | [
0.0230096057,
-0.0102735162,
-0.142605871,
0.2076648176,
0.2370128185,
0.167278409,
0.1267410368,
0.2487182319,
-0.0738425702,
-0.2155073881,
0.1673995703,
0.2824070752,
-0.2035115063,
0.4252622128,
0.1327463239,
-0.2236520052,
0.124682799,
0.0284452848,
0.0778508633,
0.3117169738,
-0.3771745563,
-0.0838787034,
-0.0995520055,
0.238972649,
-0.1983464956,
-0.1293814182,
-0.4564826488,
-0.3532524407,
-0.2593736649,
-0.8477247953,
0.3983207047,
0.3807969391,
0.1931586713,
0.1735819578,
-0.000115819,
-0.307919383,
0.2058766037,
-0.04252363,
-0.0503572747,
-0.443198204,
-0.29180637,
-0.2464247942,
0.3244949579,
-0.2901012897,
-0.2068760246,
-0.2925748229,
0.0067170095,
-0.2727672458,
0.2317004055,
0.0666810721,
0.166751653,
-0.0497689843,
-0.027442582,
-0.1615714729,
0.131834507,
0.4021802843,
-0.2371010184,
0.2846889496,
0.5595126152,
-0.3222315311,
0.2314839363,
0.0884721205,
0.0117629617,
-0.32495749,
0.6005215645,
0.0490310155,
0.2645990849,
-0.5582160354,
-0.3079172671,
0.1526242793,
0.3408334851,
-0.3584246039,
-0.5394774079,
-0.4483573437,
-0.0119226612,
-0.2346875072,
-0.0061523803,
-0.1371360123,
-0.1556821764,
0.1616318524,
0.1122617945,
-0.0731579438,
-0.2399316877,
0.1873063892,
-0.084433198,
0.1476807743,
-0.044365447,
-0.1675475389,
0.2131358534,
-0.0995403528,
0.0033084266,
-0.1552575678,
0.0259757079,
0.6048637629,
-0.2247824073,
-0.1363117397,
0.0729947686,
0.2108624876,
0.0177887529,
-0.008306751,
0.100210093,
0.0134376511,
-0.2314001918,
-0.0256742202,
0.3791762292,
0.2924356461,
0.1800874323,
-0.0863290057,
-0.128845185,
0.3581373096,
-0.0041657537,
0.0536367223,
-0.1282648444,
-0.1226121485,
-0.1723236144,
0.1386384964,
-0.070712693,
0.2454465479,
-0.0304598138,
-0.0372078195,
-0.0177116916,
-0.0263489913,
0.0152255632,
0.2628427744,
-0.0659547299,
0.0928889215,
-0.0641005933,
0.0835101455,
0.2764410079,
0.1206505597,
-0.0680940449,
0.0020983815,
-0.2102169394,
0.1605478823,
-0.0775293857,
-0.1858754158,
-0.1492253244,
-0.0344319344,
0.1192694008,
0.1085183024,
-0.2276077121,
-0.0936202854,
0.1229104847,
0.0074044154,
-0.4257947803,
0.1622293591,
0.2187953591,
-0.0850058496,
-0.3504914641,
0.1319193989,
0.1542150229,
-0.2472275198,
-0.23920542,
0.1113658696,
-0.2031581253,
0.1457659006,
-0.3309437335,
0.4760414362,
-0.0080963634,
0.0497262031,
0.2327816039,
0.0547784343,
-0.2638413906,
-0.3284742236,
-0.1710062325,
0.4522376359,
-0.3215055466,
-0.2309906036,
0.0531621315,
-0.0306911841,
0.036155507,
0.0772850364,
-0.1583129764,
0.119236365,
-0.0716489851,
0.1776525676,
0.5333982706,
-0.1753787398,
-0.1190231889,
0.3577376902,
0.0055855513,
0.1976587772,
0.2812624872,
0.2622911036,
0.1901308447,
-0.2151020765,
0.0931512266,
0.1283134222,
-0.1040708944,
-0.2355067581,
0.1491523683,
-0.330439508,
0.1761065722,
0.1930087656,
-0.0684678257,
0.2835514545,
0.074393481,
-0.3360265195,
0.3749299049,
-0.2894715965,
0.1078302935,
0.0636767447,
0.3066045046,
0.4409076571,
0.0205316953,
-0.4798708558,
-0.1257181019,
0.3306059539,
-0.0952281058,
0.1742292345,
-0.3567067981,
-0.4952705204,
-0.0397887304,
0.2739691734,
-0.2659195662,
0.1416191757,
-0.009122286,
-0.2109058201,
0.2343252301,
0.2042149752,
-0.3539478779,
0.5252866745,
-0.0650036484,
0.2001338601,
-0.2408440411,
0.1756787002,
-0.0458534136,
0.0768375844,
-0.1055376381,
0.3452533185,
-0.0846972913,
-0.1197187603,
0.2849960923,
0.0615111254,
0.2428296357,
-0.094621025,
-0.3645077646,
0.2292647958,
0.5264254808,
0.0491378531,
-0.0908040553,
-0.1106584221,
0.084387064,
-0.0667051226,
0.0166786388,
0.5823572874,
-0.4817156792,
0.2271655947,
0.1415825486,
0.1412526071,
0.0998749509,
0.2575954795,
-0.2499680966,
-0.1974931359,
0.1438123286,
0.2578537166,
-0.2217442542,
0.0667167678,
-0.1658439934,
0.1915976405,
0.3040450811,
-0.1511571109,
0.1042438075,
0.1038230807,
0.1065930277,
-0.170360297,
0.2263922393,
0.1434835196,
0.2757692337,
0.0754673481,
0.0635722429,
0.1523474008,
-0.1569282115,
-0.0685347915,
0.3644998372,
0.027278306,
-0.085646525,
-0.2118578553,
0.0292043351,
0.0771163404,
0.0599483252,
-0.3495514095,
-0.0224031359,
-0.0369301066,
-0.1203359812,
0.1702975333,
-0.1160881072,
-0.5219680071,
0.14127177,
-0.5717269778,
0.0078103691,
-0.1792369187,
-0.0621801354,
0.0817524418,
0.0744456798,
0.1140424013,
-0.3413845301,
-0.035435766,
0.1047069058,
-0.2842971385,
-0.2089478821,
-0.4510959089,
0.2566653788,
0.0999105871,
-0.0136358291,
-0.1809959859,
0.2080219984,
0.0345665291,
-0.031090185,
-0.2123506963,
-0.279039979,
0.0493884236,
-0.1699927449,
0.026470961,
0.2983823717,
-0.2530080378,
-0.1749663949,
-0.2625125647,
0.1982941329,
0.1198827177,
-0.1365109533,
-0.2098923624,
-0.1893090755,
0.0895449966,
-0.1031984016,
0.102033332,
-0.4095128775,
-0.3784289956,
0.6761474609,
0.2192849815,
0.1426875591,
0.1680995673,
-0.0058761388,
0.0730163231,
-0.0896326676,
-0.1047140434,
-0.1668846905,
0.2988792658,
0.2073049545,
-0.1608502716,
-0.3180471957,
-0.0798368454,
-0.3646600246,
0.1057947725,
0.2121287882,
-0.1410654038,
-0.3041149974,
0.0048827827,
0.5083075166,
0.1135893539,
0.0284398738,
0.2844409049,
0.2777574658,
0.0575978793,
0.1198546216,
0.0421051756,
0.0113997199,
0.1779200137,
-0.2589244843,
0.2157592922,
0.4879330099,
0.1209345162,
0.2910845578,
0.0937248617,
-0.0136831412,
0.495490849,
-0.1015296653,
0.2374846935,
-0.1354481876,
-0.2925467491,
-0.3849199414,
-0.3812012076,
0.087369144,
0.0922508314,
-0.1177487299,
-0.1566376537,
-0.2011427283,
-0.0049428493,
0.1318277121,
-0.0870388895,
-0.026894195,
-0.3554051518,
0.4113956392,
-0.0421571992,
-0.1222861707,
-0.4082294703,
-0.0289857928,
-0.0885422379,
0.1240104064,
0.3442617357,
-0.1650989652,
-0.3631923497,
-0.1629422605,
-0.4993349314,
0.2302028537,
0.6471201181,
-0.142465055,
0.0527762696,
-0.0200610384,
-0.0053712726,
0.1159426942,
0.5606773496,
-0.0718633756,
-0.337580204,
0.2686216533,
-0.4713848233,
-0.3781881332,
0.3169171214,
0.0455298573,
0.0441104062,
-0.015689645,
0.1021610573,
-0.0899454951,
-0.2031951547,
-0.0265191142,
-0.0172946341,
0.0012238696,
-0.0552296191,
0.0234432667,
-0.0369385332,
-0.1320167482,
-0.17588754,
0.4480683804,
0.1625965536,
0.0418879762,
-0.1095062047,
-0.2154186666,
0.0515595973,
0.2643418312,
-0.0085914135,
-0.0637493804,
-0.059256915,
0.1627685279,
0.0402164906,
0.0491838306,
0.1073397994,
0.2975521088,
-0.1458133608,
-0.5790701509,
-0.03947616,
0.0585529618,
0.4294226468,
0.3097957969,
-0.0215763208,
0.301440984,
-0.0531078503,
-0.0325051583,
-0.3707031608,
-0.2006964833,
0.3816449642,
-0.0846342146,
-0.3441472054,
-0.7334688902,
0.35093382,
0.1426516175,
-0.3018722534,
0.1473785639,
0.3631759882,
-0.5190954804,
0.293822825,
0.0408901721,
0.8400940895,
-0.092752777,
0.3819608688,
0.3813298345,
0.309545815,
0.7505722046,
0.27758044,
0.0987700075,
-0.3746115565,
-0.2893297076,
-0.3023834527,
-0.1062902287,
0.2692961097,
0.3666597009,
-0.3419591784,
0.3305616081,
-0.1324574798,
0.2882317305,
0.1098045334,
-0.1822686493,
-0.118922472,
-0.2034188211,
-0.725779295,
0.0480664074,
0.452559948,
-0.0365069583,
-0.2351938784,
-0.2062584609,
-0.1167572141,
0.0931269601,
-0.2607622147,
-0.2176082134,
-0.4321040213,
0.1444188654,
-0.1006665975,
0.1561187804,
0.2696221173,
0.4705724716,
0.3740429282,
-0.3842305839,
0.0270238128,
-0.1776851863,
-0.1497763991,
-0.1718023121,
-0.0811987668,
-0.0599162541,
0.4753719866,
0.033637926,
0.0971908569,
0.3608367443,
0.0548866466,
-0.2822109461,
-0.3045468628,
0.2037368268,
-0.0772629008,
-0.2618442178,
-0.2907629609,
-0.2406387925,
0.0460179001,
-0.1168738157,
0.0699091107,
0.2150896788,
-0.036970634,
0.1649360061,
-0.353825599,
0.1511006504,
-0.2132050991,
0.1253352314,
0.2026966512,
-0.0330267884,
0.107727021,
0.1574064493,
-0.2605878711,
-0.0517194532,
-0.005398497,
0.2587835193,
-0.4805978537,
0.1313625127,
0.2005819976,
-0.2650959492,
-0.127681613,
0.5652498007,
0.2517307401,
0.2128714323,
-0.0211028159,
-0.1289939582,
-0.2973943353,
0.3678024113,
0.1430946141,
0.24398911,
-0.1040261313,
0.13226524,
-0.18524912,
-0.1136159524,
-0.2206496894,
-0.0021181032,
-0.0052059069,
-0.0675509572,
0.4329421222,
0.0207921062,
0.1636744887,
0.2392760515,
0.1212814748,
0.2636921406,
-0.1925225705,
-0.1320178509,
-0.0871847421,
0.1703846902,
-0.0201828629,
-0.0114977956,
0.0209326074,
-0.0894580856,
-0.2565030158,
-0.2649770677,
0.0943428278,
0.2512863874,
0.1210569441,
-0.044832997,
0.3994724154,
-0.0346533991,
-0.1810764968,
0.3869204223,
0.0380891152,
0.4918481112,
0.0317568667,
0.0474309064,
-0.3425094485,
0.0545477271,
0.0307745598,
0.3291056156,
-0.0392378531,
-0.4100054801,
0.640925169,
0.0848435909,
-0.1820202172,
0.1968025118,
0.525116384,
0.2113812417,
0.1651662588,
0.0732802153,
0.1216477305,
0.0837515891,
-0.333778441,
-0.3142721951,
0.1178405583,
-0.1556552649,
0.171490863,
0.5202996731,
0.3050470948,
0.2484452128,
0.1247213334,
-0.1429638267,
0.2160944343,
0.1981442869,
0.0745486766,
0.3195813298,
0.3900642395,
-0.0091482699,
0.3561512828,
0.3804426789,
0.5041667819,
0.1832934767,
-0.0932920873,
0.0783443078,
0.0882703066,
-0.0172893927,
0.604726553,
0.0295800418,
0.0089014219,
0.4108659625,
0.0399432816,
-0.008216396,
-0.0557899103,
0.0577750057,
-0.1989379376,
0.0129612088,
-0.0629524663,
0.2387242019,
0.216109097,
-0.1524890214,
0.4791820645,
-0.1766899526,
-0.128983885,
-0.3476144671,
-0.1384477764,
-0.2974571586,
0.4343165755,
-0.130851984,
0.5650552511,
-0.0634815097,
0.1978338063,
-0.079186067,
0.3597237468,
-0.0898835137,
-0.0796708241,
-0.0958933383,
-0.0673904866,
0.0075929407,
0.2095281184,
0.4526984096,
0.1809923351,
-0.1542154402,
0.1083930284,
-0.426173389,
-0.2367449254,
-0.0348733291,
0.2622196376,
0.1265500635,
0.0943510532,
0.2722555399,
0.053081356,
-0.0310171098,
0.0222227406,
0.0101431422,
0.2748625875,
-0.3390368521,
0.7011252046,
-0.2181458771,
-0.235043779,
0.1938221306,
0.3255452216,
-0.3201076388,
0.0481974073,
0.0591657236,
-0.1691699624,
0.1492702365,
0.1201525033,
0.0416861959,
-0.0675470382,
0.1855738759,
0.1969930232,
0.4075562656,
0.0107359998,
-0.2408181429,
-0.4748311043,
-0.1409270167,
0.2547972202,
-0.2541851103,
0.2475463152,
-0.0389153138,
-0.023624666,
0.0910675228,
-0.081563279,
-0.3627572656,
-0.1309396923,
0.2148239762,
-0.0924036801,
-0.276786238,
0.1841605306,
0.1239571422,
-0.1801173091,
-0.2226979285,
0.0785782933,
-0.2537125945,
-0.0861013159,
-0.1989879161,
0.143423304,
-0.1959644854,
0.0455566198,
0.2694398463,
0.2227223068,
0.1807645857,
0.2077306956,
-0.169336617,
-0.0346202068,
-0.3444250226,
0.0613576137,
0.5801239014,
-0.0249673594,
0.3234555721,
-0.3697853088,
-0.275303036,
-0.0232930928,
0.2424985617,
0.2755201757,
-0.193352446,
-0.5966460705,
0.0874608532,
-0.4430661201,
-0.0298177488,
0.1695705801,
0.3679268062,
-0.1307650208,
0.3309235573,
0.1300219595,
-0.2077973038,
0.3422711492,
-0.4958491027,
-0.2886919379,
0.035095796,
-0.2006174475,
0.3782950938,
-0.1855074614,
-0.1591356397,
-0.0128235519,
0.4850820899,
0.1197654679,
-0.1062525436,
0.174354434,
-0.2196150422,
-0.1083699167,
-0.3893577456,
0.0817013681,
-0.2865291834,
-0.0213865116,
0.0420440584,
-0.4636684656
] |
https://github.com/huggingface/datasets/issues/1906 | Feature Request: Support for Pandas `Categorical` | Now that I've heard you explain ClassLabel, that makes a lot of sense! While DictionaryType for Arrow (I think) can have arbitrarily typed keys, so it won't cover all potential cases, pandas' Category is *probably* the most common use for that pyarrow type, and ClassLabel should match that perfectly?
Other thoughts:
- changing the resulting patype on ClassLabel might be backward-incompatible? I'm not totally sure if users of the `datasets` library tend to directly access the `patype` attribute (I don't think we really do, but we haven't been using it for very long yet).
- would ClassLabel's dtype change to `dict[int64, string]`? It seems like in practice a ClassLabel (when not explicitly specified) would be constructed from the DictionaryType branch of `generate_from_arrow_type`, so it's not totally clear to me that anyone ever actually accesses/uses that dtype?
- I don't quite know how `.int2str` and `.str2int` are used in practice - would those be kept? Perhaps the implementation might actually be substantially smaller if we can just delegate to pyarrow's dict methods?
Another idea that just occurred to me: add a branch in here to generate a ClassLabel if the dict key is int64 and the values are string: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L932 , and then don't touch anything else.
In practice, I don't think this would be backward-incompatible in a way anyone would care about since the current behavior just throws an exception, and this way, we could support *reading* a pandas Categorical into a `Dataset` as a ClassLabel. I *think* from there, while it would require some custom glue it wouldn't be too hard to convert the ClassLabel into a pandas Category if we want to go back - I think this would improve on the current behavior without risking changing the behavior of ClassLabel in a backward-incompat way.
Thoughts? I'm not sure if this is overly cautious. Whichever approach you think is better, I'd be happy to take it on!
| ```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints. | 319 | Feature Request: Support for Pandas `Categorical`
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints.
Now that I've heard you explain ClassLabel, that makes a lot of sense! While DictionaryType for Arrow (I think) can have arbitrarily typed keys, so it won't cover all potential cases, pandas' Category is *probably* the most common use for that pyarrow type, and ClassLabel should match that perfectly?
Other thoughts:
- changing the resulting patype on ClassLabel might be backward-incompatible? I'm not totally sure if users of the `datasets` library tend to directly access the `patype` attribute (I don't think we really do, but we haven't been using it for very long yet).
- would ClassLabel's dtype change to `dict[int64, string]`? It seems like in practice a ClassLabel (when not explicitly specified) would be constructed from the DictionaryType branch of `generate_from_arrow_type`, so it's not totally clear to me that anyone ever actually accesses/uses that dtype?
- I don't quite know how `.int2str` and `.str2int` are used in practice - would those be kept? Perhaps the implementation might actually be substantially smaller if we can just delegate to pyarrow's dict methods?
Another idea that just occurred to me: add a branch in here to generate a ClassLabel if the dict key is int64 and the values are string: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L932 , and then don't touch anything else.
In practice, I don't think this would be backward-incompatible in a way anyone would care about since the current behavior just throws an exception, and this way, we could support *reading* a pandas Categorical into a `Dataset` as a ClassLabel. I *think* from there, while it would require some custom glue it wouldn't be too hard to convert the ClassLabel into a pandas Category if we want to go back - I think this would improve on the current behavior without risking changing the behavior of ClassLabel in a backward-incompat way.
Thoughts? I'm not sure if this is overly cautious. Whichever approach you think is better, I'd be happy to take it on!
| [
0.0230096057,
-0.0102735162,
-0.142605871,
0.2076648176,
0.2370128185,
0.167278409,
0.1267410368,
0.2487182319,
-0.0738425702,
-0.2155073881,
0.1673995703,
0.2824070752,
-0.2035115063,
0.4252622128,
0.1327463239,
-0.2236520052,
0.124682799,
0.0284452848,
0.0778508633,
0.3117169738,
-0.3771745563,
-0.0838787034,
-0.0995520055,
0.238972649,
-0.1983464956,
-0.1293814182,
-0.4564826488,
-0.3532524407,
-0.2593736649,
-0.8477247953,
0.3983207047,
0.3807969391,
0.1931586713,
0.1735819578,
-0.000115819,
-0.307919383,
0.2058766037,
-0.04252363,
-0.0503572747,
-0.443198204,
-0.29180637,
-0.2464247942,
0.3244949579,
-0.2901012897,
-0.2068760246,
-0.2925748229,
0.0067170095,
-0.2727672458,
0.2317004055,
0.0666810721,
0.166751653,
-0.0497689843,
-0.027442582,
-0.1615714729,
0.131834507,
0.4021802843,
-0.2371010184,
0.2846889496,
0.5595126152,
-0.3222315311,
0.2314839363,
0.0884721205,
0.0117629617,
-0.32495749,
0.6005215645,
0.0490310155,
0.2645990849,
-0.5582160354,
-0.3079172671,
0.1526242793,
0.3408334851,
-0.3584246039,
-0.5394774079,
-0.4483573437,
-0.0119226612,
-0.2346875072,
-0.0061523803,
-0.1371360123,
-0.1556821764,
0.1616318524,
0.1122617945,
-0.0731579438,
-0.2399316877,
0.1873063892,
-0.084433198,
0.1476807743,
-0.044365447,
-0.1675475389,
0.2131358534,
-0.0995403528,
0.0033084266,
-0.1552575678,
0.0259757079,
0.6048637629,
-0.2247824073,
-0.1363117397,
0.0729947686,
0.2108624876,
0.0177887529,
-0.008306751,
0.100210093,
0.0134376511,
-0.2314001918,
-0.0256742202,
0.3791762292,
0.2924356461,
0.1800874323,
-0.0863290057,
-0.128845185,
0.3581373096,
-0.0041657537,
0.0536367223,
-0.1282648444,
-0.1226121485,
-0.1723236144,
0.1386384964,
-0.070712693,
0.2454465479,
-0.0304598138,
-0.0372078195,
-0.0177116916,
-0.0263489913,
0.0152255632,
0.2628427744,
-0.0659547299,
0.0928889215,
-0.0641005933,
0.0835101455,
0.2764410079,
0.1206505597,
-0.0680940449,
0.0020983815,
-0.2102169394,
0.1605478823,
-0.0775293857,
-0.1858754158,
-0.1492253244,
-0.0344319344,
0.1192694008,
0.1085183024,
-0.2276077121,
-0.0936202854,
0.1229104847,
0.0074044154,
-0.4257947803,
0.1622293591,
0.2187953591,
-0.0850058496,
-0.3504914641,
0.1319193989,
0.1542150229,
-0.2472275198,
-0.23920542,
0.1113658696,
-0.2031581253,
0.1457659006,
-0.3309437335,
0.4760414362,
-0.0080963634,
0.0497262031,
0.2327816039,
0.0547784343,
-0.2638413906,
-0.3284742236,
-0.1710062325,
0.4522376359,
-0.3215055466,
-0.2309906036,
0.0531621315,
-0.0306911841,
0.036155507,
0.0772850364,
-0.1583129764,
0.119236365,
-0.0716489851,
0.1776525676,
0.5333982706,
-0.1753787398,
-0.1190231889,
0.3577376902,
0.0055855513,
0.1976587772,
0.2812624872,
0.2622911036,
0.1901308447,
-0.2151020765,
0.0931512266,
0.1283134222,
-0.1040708944,
-0.2355067581,
0.1491523683,
-0.330439508,
0.1761065722,
0.1930087656,
-0.0684678257,
0.2835514545,
0.074393481,
-0.3360265195,
0.3749299049,
-0.2894715965,
0.1078302935,
0.0636767447,
0.3066045046,
0.4409076571,
0.0205316953,
-0.4798708558,
-0.1257181019,
0.3306059539,
-0.0952281058,
0.1742292345,
-0.3567067981,
-0.4952705204,
-0.0397887304,
0.2739691734,
-0.2659195662,
0.1416191757,
-0.009122286,
-0.2109058201,
0.2343252301,
0.2042149752,
-0.3539478779,
0.5252866745,
-0.0650036484,
0.2001338601,
-0.2408440411,
0.1756787002,
-0.0458534136,
0.0768375844,
-0.1055376381,
0.3452533185,
-0.0846972913,
-0.1197187603,
0.2849960923,
0.0615111254,
0.2428296357,
-0.094621025,
-0.3645077646,
0.2292647958,
0.5264254808,
0.0491378531,
-0.0908040553,
-0.1106584221,
0.084387064,
-0.0667051226,
0.0166786388,
0.5823572874,
-0.4817156792,
0.2271655947,
0.1415825486,
0.1412526071,
0.0998749509,
0.2575954795,
-0.2499680966,
-0.1974931359,
0.1438123286,
0.2578537166,
-0.2217442542,
0.0667167678,
-0.1658439934,
0.1915976405,
0.3040450811,
-0.1511571109,
0.1042438075,
0.1038230807,
0.1065930277,
-0.170360297,
0.2263922393,
0.1434835196,
0.2757692337,
0.0754673481,
0.0635722429,
0.1523474008,
-0.1569282115,
-0.0685347915,
0.3644998372,
0.027278306,
-0.085646525,
-0.2118578553,
0.0292043351,
0.0771163404,
0.0599483252,
-0.3495514095,
-0.0224031359,
-0.0369301066,
-0.1203359812,
0.1702975333,
-0.1160881072,
-0.5219680071,
0.14127177,
-0.5717269778,
0.0078103691,
-0.1792369187,
-0.0621801354,
0.0817524418,
0.0744456798,
0.1140424013,
-0.3413845301,
-0.035435766,
0.1047069058,
-0.2842971385,
-0.2089478821,
-0.4510959089,
0.2566653788,
0.0999105871,
-0.0136358291,
-0.1809959859,
0.2080219984,
0.0345665291,
-0.031090185,
-0.2123506963,
-0.279039979,
0.0493884236,
-0.1699927449,
0.026470961,
0.2983823717,
-0.2530080378,
-0.1749663949,
-0.2625125647,
0.1982941329,
0.1198827177,
-0.1365109533,
-0.2098923624,
-0.1893090755,
0.0895449966,
-0.1031984016,
0.102033332,
-0.4095128775,
-0.3784289956,
0.6761474609,
0.2192849815,
0.1426875591,
0.1680995673,
-0.0058761388,
0.0730163231,
-0.0896326676,
-0.1047140434,
-0.1668846905,
0.2988792658,
0.2073049545,
-0.1608502716,
-0.3180471957,
-0.0798368454,
-0.3646600246,
0.1057947725,
0.2121287882,
-0.1410654038,
-0.3041149974,
0.0048827827,
0.5083075166,
0.1135893539,
0.0284398738,
0.2844409049,
0.2777574658,
0.0575978793,
0.1198546216,
0.0421051756,
0.0113997199,
0.1779200137,
-0.2589244843,
0.2157592922,
0.4879330099,
0.1209345162,
0.2910845578,
0.0937248617,
-0.0136831412,
0.495490849,
-0.1015296653,
0.2374846935,
-0.1354481876,
-0.2925467491,
-0.3849199414,
-0.3812012076,
0.087369144,
0.0922508314,
-0.1177487299,
-0.1566376537,
-0.2011427283,
-0.0049428493,
0.1318277121,
-0.0870388895,
-0.026894195,
-0.3554051518,
0.4113956392,
-0.0421571992,
-0.1222861707,
-0.4082294703,
-0.0289857928,
-0.0885422379,
0.1240104064,
0.3442617357,
-0.1650989652,
-0.3631923497,
-0.1629422605,
-0.4993349314,
0.2302028537,
0.6471201181,
-0.142465055,
0.0527762696,
-0.0200610384,
-0.0053712726,
0.1159426942,
0.5606773496,
-0.0718633756,
-0.337580204,
0.2686216533,
-0.4713848233,
-0.3781881332,
0.3169171214,
0.0455298573,
0.0441104062,
-0.015689645,
0.1021610573,
-0.0899454951,
-0.2031951547,
-0.0265191142,
-0.0172946341,
0.0012238696,
-0.0552296191,
0.0234432667,
-0.0369385332,
-0.1320167482,
-0.17588754,
0.4480683804,
0.1625965536,
0.0418879762,
-0.1095062047,
-0.2154186666,
0.0515595973,
0.2643418312,
-0.0085914135,
-0.0637493804,
-0.059256915,
0.1627685279,
0.0402164906,
0.0491838306,
0.1073397994,
0.2975521088,
-0.1458133608,
-0.5790701509,
-0.03947616,
0.0585529618,
0.4294226468,
0.3097957969,
-0.0215763208,
0.301440984,
-0.0531078503,
-0.0325051583,
-0.3707031608,
-0.2006964833,
0.3816449642,
-0.0846342146,
-0.3441472054,
-0.7334688902,
0.35093382,
0.1426516175,
-0.3018722534,
0.1473785639,
0.3631759882,
-0.5190954804,
0.293822825,
0.0408901721,
0.8400940895,
-0.092752777,
0.3819608688,
0.3813298345,
0.309545815,
0.7505722046,
0.27758044,
0.0987700075,
-0.3746115565,
-0.2893297076,
-0.3023834527,
-0.1062902287,
0.2692961097,
0.3666597009,
-0.3419591784,
0.3305616081,
-0.1324574798,
0.2882317305,
0.1098045334,
-0.1822686493,
-0.118922472,
-0.2034188211,
-0.725779295,
0.0480664074,
0.452559948,
-0.0365069583,
-0.2351938784,
-0.2062584609,
-0.1167572141,
0.0931269601,
-0.2607622147,
-0.2176082134,
-0.4321040213,
0.1444188654,
-0.1006665975,
0.1561187804,
0.2696221173,
0.4705724716,
0.3740429282,
-0.3842305839,
0.0270238128,
-0.1776851863,
-0.1497763991,
-0.1718023121,
-0.0811987668,
-0.0599162541,
0.4753719866,
0.033637926,
0.0971908569,
0.3608367443,
0.0548866466,
-0.2822109461,
-0.3045468628,
0.2037368268,
-0.0772629008,
-0.2618442178,
-0.2907629609,
-0.2406387925,
0.0460179001,
-0.1168738157,
0.0699091107,
0.2150896788,
-0.036970634,
0.1649360061,
-0.353825599,
0.1511006504,
-0.2132050991,
0.1253352314,
0.2026966512,
-0.0330267884,
0.107727021,
0.1574064493,
-0.2605878711,
-0.0517194532,
-0.005398497,
0.2587835193,
-0.4805978537,
0.1313625127,
0.2005819976,
-0.2650959492,
-0.127681613,
0.5652498007,
0.2517307401,
0.2128714323,
-0.0211028159,
-0.1289939582,
-0.2973943353,
0.3678024113,
0.1430946141,
0.24398911,
-0.1040261313,
0.13226524,
-0.18524912,
-0.1136159524,
-0.2206496894,
-0.0021181032,
-0.0052059069,
-0.0675509572,
0.4329421222,
0.0207921062,
0.1636744887,
0.2392760515,
0.1212814748,
0.2636921406,
-0.1925225705,
-0.1320178509,
-0.0871847421,
0.1703846902,
-0.0201828629,
-0.0114977956,
0.0209326074,
-0.0894580856,
-0.2565030158,
-0.2649770677,
0.0943428278,
0.2512863874,
0.1210569441,
-0.044832997,
0.3994724154,
-0.0346533991,
-0.1810764968,
0.3869204223,
0.0380891152,
0.4918481112,
0.0317568667,
0.0474309064,
-0.3425094485,
0.0545477271,
0.0307745598,
0.3291056156,
-0.0392378531,
-0.4100054801,
0.640925169,
0.0848435909,
-0.1820202172,
0.1968025118,
0.525116384,
0.2113812417,
0.1651662588,
0.0732802153,
0.1216477305,
0.0837515891,
-0.333778441,
-0.3142721951,
0.1178405583,
-0.1556552649,
0.171490863,
0.5202996731,
0.3050470948,
0.2484452128,
0.1247213334,
-0.1429638267,
0.2160944343,
0.1981442869,
0.0745486766,
0.3195813298,
0.3900642395,
-0.0091482699,
0.3561512828,
0.3804426789,
0.5041667819,
0.1832934767,
-0.0932920873,
0.0783443078,
0.0882703066,
-0.0172893927,
0.604726553,
0.0295800418,
0.0089014219,
0.4108659625,
0.0399432816,
-0.008216396,
-0.0557899103,
0.0577750057,
-0.1989379376,
0.0129612088,
-0.0629524663,
0.2387242019,
0.216109097,
-0.1524890214,
0.4791820645,
-0.1766899526,
-0.128983885,
-0.3476144671,
-0.1384477764,
-0.2974571586,
0.4343165755,
-0.130851984,
0.5650552511,
-0.0634815097,
0.1978338063,
-0.079186067,
0.3597237468,
-0.0898835137,
-0.0796708241,
-0.0958933383,
-0.0673904866,
0.0075929407,
0.2095281184,
0.4526984096,
0.1809923351,
-0.1542154402,
0.1083930284,
-0.426173389,
-0.2367449254,
-0.0348733291,
0.2622196376,
0.1265500635,
0.0943510532,
0.2722555399,
0.053081356,
-0.0310171098,
0.0222227406,
0.0101431422,
0.2748625875,
-0.3390368521,
0.7011252046,
-0.2181458771,
-0.235043779,
0.1938221306,
0.3255452216,
-0.3201076388,
0.0481974073,
0.0591657236,
-0.1691699624,
0.1492702365,
0.1201525033,
0.0416861959,
-0.0675470382,
0.1855738759,
0.1969930232,
0.4075562656,
0.0107359998,
-0.2408181429,
-0.4748311043,
-0.1409270167,
0.2547972202,
-0.2541851103,
0.2475463152,
-0.0389153138,
-0.023624666,
0.0910675228,
-0.081563279,
-0.3627572656,
-0.1309396923,
0.2148239762,
-0.0924036801,
-0.276786238,
0.1841605306,
0.1239571422,
-0.1801173091,
-0.2226979285,
0.0785782933,
-0.2537125945,
-0.0861013159,
-0.1989879161,
0.143423304,
-0.1959644854,
0.0455566198,
0.2694398463,
0.2227223068,
0.1807645857,
0.2077306956,
-0.169336617,
-0.0346202068,
-0.3444250226,
0.0613576137,
0.5801239014,
-0.0249673594,
0.3234555721,
-0.3697853088,
-0.275303036,
-0.0232930928,
0.2424985617,
0.2755201757,
-0.193352446,
-0.5966460705,
0.0874608532,
-0.4430661201,
-0.0298177488,
0.1695705801,
0.3679268062,
-0.1307650208,
0.3309235573,
0.1300219595,
-0.2077973038,
0.3422711492,
-0.4958491027,
-0.2886919379,
0.035095796,
-0.2006174475,
0.3782950938,
-0.1855074614,
-0.1591356397,
-0.0128235519,
0.4850820899,
0.1197654679,
-0.1062525436,
0.174354434,
-0.2196150422,
-0.1083699167,
-0.3893577456,
0.0817013681,
-0.2865291834,
-0.0213865116,
0.0420440584,
-0.4636684656
] |
https://github.com/huggingface/datasets/issues/1906 | Feature Request: Support for Pandas `Categorical` | I think we can first keep the int64 precision but with an arrow Dictionary for ClassLabel, and focus on the connection with arrow and pandas.
In this scope, I really like the idea of checking for the dictionary type:
> Another idea that just occurred to me: add a branch in here to generate a ClassLabel if the dict key is int64 and the values are string: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L932 , and then don't touch anything else.
This looks like a great start.
Then as you said we'd have to add the conversion from classlabel to the correct arrow dictionary type. Arrow is already able to convert from arrow Dictionary to pandas Categorical so it should be enough.
I can see two things that we must take case of to make this change backward compatible:
- first we must still be able to load an arrow file with arrow int64 dtype and `datasets` ClassLabel type without crashing. This can be fixed by casting the arrow int64 array to an arrow Dictionary array on-the-fly when loading the table in the ArrowReader.
- then we still have to return integers when accessing examples from a ClassLabel column. Currently it would return the strings values since it's based on the pandas behavior for converting from pandas to python/numpy. To do so we just have to adapt the python/numpy extractors in formatting.py (it takes care of converting an arrow table to a dictionary of python objects by doing arrow table -> pandas dataframe -> python dictionary)
Any help on this matter is very much welcome :) | ```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints. | 260 | Feature Request: Support for Pandas `Categorical`
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.Series(["a", "b", "c", "a"], dtype="category"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws NotImplementedError
# TODO(thom) this will need access to the dictionary as well (for labels). I.e. to the py_table
```
I'm curious if https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L796 could be built out in a way similar to `Sequence`?
e.g. a `Map` class (or whatever name the maintainers might prefer) that can accept:
```
index_type = generate_from_arrow_type(pa_type.index_type)
value_type = generate_from_arrow_type(pa_type.value_type)
```
and then additional code points to modify:
- FeatureType: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L694
- A branch to handle Map in get_nested_type: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L719
- I don't quite understand what `encode_nested_example` does but perhaps a branch there? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L755
- Similarly, I don't quite understand why `Sequence` is used this way in `generate_from_dict`, but perhaps a branch here? https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L775
I couldn't find other usages of `Sequence` outside of defining specific datasets, so I'm not sure if that's a comprehensive set of touchpoints.
I think we can first keep the int64 precision but with an arrow Dictionary for ClassLabel, and focus on the connection with arrow and pandas.
In this scope, I really like the idea of checking for the dictionary type:
> Another idea that just occurred to me: add a branch in here to generate a ClassLabel if the dict key is int64 and the values are string: https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L932 , and then don't touch anything else.
This looks like a great start.
Then as you said we'd have to add the conversion from classlabel to the correct arrow dictionary type. Arrow is already able to convert from arrow Dictionary to pandas Categorical so it should be enough.
I can see two things that we must take case of to make this change backward compatible:
- first we must still be able to load an arrow file with arrow int64 dtype and `datasets` ClassLabel type without crashing. This can be fixed by casting the arrow int64 array to an arrow Dictionary array on-the-fly when loading the table in the ArrowReader.
- then we still have to return integers when accessing examples from a ClassLabel column. Currently it would return the strings values since it's based on the pandas behavior for converting from pandas to python/numpy. To do so we just have to adapt the python/numpy extractors in formatting.py (it takes care of converting an arrow table to a dictionary of python objects by doing arrow table -> pandas dataframe -> python dictionary)
Any help on this matter is very much welcome :) | [
0.0230096057,
-0.0102735162,
-0.142605871,
0.2076648176,
0.2370128185,
0.167278409,
0.1267410368,
0.2487182319,
-0.0738425702,
-0.2155073881,
0.1673995703,
0.2824070752,
-0.2035115063,
0.4252622128,
0.1327463239,
-0.2236520052,
0.124682799,
0.0284452848,
0.0778508633,
0.3117169738,
-0.3771745563,
-0.0838787034,
-0.0995520055,
0.238972649,
-0.1983464956,
-0.1293814182,
-0.4564826488,
-0.3532524407,
-0.2593736649,
-0.8477247953,
0.3983207047,
0.3807969391,
0.1931586713,
0.1735819578,
-0.000115819,
-0.307919383,
0.2058766037,
-0.04252363,
-0.0503572747,
-0.443198204,
-0.29180637,
-0.2464247942,
0.3244949579,
-0.2901012897,
-0.2068760246,
-0.2925748229,
0.0067170095,
-0.2727672458,
0.2317004055,
0.0666810721,
0.166751653,
-0.0497689843,
-0.027442582,
-0.1615714729,
0.131834507,
0.4021802843,
-0.2371010184,
0.2846889496,
0.5595126152,
-0.3222315311,
0.2314839363,
0.0884721205,
0.0117629617,
-0.32495749,
0.6005215645,
0.0490310155,
0.2645990849,
-0.5582160354,
-0.3079172671,
0.1526242793,
0.3408334851,
-0.3584246039,
-0.5394774079,
-0.4483573437,
-0.0119226612,
-0.2346875072,
-0.0061523803,
-0.1371360123,
-0.1556821764,
0.1616318524,
0.1122617945,
-0.0731579438,
-0.2399316877,
0.1873063892,
-0.084433198,
0.1476807743,
-0.044365447,
-0.1675475389,
0.2131358534,
-0.0995403528,
0.0033084266,
-0.1552575678,
0.0259757079,
0.6048637629,
-0.2247824073,
-0.1363117397,
0.0729947686,
0.2108624876,
0.0177887529,
-0.008306751,
0.100210093,
0.0134376511,
-0.2314001918,
-0.0256742202,
0.3791762292,
0.2924356461,
0.1800874323,
-0.0863290057,
-0.128845185,
0.3581373096,
-0.0041657537,
0.0536367223,
-0.1282648444,
-0.1226121485,
-0.1723236144,
0.1386384964,
-0.070712693,
0.2454465479,
-0.0304598138,
-0.0372078195,
-0.0177116916,
-0.0263489913,
0.0152255632,
0.2628427744,
-0.0659547299,
0.0928889215,
-0.0641005933,
0.0835101455,
0.2764410079,
0.1206505597,
-0.0680940449,
0.0020983815,
-0.2102169394,
0.1605478823,
-0.0775293857,
-0.1858754158,
-0.1492253244,
-0.0344319344,
0.1192694008,
0.1085183024,
-0.2276077121,
-0.0936202854,
0.1229104847,
0.0074044154,
-0.4257947803,
0.1622293591,
0.2187953591,
-0.0850058496,
-0.3504914641,
0.1319193989,
0.1542150229,
-0.2472275198,
-0.23920542,
0.1113658696,
-0.2031581253,
0.1457659006,
-0.3309437335,
0.4760414362,
-0.0080963634,
0.0497262031,
0.2327816039,
0.0547784343,
-0.2638413906,
-0.3284742236,
-0.1710062325,
0.4522376359,
-0.3215055466,
-0.2309906036,
0.0531621315,
-0.0306911841,
0.036155507,
0.0772850364,
-0.1583129764,
0.119236365,
-0.0716489851,
0.1776525676,
0.5333982706,
-0.1753787398,
-0.1190231889,
0.3577376902,
0.0055855513,
0.1976587772,
0.2812624872,
0.2622911036,
0.1901308447,
-0.2151020765,
0.0931512266,
0.1283134222,
-0.1040708944,
-0.2355067581,
0.1491523683,
-0.330439508,
0.1761065722,
0.1930087656,
-0.0684678257,
0.2835514545,
0.074393481,
-0.3360265195,
0.3749299049,
-0.2894715965,
0.1078302935,
0.0636767447,
0.3066045046,
0.4409076571,
0.0205316953,
-0.4798708558,
-0.1257181019,
0.3306059539,
-0.0952281058,
0.1742292345,
-0.3567067981,
-0.4952705204,
-0.0397887304,
0.2739691734,
-0.2659195662,
0.1416191757,
-0.009122286,
-0.2109058201,
0.2343252301,
0.2042149752,
-0.3539478779,
0.5252866745,
-0.0650036484,
0.2001338601,
-0.2408440411,
0.1756787002,
-0.0458534136,
0.0768375844,
-0.1055376381,
0.3452533185,
-0.0846972913,
-0.1197187603,
0.2849960923,
0.0615111254,
0.2428296357,
-0.094621025,
-0.3645077646,
0.2292647958,
0.5264254808,
0.0491378531,
-0.0908040553,
-0.1106584221,
0.084387064,
-0.0667051226,
0.0166786388,
0.5823572874,
-0.4817156792,
0.2271655947,
0.1415825486,
0.1412526071,
0.0998749509,
0.2575954795,
-0.2499680966,
-0.1974931359,
0.1438123286,
0.2578537166,
-0.2217442542,
0.0667167678,
-0.1658439934,
0.1915976405,
0.3040450811,
-0.1511571109,
0.1042438075,
0.1038230807,
0.1065930277,
-0.170360297,
0.2263922393,
0.1434835196,
0.2757692337,
0.0754673481,
0.0635722429,
0.1523474008,
-0.1569282115,
-0.0685347915,
0.3644998372,
0.027278306,
-0.085646525,
-0.2118578553,
0.0292043351,
0.0771163404,
0.0599483252,
-0.3495514095,
-0.0224031359,
-0.0369301066,
-0.1203359812,
0.1702975333,
-0.1160881072,
-0.5219680071,
0.14127177,
-0.5717269778,
0.0078103691,
-0.1792369187,
-0.0621801354,
0.0817524418,
0.0744456798,
0.1140424013,
-0.3413845301,
-0.035435766,
0.1047069058,
-0.2842971385,
-0.2089478821,
-0.4510959089,
0.2566653788,
0.0999105871,
-0.0136358291,
-0.1809959859,
0.2080219984,
0.0345665291,
-0.031090185,
-0.2123506963,
-0.279039979,
0.0493884236,
-0.1699927449,
0.026470961,
0.2983823717,
-0.2530080378,
-0.1749663949,
-0.2625125647,
0.1982941329,
0.1198827177,
-0.1365109533,
-0.2098923624,
-0.1893090755,
0.0895449966,
-0.1031984016,
0.102033332,
-0.4095128775,
-0.3784289956,
0.6761474609,
0.2192849815,
0.1426875591,
0.1680995673,
-0.0058761388,
0.0730163231,
-0.0896326676,
-0.1047140434,
-0.1668846905,
0.2988792658,
0.2073049545,
-0.1608502716,
-0.3180471957,
-0.0798368454,
-0.3646600246,
0.1057947725,
0.2121287882,
-0.1410654038,
-0.3041149974,
0.0048827827,
0.5083075166,
0.1135893539,
0.0284398738,
0.2844409049,
0.2777574658,
0.0575978793,
0.1198546216,
0.0421051756,
0.0113997199,
0.1779200137,
-0.2589244843,
0.2157592922,
0.4879330099,
0.1209345162,
0.2910845578,
0.0937248617,
-0.0136831412,
0.495490849,
-0.1015296653,
0.2374846935,
-0.1354481876,
-0.2925467491,
-0.3849199414,
-0.3812012076,
0.087369144,
0.0922508314,
-0.1177487299,
-0.1566376537,
-0.2011427283,
-0.0049428493,
0.1318277121,
-0.0870388895,
-0.026894195,
-0.3554051518,
0.4113956392,
-0.0421571992,
-0.1222861707,
-0.4082294703,
-0.0289857928,
-0.0885422379,
0.1240104064,
0.3442617357,
-0.1650989652,
-0.3631923497,
-0.1629422605,
-0.4993349314,
0.2302028537,
0.6471201181,
-0.142465055,
0.0527762696,
-0.0200610384,
-0.0053712726,
0.1159426942,
0.5606773496,
-0.0718633756,
-0.337580204,
0.2686216533,
-0.4713848233,
-0.3781881332,
0.3169171214,
0.0455298573,
0.0441104062,
-0.015689645,
0.1021610573,
-0.0899454951,
-0.2031951547,
-0.0265191142,
-0.0172946341,
0.0012238696,
-0.0552296191,
0.0234432667,
-0.0369385332,
-0.1320167482,
-0.17588754,
0.4480683804,
0.1625965536,
0.0418879762,
-0.1095062047,
-0.2154186666,
0.0515595973,
0.2643418312,
-0.0085914135,
-0.0637493804,
-0.059256915,
0.1627685279,
0.0402164906,
0.0491838306,
0.1073397994,
0.2975521088,
-0.1458133608,
-0.5790701509,
-0.03947616,
0.0585529618,
0.4294226468,
0.3097957969,
-0.0215763208,
0.301440984,
-0.0531078503,
-0.0325051583,
-0.3707031608,
-0.2006964833,
0.3816449642,
-0.0846342146,
-0.3441472054,
-0.7334688902,
0.35093382,
0.1426516175,
-0.3018722534,
0.1473785639,
0.3631759882,
-0.5190954804,
0.293822825,
0.0408901721,
0.8400940895,
-0.092752777,
0.3819608688,
0.3813298345,
0.309545815,
0.7505722046,
0.27758044,
0.0987700075,
-0.3746115565,
-0.2893297076,
-0.3023834527,
-0.1062902287,
0.2692961097,
0.3666597009,
-0.3419591784,
0.3305616081,
-0.1324574798,
0.2882317305,
0.1098045334,
-0.1822686493,
-0.118922472,
-0.2034188211,
-0.725779295,
0.0480664074,
0.452559948,
-0.0365069583,
-0.2351938784,
-0.2062584609,
-0.1167572141,
0.0931269601,
-0.2607622147,
-0.2176082134,
-0.4321040213,
0.1444188654,
-0.1006665975,
0.1561187804,
0.2696221173,
0.4705724716,
0.3740429282,
-0.3842305839,
0.0270238128,
-0.1776851863,
-0.1497763991,
-0.1718023121,
-0.0811987668,
-0.0599162541,
0.4753719866,
0.033637926,
0.0971908569,
0.3608367443,
0.0548866466,
-0.2822109461,
-0.3045468628,
0.2037368268,
-0.0772629008,
-0.2618442178,
-0.2907629609,
-0.2406387925,
0.0460179001,
-0.1168738157,
0.0699091107,
0.2150896788,
-0.036970634,
0.1649360061,
-0.353825599,
0.1511006504,
-0.2132050991,
0.1253352314,
0.2026966512,
-0.0330267884,
0.107727021,
0.1574064493,
-0.2605878711,
-0.0517194532,
-0.005398497,
0.2587835193,
-0.4805978537,
0.1313625127,
0.2005819976,
-0.2650959492,
-0.127681613,
0.5652498007,
0.2517307401,
0.2128714323,
-0.0211028159,
-0.1289939582,
-0.2973943353,
0.3678024113,
0.1430946141,
0.24398911,
-0.1040261313,
0.13226524,
-0.18524912,
-0.1136159524,
-0.2206496894,
-0.0021181032,
-0.0052059069,
-0.0675509572,
0.4329421222,
0.0207921062,
0.1636744887,
0.2392760515,
0.1212814748,
0.2636921406,
-0.1925225705,
-0.1320178509,
-0.0871847421,
0.1703846902,
-0.0201828629,
-0.0114977956,
0.0209326074,
-0.0894580856,
-0.2565030158,
-0.2649770677,
0.0943428278,
0.2512863874,
0.1210569441,
-0.044832997,
0.3994724154,
-0.0346533991,
-0.1810764968,
0.3869204223,
0.0380891152,
0.4918481112,
0.0317568667,
0.0474309064,
-0.3425094485,
0.0545477271,
0.0307745598,
0.3291056156,
-0.0392378531,
-0.4100054801,
0.640925169,
0.0848435909,
-0.1820202172,
0.1968025118,
0.525116384,
0.2113812417,
0.1651662588,
0.0732802153,
0.1216477305,
0.0837515891,
-0.333778441,
-0.3142721951,
0.1178405583,
-0.1556552649,
0.171490863,
0.5202996731,
0.3050470948,
0.2484452128,
0.1247213334,
-0.1429638267,
0.2160944343,
0.1981442869,
0.0745486766,
0.3195813298,
0.3900642395,
-0.0091482699,
0.3561512828,
0.3804426789,
0.5041667819,
0.1832934767,
-0.0932920873,
0.0783443078,
0.0882703066,
-0.0172893927,
0.604726553,
0.0295800418,
0.0089014219,
0.4108659625,
0.0399432816,
-0.008216396,
-0.0557899103,
0.0577750057,
-0.1989379376,
0.0129612088,
-0.0629524663,
0.2387242019,
0.216109097,
-0.1524890214,
0.4791820645,
-0.1766899526,
-0.128983885,
-0.3476144671,
-0.1384477764,
-0.2974571586,
0.4343165755,
-0.130851984,
0.5650552511,
-0.0634815097,
0.1978338063,
-0.079186067,
0.3597237468,
-0.0898835137,
-0.0796708241,
-0.0958933383,
-0.0673904866,
0.0075929407,
0.2095281184,
0.4526984096,
0.1809923351,
-0.1542154402,
0.1083930284,
-0.426173389,
-0.2367449254,
-0.0348733291,
0.2622196376,
0.1265500635,
0.0943510532,
0.2722555399,
0.053081356,
-0.0310171098,
0.0222227406,
0.0101431422,
0.2748625875,
-0.3390368521,
0.7011252046,
-0.2181458771,
-0.235043779,
0.1938221306,
0.3255452216,
-0.3201076388,
0.0481974073,
0.0591657236,
-0.1691699624,
0.1492702365,
0.1201525033,
0.0416861959,
-0.0675470382,
0.1855738759,
0.1969930232,
0.4075562656,
0.0107359998,
-0.2408181429,
-0.4748311043,
-0.1409270167,
0.2547972202,
-0.2541851103,
0.2475463152,
-0.0389153138,
-0.023624666,
0.0910675228,
-0.081563279,
-0.3627572656,
-0.1309396923,
0.2148239762,
-0.0924036801,
-0.276786238,
0.1841605306,
0.1239571422,
-0.1801173091,
-0.2226979285,
0.0785782933,
-0.2537125945,
-0.0861013159,
-0.1989879161,
0.143423304,
-0.1959644854,
0.0455566198,
0.2694398463,
0.2227223068,
0.1807645857,
0.2077306956,
-0.169336617,
-0.0346202068,
-0.3444250226,
0.0613576137,
0.5801239014,
-0.0249673594,
0.3234555721,
-0.3697853088,
-0.275303036,
-0.0232930928,
0.2424985617,
0.2755201757,
-0.193352446,
-0.5966460705,
0.0874608532,
-0.4430661201,
-0.0298177488,
0.1695705801,
0.3679268062,
-0.1307650208,
0.3309235573,
0.1300219595,
-0.2077973038,
0.3422711492,
-0.4958491027,
-0.2886919379,
0.035095796,
-0.2006174475,
0.3782950938,
-0.1855074614,
-0.1591356397,
-0.0128235519,
0.4850820899,
0.1197654679,
-0.1062525436,
0.174354434,
-0.2196150422,
-0.1083699167,
-0.3893577456,
0.0817013681,
-0.2865291834,
-0.0213865116,
0.0420440584,
-0.4636684656
] |
https://github.com/huggingface/datasets/issues/1898 | ALT dataset has repeating instances in all splits | I just merged a fix, we'll do a patch release soon. Thanks again for reporting, and sorry for the inconvenience.
In the meantime you can load `ALT` using `datasets` from the master branch | The [ALT](https://huggingface.co/datasets/alt) dataset has all the same instances within each split :/
Seemed like a great dataset for some experiments I wanted to carry out, especially since its medium-sized, and has all splits.
Would be great if this could be fixed :)
Added a snapshot of the contents from `explore-datset` feature, for quick reference.
| 33 | ALT dataset has repeating instances in all splits
The [ALT](https://huggingface.co/datasets/alt) dataset has all the same instances within each split :/
Seemed like a great dataset for some experiments I wanted to carry out, especially since its medium-sized, and has all splits.
Would be great if this could be fixed :)
Added a snapshot of the contents from `explore-datset` feature, for quick reference.
I just merged a fix, we'll do a patch release soon. Thanks again for reporting, and sorry for the inconvenience.
In the meantime you can load `ALT` using `datasets` from the master branch | [
-0.2530450821,
-0.7312533259,
-0.0555955917,
0.4320778251,
0.317453742,
-0.1736769378,
0.3799054623,
0.1424698532,
0.2794250548,
0.331989795,
-0.1614305377,
0.0980521291,
0.0777061582,
-0.0084830374,
-0.0531924106,
-0.0700329691,
-0.0420136601,
0.2138656378,
-0.0015912578,
-0.2287293226,
-0.1361037195,
0.1931892633,
-0.3067331314,
0.0349793583,
-0.1967668682,
-0.2158622444,
-0.0085894614,
0.0501791686,
0.0790566429,
-0.2780336142,
0.0863048285,
0.1110691428,
0.0143351555,
0.4517410994,
-0.0001077245,
0.1526557505,
0.4967344403,
0.033245787,
-0.2704988122,
-0.2112118006,
-0.3161487579,
0.2112976909,
0.2924029827,
-0.1687503159,
-0.0563594252,
-0.2315427661,
-0.1243323833,
-0.402652204,
0.334961921,
-0.0742318332,
0.2620728016,
0.4766895771,
-0.3479263783,
-0.4059587121,
0.4381238222,
0.5822083354,
-0.1347048879,
-0.1824560761,
0.3820550144,
0.4519360363,
0.0182801206,
0.3951461315,
0.1519045234,
0.2581331134,
0.1004770249,
0.1850467622,
0.1533163488,
-0.1135971695,
0.0046807164,
0.2672683597,
0.3437399864,
-0.0407963656,
-0.0840745121,
-0.2082262635,
0.0375720263,
0.0121966004,
0.242644608,
0.0158099905,
0.0203863084,
0.1143072993,
0.2770201862,
0.2895072997,
0.0548392981,
-0.0553698465,
0.0077199414,
-0.0340843946,
-0.1891285926,
0.0874742419,
0.1575710028,
-0.1861014366,
0.3643368483,
-0.0746587589,
-0.141969502,
-0.2027995884,
-0.3039720953,
-0.0242451802,
-0.1317782551,
-0.1549827307,
0.1851766258,
0.3924608231,
0.3807052672,
0.2239826173,
-0.3758053482,
0.0164200589,
0.263672173,
0.0733942688,
0.303794235,
-0.0638524964,
0.1854225844,
0.0785439461,
-0.1949548423,
0.0184876472,
-0.2299308479,
-0.0099539272,
-0.23035191,
-0.0250954442,
0.2488904595,
-0.1194887608,
-0.4183261395,
-0.1163930744,
0.1127994061,
-0.1395734251,
0.0736997426,
0.3457527459,
0.3099733293,
0.1333233565,
-0.0907324106,
0.0623500198,
-0.1016263664,
-0.4191182256,
-0.2110075355,
-0.2572088838,
-0.3241527379,
-0.0920961797,
-0.0299138166,
-0.1140045673,
0.2681919634,
0.2965402007,
0.0488087088,
-0.125448972,
0.2154405415,
0.0645246655,
0.1204702705,
-0.0578569248,
0.103672877,
0.3334431052,
0.153557688,
-0.0137165338,
-0.0288757011,
0.063165307,
-0.111020796,
-0.0017042682,
0.1181944609,
0.2778227329,
-0.0948352292,
0.1378462911,
0.0151471421,
0.294777751,
-0.1778621823,
0.0043671876,
-0.1145089567,
0.0638621226,
0.0688560084,
-0.0965424776,
0.2541606426,
0.4378032088,
-0.2287375331,
-0.3377066851,
-0.5115398169,
-0.072174713,
0.12817204,
0.5561010838,
-0.1574354321,
-0.1233195215,
-0.399990648,
0.2835873365,
0.1919440031,
0.2184077352,
-0.2081332356,
0.1348675489,
-0.1094419733,
-0.0671294406,
0.0755431056,
0.1349572688,
0.3912535906,
0.0080073737,
0.0901446119,
0.2794890702,
0.287633121,
-0.0885437801,
-0.2506302595,
-0.0650309548,
0.0467614345,
-0.0413872227,
-0.2071635872,
-0.322815001,
0.1217944697,
0.0372690111,
0.1147335842,
-0.0210301802,
-0.0985307172,
0.0281674042,
0.2101827413,
0.2042815536,
0.3248485029,
-0.0377074704,
-0.2207814902,
0.047752738,
-0.173679024,
-0.1478262246,
0.2951382399,
-0.4183107018,
-0.2752802372,
0.119378686,
-0.350612253,
-0.0463986322,
0.190340966,
0.2829101682,
-0.1916193068,
0.0162031427,
-0.0466970019,
-0.23415488,
0.1645259559,
0.0080303326,
-0.1809084564,
0.2991644442,
-0.1965729743,
0.018705789,
0.0256754085,
0.108078599,
0.1730460972,
-0.2686106563,
-0.038383022,
0.4057911336,
0.3260444403,
0.1635498852,
0.0293143559,
-0.4810566604,
0.412296325,
-0.234893322,
0.1248608008,
0.2347396314,
0.1278835833,
-0.2357154489,
-0.155087173,
0.1593184471,
0.1920385659,
0.2863156796,
-0.033364974,
0.2085597217,
0.3136180043,
-0.1635527909,
0.2406592965,
-0.5274127722,
0.1388352513,
-0.0694964826,
0.0954828709,
0.1965304911,
-0.2810155153,
0.3699356019,
0.1038463712,
0.1075507626,
0.2117596865,
0.0147761656,
-0.5166522264,
-0.0949607342,
-0.0444389433,
0.0235480443,
0.1444617212,
0.3443718553,
0.0418832675,
-0.1660428494,
0.3011956513,
-0.1309050024,
0.3186791241,
0.1758502126,
-0.1268956512,
0.0922172815,
0.123243846,
-0.1161587387,
-0.5793970823,
-0.1676858366,
0.3131717443,
0.0630044937,
-0.2004716992,
-0.2201201618,
-0.5879808068,
-0.0139657781,
-0.1368467212,
0.0108069591,
-0.1271449476,
-0.1715042293,
0.2827455401,
0.0558801331,
0.0002114922,
0.3667479753,
0.3949828446,
0.4391367137,
-0.0003227219,
0.442353338,
-0.3167382479,
-0.1895714998,
-0.3355870247,
0.1189157367,
0.4297170043,
0.3283495307,
0.4452003837,
-0.0780441463,
-0.0210394338,
-0.2582357824,
-0.3862938881,
0.1428807676,
0.0395932384,
-0.0219461769,
0.1286535859,
-0.1326856613,
0.0025741979,
-0.1702553332,
0.1888429224,
-0.400467962,
-0.4218912423,
0.1067009345,
-0.1009612679,
-0.2328897864,
-0.1279153079,
-0.296222955,
0.062943399,
-0.3463541269,
0.1321720183,
0.0558892265,
0.0171966497,
0.0587165542,
-0.1139380038,
-0.0360045582,
-0.0893378779,
-0.1253001094,
-0.5492932796,
0.06732703,
0.0928751677,
-0.2612451315,
-0.1331634521,
-0.0414443649,
-0.219979763,
-0.141622588,
0.032470841,
-0.4009284973,
0.1879467368,
-0.4680109322,
-0.0788507983,
0.1960983574,
0.0292477887,
0.2848636806,
-0.0054308437,
-0.2143378258,
-0.0154975802,
-0.2960484028,
0.1560645998,
0.0158260092,
0.1873741895,
-0.0449536741,
0.0046862373,
0.0418865383,
0.3934040368,
0.318161577,
-0.1329948008,
0.0471849889,
-0.1528193653,
0.1869280934,
-0.4131633341,
-0.1799940467,
0.0803353637,
0.0075402334,
-0.1402645707,
0.386606127,
0.0220706947,
-0.0186825246,
0.1082530171,
0.0504921004,
0.0193110257,
-0.3562464714,
-0.1019512415,
-0.1003297269,
-0.0155535862,
-0.0355559178,
0.036145322,
-0.1161255166,
-0.0161619373,
0.1786687523,
0.1157961786,
-0.2331679016,
-0.1852467805,
-0.6297690272,
0.1757915169,
-0.2929063439,
0.345325619,
-0.2199639976,
0.1585962623,
0.0163654611,
-0.2085307091,
0.3482231796,
0.1503783613,
0.7905904055,
-0.2985008359,
0.018055981,
-0.0152584687,
0.0137589872,
-0.5073357224,
-0.1102744788,
-0.2534875274,
0.0568083711,
0.084810555,
0.5961065888,
-0.0769747943,
-0.0178146269,
0.256926775,
0.0763162598,
-0.2049286664,
-0.1959681213,
-0.0262700841,
-0.1808042228,
-0.2594664097,
-0.2019483298,
0.0664618313,
-0.0534826703,
0.1716032624,
-0.1570572555,
-0.3182402551,
-0.0022013262,
0.1731119752,
0.3088166416,
0.5221158862,
0.0686084479,
0.1972514689,
0.4771072567,
0.1118465737,
0.1605861932,
0.7080814838,
-0.1246948093,
-0.0885444134,
0.1793955564,
-0.4542677999,
0.0541721284,
0.310457319,
0.208019197,
0.2227959335,
-0.1944971085,
0.0548254214,
-0.1967511475,
0.155136615,
0.1092845947,
0.0995220318,
0.0047361106,
-0.3371591568,
0.1542251855,
-0.1057450622,
-0.0794186741,
0.4623818398,
0.0169355143,
0.0483622923,
0.5934587121,
-0.1379653364,
0.7147191167,
0.0982608423,
-0.0499799103,
0.1392764598,
0.0757336244,
0.1597910821,
0.2165878117,
0.0462793037,
-0.5335630178,
-0.4241667688,
0.1115131676,
-0.1202279925,
0.008577018,
-0.0140784476,
-0.4247623086,
-0.2648170292,
0.0503542572,
0.3281959593,
-0.0736604333,
0.1685639024,
-0.1620361358,
-0.05687223,
0.4204364121,
0.1702889055,
-0.0780798048,
0.2189978659,
0.1155496389,
-0.0110786818,
0.1699694097,
-0.1600705087,
-0.3022379875,
0.0105894953,
-0.3480477929,
0.4214712977,
-0.1170585006,
-0.5092540979,
0.2436767519,
-0.2589382827,
0.35779351,
0.0583618954,
0.0356461257,
0.3490626514,
0.3187751472,
0.0343276933,
-0.113534309,
-0.1730252206,
-0.0191466585,
0.2346480936,
-0.285566777,
-0.2195286453,
0.3488926888,
0.1695675701,
-0.372600615,
0.0135491248,
0.0414604843,
-0.1194157451,
-0.2098166943,
0.136293456,
-0.1320104599,
-0.3592993021,
0.1538861543,
-0.3178062141,
-0.0152982473,
0.0411127098,
0.0042620366,
-0.0640338212,
0.0624862984,
0.0726291835,
0.3321678936,
0.0711813197,
0.2105047256,
0.2428442687,
-0.189086616,
-0.4889021516,
0.55479002,
0.2250995338,
-0.619346559,
0.3700903952,
-0.3495392203,
-0.1074077711,
0.0922145769,
0.1170549691,
0.1182191744,
0.0885552615,
-0.3329209685,
-0.1235655993,
-0.0467917547,
0.1392304599,
-0.36506477,
0.3558062613,
-0.0435730405,
0.1386946142,
-0.0411707312,
0.1824301183,
-0.3591293991,
0.2948114276,
0.1066084579,
0.195819363,
-0.5335171819,
-0.3546473682,
0.1680223793,
-0.1661372781,
0.1139766797,
0.0865006223,
-0.0731940269,
-0.2709961534,
-0.1109671146,
0.0990341306,
-0.293828845,
0.1494211257,
-0.0456149876,
-0.0791162401,
0.1114270687,
-0.3093270659,
0.2880663872,
0.1603247821,
-0.0819973573,
0.0103600733,
0.1283070743,
-0.2510235608,
-0.1187814623,
0.2032211125,
0.0138758626,
0.0407719575,
0.2367193848,
0.0036918772,
-0.0783745944,
-0.1281886697,
-0.1712077856,
0.1973726898,
-0.1173792183,
0.1716748774,
0.0816532969,
-0.1096391454,
-0.152700156,
0.1485709995,
0.3307650983,
0.0729482621,
-0.084485516,
0.0714307576,
0.0353540443,
0.2475075722,
-0.4756089449,
-0.2290025949,
0.3186690509,
0.1778694391,
0.1701246649,
0.209905073,
0.1794620901,
-0.1407877505,
-0.2270627767,
0.1378558278,
0.4304170609,
-0.1530753821,
0.499303937,
0.4427482784,
0.0093533248,
-0.2456500381,
0.3271172345,
0.0901499838,
0.139837876,
0.3966962099,
0.0001597963,
0.2757714987,
0.3659014702,
-0.0076316055,
0.3976855278,
-0.3081563711,
-0.0114764767,
0.0781330019,
0.053365849,
0.1364763826,
0.3035895526,
0.2825963497,
-0.3126277626,
-0.2198748738,
-0.2563143373,
-0.3658165634,
-0.1523785144,
0.0193886384,
-0.3127083182,
-0.0757382438,
-0.4050968885,
-0.1092406958,
-0.0566346794,
0.0324508287,
0.3366051018,
0.010239033,
-0.0523798689,
-0.4539519548,
-0.1849161386,
0.1961034983,
0.1046568453,
-0.3900808692,
0.174012661,
0.2441932261,
-0.0297509581,
0.3311134875,
0.4414887428,
0.2335391939,
0.1364922673,
-0.2641116083,
-0.0189033616,
-0.0341616645,
-0.2060284615,
-0.1551760435,
-0.1845019758,
-0.0049259998,
-0.0138343703,
0.3606175482,
0.148945421,
-0.1105719507,
-0.0897163823,
0.3804370761,
0.0569639318,
-0.1910246909,
0.0530698933,
-0.2649301589,
-0.0524506196,
-0.1745593846,
0.0773496777,
-0.5230467319,
-0.3014148176,
0.3586502373,
-0.4063143432,
0.0645209625,
-0.1936036199,
0.1006171927,
-0.2720005512,
0.2932376266,
0.1757777929,
-0.4867838621,
-0.061439015,
-0.2895148695,
-0.4792464375,
0.2440947145,
-0.1913364232,
-0.1096363217,
0.3604549468,
0.2841496766,
0.0968425423,
-0.0035267477,
0.4418772757,
0.215323329,
-0.2735733092,
0.1427684724,
-0.5009114146,
-0.0092661232,
0.0609096475,
0.0025332049,
-0.0517023094,
-0.1849664599,
0.3137893081,
0.0846939608,
0.0423716977,
0.0779923946,
-0.0587279275,
0.151083529,
-0.1502036154,
0.1557244658,
0.1671285778,
0.0088139102,
-0.1092446297,
0.2618470192,
-0.1659753323,
-0.5043037534,
-0.1913361996,
0.18926768,
0.5310876369,
0.2051564008,
-0.1579914987,
-0.3544665277,
-0.1397816688,
-0.0417206921,
0.2464243323,
0.3472801149,
-0.3106598258,
0.3282320201,
-0.1817418486,
0.0989635438,
0.1563435495,
0.2996596098,
-0.0352364406,
-0.2530743778,
0.1073725671,
-0.5553369522,
0.4459261596,
0.0170266815,
-0.301894933,
-0.146708414,
0.2151525766,
0.3208591342,
0.0979312584,
-0.3039987981,
-0.1042319536,
0.3273186088,
-0.1401027143,
-0.0631003082,
0.108116433,
0.0052376613,
-0.1605119407,
-0.0747159868,
0.1002992168,
0.4923507273,
-0.2181967199,
0.2114474475,
-0.2496320754
] |
https://github.com/huggingface/datasets/issues/1895 | Bug Report: timestamp[ns] not recognized | Thanks for reporting !
You're right, `string_to_arrow` should be able to take `"timestamp[ns]"` as input and return the right pyarrow timestamp type.
Feel free to suggest a fix for `string_to_arrow` and open a PR if you want to contribute ! This would be very appreciated :)
To give you more context:
As you may know we define the features types of a dataset using the `Features` object in combination with feature types like `Value`. For example
```python
features = Features({
"age": Value("int32")
})
```
However under the hood we are actually using pyarrow to store the data, and so we have a mapping between the feature types of `datasets` and the types of pyarrow.
For example, the `Value` feature types are created from a pyarrow type with `Value(str(pa_type))`.
However it looks like the conversion back to a pyarrow type doesn't work with `"timestamp[ns]"`.
This is the `string_to_arrow` function you highlighted that does this conversion, so we should fix that.
| Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
``` | 159 | Bug Report: timestamp[ns] not recognized
Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
```
Thanks for reporting !
You're right, `string_to_arrow` should be able to take `"timestamp[ns]"` as input and return the right pyarrow timestamp type.
Feel free to suggest a fix for `string_to_arrow` and open a PR if you want to contribute ! This would be very appreciated :)
To give you more context:
As you may know we define the features types of a dataset using the `Features` object in combination with feature types like `Value`. For example
```python
features = Features({
"age": Value("int32")
})
```
However under the hood we are actually using pyarrow to store the data, and so we have a mapping between the feature types of `datasets` and the types of pyarrow.
For example, the `Value` feature types are created from a pyarrow type with `Value(str(pa_type))`.
However it looks like the conversion back to a pyarrow type doesn't work with `"timestamp[ns]"`.
This is the `string_to_arrow` function you highlighted that does this conversion, so we should fix that.
| [
-0.1527936608,
0.3287224174,
0.0040783435,
0.0871676952,
0.0947729796,
-0.0773304105,
0.4513310194,
0.3623485565,
-0.4917006195,
-0.2842441797,
0.2675410509,
0.5988479257,
-0.1803752333,
0.0181879103,
0.0292804688,
-0.1218025833,
0.1887535006,
0.2048997879,
0.0169893932,
0.2929655313,
-0.4117574096,
0.0337460786,
-0.058228001,
0.4490805864,
-0.2121626586,
-0.0986314118,
-0.060336519,
0.0371376202,
-0.1723989546,
-0.6164935827,
-0.0294911899,
-0.292327255,
0.2255379558,
0.4583447576,
-0.0001204796,
-0.0532898307,
0.3580793738,
-0.0033799857,
-0.1952472031,
0.2531561255,
0.2438883036,
-0.2023749053,
0.2693036795,
-0.2410609424,
0.0408184156,
-0.2710428536,
-0.0724562034,
-0.0755449682,
0.0478874929,
0.5628814697,
0.1174847707,
0.3736269772,
0.4515483379,
0.2590076923,
0.8312050104,
0.269094646,
-0.2306952029,
0.0563388094,
0.4777922332,
0.0477019623,
0.2724706531,
0.0612959601,
0.018426463,
-0.2222912908,
0.3558759391,
0.1927735806,
0.2080835849,
-0.2612552345,
-0.2831192911,
0.0733709261,
0.5608475804,
-0.4021898806,
-0.3237974346,
-0.2525912225,
-0.3008704185,
-0.3398885429,
0.1143041775,
-0.0021666437,
0.0236199722,
0.1635646522,
0.2240137607,
0.1570544243,
-0.253872484,
0.3059595525,
0.0345588177,
0.1951956004,
0.036505945,
0.2330447733,
-0.1538335979,
-0.1633094251,
0.4218268692,
-0.0682977885,
0.2391828448,
0.2117651254,
0.0548947938,
-0.002604261,
0.0709579289,
0.0192173552,
-0.0629314482,
0.2086342722,
-0.3227856755,
-0.0552375652,
-0.0514088199,
0.047479216,
0.1782120466,
0.1103379801,
-0.04657083,
0.196129024,
0.169195652,
0.0495368168,
0.2670125365,
-0.0758633986,
0.0448121391,
-0.3755870759,
0.2224443704,
0.1253213137,
0.5929381847,
-0.1710084081,
-0.1433097422,
0.0585861802,
-0.3615441322,
-0.1077892333,
0.1066296995,
0.1574503481,
-0.1743688285,
0.4042027295,
0.220969677,
0.1066838056,
-0.0398551151,
-0.1873345226,
-0.1595071107,
0.0632437021,
-0.3129244447,
-0.0837126821,
-0.052570492,
0.0856094584,
-0.1687336862,
0.1228703409,
-0.2593176067,
-0.0080448911,
-0.2724781632,
-0.0224867649,
-0.0080478191,
0.2598342299,
-0.3691706061,
0.1590272039,
0.0179570932,
-0.2260667086,
-0.3272782564,
0.2760612369,
-0.3891296685,
-0.3840899467,
-0.2853717804,
0.0803796798,
-0.2141952962,
-0.0022991276,
-0.120146066,
0.0653201118,
0.3260179162,
-0.2303648591,
0.1859743744,
0.0410008691,
-0.0157307312,
-0.4460173249,
-0.0160409585,
-0.0757953152,
-0.5357326269,
-0.0214965716,
-0.0532198995,
-0.18646954,
0.2656604946,
0.1523072422,
0.0214726105,
0.1077252179,
-0.0525945723,
0.1655794829,
0.1830978394,
-0.0440339893,
-0.100068897,
-0.062841326,
0.1198739037,
-0.1564227045,
-0.0947116613,
0.1092775613,
-0.0736897215,
-0.3927890062,
0.0568927974,
0.154198885,
-0.1479738206,
-0.1493550241,
-0.2447481602,
-0.2944936156,
0.3030257523,
0.1998862177,
0.2393239886,
-0.0407983921,
0.0045120418,
-0.5796672106,
0.1131131947,
-0.0502080284,
0.1263217628,
-0.054815121,
0.5200362802,
0.214190051,
-0.0707634464,
-0.3474245667,
-0.4297999144,
0.1461999565,
-0.0691068694,
0.1610941589,
-0.4859772027,
-0.3641881645,
-0.2271836996,
0.2800624073,
0.1993658692,
0.1383029521,
-0.0436112918,
-0.0297361203,
0.1434595883,
0.333236903,
-0.1515098512,
0.0788407326,
-0.1270582974,
0.2203105688,
-0.2441529036,
0.296492219,
-0.2026635855,
-0.2622420788,
-0.1208605021,
0.3751483262,
0.1991806477,
0.0832861662,
0.0865879655,
0.2168285698,
0.1563479602,
0.0290441047,
-0.4961469173,
0.2458916605,
0.1520672292,
-0.2512366772,
-0.2621864974,
0.355476588,
0.2291230261,
-0.0469217002,
0.0833045691,
0.4930121303,
-0.2353566885,
0.0319038779,
-0.1030334681,
0.1757521629,
-0.0606437139,
0.0314013138,
-0.2132782787,
0.0498018786,
-0.2454683334,
0.1282762587,
0.170557797,
0.0274255257,
-0.4170606732,
0.2885538638,
0.4863221645,
-0.1517169476,
0.1151833981,
0.1635687351,
-0.2545302212,
-0.1554028243,
0.0685560852,
0.015366286,
0.173956424,
0.1825348884,
-0.027013965,
-0.0015857071,
-0.3035329878,
0.0931104869,
0.2514899969,
0.1319378912,
0.1806955636,
-0.0920818225,
0.1278416812,
0.1615656912,
0.0097598881,
-0.1166410223,
-0.0649032816,
0.3374014497,
-0.6451542974,
0.1693529189,
-0.2670384645,
-0.2664467692,
-0.0653510615,
-0.5234566927,
-0.2097802758,
-0.4976218641,
0.2060505301,
0.0130745135,
-0.1290356964,
0.1445627958,
-0.2396760434,
0.329580307,
0.0367026776,
-0.0488290936,
-0.1746849567,
-0.329413265,
-0.3490489721,
0.0426957682,
0.122763887,
0.1496357024,
0.238105461,
-0.2039759904,
0.2300888002,
-0.1850443333,
-0.5263569951,
0.0271415524,
-0.2452928722,
0.5031445622,
0.1099937409,
0.128012687,
0.155254662,
-0.1925951838,
0.1817749441,
-0.245816499,
-0.3332669437,
0.0371793061,
-0.157066524,
0.147702828,
-0.1943174005,
-0.0874928162,
-0.5084862113,
-0.3264510036,
0.2179850638,
-0.1214898005,
0.2835151553,
0.4560056329,
-0.1136699468,
0.0235684179,
-0.0690424517,
0.0206421949,
-0.245624423,
-0.0137550905,
0.4213277698,
-0.1864164621,
-0.3269936144,
-0.0897911638,
-0.1323458701,
-0.4214050472,
0.0435696058,
-0.1114261448,
-0.5367971659,
-0.0529440269,
0.4217866659,
0.1056531295,
0.0251025651,
0.3719082773,
0.1785486788,
0.063028425,
-0.1003422737,
-0.2586352527,
0.0566187166,
0.1510968953,
0.1743384451,
0.2761528492,
0.2530387938,
-0.0747739375,
0.9659366608,
0.1138235629,
-0.4457322359,
0.3898625374,
-0.3384913504,
0.1157401651,
-0.0298456848,
-0.0037475228,
-0.283567965,
-0.2376618683,
-0.0190819725,
0.1120849997,
-0.0420456305,
-0.2181745321,
-0.0301217698,
0.1389136761,
-0.6106888056,
-0.3204310238,
0.1585178524,
0.1101705581,
0.2406479269,
-0.110891901,
-0.0829684511,
-0.305722177,
-0.2346668988,
0.0578369424,
0.1387051791,
0.0329603553,
0.064321667,
-0.4295960665,
-0.0718573034,
-0.4033311903,
0.2331345677,
0.0878610685,
0.2045991272,
0.086094372,
-0.4413841963,
-0.0051742643,
-0.0607056692,
0.3204252124,
-0.1267915219,
0.0915326774,
0.2419283539,
0.1802394986,
-0.4087588489,
-0.1198683977,
-0.0064535215,
-0.0864057094,
-0.1325340569,
0.2273883373,
-0.6342526674,
-0.1165587455,
0.1460366845,
-0.0613227859,
0.0395365208,
0.1096672863,
0.0564646348,
-0.1275828481,
-0.074372001,
-0.0667923018,
0.3582877517,
0.3050358891,
0.0262299627,
0.1235985905,
-0.2668555975,
0.180650413,
0.1099920049,
0.1013184935,
0.2407739013,
-0.458218962,
0.2066278458,
-0.1807591468,
0.0120182913,
0.4070290327,
0.7095594406,
0.0807233229,
-0.3991870284,
-0.0225861501,
-0.0491637588,
0.2766856849,
0.4198621511,
-0.1451696157,
-0.0110798217,
0.0248730071,
0.034269657,
-0.3001813293,
-0.139523685,
0.2934096754,
-0.0880784839,
-0.3187105656,
-0.0402576327,
0.3679548204,
0.2904570997,
-0.1857379973,
0.2262471914,
0.3401791155,
-0.1935940981,
0.2715736032,
0.1173752025,
1.1871391535,
0.0625182092,
0.2902382016,
0.6918599606,
-0.2262695134,
0.3874614537,
0.1182626635,
-0.1172609627,
-0.4156976044,
0.0445863083,
-0.0881239548,
-0.0749850795,
0.1273634583,
0.1820989549,
-0.2661907077,
0.127476871,
-0.109205097,
0.5751079321,
0.1092149392,
0.0284857452,
0.3513975441,
0.0581370704,
-0.3950576782,
0.0595080629,
-0.1463334113,
-0.066967532,
-0.1203709394,
-0.3600037396,
-0.4582648277,
-0.2313316911,
-0.3740947843,
-0.141871497,
-0.1527080238,
0.1279592067,
0.3998520374,
-0.2702070773,
-0.1875248104,
-0.0428726859,
0.1339923888,
-0.4898432791,
0.0362874269,
0.0534932353,
0.1385067701,
0.0911613032,
-0.0083293952,
-0.0276890472,
0.4077920616,
-0.1116942316,
0.0431236327,
0.1187886223,
-0.168257907,
-0.4002342224,
0.0978208184,
0.1089321077,
-0.3288659155,
-0.1873814762,
-0.3726101518,
0.0175437257,
0.1215214431,
-0.2492666692,
0.0417998172,
-0.0047182869,
0.1482403427,
0.3003706932,
-0.2605713308,
-0.1535680294,
-0.1635640711,
-0.0087851733,
-0.1962372661,
0.2959886193,
0.0721458197,
0.44993788,
-0.2437379658,
0.0095925778,
0.22667934,
0.2774643898,
-1.1125528812,
0.1213630661,
-0.217654869,
-0.0115720034,
0.1735292375,
0.4090936184,
0.2185938805,
0.0233567487,
0.0249560773,
-0.3650183976,
-0.2235945761,
0.4313727915,
0.029100623,
0.1161477417,
-0.1057687402,
0.1679747105,
-0.0146392845,
-0.1159443483,
-0.2461661398,
-0.0055376664,
-0.073109515,
0.3581995666,
-0.1237031221,
-0.1004959121,
0.0893468559,
-0.1796739548,
0.001819592,
0.0249230489,
-0.1933642775,
-0.0921413004,
-0.1605199277,
0.2176163048,
0.0471407771,
-0.2346195281,
0.0659840405,
0.0065068901,
-0.1924471408,
0.0714405477,
-0.0881934762,
0.3197441697,
-0.118250683,
0.3038514256,
0.5535710454,
-0.040660575,
-0.0375195593,
0.0780659616,
-0.0039794445,
0.5921233296,
0.0860235393,
0.3497181535,
-0.1798066497,
0.0273920447,
-0.2085114419,
0.298938632,
-0.0330961421,
-0.2361285388,
0.2767896056,
-0.2176705748,
-0.1418225467,
0.0069900416,
0.3088606894,
0.3089556098,
-0.161372751,
-0.0572144315,
0.1093277708,
0.0551286303,
-0.3600951433,
0.0201886445,
0.1924484372,
-0.009106826,
0.0326310694,
0.3184308708,
0.2696882188,
0.4179320335,
0.081450358,
0.0487552881,
0.1866779625,
0.0920561552,
0.0785364062,
0.5822567344,
-0.1735045612,
0.1097607538,
0.4740929306,
0.2046062052,
0.2165539563,
-0.0329337344,
0.1398325711,
-0.0628830567,
0.5334262252,
0.1327576041,
0.3355413377,
-0.1961827278,
-0.2704680264,
0.0710256025,
0.0618192181,
-0.0300510898,
0.4114813209,
0.4589701891,
0.1298113018,
-0.0805712789,
-0.00946711,
0.3307629824,
0.1050334275,
-0.1596031487,
0.1815198958,
-0.1845351011,
-0.3742823005,
-0.2979974151,
-0.3258636892,
-0.2028909028,
0.5982096791,
-0.0280541852,
0.0776660219,
-0.283575207,
0.2046012878,
0.0635139048,
0.4366499782,
0.0131073818,
0.0410351679,
-0.0268680975,
-0.2891231775,
0.2568718791,
0.3931130767,
0.6102509499,
0.2314990759,
-0.3602035046,
0.0619788617,
-0.1825956851,
-0.0692615658,
0.1624371856,
0.1441892385,
0.2920225859,
-0.0268784761,
0.325353533,
0.0861776844,
0.0038222075,
0.0794859678,
-0.2405526191,
0.2377732545,
-0.0654831082,
0.6614518166,
-0.2060134858,
-0.17969051,
0.0107789859,
0.171962291,
0.0381425992,
-0.0641421452,
0.3810345232,
0.2532756925,
0.1315001249,
-0.1354816109,
0.0227105729,
-0.0019414928,
0.2765327394,
0.3934855461,
0.5602558255,
-0.0075215045,
-0.0518900231,
-0.5826061964,
0.0651909485,
-0.2603630722,
0.0356153995,
0.5112120509,
0.0183603317,
0.2061680257,
0.2826226652,
0.3524315953,
-0.0367423221,
-0.2190715969,
0.2220224142,
-0.2706004679,
0.1071839631,
0.0888692886,
0.0353246331,
-0.0358550772,
-0.2874417007,
0.2049768567,
-0.1357735991,
-0.0541670322,
-0.2018859684,
-0.2639065087,
0.0873278826,
0.0622770935,
0.6423354149,
0.2027245164,
0.2413538247,
0.0302190706,
-0.136701107,
-0.1784454882,
-0.2873968482,
-0.381934315,
0.3547769785,
-0.3083920479,
0.1692907512,
-0.229847312,
-0.0136591885,
-0.2985233366,
0.1240805611,
-0.071254231,
-0.0334208794,
-0.2776943147,
0.2480270267,
-0.2848936915,
0.0948364288,
0.1316544116,
0.0741192251,
-0.0224251784,
0.3093958199,
-0.1988306344,
-0.6252143979,
0.4662373066,
-0.0962075591,
-0.1410777867,
0.1713167876,
-0.0737686232,
0.2998207211,
-0.0596598163,
-0.4555013776,
-0.3096268773,
0.3819793463,
-0.060506884,
-0.2291326374,
-0.1350295395,
-0.1367223412,
-0.120448634,
-0.1010308638,
0.351950407,
-0.2100378275,
0.1753137112,
-0.0738834888,
-0.4196283519
] |
https://github.com/huggingface/datasets/issues/1895 | Bug Report: timestamp[ns] not recognized | Thanks for the clarification @lhoestq !
This may be a little bit of a stupid question, but I wanted to clarify one more thing before I took a stab at this:
When the features get inferred, I believe they already have a pyarrow schema (https://github.com/huggingface/datasets/blob/master/src/datasets/arrow_dataset.py#L234).
We then convert it to a string (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L778) only to convert it back into the arrow type (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L143, and https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L35). Is there a reason for this round-trip?
I'll open a PR later to add `timestamp` support to `string_to_arrow`, but I'd be curious to understand since it feels like there may be some opportunities to simplify! | Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
``` | 100 | Bug Report: timestamp[ns] not recognized
Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
```
Thanks for the clarification @lhoestq !
This may be a little bit of a stupid question, but I wanted to clarify one more thing before I took a stab at this:
When the features get inferred, I believe they already have a pyarrow schema (https://github.com/huggingface/datasets/blob/master/src/datasets/arrow_dataset.py#L234).
We then convert it to a string (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L778) only to convert it back into the arrow type (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L143, and https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L35). Is there a reason for this round-trip?
I'll open a PR later to add `timestamp` support to `string_to_arrow`, but I'd be curious to understand since it feels like there may be some opportunities to simplify! | [
-0.0739532709,
0.2633906901,
0.0234289877,
0.1172331795,
0.0697292835,
-0.1028006822,
0.348251909,
0.2648755312,
-0.5368142724,
-0.2700379193,
0.2399615347,
0.4913213849,
-0.1497713774,
-0.0816282257,
0.0596875772,
-0.1228537261,
0.1500018686,
0.1810297519,
0.0497960262,
0.276060611,
-0.4647657275,
0.026809046,
-0.117803432,
0.4554686844,
-0.1349895746,
-0.0517359823,
-0.1678007841,
0.1312937289,
-0.2226994038,
-0.6291897297,
-0.0068090633,
-0.2124489099,
0.1671675146,
0.5190417171,
-0.0001214839,
-0.0243570209,
0.3058463633,
-0.081429705,
-0.0674783066,
0.2833606303,
0.3476235271,
-0.0902879089,
0.2400756478,
-0.2225872278,
0.0005370975,
-0.2519569397,
-0.0341049656,
0.0672974735,
0.1444161534,
0.4907273054,
0.1079130173,
0.3850308657,
0.4924411476,
0.2931484878,
0.8187013268,
0.2305019498,
-0.232706055,
0.2020153701,
0.4911268353,
-0.0119445026,
0.3040031493,
0.1110464782,
-0.020267956,
-0.3036398888,
0.4174170494,
0.2343474329,
0.2308803201,
-0.2514508665,
-0.3207890391,
0.0124455476,
0.4684600532,
-0.4220658243,
-0.2735633552,
-0.2206661105,
-0.2268996537,
-0.3210523129,
0.0590609312,
0.0244469866,
-0.0944184214,
0.1667209268,
0.1804579794,
0.1114423871,
-0.2595478296,
0.1966617256,
0.0458393097,
0.2566615641,
0.0918179974,
0.2258581966,
-0.1358293146,
-0.0887858346,
0.3104706407,
0.0069872402,
0.1979390681,
0.2608563602,
0.1146382317,
-0.0238551944,
0.1147633567,
0.1258626729,
-0.0543546602,
0.2180196047,
-0.275297761,
-0.1187838912,
0.0749940127,
0.0247222316,
0.1478240043,
0.1771612465,
-0.0778385103,
0.1958504617,
0.1620577723,
0.1409116983,
0.2483190894,
-0.0972782969,
0.0117753968,
-0.3109064102,
0.0915320516,
0.036988385,
0.4998414814,
-0.1672777534,
-0.0820687115,
0.0809870809,
-0.3534942567,
-0.0999072492,
0.1619464308,
0.2031746209,
-0.175402835,
0.2231133878,
0.297590673,
0.0646354407,
-0.0271321721,
-0.1359475255,
-0.1771474779,
0.0740087181,
-0.3025512099,
0.041939117,
-0.1343197227,
0.1324588507,
-0.1791501343,
0.1216743588,
-0.2732576728,
0.0485025346,
-0.2948606908,
-0.0084001347,
-0.0298532546,
0.2891800404,
-0.3410618901,
0.0869280696,
0.0180823579,
-0.2282739133,
-0.2970761657,
0.2654173374,
-0.261515975,
-0.4053732157,
-0.3469693363,
0.0767560229,
-0.2768324018,
0.0730394199,
-0.0575079918,
0.0426733606,
0.3180197477,
-0.2170474678,
0.1977299601,
0.1241653636,
0.0056942254,
-0.4275838137,
0.0414002761,
0.0141741484,
-0.4683755338,
-0.0072705597,
0.0773428977,
-0.203006655,
0.2537609935,
0.104912661,
-0.0016871169,
0.0774713457,
-0.0125499032,
0.2026351392,
-0.0127513483,
-0.0462621339,
-0.0595333576,
-0.0833552852,
0.0376258045,
-0.1311549544,
-0.0338490717,
0.0006080084,
-0.1404958665,
-0.3639113009,
0.0298568644,
-0.0254538972,
-0.0889191404,
-0.2139907181,
-0.3913457692,
-0.30402565,
0.1618695855,
0.2460024506,
0.2437742651,
-0.050209377,
-0.0592441522,
-0.5101995468,
0.2055174857,
-0.024705939,
0.0671348646,
0.0306329988,
0.4889729619,
0.2637949288,
-0.0420462005,
-0.2111524045,
-0.3733324707,
0.1900897324,
-0.1035774127,
0.2703512311,
-0.5419313312,
-0.4424498975,
-0.2253597826,
0.3193967342,
0.1861851066,
0.040704757,
-0.0255179219,
-0.0455300957,
0.2115220129,
0.3166397214,
-0.0906741843,
0.0358714014,
-0.1910473108,
0.218768388,
-0.3219425082,
0.3889368773,
-0.2348879129,
-0.2647876143,
-0.1389499605,
0.3878715634,
0.2070740908,
0.0228758119,
0.1657654643,
0.2556868494,
0.1249754578,
0.0624820031,
-0.4905864596,
0.2075887471,
0.1242333874,
-0.1908749342,
-0.3281772137,
0.2835144401,
0.1718750894,
-0.0430424884,
0.0834822208,
0.473420918,
-0.238683939,
-0.0334386379,
-0.0330952406,
0.1673657298,
-0.0946147814,
-0.0246203542,
-0.1773063987,
0.0732665807,
-0.2078604549,
0.1559967995,
0.134933576,
0.0551785752,
-0.3709950149,
0.2192386538,
0.4112531543,
-0.1022618636,
-0.0292533264,
0.0717291832,
-0.2304463387,
-0.0914865285,
0.1116987765,
-0.067477189,
0.1516864151,
0.1919139177,
-0.0590075664,
0.1059635207,
-0.2441173792,
0.0980610996,
0.2665566206,
0.1425212026,
0.2044154704,
-0.1282247156,
0.0713002607,
0.1824332625,
-0.0108704865,
-0.1943415254,
-0.0333505794,
0.3838221729,
-0.6351844072,
0.1070172936,
-0.1585381329,
-0.3386042118,
-0.0780427158,
-0.5197442174,
-0.2630253732,
-0.4474777281,
0.2224145383,
-0.0186962932,
-0.1537582874,
0.0871336982,
-0.2732032239,
0.2935720682,
-0.024544768,
-0.0188588388,
-0.2927139997,
-0.3856007457,
-0.2980496883,
0.0385947563,
0.1031717956,
0.1414163113,
0.186309129,
-0.2234017849,
0.1416749954,
-0.0752854347,
-0.580976367,
0.0517928153,
-0.2343017757,
0.4214695394,
0.0943550467,
0.1354866624,
-0.00835675,
-0.2265823781,
0.2086319923,
-0.1833893657,
-0.2840285897,
0.0269484073,
-0.1109138131,
0.077455759,
-0.2562033534,
0.0519673154,
-0.4750353396,
-0.3110540509,
0.3278079927,
-0.0538621619,
0.244488284,
0.4900135398,
-0.2033355236,
0.0783637911,
-0.186273247,
-0.0581390262,
-0.1955920458,
-0.0609746203,
0.4027331471,
-0.1493616104,
-0.3285056651,
-0.1424712837,
-0.0957722813,
-0.3075482249,
0.0162442364,
-0.2464180887,
-0.5779741406,
0.0321431458,
0.3910687268,
0.0669245198,
0.0162401367,
0.3755380809,
0.0941713005,
0.0749979764,
-0.0748712122,
-0.139576003,
0.0567207485,
0.1027209312,
0.1628658772,
0.1843211204,
0.228666991,
-0.1184220463,
1.0252785683,
0.1548279524,
-0.3985678852,
0.2973777056,
-0.4068079293,
0.1511543095,
0.1008079052,
0.0101967379,
-0.2597641051,
-0.1983231306,
-0.0663354248,
0.1671135426,
-0.0206994656,
-0.0396875665,
-0.1277154088,
0.1601395309,
-0.5489756465,
-0.3292186856,
0.1012685522,
0.1339628696,
0.2427388579,
-0.0709932446,
-0.0134774894,
-0.2422406077,
-0.2760464847,
-0.007136181,
0.1489181221,
0.1206923872,
0.0424135849,
-0.3474947214,
-0.07462053,
-0.5400822759,
0.200055182,
0.09041325,
0.104606688,
0.1293871105,
-0.440654099,
0.0726008117,
-0.0931582302,
0.3412595689,
-0.0809386745,
0.0096040489,
0.259365797,
0.2304244936,
-0.4890091419,
-0.1646180451,
0.0079613551,
-0.1361945719,
-0.0026926138,
0.2836538851,
-0.6594839096,
-0.0884753913,
0.1195068508,
-0.081374377,
0.0034234077,
0.1037500948,
-0.0359478891,
-0.2243494391,
-0.0199495926,
-0.0011441559,
0.344655931,
0.2038824707,
0.0385995209,
0.2203024328,
-0.1765737534,
0.2099398375,
0.0870146751,
0.2514796853,
0.237882182,
-0.4663799405,
0.2183128297,
-0.1073824167,
-0.0237034131,
0.4341365695,
0.8395715356,
0.0945456401,
-0.3643737137,
-0.0120035587,
-0.106465295,
0.2241787612,
0.4501791894,
-0.1059036404,
0.0466722436,
0.0237453692,
0.1539506614,
-0.2946676612,
-0.2546076179,
0.3219578266,
-0.261338979,
-0.2040187567,
0.0996509865,
0.4065248072,
0.2346212864,
-0.2085292041,
0.1910949051,
0.3937910199,
-0.1838940978,
0.3359523416,
0.1511057615,
1.270280242,
0.0139771914,
0.2444763482,
0.8142850995,
-0.2683042288,
0.4792146683,
0.0589139014,
-0.1071833074,
-0.4414277673,
-0.0196417514,
-0.05120809,
-0.0609590374,
0.0706432164,
0.2082357705,
-0.298569262,
-0.0040788949,
-0.0550605357,
0.6143824458,
0.1036399975,
0.0107558519,
0.4065077305,
0.1416152716,
-0.4835610092,
0.0149957091,
-0.0665135384,
-0.0387882218,
-0.2164725363,
-0.3448265493,
-0.4433392882,
-0.2966755331,
-0.3549309671,
-0.0789484382,
-0.2381044775,
0.0777386576,
0.3928999901,
-0.1791934669,
-0.2161376476,
-0.0108063752,
0.1059907228,
-0.468914181,
-0.0151649835,
0.096690163,
0.241517663,
0.0263712965,
-0.0251892544,
0.0395318642,
0.4224718213,
-0.1926265955,
0.0911814123,
0.0659669265,
-0.227188915,
-0.4013901949,
-0.0803044885,
0.1792919785,
-0.2816669941,
-0.1221653894,
-0.2321113348,
0.0369483232,
0.1103407592,
-0.222700268,
0.0303715989,
0.0000719912,
0.1590046138,
0.3048903048,
-0.3273972273,
-0.170645684,
-0.1958442479,
0.0490123555,
-0.1522260159,
0.1817831993,
0.1040559709,
0.4240527153,
-0.2806767225,
0.0272602066,
0.1039106697,
0.3103793859,
-1.2080916166,
0.0783216655,
-0.1443828046,
0.1256977171,
0.1736297309,
0.3578544557,
0.350212574,
-0.0629296303,
0.0975495949,
-0.3580777049,
-0.1638458371,
0.5015026331,
0.1614504009,
0.0878616199,
-0.0682929009,
0.3125722706,
-0.0154463984,
-0.0080321711,
-0.2533616424,
0.0375947952,
-0.0609769747,
0.3026731014,
-0.0879840031,
-0.0993151367,
0.0818225294,
-0.2171733677,
0.0216178969,
0.1100185215,
-0.1998911649,
-0.0590152778,
-0.1486637294,
0.217329815,
-0.0103565967,
-0.2744112015,
0.1058252007,
0.0243902281,
-0.1482447088,
0.1286579221,
-0.0399121419,
0.3466379046,
-0.0478721745,
0.1967079639,
0.629301846,
0.0310006849,
-0.2173806131,
0.0282815136,
0.0387977883,
0.6772252917,
0.1184673458,
0.3900328875,
-0.2114503682,
0.0099746585,
-0.149438113,
0.2731032968,
-0.1109384596,
-0.2645557523,
0.3118438423,
-0.1926223338,
-0.1457065046,
-0.1146255732,
0.3807466626,
0.3744280934,
-0.1657449454,
-0.1251213849,
0.1148408577,
0.0503028706,
-0.3930955529,
0.068607606,
0.1412354261,
0.0108563676,
0.0223657042,
0.3505512476,
0.2716421187,
0.3389985561,
0.0833525658,
0.0825151801,
0.2047493011,
0.2121265233,
0.0339286327,
0.6880730391,
-0.2037144005,
0.067374289,
0.4253975749,
0.2280193418,
0.2544557452,
-0.1298795193,
0.1423653662,
-0.0018904191,
0.2977820635,
0.1041296944,
0.2363479882,
-0.1910445094,
-0.2957144976,
0.1016254351,
0.0756593645,
-0.0666562989,
0.2682634294,
0.5727298856,
0.1583224237,
-0.0059439652,
-0.0279539712,
0.3884643614,
0.127522856,
-0.0853435919,
0.1714217961,
-0.2963219285,
-0.412691772,
-0.2512806654,
-0.2750641406,
-0.1715149879,
0.5290327668,
-0.0194255076,
0.1003439873,
-0.1836781204,
0.1278780997,
0.0605847687,
0.4979171753,
-0.0662817806,
0.0501496717,
0.0980921388,
-0.2756038904,
0.2209878713,
0.3142208159,
0.6572130322,
0.172727704,
-0.2627298832,
0.053960599,
-0.1522947252,
-0.0122875981,
0.1789769232,
0.1686478555,
0.3814448714,
-0.0905968472,
0.3732958436,
0.0567316785,
-0.0017730556,
0.0363790058,
-0.2228272855,
0.2455722392,
-0.018854104,
0.6408085227,
-0.2512755394,
-0.2848408222,
0.0452857204,
0.0861149281,
0.0581636652,
0.0230360888,
0.0873262212,
0.2503396273,
0.1181758791,
-0.15176256,
0.0167905465,
0.0157951191,
0.3172743917,
0.4708020985,
0.5868932009,
-0.0127282273,
-0.2436190397,
-0.618245244,
0.137902081,
-0.2962486744,
-0.0406313799,
0.5741755962,
-0.0034812093,
0.1780951321,
0.2483142465,
0.2168858498,
-0.146063298,
-0.404574275,
0.2246894538,
-0.3030554652,
0.1460397243,
-0.0170188993,
-0.029210953,
-0.0141216815,
-0.1510021687,
0.2226449698,
-0.160037756,
-0.0567626655,
-0.2143971324,
-0.2009408474,
0.1222720146,
0.0438150279,
0.6215817332,
0.1074526608,
0.2562090158,
0.0313697383,
-0.1709303558,
-0.1546257436,
-0.2967072725,
-0.3674126267,
0.3668495715,
-0.2256983668,
0.1504354179,
-0.1462483108,
-0.2550419569,
-0.3340623081,
0.1019861475,
-0.0748039633,
-0.0062332675,
-0.1942545772,
0.1319134384,
-0.2435485721,
0.0780954808,
0.1977066994,
0.0747335479,
-0.0230464377,
0.3005979657,
-0.2489378154,
-0.6309580207,
0.6126273274,
-0.0882748812,
-0.0696325004,
0.1810093522,
-0.0121725127,
0.1211144328,
-0.1286633164,
-0.4617308378,
-0.2831816077,
0.3873357177,
-0.0093718208,
-0.2162743211,
-0.1066666394,
-0.1331150681,
-0.1938742399,
-0.1164907441,
0.2906968296,
-0.2417821735,
0.1810141504,
-0.0667145923,
-0.3776303232
] |
https://github.com/huggingface/datasets/issues/1895 | Bug Report: timestamp[ns] not recognized | The objective in terms of design is to make it easy to create Features in a pythonic way. So for example we use a string to define a Value type.
That's why when inferring the Features from an arrow schema we have to find the right string definitions for Value types. I guess we could also have a constructor `Value.from_arrow_type` to avoid recreating the arrow type, but this could create silent errors if the pyarrow type doesn't have a valid mapping with the string definition. The "round-trip" is used to enforce that the ground truth is the string definition, not the pyarrow type, and also as a sanity check.
Let me know if that makes sense | Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
``` | 116 | Bug Report: timestamp[ns] not recognized
Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
```
The objective in terms of design is to make it easy to create Features in a pythonic way. So for example we use a string to define a Value type.
That's why when inferring the Features from an arrow schema we have to find the right string definitions for Value types. I guess we could also have a constructor `Value.from_arrow_type` to avoid recreating the arrow type, but this could create silent errors if the pyarrow type doesn't have a valid mapping with the string definition. The "round-trip" is used to enforce that the ground truth is the string definition, not the pyarrow type, and also as a sanity check.
Let me know if that makes sense | [
-0.1146751493,
0.1922075748,
0.0449614152,
0.1444823295,
-0.0285735503,
-0.122597672,
0.444470793,
0.2822775543,
-0.5322682261,
-0.3254838884,
0.328125,
0.565228343,
-0.2068838179,
0.0883833691,
0.0604490116,
-0.1087239757,
0.1322406828,
0.1888606548,
0.0694064349,
0.302650094,
-0.4672091305,
-0.0324613564,
-0.0768621117,
0.5418872237,
-0.2132998556,
-0.1348687857,
-0.0447073542,
0.1213393435,
-0.1925988495,
-0.6285111308,
-0.0460074693,
-0.251362592,
0.1139964461,
0.4543079138,
-0.0001248188,
-0.061076086,
0.319085896,
-0.0606154911,
-0.1071546376,
0.3047243655,
0.2991988659,
-0.1720588803,
0.2267457247,
-0.2034840882,
0.0648340881,
-0.2897109985,
-0.0865032151,
0.0162863433,
0.1138692498,
0.5138629675,
0.0579913035,
0.3322010934,
0.4559004307,
0.3164355755,
0.7315241098,
0.2873693407,
-0.3010264635,
0.1178454012,
0.4375963211,
-0.0148453079,
0.2742496431,
0.0601226501,
-0.0338257886,
-0.2455759645,
0.4660461247,
0.1590059698,
0.2337597609,
-0.2711874843,
-0.2636853456,
0.0157379806,
0.6011712551,
-0.3657228053,
-0.3060898781,
-0.2891941965,
-0.2802903652,
-0.3191670477,
0.17801404,
0.0389162228,
-0.0652728155,
0.1685670465,
0.2455637753,
0.0536339171,
-0.2926947474,
0.2109454572,
-0.0086206198,
0.1789676845,
0.1051899493,
0.251760602,
-0.1219964325,
-0.1045324206,
0.4222157598,
-0.0869215205,
0.2680027485,
0.2776352763,
0.1601044238,
-0.0247630104,
0.2341123223,
0.0900534838,
-0.0380095541,
0.2515921593,
-0.2591481805,
-0.1095593721,
-0.0133118443,
0.0278954525,
0.1763332188,
0.091573149,
-0.0460613258,
0.254768014,
0.2294875681,
0.0362675861,
0.2801785767,
-0.0112604108,
0.0494236536,
-0.3504871726,
0.1208941638,
0.0874276608,
0.5513425469,
-0.1017851159,
-0.0562897101,
0.0878305808,
-0.4067277014,
-0.1900676787,
0.1278893799,
0.1757335067,
-0.106148228,
0.3503421843,
0.1927406639,
0.035435535,
-0.0859620497,
-0.1422041506,
-0.12475086,
0.0912907869,
-0.3831537962,
-0.0046768803,
-0.1261663735,
0.149795562,
-0.1982554197,
0.1039196327,
-0.3030175865,
0.0053041875,
-0.2102535963,
-0.0027508028,
-0.0778624639,
0.2609840333,
-0.4186575413,
0.041618444,
-0.0050418708,
-0.147532925,
-0.2964427471,
0.3294449449,
-0.3006253541,
-0.4777574539,
-0.3351680636,
0.0246195197,
-0.2629361749,
0.0942771658,
-0.0529214665,
0.0758387968,
0.3659577072,
-0.2005757391,
0.2076980919,
0.1488825679,
0.0198463053,
-0.4542136788,
-0.0389558971,
0.0858146697,
-0.4874393344,
0.0180766061,
0.0117788985,
-0.1783303022,
0.2273165286,
0.1492523104,
-0.0544849783,
0.2853113413,
-0.0690123737,
0.0304844975,
0.1737557352,
-0.0537424572,
-0.0090087857,
-0.0453505591,
0.0166325718,
-0.1620358825,
-0.0546854436,
0.0518297888,
-0.1548092961,
-0.3739095628,
0.1024481356,
-0.027269436,
-0.1559442133,
-0.1869694293,
-0.3173422515,
-0.2053686976,
0.2707275152,
0.1471915245,
0.2972133756,
-0.0741159618,
-0.0182959177,
-0.5894555449,
0.0808067024,
-0.0670731664,
0.1290231645,
-0.1090822965,
0.5267506242,
0.2543953359,
-0.0195591561,
-0.3033370078,
-0.4691958725,
0.1485734582,
-0.0458119735,
0.2215670198,
-0.467838347,
-0.3387564719,
-0.2100583464,
0.316087544,
0.1810049117,
0.0692278817,
-0.1275725961,
-0.0425603427,
0.1440313309,
0.317738533,
-0.2035168558,
0.0742748901,
-0.2066330314,
0.2454379797,
-0.229748562,
0.2956590056,
-0.2230956703,
-0.2008727491,
-0.1884638667,
0.4087279141,
0.1667343825,
0.0398217924,
0.1125772595,
0.1886122674,
0.2222638875,
0.0503241979,
-0.4616858661,
0.2739303112,
0.1825553924,
-0.2483139336,
-0.2325449735,
0.2102225423,
0.2238047123,
-0.0391979516,
0.0242675096,
0.5012862086,
-0.2386717498,
-0.038110055,
-0.0554042943,
0.1768051684,
-0.0629529059,
-0.0053470582,
-0.2213297635,
0.0950105265,
-0.1456542015,
0.1314736456,
0.1823021472,
0.098871991,
-0.419465363,
0.3065634966,
0.441277951,
-0.184552744,
0.065700449,
0.1669249088,
-0.1256086528,
-0.1821096539,
0.0243959036,
0.0633509755,
0.1932552308,
0.1093817651,
-0.0737956539,
0.0870249048,
-0.3350421786,
0.1240946501,
0.3025605083,
0.1057849526,
0.1506130844,
-0.0427633934,
0.0878643766,
0.1648916304,
0.0059596524,
-0.2358995676,
0.0087932348,
0.3616996706,
-0.6444138288,
0.0834590495,
-0.1029472798,
-0.2071098983,
-0.113999635,
-0.4375004172,
-0.1725717485,
-0.434993118,
0.2647048533,
-0.0309934244,
-0.2029866874,
0.1580699831,
-0.2468388826,
0.4031080604,
-0.0202231035,
-0.0244633202,
-0.189421922,
-0.3819472194,
-0.2783703208,
-0.0091069005,
0.1031071097,
0.1616317183,
0.2546699941,
-0.1953089833,
0.1725183874,
-0.255940944,
-0.5900085568,
0.1090603173,
-0.1908491701,
0.4780799747,
0.2302552164,
0.0542721972,
0.1150304601,
-0.1919829398,
0.1805396676,
-0.2277274728,
-0.3124534786,
0.00685063,
-0.1336706877,
0.1433005035,
-0.1865098923,
-0.0971059799,
-0.4087035656,
-0.3184027076,
0.3017267585,
-0.1244101599,
0.2925083041,
0.3748559058,
-0.1885422915,
0.0390924513,
-0.0794435814,
-0.0317176916,
-0.2190651894,
-0.0592358522,
0.3438012004,
-0.2040378451,
-0.271707505,
-0.07155478,
-0.2190255523,
-0.3240417242,
0.1017379761,
-0.1886049211,
-0.5865525007,
0.0506292954,
0.4496215284,
0.0195625611,
0.057449922,
0.3803962469,
0.2366750985,
0.1188608781,
-0.0545690618,
-0.2511973083,
0.0852107406,
0.1458353698,
0.1823967993,
0.296367228,
0.2636639476,
-0.1237483323,
1.0641583204,
0.1239251196,
-0.3048209846,
0.3154622316,
-0.3813146949,
0.127888456,
0.0245103091,
0.0315708816,
-0.288428545,
-0.1301996261,
-0.1269647926,
0.0954814851,
-0.0099105462,
-0.1974672228,
-0.0531250834,
0.114380978,
-0.551130712,
-0.3061093688,
0.1621592939,
0.0865290835,
0.3027019799,
-0.1337640285,
-0.0555713847,
-0.3662067056,
-0.1868462712,
-0.0149498247,
0.1488547623,
0.1591362655,
-0.0144724436,
-0.4274675548,
-0.1349992901,
-0.4412767887,
0.1400614232,
0.1244694218,
0.1612965614,
0.0797284693,
-0.3914799392,
0.0157235935,
-0.0967815071,
0.3329073787,
-0.2039230913,
0.0405735523,
0.1984585226,
0.2034589499,
-0.3924632967,
-0.1231595725,
0.0506472439,
-0.077662006,
-0.1415316164,
0.2285327762,
-0.6666915417,
-0.083203733,
0.1930224597,
-0.0938977525,
-0.0049131885,
0.1773254722,
0.0341728628,
-0.2593522668,
-0.0734565035,
-0.0109665319,
0.3833388984,
0.1678201556,
0.1851973832,
0.2016892433,
-0.1962920874,
0.2447927743,
0.0863718987,
0.2237507403,
0.168403253,
-0.5131531358,
0.2636956573,
-0.1476709545,
-0.0958306566,
0.4380193651,
0.7881330252,
0.0549324602,
-0.3673034012,
-0.019252073,
-0.1671002954,
0.3739964962,
0.4537482858,
-0.1495277882,
-0.0694998354,
-0.0603553839,
0.1180330068,
-0.36181584,
-0.2566978335,
0.2628569603,
-0.1266414225,
-0.2913715541,
-0.0681965649,
0.337643683,
0.2847601175,
-0.2532697916,
0.3074702919,
0.351467818,
-0.1588598341,
0.294798404,
0.1517976671,
1.2444746494,
0.0614706613,
0.2472411692,
0.7527224422,
-0.1919377446,
0.426687181,
0.0509968698,
-0.154717043,
-0.3975501359,
0.0600553565,
-0.1092935503,
-0.1424698234,
0.1742727309,
0.223943308,
-0.2190534174,
0.0858550072,
-0.0274108797,
0.5473248959,
0.1140354127,
0.0093627423,
0.3956254125,
0.0952996537,
-0.3970621824,
-0.0251928233,
-0.1131136864,
-0.1045868844,
-0.1930862963,
-0.2934207916,
-0.3712640405,
-0.3107536137,
-0.4376636446,
-0.1997317821,
-0.1791953295,
0.1093508005,
0.379430145,
-0.1861805469,
-0.1206923276,
-0.0383758508,
0.0713255703,
-0.4952209294,
0.0719387457,
0.0182911213,
0.1696642935,
0.0685540438,
-0.0681992769,
0.0092354976,
0.4175358117,
-0.1730700731,
0.0828035474,
0.1338685155,
-0.2195999622,
-0.4088667929,
0.0619024783,
0.1559413224,
-0.2963090837,
-0.1515338719,
-0.2595838308,
0.0117758252,
0.0861447901,
-0.2296880335,
-0.0149544487,
0.0127764046,
0.1038632095,
0.3013721108,
-0.3263357282,
-0.1554419398,
-0.1420364529,
-0.1136324033,
-0.1120788157,
0.2243079841,
0.0475021303,
0.398650527,
-0.2858335972,
-0.0046175271,
0.2311016619,
0.1710650325,
-1.1378571987,
0.155197233,
-0.1599297822,
-0.078761667,
0.1355423927,
0.3849827647,
0.37261343,
0.0192533452,
0.0697639883,
-0.3278524876,
-0.1900338531,
0.4407494366,
0.112959072,
0.1254667044,
-0.1644764394,
0.3093412519,
0.0484090149,
0.0023711622,
-0.2117621303,
-0.0141785908,
-0.0780037045,
0.3418264389,
-0.0410258621,
-0.0964815021,
-0.0091722757,
-0.2040373981,
-0.0508273095,
0.0441774391,
-0.2312681377,
-0.010058444,
-0.1532643586,
0.2547922134,
-0.007160564,
-0.2541337907,
0.1426767409,
0.0120171234,
-0.226139769,
0.1298966259,
-0.0766760334,
0.3577627242,
-0.1088047177,
0.2292369455,
0.5094118118,
-0.0565431453,
-0.1564702839,
0.1130512655,
0.0622369424,
0.6591525078,
0.1150817126,
0.3603151143,
-0.1920642853,
-0.0039917454,
-0.2051601559,
0.2868704796,
-0.1598636508,
-0.1706919968,
0.2907279134,
-0.116799742,
-0.1360354275,
-0.0547790006,
0.3114767969,
0.3722765148,
-0.136141479,
-0.0720524788,
0.1595932841,
-0.0159566663,
-0.4048152864,
-0.0187317319,
0.2147328407,
-0.0864782706,
0.0691228285,
0.4023577571,
0.2419648767,
0.3705166578,
-0.0420314483,
0.0478396043,
0.2017495632,
0.045004636,
0.1762165874,
0.6065783501,
-0.2892412841,
0.1085500717,
0.5195783377,
0.2501334548,
0.2066776007,
-0.0102569284,
0.2667909265,
-0.0877648443,
0.4458204806,
0.188316524,
0.3032173216,
-0.18757689,
-0.1511213779,
0.1049320623,
0.0029153191,
-0.0958706513,
0.3808428943,
0.3216403723,
0.0992826372,
-0.0064416751,
-0.0491467305,
0.3589021862,
0.1302799284,
-0.1068480313,
0.1717433333,
-0.3093047738,
-0.3062059283,
-0.2060708702,
-0.2572781146,
-0.1837343872,
0.6696664095,
-0.0245141834,
0.1644705236,
-0.1349966377,
0.2490407079,
0.0209420566,
0.468937695,
-0.0211452097,
0.0509723499,
0.0349875428,
-0.2753678858,
0.146172449,
0.3481436372,
0.5539655685,
0.2513148785,
-0.3344511092,
0.0559432916,
-0.2266015708,
-0.0308700018,
0.1379647553,
0.2152819932,
0.3123206794,
-0.0863273144,
0.3279179931,
0.0120885745,
0.0131510235,
0.1103929281,
-0.1746349633,
0.1903256476,
-0.0719168708,
0.6846841574,
-0.2405489683,
-0.2191604078,
0.0798334554,
0.2028457969,
0.1270464808,
-0.0655572191,
0.3195227087,
0.2477748394,
0.1120903641,
-0.1479396373,
-0.0037352145,
0.084995836,
0.335862875,
0.4570593536,
0.5510883331,
0.0002884716,
-0.0565856621,
-0.6022316217,
0.1106291637,
-0.2849781215,
-0.0564050861,
0.5855396986,
0.0052763969,
0.1396756321,
0.263587743,
0.3131504655,
0.0294617936,
-0.2463750243,
0.2116759121,
-0.2150539458,
0.0883641541,
0.1211811006,
0.0210586581,
-0.0386589468,
-0.2935078144,
0.1734873503,
-0.0886842981,
-0.1207084954,
-0.1904479712,
-0.2710630894,
0.078265585,
0.0362032503,
0.6241985559,
0.2537047565,
0.2888673842,
0.0458206385,
-0.2056273818,
-0.1399154812,
-0.3122879565,
-0.363578707,
0.3949232399,
-0.2031982839,
0.128341794,
-0.2055516839,
-0.0619658381,
-0.3853030205,
0.067685768,
-0.0603736937,
0.039914269,
-0.2818532586,
0.2246339917,
-0.2620016932,
0.0083149262,
0.1375734806,
0.126462698,
-0.0457654297,
0.2847644091,
-0.1562809497,
-0.5460676551,
0.5290355086,
-0.0651914403,
-0.1494638473,
0.2081086636,
-0.0053829476,
0.2078641951,
-0.1047977284,
-0.5080914497,
-0.3518576622,
0.3884011805,
-0.0272089168,
-0.249804765,
-0.1115683243,
-0.187611416,
-0.2963374853,
-0.0898064002,
0.3477852046,
-0.3409021497,
0.1478937268,
-0.0169024169,
-0.4204327464
] |
https://github.com/huggingface/datasets/issues/1895 | Bug Report: timestamp[ns] not recognized | OK I think I understand now:
Features are datasets' internal representation of a schema type, distinct from pyarrow's schema.
Value() corresponds to pyarrow's "primitive" types (e.g. `int` or `string`, but not things like `list` or `dict`).
`get_nested_type()` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L698) and `generate_from_arrow_type()` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L778) *should* be inverses of each other, and similarly, for the primitive values, `string_to_arrow()` and `Value.__call__` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L146) should be inverses of each other?
Thanks for taking the time to answer - I just wanted to make sure I understood before opening a PR so I'm not disrupting anything about how the codebase is expected to work! | Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
``` | 96 | Bug Report: timestamp[ns] not recognized
Repro:
```
from datasets import Dataset
import pandas as pd
import pyarrow
df = pd.DataFrame(pd.date_range("2018-01-01", periods=3, freq="H"))
pyarrow.Table.from_pandas(df)
Dataset.from_pandas(df)
# Throws ValueError: Neither timestamp[ns] nor timestamp[ns]_ seems to be a pyarrow data type.
```
The factory function seems to be just "timestamp": https://arrow.apache.org/docs/python/generated/pyarrow.timestamp.html#pyarrow.timestamp
It seems like https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L36-L43 could have a little bit of additional structure for handling these cases? I'd be happy to take a shot at opening a PR if I could receive some guidance on whether parsing something like `timestamp[ns]` and resolving it to timestamp('ns') is the goal of this method.
Alternatively, if I'm using this incorrectly (e.g. is the expectation that we always provide a schema when timestamps are involved?), that would be very helpful to know as well!
```
$ pip list # only the relevant libraries/versions
datasets 1.2.1
pandas 1.0.3
pyarrow 3.0.0
```
OK I think I understand now:
Features are datasets' internal representation of a schema type, distinct from pyarrow's schema.
Value() corresponds to pyarrow's "primitive" types (e.g. `int` or `string`, but not things like `list` or `dict`).
`get_nested_type()` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L698) and `generate_from_arrow_type()` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L778) *should* be inverses of each other, and similarly, for the primitive values, `string_to_arrow()` and `Value.__call__` (https://github.com/huggingface/datasets/blob/master/src/datasets/features.py#L146) should be inverses of each other?
Thanks for taking the time to answer - I just wanted to make sure I understood before opening a PR so I'm not disrupting anything about how the codebase is expected to work! | [
-0.1425736248,
0.3246959746,
-0.0001188777,
0.1199892014,
0.0674354658,
-0.0765271783,
0.3947216868,
0.2770197093,
-0.4754168689,
-0.2832738161,
0.2034532428,
0.5407938957,
-0.1618388891,
0.0006555095,
0.0657190084,
-0.1031980067,
0.1609962434,
0.2012022734,
0.074615702,
0.2967379391,
-0.5015547276,
0.0138930138,
-0.0977717116,
0.4974122047,
-0.1124669909,
-0.1238010973,
-0.1548793912,
0.158641398,
-0.2299350202,
-0.6235613227,
0.0846158564,
-0.2241215259,
0.2440820336,
0.59282583,
-0.0001198721,
0.0047002062,
0.3186535835,
-0.0295717046,
-0.0745581314,
0.2892122269,
0.3250997066,
-0.1471632123,
0.161763981,
-0.249082312,
0.0260684043,
-0.3570151925,
-0.0386648066,
0.0843393654,
0.0631872192,
0.5538396239,
0.1039332151,
0.3807547688,
0.5164234042,
0.2587672174,
0.791349411,
0.2428606153,
-0.241189152,
0.1639855355,
0.4420026541,
-0.0212106667,
0.3544450104,
0.0041860491,
-0.0741582513,
-0.3006657362,
0.4204320312,
0.1880339682,
0.1718085557,
-0.270450443,
-0.3142209351,
0.029743053,
0.560644567,
-0.4373674691,
-0.2953949571,
-0.1770097613,
-0.2630363107,
-0.3189738989,
0.1110462397,
0.0432382375,
-0.0600383691,
0.1891307235,
0.1954860389,
0.1987186223,
-0.2890895009,
0.1917300075,
0.0323539972,
0.1801726222,
0.0781725049,
0.2091455907,
-0.1302070022,
-0.1141496152,
0.3688924015,
-0.0431591719,
0.1968962252,
0.2251401395,
0.0955171138,
-0.0600449406,
0.1935307533,
0.0480223075,
-0.0821162462,
0.194998771,
-0.3199302554,
-0.1341320574,
0.0023785271,
0.0736260861,
0.1694235057,
0.0672126114,
-0.0499916561,
0.2378935516,
0.2057166696,
0.1243256405,
0.2681445479,
-0.0881943628,
-0.0147984549,
-0.2870157957,
0.1352505386,
0.0673967302,
0.4890877306,
-0.141748637,
-0.1287477762,
0.0631196573,
-0.3517039418,
-0.1332252175,
0.1493441463,
0.2231166214,
-0.173820883,
0.3179600835,
0.2768837512,
0.0402336568,
-0.0383711755,
-0.1485456526,
-0.1716763079,
0.1480691731,
-0.2860535681,
-0.0099424813,
-0.1201553419,
0.1745958775,
-0.241594553,
0.2068474591,
-0.3203714192,
0.0915216506,
-0.2103436738,
-0.0939714238,
-0.0475955121,
0.3335159719,
-0.3074399233,
0.090660803,
-0.0102370325,
-0.2213968635,
-0.3277478218,
0.2611619532,
-0.2281771153,
-0.4078333974,
-0.3511123359,
0.0891688615,
-0.1868233979,
0.099014543,
-0.1131185517,
0.0563397743,
0.3458617926,
-0.2626353502,
0.1782420278,
0.0927587152,
-0.0486756787,
-0.4944673479,
0.0261732787,
-0.0464598313,
-0.5584203005,
0.0517709851,
0.0682190806,
-0.2407458127,
0.2964721024,
0.113672778,
0.0055101924,
0.1590034813,
-0.0400642455,
0.0783469081,
0.0067199469,
-0.0000017583,
-0.0172603149,
-0.0826758742,
0.1535086632,
-0.1110582054,
-0.0318771079,
-0.004098434,
-0.1063612849,
-0.4134666324,
0.0506182462,
0.043564532,
-0.1433008611,
-0.2232988477,
-0.3076543212,
-0.2335508764,
0.2190195173,
0.2242190093,
0.2780327499,
-0.0184107795,
-0.0233397968,
-0.4954234958,
0.1061689556,
-0.0543603823,
0.0125321858,
0.0036807209,
0.4620351493,
0.2442010343,
-0.0523250885,
-0.3410774767,
-0.3519194722,
0.2165430486,
-0.1543736309,
0.2665564418,
-0.4333076179,
-0.4322145879,
-0.246998921,
0.2843511701,
0.2359924316,
0.0394702666,
-0.0127506331,
-0.0075017139,
0.1634653211,
0.3137506247,
-0.1070329472,
0.0478355885,
-0.2006305158,
0.1544550657,
-0.2843419313,
0.3436804414,
-0.2499449998,
-0.2929553986,
-0.1658714712,
0.370357126,
0.1957434416,
0.031840533,
0.1319422722,
0.2155216485,
0.1618632525,
0.0285044108,
-0.476075083,
0.2356524169,
0.1555478275,
-0.1008736044,
-0.3313787282,
0.246457234,
0.185911566,
-0.0580177344,
0.0594785288,
0.4856965244,
-0.1908368766,
-0.0566005334,
-0.0343561098,
0.1692692339,
-0.0513295569,
-0.001323767,
-0.1011033207,
0.1046522409,
-0.2391402572,
0.2305874228,
0.000136286,
0.056586206,
-0.3509813249,
0.2909887731,
0.4456031322,
-0.1674014479,
0.0147181563,
0.0834932923,
-0.1784944832,
-0.1631644666,
0.0915302262,
-0.0366802886,
0.1846927404,
0.1808862686,
-0.0249651335,
0.0427773148,
-0.3133360445,
0.1275975406,
0.2737650275,
0.1087989211,
0.1261410713,
-0.1336374879,
0.0752123892,
0.2009641826,
0.0007416904,
-0.2132478207,
-0.0554304458,
0.4063255787,
-0.6351137757,
0.1275180578,
-0.1732439846,
-0.2968249321,
0.0163702331,
-0.400411278,
-0.1705919206,
-0.4447684884,
0.1812201291,
0.0509732105,
-0.1545253843,
0.0874487311,
-0.2516879439,
0.3074043393,
0.0014128461,
0.0575638935,
-0.258864522,
-0.353238225,
-0.308522135,
0.0650168359,
0.1218186766,
0.1679166555,
0.28746593,
-0.2560242712,
0.156314075,
-0.1229113191,
-0.5098376274,
0.0441637337,
-0.2387232184,
0.4635877907,
0.1750923842,
0.1402406543,
0.0216936916,
-0.2076885998,
0.2008842528,
-0.1902019382,
-0.2915955782,
0.0148492083,
-0.1831305027,
0.1171561778,
-0.2185304761,
0.0000782758,
-0.5082907677,
-0.337680161,
0.2521589994,
-0.0579727516,
0.3331134021,
0.4523908496,
-0.2341369987,
0.0739860535,
-0.1119886488,
-0.0105879623,
-0.2284048647,
-0.0708663613,
0.3865042627,
-0.1658468544,
-0.3249437213,
-0.1007073,
-0.1360385418,
-0.3380486965,
0.009703882,
-0.1703652292,
-0.6069760919,
-0.0062825847,
0.4324369431,
0.0875602216,
0.0493193716,
0.3641598523,
0.0953378752,
0.0434646942,
-0.0619656071,
-0.2019961625,
-0.016594559,
0.1031661779,
0.1257826537,
0.2682312727,
0.2443551719,
-0.1529219747,
0.9732431173,
0.1925331056,
-0.425270021,
0.2978717685,
-0.3933078945,
0.1986406595,
0.1053329557,
-0.0577906221,
-0.2490447909,
-0.1892093718,
-0.0495349802,
0.1236296743,
-0.0490291119,
-0.1206423864,
-0.0839372277,
0.1164635271,
-0.5242835283,
-0.3016832769,
0.1639481634,
0.1452861428,
0.2797558904,
-0.1104722098,
-0.0359599739,
-0.3287212253,
-0.236114502,
-0.0150082596,
0.1915812194,
0.1116149575,
0.0231648535,
-0.4009039998,
-0.0878892094,
-0.4970326424,
0.1842300445,
0.0980046988,
0.1478227824,
0.1063741967,
-0.3894720078,
0.0219027828,
-0.0667377263,
0.4054003358,
-0.2339542359,
-0.0019956275,
0.3085433543,
0.1621927917,
-0.4127203226,
-0.1350302398,
-0.0340221226,
-0.0879735574,
-0.0286876932,
0.2912922204,
-0.6167087555,
-0.1203344315,
0.1549595594,
-0.0830820203,
0.0244567841,
0.0715193003,
0.0052369907,
-0.2415817827,
-0.0116972625,
0.0000383705,
0.3617350459,
0.2321662009,
0.0151413605,
0.1140033305,
-0.2656274438,
0.2865318656,
0.1159244552,
0.2600770593,
0.241989851,
-0.4722296894,
0.1895930916,
-0.1549285054,
-0.0320109352,
0.4399716556,
0.8017579913,
0.0013363529,
-0.3774530292,
0.0692612231,
-0.0825816914,
0.1664777398,
0.3741439879,
-0.1755115241,
-0.0466295555,
-0.0018875953,
0.0905060917,
-0.3189401031,
-0.2506577373,
0.3491077423,
-0.2284454554,
-0.2749235332,
-0.0753931776,
0.3444893658,
0.3377459645,
-0.2223607004,
0.1533694863,
0.2905140221,
-0.1511126012,
0.2589246035,
0.1570721865,
1.206406951,
-0.0153218769,
0.2329270691,
0.8293605447,
-0.2247520983,
0.4708927572,
-0.015407607,
-0.1143921614,
-0.4183697104,
-0.0701898038,
-0.0909998566,
-0.0843941122,
0.1406826377,
0.2364741564,
-0.3014637828,
0.0941989198,
-0.0105564594,
0.6234747767,
0.145023495,
0.0283701718,
0.3794139326,
0.138644442,
-0.4682631195,
0.0396067426,
-0.0489864014,
-0.1091411039,
-0.186329335,
-0.296118468,
-0.3392007947,
-0.2683620155,
-0.4254040122,
-0.1056795418,
-0.1911342293,
0.1342066526,
0.3609836996,
-0.1450232416,
-0.1503578573,
-0.0021079713,
0.1601799428,
-0.469032228,
0.0325322524,
0.0656603128,
0.2302645445,
0.0883295834,
-0.0744974166,
0.0340187326,
0.457460314,
-0.1669486463,
0.0878555179,
0.097376518,
-0.2251965404,
-0.373642832,
-0.0168208331,
0.1957817376,
-0.3043115735,
-0.1824149489,
-0.2569030523,
-0.0583668984,
0.1101160496,
-0.2604665756,
0.0407189094,
0.0249344911,
0.1772093177,
0.364712745,
-0.4047938883,
-0.1612393856,
-0.1925475448,
-0.009280283,
-0.0929726437,
0.2434476018,
0.0639284924,
0.4381603003,
-0.312268436,
-0.0079386309,
0.1511161178,
0.2893584073,
-1.1213923693,
0.173657909,
-0.1421821117,
0.0422295332,
0.2248422056,
0.3096893728,
0.3495359123,
-0.0235745572,
0.069869563,
-0.3813154399,
-0.1188302487,
0.4501493573,
0.1593355536,
0.0845623463,
0.0077528059,
0.2527997792,
-0.0186623484,
-0.0796030983,
-0.2650854588,
0.0204531103,
-0.1176595837,
0.2741378844,
-0.1268700212,
-0.0997328684,
0.0859550163,
-0.2266237885,
0.0168995,
0.0621908605,
-0.1944763511,
-0.0668080077,
-0.1436805129,
0.2118017375,
-0.0286763608,
-0.2419744134,
0.1111502647,
0.0409473553,
-0.2007032037,
0.1260417402,
-0.0283883922,
0.3093132377,
-0.0737163126,
0.2115764767,
0.5683372021,
0.0224352032,
-0.2005321085,
0.0826722458,
-0.026908651,
0.6481399536,
0.0997173339,
0.4128547609,
-0.1809681356,
-0.0154884979,
-0.2070540786,
0.3070142269,
-0.0667618215,
-0.2506662011,
0.3473500609,
-0.1496489048,
-0.1441102773,
-0.0238026008,
0.3532019556,
0.3636677563,
-0.163669765,
-0.0987512991,
0.1394598186,
0.0525337309,
-0.3912109733,
0.0406722836,
0.1080646068,
-0.0568071306,
0.0505806804,
0.3678474426,
0.2521849275,
0.3781633377,
0.0158222169,
0.0582318828,
0.1898457557,
0.1869030446,
0.1009848416,
0.6963256001,
-0.1746328771,
0.0840023011,
0.4598167539,
0.246266976,
0.2071758807,
-0.0805820897,
0.234916389,
-0.0046839509,
0.3384930491,
0.0963290334,
0.3023402393,
-0.2732301354,
-0.187320888,
0.1139710546,
0.0383938588,
-0.0926936567,
0.3359112442,
0.4884850979,
0.0057588443,
-0.0479987264,
-0.0669932067,
0.3490698338,
0.1283696294,
-0.0297105592,
0.1874416918,
-0.2616686821,
-0.3701867461,
-0.2805121541,
-0.2940531373,
-0.1471426636,
0.477112174,
-0.0832234472,
0.112422429,
-0.228641957,
0.2214784026,
0.004658537,
0.4474373162,
-0.0118358061,
0.0205222834,
0.0879584104,
-0.2916252017,
0.1603858173,
0.3764313757,
0.6810595393,
0.2150316536,
-0.2994111776,
0.0295353681,
-0.1387147009,
-0.031643752,
0.174583286,
0.1355930269,
0.261788249,
-0.0924993455,
0.2974662483,
0.0751303881,
0.0038917996,
0.0230820961,
-0.2199762613,
0.2572870851,
-0.0387227386,
0.7626438141,
-0.2543909252,
-0.2159517109,
0.1172539145,
0.1648070216,
0.0974279791,
-0.0208561197,
0.1874940097,
0.2575360835,
0.1307535022,
-0.1604882032,
0.0241172872,
0.0179903917,
0.2956249118,
0.416070044,
0.5409516692,
0.0188522153,
-0.0909028426,
-0.5557604432,
0.1335930377,
-0.2954983115,
-0.1014702544,
0.5853030682,
-0.0006330088,
0.1555575132,
0.2369495332,
0.2687373757,
-0.1316556185,
-0.3562445939,
0.2115816772,
-0.2996394932,
0.1014941037,
-0.0165900998,
-0.0062037315,
-0.0589288771,
-0.2343900204,
0.2052752078,
-0.1019257158,
-0.0483352393,
-0.2327218801,
-0.2531911433,
0.1642947495,
0.0329700559,
0.5891048908,
0.1821758896,
0.2321146131,
0.0006470233,
-0.1848766506,
-0.1043764874,
-0.2903870642,
-0.3705576658,
0.4406277239,
-0.2583461702,
0.1234625727,
-0.1702501774,
-0.1884606779,
-0.3143101335,
0.1531975865,
-0.1151577532,
-0.0582043417,
-0.2314801216,
0.1538599283,
-0.2495626807,
0.025083214,
0.1734545678,
0.0568797365,
-0.0339187793,
0.2845540047,
-0.182142511,
-0.6252740622,
0.5489288568,
-0.0608155467,
-0.1296713799,
0.1641037464,
-0.044765234,
0.0542821437,
-0.0908271149,
-0.3985229135,
-0.2870024443,
0.3825752139,
-0.0081694964,
-0.1730784476,
-0.0893268585,
-0.1535513997,
-0.1996260583,
-0.131586507,
0.2530611753,
-0.2761400938,
0.1293632686,
0.0007358864,
-0.3553195298
] |
https://github.com/huggingface/datasets/issues/1894 | benchmarking against MMapIndexedDataset | Hi sam !
Indeed we can expect the performances to be very close since both MMapIndexedDataset and the `datasets` implem use memory mapping. With memory mapping what determines the I/O performance is the speed of your hard drive/SSD.
In terms of performance we're pretty close to the optimal speed for reading text, even though I found recently that we could still slightly improve speed for big datasets (see [here](https://github.com/huggingface/datasets/issues/1803)).
In terms of number of examples and example sizes, the only limit is the available disk space you have.
I haven't used `psrecord` yet but it seems to be a very interesting tool for benchmarking. Currently for benchmarks we only have github actions to avoid regressions in terms of speed. But it would be cool to have benchmarks with comparisons with other dataset tools ! This would be useful to many people | I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam | 141 | benchmarking against MMapIndexedDataset
I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam
Hi sam !
Indeed we can expect the performances to be very close since both MMapIndexedDataset and the `datasets` implem use memory mapping. With memory mapping what determines the I/O performance is the speed of your hard drive/SSD.
In terms of performance we're pretty close to the optimal speed for reading text, even though I found recently that we could still slightly improve speed for big datasets (see [here](https://github.com/huggingface/datasets/issues/1803)).
In terms of number of examples and example sizes, the only limit is the available disk space you have.
I haven't used `psrecord` yet but it seems to be a very interesting tool for benchmarking. Currently for benchmarks we only have github actions to avoid regressions in terms of speed. But it would be cool to have benchmarks with comparisons with other dataset tools ! This would be useful to many people | [
-0.3810417652,
-0.0305888355,
-0.1813576221,
0.2504057884,
-0.2935996652,
-0.0467079058,
0.0398221128,
0.1909919679,
-0.1505497545,
-0.0903935432,
-0.1634378731,
0.3356113136,
0.0311311819,
-0.4226369858,
0.2201002836,
-0.1310539544,
0.3650215864,
0.0921236426,
-0.2799270153,
0.0275713429,
-0.0007041097,
0.056951426,
-0.2673868537,
0.0297653116,
-0.2419723719,
0.0386082679,
-0.0066018403,
0.0845294148,
-0.0699153841,
-0.2794291973,
0.2004750371,
0.2412212193,
0.0896890461,
0.5924608111,
-0.0001197653,
-0.1850561202,
-0.0232875533,
0.1665840745,
0.0442656279,
0.3798986971,
-0.334876895,
-0.2034230381,
0.1658642441,
-0.1334244609,
0.1651627719,
-0.2025375217,
-0.0441732295,
-0.7676187754,
0.0728935748,
0.1517504454,
0.0765773058,
0.4518502951,
-0.4108698368,
0.0904259235,
0.0827237442,
-0.0421302095,
-0.1984525323,
0.2391929626,
0.5925349593,
-0.0382883325,
-0.0415332727,
-0.0342433341,
-0.2421228886,
0.0677973032,
0.1566680372,
-0.054566212,
-0.2205549031,
-0.0158526637,
0.1271638274,
0.538734436,
0.2411511391,
-0.4278272092,
-0.2604742944,
-0.3222858906,
-0.2503652573,
0.0612294003,
0.134578228,
-0.2026057988,
0.0095427483,
0.0385259353,
-0.5764119625,
0.1521899998,
0.1874470413,
-0.0834676251,
0.0837979615,
0.2970575392,
0.25200665,
0.0690652728,
0.4780254364,
-0.2811427414,
0.2511442304,
-0.3926385939,
-0.1077354476,
0.0773698092,
-0.5279676914,
-0.2925717235,
-0.100003086,
-0.0710348338,
0.1268396378,
0.3535186648,
-0.0782524645,
0.401581198,
0.4423338473,
0.0821369886,
0.1381017119,
0.3148057461,
0.1134951711,
0.1276744008,
0.0943862721,
-0.1098216921,
-0.3615860343,
0.1438648701,
0.2758622169,
0.0285089239,
-0.1372570843,
-0.2606769204,
-0.435425818,
-0.5328080654,
-0.2650945187,
0.1021635681,
0.1557457596,
-0.3873811066,
-0.0231652446,
0.1626878083,
-0.0804627761,
0.1054965705,
-0.4629504085,
0.1284534633,
-0.2905659378,
-0.2079002261,
-0.0178970285,
0.1329869926,
-0.0984139815,
0.4099319279,
0.112796776,
-0.0160042979,
0.2330068648,
-0.0401803814,
0.1472287178,
0.0203947946,
0.608997643,
-0.3896601796,
-0.0092050917,
-0.050961189,
-0.0262422003,
0.1041253209,
0.0037976224,
0.2796249986,
-0.3651384711,
-0.0016902387,
-0.0569005869,
-0.1275284737,
0.3806733489,
0.0101187425,
-0.4040935934,
-0.3670143783,
-0.6880219579,
0.4135318995,
-0.1943508387,
-0.2400918007,
0.1414817572,
-0.0057908269,
-0.2251418084,
-0.1442047358,
0.0685542524,
-0.0652071908,
-0.2823385,
0.0577857494,
-0.2562698126,
0.096420303,
-0.1205663234,
0.4209594429,
-0.3417024612,
0.2553826571,
0.0339744017,
0.1158020496,
0.2943271101,
-0.5225214958,
-0.4324802458,
0.0042806678,
0.2473975718,
0.2319125235,
0.01300437,
0.1114885062,
0.3896989822,
-0.2509155571,
-0.0663527548,
-0.0181876775,
0.1446330398,
0.2138110697,
-0.1329522133,
-0.377420783,
0.1667663306,
0.1763425767,
-0.2033472061,
-0.3884896934,
-0.0828034729,
-0.2697423995,
-0.0737745017,
-0.1108154058,
-0.2474599183,
0.0625224784,
-0.0462348722,
-0.2136874497,
-0.0365386717,
-0.3376495242,
-0.0658711195,
0.2914369106,
-0.2398123145,
0.020189926,
0.3241516054,
0.0338925645,
0.0636359006,
-0.1058216393,
0.0409151539,
-0.2654238939,
-0.0032365322,
-0.1728929877,
0.3089709282,
0.1328316927,
-0.3076333404,
-0.1796546876,
0.173801139,
0.0090149865,
0.1496797502,
-0.0337290317,
0.3308539689,
-0.1629498303,
-0.1025734618,
0.1462417543,
-0.2228515148,
0.0144665949,
-0.1926181018,
0.0757530183,
0.2122279108,
0.330645442,
0.5596897602,
0.459672153,
-0.0421851017,
-0.2652946711,
0.2174793929,
0.0666739047,
0.1624611616,
-0.1951559186,
-0.1541748047,
0.602868855,
0.3290906847,
0.4167497456,
0.0896756202,
-0.4029941559,
-0.2495695353,
0.1632982492,
0.1910123229,
0.2234081328,
0.1918012202,
0.5014185309,
0.3815573752,
0.1069944724,
0.2281944603,
0.1957741976,
0.4031579494,
-0.0130919591,
-0.0780945718,
0.0731680393,
-0.4537007213,
-0.2498406619,
0.358250618,
-0.0080586448,
0.6080089808,
0.0452939793,
0.3489488959,
-0.0202741418,
0.0384118259,
0.2139678299,
0.2788414061,
0.1956565082,
-0.3231141865,
-0.2122598588,
0.0817538053,
-0.099622719,
0.0957183614,
-0.3293739557,
-0.1609759182,
0.1362849623,
-0.0265139565,
0.1889923215,
-0.1423367858,
0.2636855245,
0.1514910758,
-0.091987744,
-0.2330504656,
-0.2190004885,
0.029167939,
0.0568969361,
0.1278788447,
0.0836858302,
-0.0180446543,
0.2556388974,
-0.5038366318,
-0.3434745967,
0.1368795931,
-0.1046151966,
-0.4280408323,
0.0491807051,
0.4018224478,
0.2501461506,
0.3435858786,
-0.0171822086,
0.1546420306,
-0.0059223156,
-0.054626517,
0.0654992461,
-0.0305259619,
0.2342978865,
-0.2400769144,
-0.2851138711,
-0.3815673292,
-0.2068901062,
-0.019485971,
-0.2105272561,
-0.0706487,
-0.3994477391,
0.042696137,
-0.0061723366,
0.3142235279,
-0.0383282788,
-0.1715775728,
0.0247890949,
0.1571283787,
-0.1520950794,
0.0684839115,
-0.6228305101,
-0.0046544187,
0.3014577031,
-0.3341757655,
0.2447915226,
-0.0152125247,
-0.6034365296,
0.1193516254,
-0.0281743091,
-0.1584166884,
-0.0312687308,
0.0633481368,
0.2792359889,
0.1650235504,
-0.2042659521,
0.0022455528,
-0.1447731405,
0.2476503402,
0.205782786,
0.1236647889,
0.3462876976,
-0.0707706958,
0.0550053418,
0.0003555566,
-0.1202046424,
0.0172312092,
-0.2281661928,
0.0850013867,
-0.0106466413,
0.242855534,
-0.1317145973,
0.9007660747,
0.308170259,
-0.2566816509,
-0.0362227373,
-0.1160543114,
-0.0649502054,
-0.2022508234,
-0.0032778978,
-0.1313832849,
0.0836991444,
0.0835964233,
0.4366496205,
0.0672883391,
-0.1921498775,
-0.2168180346,
0.2444776595,
-0.1167562678,
-0.1476194561,
0.142224893,
0.0472921357,
0.0865885615,
0.0697299764,
0.2408089042,
-0.4479860067,
-0.3257125318,
0.1657263935,
-0.1058406383,
-0.2687660456,
-0.2080680877,
-0.4598978162,
-0.2253923714,
-0.4417968094,
0.3698552549,
0.2283752114,
0.3858379126,
0.0419384688,
-0.049347546,
0.1058859602,
-0.0716942921,
-0.1567471325,
-0.1842283756,
-0.2512067854,
0.1228974983,
-0.1194101721,
-0.2804477513,
0.0593972653,
-0.1287489533,
0.0730538219,
0.1754039973,
0.0934486091,
-0.1135773063,
-0.1453135908,
-0.1706187725,
0.1141076982,
0.0440542549,
-0.1558130831,
-0.149545148,
0.02692727,
0.2580503523,
0.1313979924,
-0.3343251944,
0.0245794244,
-0.0810633376,
0.0945233405,
-0.2703288198,
-0.0469454005,
0.4475468099,
0.5082720518,
0.2300993353,
-0.4144334495,
0.4280779362,
0.1398909837,
-0.2195222825,
0.285253644,
0.2222191989,
0.1147794724,
0.1803375036,
0.0512734726,
0.0061853044,
0.1618947387,
0.0543432906,
0.0615260415,
0.159607932,
-0.1863074154,
-0.0937587246,
-0.2137927711,
0.0187278055,
0.251319021,
0.3744756281,
-0.3786368072,
-0.3073333502,
0.6385366917,
0.5235547423,
-0.0965764076,
0.0525797233,
0.012998743,
-0.1762462258,
0.0544871949,
0.2187593132,
0.8930150867,
-0.1143618524,
0.1361536384,
-0.0392667837,
-0.0860791802,
0.3093454838,
-0.3286457658,
-0.1282598972,
-0.1329839677,
-0.3400752246,
0.0043990016,
-0.1569236964,
0.3266392648,
0.2135234326,
-0.0005690902,
0.3174335063,
0.3206453919,
-0.1586578935,
-0.0454048775,
0.3780542016,
0.1787363887,
-0.1435992718,
-0.1427613199,
0.0819605961,
0.0628941655,
-0.173657015,
0.0926643312,
0.108484596,
-0.4816479981,
0.3293640316,
-0.4015114307,
-0.2510000467,
0.007993415,
0.239618361,
-0.0797397718,
-0.1534738392,
0.1838404536,
-0.0968569219,
0.1813537329,
-0.1740345657,
-0.1642562151,
0.1746467352,
0.06997329,
0.3781453073,
0.4506776631,
-0.2396116406,
-0.0631692857,
-0.1362428367,
-0.1378514916,
0.245946154,
0.0079498738,
-0.4741999507,
-0.2062947899,
0.0371297672,
0.1810426414,
-0.0923270285,
-0.1455953121,
0.1721224785,
-0.0562837794,
0.1244475394,
0.0392775685,
0.1677744985,
0.0337304212,
0.837597549,
-0.2117082179,
-0.3055509329,
-0.1786485761,
0.5786249638,
0.0612994432,
-0.3443078101,
0.1897520423,
-0.0601633415,
-0.202663362,
-0.1124158949,
0.1679847986,
0.1435305178,
-0.1207109839,
-0.0137460753,
-0.4844606817,
-0.135461539,
0.3008345366,
0.188630566,
0.3548611701,
-0.2195271999,
-0.0635789335,
0.1740077734,
-0.284173876,
0.2056916058,
0.1625823826,
0.2614246309,
-0.1952730864,
-0.2161902934,
-0.2001567185,
-0.0141646955,
-0.1311892271,
0.1216101348,
0.0355174392,
0.0125302039,
-0.0895134062,
-0.0025197465,
0.2382043004,
0.1458500922,
-0.1338467598,
0.11993213,
-0.2701045871,
-0.0916060954,
-0.0081749782,
0.1948790401,
-0.1186746582,
-0.4352927208,
-0.0065527931,
0.065830946,
0.044383239,
-0.2980177999,
-0.1076629683,
-0.3384865522,
-0.1209913194,
-0.2123243511,
0.4334197938,
-0.0772561431,
-0.1824726909,
0.3083029985,
-0.0912149251,
0.1343856305,
0.1567107737,
0.3019173145,
0.1051746458,
-0.2173038721,
0.2668793499,
0.2612126768,
0.2187346518,
-0.1041396558,
0.1611050069,
-0.1335746646,
0.0120309666,
0.3733326793,
0.1930890083,
0.7797373533,
-0.0087354966,
-0.4619468749,
0.2343358397,
0.0584148504,
0.2420034111,
-0.2574579716,
0.3182438314,
0.1881201863,
0.109991625,
0.1034224331,
0.3604651392,
0.1181779653,
0.2291564345,
-0.1195740104,
0.4428495765,
0.0472077802,
0.1710547358,
-0.4130826592,
0.1040694416,
-0.0341942012,
0.1235076636,
0.2168707848,
0.258546412,
0.169354707,
-0.2604326308,
0.6944690943,
0.4254677296,
0.2541796267,
-0.0289786458,
-0.202191174,
0.3091745079,
0.3924273849,
0.1401470155,
0.2696555257,
0.2136563361,
0.3649649918,
-0.2042587101,
-0.0972289443,
-0.17676875,
-0.0155597366,
0.1200893596,
-0.428458482,
0.3444975019,
-0.1480667591,
0.0189350843,
0.177025184,
-0.2406120896,
-0.1653128564,
0.1838983446,
-0.1320495009,
0.084463127,
-0.1065875143,
-0.1226844043,
0.056737911,
0.16785869,
0.0700049549,
0.165328294,
0.0613428578,
-0.0631595254,
0.2499225736,
0.3911221921,
0.0704721585,
-0.1490615606,
-0.579934895,
-0.0442051701,
-0.2540735006,
0.2481927872,
0.1321779788,
0.2059018314,
-0.3404233754,
-0.1203940287,
0.2132734209,
0.0382157043,
0.0154141784,
-0.5190164447,
-0.2173963785,
-0.0050256737,
-0.3723582029,
-0.3217619359,
0.0190564618,
0.1719228774,
0.1021029651,
0.1325764656,
-0.0789245665,
-0.3048039675,
0.141880855,
-0.2075068504,
-0.0879500806,
0.114605166,
0.0292846337,
-0.1355268955,
0.6446290016,
0.2345736921,
0.1028292328,
-0.1635217667,
-0.4154815972,
-0.351272285,
0.0232469626,
-0.1005348936,
-0.1174557358,
0.145858407,
0.2632142305,
-0.1220303774,
0.3209250867,
0.3944303095,
0.5208832622,
-0.1863095462,
0.1197667792,
-0.4308365881,
-0.0069042072,
-0.3667603731,
-0.2619172931,
-0.0267944671,
-0.0647681057,
0.0552322119,
-0.1110080779,
-0.1463377625,
-0.1283795834,
-0.0659105256,
0.0990434885,
0.0858682543,
0.5319931507,
0.2502122223,
0.5302526951,
0.3108709455,
-0.0715050697,
-0.070661135,
0.1376151145,
-0.068052955,
0.1617411226,
-0.0666430593,
0.047035858,
-0.1838990748,
0.3131072521,
0.1243217066,
0.6157351732,
0.1630043685,
-0.3072163165,
-0.1744778007,
-0.2664985061,
0.2768175602,
0.0297354087,
-0.0782980695,
0.3815820217,
0.1174020916,
0.4984681606,
-0.135516122,
0.1612101644,
0.139135614,
-0.4457459152,
0.0567013323,
-0.1604024768,
0.0219392329,
0.1593650877,
-0.0639596432,
-0.5078742504,
-0.3242853284,
0.392013371,
-0.0146356896,
0.1442725658,
0.2979818285,
0.003414087,
0.1368669569,
-0.1357824504,
0.0685592145,
-0.3417793512,
-0.0102835,
-0.042737063,
-0.0472321771
] |
https://github.com/huggingface/datasets/issues/1894 | benchmarking against MMapIndexedDataset | Also I would be interested to know what data types `MMapIndexedDataset` supports. Is there some documentation somewhere ? | I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam | 18 | benchmarking against MMapIndexedDataset
I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam
Also I would be interested to know what data types `MMapIndexedDataset` supports. Is there some documentation somewhere ? | [
-0.2623206377,
-0.0587808043,
-0.189609617,
0.2777881026,
-0.227725774,
-0.0737619847,
0.0111634843,
0.1357610226,
-0.2933738232,
-0.1568025053,
-0.1532005966,
0.3774221539,
0.0027406253,
-0.3708999753,
0.2234068662,
-0.1318498254,
0.3597463369,
0.102727592,
-0.2230794132,
0.0388148576,
0.0047571212,
0.0907567516,
-0.3878265917,
-0.0007551238,
-0.308994174,
0.160897553,
0.0237746835,
0.125249356,
-0.0787538886,
-0.3271166086,
0.2054045349,
0.2334436327,
0.0119259283,
0.6243492365,
-0.0001231747,
-0.2159654498,
-0.0060990117,
0.1145242378,
0.1292814016,
0.4733528495,
-0.3553677201,
-0.1873499155,
0.1760194451,
-0.0643594265,
0.1672020704,
-0.1405492723,
-0.1546870619,
-0.7546443343,
0.0663228631,
0.1668344736,
0.0619084537,
0.3860283792,
-0.4022425711,
0.0922336131,
0.0853737742,
-0.0240260884,
-0.2072672695,
0.1890346706,
0.5677086115,
-0.0037733875,
-0.0012581721,
-0.0574583896,
-0.1934127808,
0.0176084191,
0.1063681394,
-0.0935724974,
-0.1897952259,
-0.0806949139,
0.1602993309,
0.57144171,
0.3033568263,
-0.3853617013,
-0.1944887191,
-0.3469294906,
-0.2907652855,
0.1559841037,
0.1440190971,
-0.3041817546,
-0.0016853139,
0.0526976958,
-0.5018878579,
0.1885734797,
0.1384050697,
-0.1666801274,
0.0034873709,
0.4123750627,
0.2059562355,
0.0534099564,
0.4638535976,
-0.2791366279,
0.2155863494,
-0.2781586051,
-0.0903227925,
0.1010187715,
-0.5098446012,
-0.2299692929,
-0.0520546474,
-0.0863871947,
0.1356038004,
0.3474751711,
-0.2032876462,
0.5046435595,
0.4298179448,
0.0884595364,
0.1067868397,
0.2343622744,
0.1511717439,
0.0173080526,
0.0804941058,
-0.1772746742,
-0.3728793263,
0.096012935,
0.1723683923,
0.0604583472,
-0.2105669081,
-0.2381732166,
-0.3924919069,
-0.4892420471,
-0.2800406516,
0.0489859097,
0.3160378933,
-0.3449775279,
0.045167435,
0.204076916,
0.0596401095,
0.1809967458,
-0.4886461794,
0.0553050004,
-0.2804616392,
-0.2519147396,
0.0053387992,
0.0612702519,
-0.0932264775,
0.33040604,
0.1170455739,
-0.0538371988,
0.2595548928,
-0.0688251331,
0.2348470688,
0.0566518009,
0.5120324492,
-0.3035808206,
-0.0164029114,
0.0229026526,
-0.1128178388,
0.1444652081,
0.0295830257,
0.3873949349,
-0.3637988269,
0.0013560141,
-0.12069913,
-0.1722417921,
0.344529748,
-0.0133162672,
-0.3269227445,
-0.373624295,
-0.575011909,
0.3550664783,
-0.1666037291,
-0.2867625356,
0.2065923661,
-0.0011881441,
-0.1739469767,
-0.238293916,
0.0253137201,
-0.1241655275,
-0.235348627,
0.0866834819,
-0.3662694693,
-0.0079208687,
-0.1065403521,
0.357750684,
-0.3426888585,
0.211458087,
0.0139791816,
0.1618291438,
0.3885025978,
-0.509039402,
-0.4399955273,
-0.0519216619,
0.2221055627,
0.2691321075,
0.0002894998,
0.1311460435,
0.2970204055,
-0.275364697,
-0.0625801906,
0.0255433097,
0.1016711965,
0.2055302262,
-0.120686546,
-0.2674798965,
0.2851321399,
0.1665534228,
-0.2390080243,
-0.3818078041,
-0.1458150744,
-0.25947088,
-0.023030702,
-0.0676742867,
-0.2381089479,
0.0109961294,
0.0580348372,
-0.1880510896,
-0.0189225636,
-0.2777181864,
-0.0237311386,
0.290214777,
-0.2565104961,
-0.0430309996,
0.3490851223,
-0.0300248712,
0.0151802823,
-0.1060821265,
0.0208045729,
-0.2491452843,
-0.0057929531,
-0.1321572512,
0.2593688369,
0.1232626438,
-0.3601058722,
-0.2201322615,
0.2696444988,
-0.0733441561,
0.2368862927,
-0.1186156496,
0.3427736461,
-0.1157046929,
-0.0664256215,
0.2260989398,
-0.1223699301,
-0.0454597101,
-0.1864117384,
0.104365319,
0.3241364658,
0.2530104816,
0.5111230612,
0.4740772843,
-0.1185913906,
-0.1677383184,
0.1293277144,
0.1104747206,
0.1646761,
-0.2225419879,
-0.093173027,
0.6572034359,
0.3265314698,
0.3121196628,
0.0560513511,
-0.3682840168,
-0.1889241189,
0.2153059244,
0.1976002306,
0.1536180675,
0.1227065921,
0.4277951717,
0.3569440246,
0.0901559815,
0.2018397748,
0.3611365855,
0.4367325604,
0.0122095346,
-0.0282974169,
0.1319125593,
-0.408421278,
-0.2810288072,
0.3035345376,
-0.0571419261,
0.6194139719,
-0.0001338385,
0.3581352234,
0.0060080625,
0.116190806,
0.1531036794,
0.3110364676,
0.1907283068,
-0.3497515321,
-0.1504903287,
0.1073557511,
-0.0877884775,
0.1404765844,
-0.2416681498,
-0.1674520969,
0.1516308039,
-0.0311499164,
0.2706887126,
-0.2101575434,
0.2350761741,
0.1818675697,
-0.1810311377,
-0.2406477928,
-0.2521363497,
0.0457765572,
-0.0088929385,
0.1707539409,
0.0854836032,
-0.018895831,
0.3148319721,
-0.3668530583,
-0.3901953101,
0.1375720054,
-0.0929023474,
-0.4670562744,
0.0459490083,
0.4081195593,
0.1970103234,
0.3088480532,
0.0678155124,
0.2102431804,
-0.0006256923,
-0.1477909237,
0.0600791536,
-0.1121852398,
0.1130770892,
-0.1586785913,
-0.3379223347,
-0.3725428283,
-0.1332513094,
-0.0509336069,
-0.2197091579,
-0.0060840398,
-0.3206319809,
0.054592181,
-0.0176923685,
0.1768093109,
0.0380965658,
-0.1530308723,
0.0944232717,
0.1511808783,
-0.1855601221,
0.1327318251,
-0.627515316,
-0.0694226176,
0.2902388275,
-0.4144510627,
0.2188191712,
-0.1059565991,
-0.6133034825,
0.1753401756,
-0.0784829855,
-0.2060005367,
0.0012250021,
0.1281180084,
0.292330265,
0.1140853539,
-0.1864192784,
-0.0077262186,
-0.024829654,
0.2191716582,
0.173627004,
0.0818924457,
0.3506795168,
-0.127230987,
0.0651780665,
-0.0971100703,
-0.1086508483,
-0.0215859003,
-0.2666110992,
0.1335425526,
-0.0696088225,
0.2099892348,
-0.1552821398,
0.7983129621,
0.2943355739,
-0.1711912155,
-0.0251083449,
-0.1154659688,
-0.0341904946,
-0.2214474082,
0.0346540958,
-0.0982865244,
0.1160929129,
0.129555285,
0.5040053129,
0.0420986414,
-0.2055518627,
-0.237073034,
0.2103697956,
-0.1381927133,
-0.1255425662,
0.1617174745,
0.0916194767,
0.1419387609,
0.0423239768,
0.2302961051,
-0.4181652069,
-0.2926328182,
0.1091583371,
-0.0408108085,
-0.2304728329,
-0.2813102007,
-0.3877125978,
-0.297351867,
-0.4328522086,
0.3820944428,
0.1841482073,
0.490147382,
0.0667017698,
-0.0495631471,
0.1984016001,
-0.0336161777,
-0.1258167922,
-0.1876070499,
-0.26991117,
0.1920618415,
-0.1328730136,
-0.3253296614,
0.0336238965,
-0.0957037136,
0.1680743098,
0.1610638499,
0.0147236222,
-0.1274619848,
-0.1373986751,
-0.2789734304,
0.0331874527,
0.0262859687,
-0.1226212084,
-0.2699666619,
-0.01818645,
0.3601047993,
0.0197658502,
-0.3271104693,
0.049405586,
-0.0690032691,
0.1861638874,
-0.2409916371,
-0.0343741328,
0.3886443973,
0.545268774,
0.2817430496,
-0.4422085583,
0.3140861094,
0.1073640138,
-0.2230246961,
0.2349051386,
0.1926436275,
0.1615495682,
0.2590869069,
0.1178107411,
-0.0231583156,
0.1559977531,
0.1223506853,
0.0014794655,
0.1502125859,
-0.232181102,
-0.1356612593,
-0.1750068218,
0.0024256594,
0.2685618997,
0.2477086335,
-0.4353986382,
-0.3170286417,
0.6888304949,
0.4300855398,
-0.0938145965,
0.034357015,
0.0793815255,
-0.2249602675,
0.0344738849,
0.2207664996,
0.9849885106,
0.0195465088,
0.1792132854,
0.0678452551,
-0.0657154471,
0.2816608846,
-0.3302944899,
-0.1168103144,
-0.155659005,
-0.2901773751,
-0.0779816061,
-0.1615458429,
0.2968317568,
0.1133744419,
-0.0163854882,
0.2415138781,
0.170519501,
-0.1562666893,
-0.0078400504,
0.3491710126,
0.2862443924,
-0.0430461355,
-0.2080046833,
0.0738809332,
0.1003975123,
-0.1788125932,
0.0935539901,
0.0736099929,
-0.4495073557,
0.2725327909,
-0.4372515678,
-0.3201458454,
0.109545067,
0.2533119917,
0.0155766513,
-0.1597277671,
0.1760913581,
-0.2382446527,
0.2436284572,
-0.1943280995,
-0.2094152272,
0.1973571628,
0.0914248228,
0.385846585,
0.3977943957,
-0.2459922433,
-0.1021099538,
-0.1435466707,
-0.1216748208,
0.2774525881,
-0.0152428858,
-0.5284810662,
-0.2779018581,
0.1535318494,
0.2374777943,
-0.0695358217,
-0.2080256939,
0.1414766759,
-0.0295076147,
0.0897427946,
0.0274558328,
0.1424981654,
-0.0217180457,
0.8859041333,
-0.1997340918,
-0.2523297071,
-0.1323384941,
0.5498284101,
0.0927195251,
-0.3913940787,
0.151319921,
-0.1264463663,
-0.2121487856,
-0.0967690647,
0.1890246272,
0.1080343425,
-0.0423378274,
-0.0924681723,
-0.4025081694,
-0.1561018527,
0.3226422668,
0.2322565317,
0.2806734443,
-0.1581172496,
-0.1462304741,
0.1624289304,
-0.2486476153,
0.1582393944,
0.0488611162,
0.210731566,
-0.2525597215,
-0.1154351979,
-0.1653578877,
-0.0204236619,
-0.1105907932,
0.1441792846,
0.108857438,
0.0445204712,
-0.1210548133,
0.0115024261,
0.2281007916,
0.1904029548,
-0.1649169624,
0.2522321641,
-0.30120489,
-0.0569662228,
-0.0036378503,
0.2221194953,
-0.0319837928,
-0.4015704691,
-0.0436999016,
0.13581267,
-0.0425886028,
-0.1925135255,
-0.1190372854,
-0.3740130365,
-0.0797030032,
-0.2223742008,
0.4970264435,
-0.0573127568,
-0.0983825475,
0.2753205597,
-0.0541338697,
0.1241819113,
0.2635471523,
0.2435033321,
0.0375811309,
-0.2240671217,
0.311183542,
0.3066112995,
0.2467313111,
-0.1525157392,
0.16617544,
-0.1351744831,
0.0642868876,
0.3267385066,
0.1774222553,
0.8196069598,
-0.0564900115,
-0.4774509966,
0.2959795296,
0.0170353521,
0.1751724184,
-0.3129903674,
0.3569062352,
0.2060687542,
0.0312647372,
0.0897602662,
0.4130364656,
0.1084636599,
0.2423647642,
-0.0868823081,
0.5108747482,
0.1430851221,
0.1881142557,
-0.4495840669,
-0.0083315372,
-0.0644417405,
0.1368060261,
0.2627198696,
0.2217084765,
0.2480023801,
-0.2684150934,
0.6961632967,
0.4385245442,
0.1038485765,
-0.1020707935,
-0.1935293674,
0.2840628624,
0.4061127007,
0.173038438,
0.2769916654,
0.1943645924,
0.3954160511,
-0.2286614329,
-0.0919966102,
-0.1883070767,
0.0622649752,
0.1141958088,
-0.4302680194,
0.4157556891,
-0.1572194993,
0.0111825764,
0.0721840262,
-0.2298716009,
-0.1569694877,
0.2616045773,
-0.0762799084,
0.1084907874,
-0.093172282,
-0.1474352777,
0.0843064934,
0.1815418452,
0.1187274605,
0.1750240475,
0.1565624475,
-0.0186476111,
0.3655470014,
0.2379736453,
0.0692331046,
-0.0609861724,
-0.6656449437,
-0.0637141019,
-0.332442224,
0.2670285702,
0.1411356777,
0.2664565742,
-0.2655003667,
-0.0878867805,
0.1292440891,
0.0156822242,
0.090010345,
-0.4702933133,
-0.1247411817,
0.0434832722,
-0.3978412747,
-0.2820217609,
-0.0617982149,
0.2348823249,
0.042859219,
0.1404764056,
-0.0633877814,
-0.3730197251,
0.0760592073,
-0.1049857363,
-0.071655266,
0.1470343173,
-0.0024820715,
-0.0942308903,
0.6994122863,
0.0653570667,
0.0618920922,
-0.0799617991,
-0.53215307,
-0.4118785262,
0.066762574,
-0.0329950936,
-0.1383030266,
0.1941888928,
0.257925868,
-0.1062396765,
0.2881214321,
0.4280997217,
0.5550519228,
-0.2165678144,
0.0425790697,
-0.4298309386,
0.0392040871,
-0.3841673434,
-0.2963056266,
-0.0905962512,
-0.0491231717,
0.1268495172,
-0.129475981,
-0.1205533147,
-0.1357968897,
-0.1788153648,
0.0664918721,
0.1710091233,
0.5717979074,
0.319772929,
0.4699349701,
0.3191301823,
-0.1181834936,
-0.159260571,
0.119049713,
-0.0503677353,
0.1566698849,
-0.0535301231,
0.0268943459,
-0.1764747947,
0.3074465394,
0.0814194828,
0.6996272206,
0.1489480138,
-0.3488850594,
-0.18896918,
-0.2172176242,
0.2014443427,
0.0730849653,
-0.1378267407,
0.3830972612,
0.141288355,
0.5235021114,
-0.1728390753,
0.1528406143,
0.3073650897,
-0.486073494,
-0.0532759205,
-0.1688017547,
0.0878733993,
0.1353638172,
-0.0636117831,
-0.5485002995,
-0.3553925455,
0.3390141129,
0.0616857074,
0.1476612836,
0.2592873275,
-0.0902004689,
0.097938776,
-0.1953220665,
0.0274968464,
-0.3222488165,
0.0253425017,
-0.0194340125,
-0.062136583
] |
https://github.com/huggingface/datasets/issues/1894 | benchmarking against MMapIndexedDataset | no docs haha, it's written to support integer numpy arrays.
You can build one in fairseq with, roughly:
```bash
wget https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-raw-v1.zip
unzip wikitext-103-raw-v1.zip
export dd=$HOME/fairseq-py/wikitext-103-raw
export mm_dir=$HOME/mmap_wikitext2
mkdir -p gpt2_bpe
wget -O gpt2_bpe/encoder.json https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/encoder.json
wget -O gpt2_bpe/vocab.bpe https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/vocab.bpe
wget -O gpt2_bpe/dict.txt https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/dict.txt
for SPLIT in train valid; do \
python -m examples.roberta.multiprocessing_bpe_encoder \
--encoder-json gpt2_bpe/encoder.json \
--vocab-bpe gpt2_bpe/vocab.bpe \
--inputs /scratch/stories_small/${SPLIT}.txt \
--outputs /scratch/stories_small/${SPLIT}.bpe \
--keep-empty \
--workers 60; \
done
mkdir -p $mm_dir
fairseq-preprocess \
--only-source \
--srcdict gpt2_bpe/dict.txt \
--trainpref $dd/wiki.train.bpe \
--validpref $dd/wiki.valid.bpe \
--destdir $mm_dir \
--workers 60 \
--dataset-impl mmap
```
I'm noticing in my benchmarking that it's much smaller on disk than arrow (200mb vs 900mb), and that both incur significant cost by increasing the number of data loader workers.
This somewhat old [post](https://ray-project.github.io/2017/10/15/fast-python-serialization-with-ray-and-arrow.html) suggests there are some gains to be had from using `pyarrow.serialize(array).tobuffer()`. I haven't yet figured out how much of this stuff `pa.Table` does under the hood.
The `MMapIndexedDataset` bottlenecks we are working on improving (by using arrow) are:
1) `MMapIndexedDataset`'s index, which stores offsets, basically gets read in its entirety by each dataloading process.
2) we have separate, identical, `MMapIndexedDatasets` on each dataloading worker, so there's redundancy there; we wonder if there is a way that arrow can somehow dedupe these in shared memory.
It will take me a few hours to get `MMapIndexedDataset` benchmarks out of `fairseq`/onto a branch in this repo, but I'm happy to invest the time if you're interested in collaborating on some performance hacking. | I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam | 249 | benchmarking against MMapIndexedDataset
I am trying to benchmark my datasets based implementation against fairseq's [`MMapIndexedDataset`](https://github.com/pytorch/fairseq/blob/master/fairseq/data/indexed_dataset.py#L365) and finding that, according to psrecord, my `datasets` implem uses about 3% more CPU memory and runs 1% slower for `wikitext103` (~1GB of tokens).
Questions:
1) Is this (basically identical) performance expected?
2) Is there a scenario where this library will outperform `MMapIndexedDataset`? (maybe more examples/larger examples?)
3) Should I be using different benchmarking tools than `psrecord`/how do you guys do benchmarks?
Thanks in advance! Sam
no docs haha, it's written to support integer numpy arrays.
You can build one in fairseq with, roughly:
```bash
wget https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-raw-v1.zip
unzip wikitext-103-raw-v1.zip
export dd=$HOME/fairseq-py/wikitext-103-raw
export mm_dir=$HOME/mmap_wikitext2
mkdir -p gpt2_bpe
wget -O gpt2_bpe/encoder.json https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/encoder.json
wget -O gpt2_bpe/vocab.bpe https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/vocab.bpe
wget -O gpt2_bpe/dict.txt https://dl.fbaipublicfiles.com/fairseq/gpt2_bpe/dict.txt
for SPLIT in train valid; do \
python -m examples.roberta.multiprocessing_bpe_encoder \
--encoder-json gpt2_bpe/encoder.json \
--vocab-bpe gpt2_bpe/vocab.bpe \
--inputs /scratch/stories_small/${SPLIT}.txt \
--outputs /scratch/stories_small/${SPLIT}.bpe \
--keep-empty \
--workers 60; \
done
mkdir -p $mm_dir
fairseq-preprocess \
--only-source \
--srcdict gpt2_bpe/dict.txt \
--trainpref $dd/wiki.train.bpe \
--validpref $dd/wiki.valid.bpe \
--destdir $mm_dir \
--workers 60 \
--dataset-impl mmap
```
I'm noticing in my benchmarking that it's much smaller on disk than arrow (200mb vs 900mb), and that both incur significant cost by increasing the number of data loader workers.
This somewhat old [post](https://ray-project.github.io/2017/10/15/fast-python-serialization-with-ray-and-arrow.html) suggests there are some gains to be had from using `pyarrow.serialize(array).tobuffer()`. I haven't yet figured out how much of this stuff `pa.Table` does under the hood.
The `MMapIndexedDataset` bottlenecks we are working on improving (by using arrow) are:
1) `MMapIndexedDataset`'s index, which stores offsets, basically gets read in its entirety by each dataloading process.
2) we have separate, identical, `MMapIndexedDatasets` on each dataloading worker, so there's redundancy there; we wonder if there is a way that arrow can somehow dedupe these in shared memory.
It will take me a few hours to get `MMapIndexedDataset` benchmarks out of `fairseq`/onto a branch in this repo, but I'm happy to invest the time if you're interested in collaborating on some performance hacking. | [
-0.2432023138,
0.0233523995,
-0.1691764891,
0.1820508838,
-0.1022190005,
-0.0226156935,
0.1294505298,
0.2604652345,
-0.2088270187,
-0.0699171424,
-0.1556990147,
0.503326416,
-0.0702500939,
-0.4188562036,
0.2817453444,
-0.0300165247,
0.3354423642,
0.1515544355,
-0.1571373641,
0.0486280471,
-0.1405940056,
0.1506154239,
-0.3253052533,
0.0204206184,
-0.3509260416,
0.1225453317,
-0.0095219538,
0.2243739963,
-0.0806349441,
-0.3892319798,
0.2625466883,
0.13121292,
-0.0679285675,
0.5093363523,
-0.0001197014,
-0.1623321176,
-0.0542120188,
0.107866697,
0.0268324949,
0.5400540233,
-0.1131935641,
-0.3049708009,
0.0386115462,
-0.1792384386,
0.0806343704,
-0.1571836472,
-0.1240352169,
-0.6968594193,
0.0376431197,
0.1124981791,
0.0433409251,
0.3202776909,
-0.2873578668,
0.1792376786,
0.0780042708,
-0.087825641,
-0.2255394161,
0.203054592,
0.60388273,
-0.066937387,
-0.0622576103,
0.0722513199,
-0.1325342953,
0.07743112,
0.0730091035,
0.0020604804,
-0.115637064,
-0.08595635,
0.0608330034,
0.4867010415,
0.1345414817,
-0.354642719,
-0.2642899156,
-0.4171715081,
-0.3238387406,
0.0939711779,
0.2001512647,
-0.2242228538,
-0.0451186486,
0.1037717015,
-0.5512922406,
0.2541920841,
0.1437451243,
-0.0920720249,
-0.0109138712,
0.4918264449,
0.3039821982,
0.0423973724,
0.4399321675,
-0.2640121579,
0.1324423254,
-0.2057049423,
0.0011937856,
0.1028620824,
-0.4656967819,
-0.280341804,
-0.0441009998,
-0.2504604459,
0.1140583456,
0.2019063681,
-0.142162174,
0.5152267218,
0.4101499319,
0.0859820992,
0.0756977946,
0.2415623665,
0.0921154097,
0.0314105451,
0.0018225927,
-0.0955255181,
-0.2015071064,
0.1789546013,
0.1996316016,
0.0112765543,
-0.2609249949,
-0.1755255014,
-0.380125463,
-0.3332431018,
-0.3777091503,
0.0147045553,
0.3565778136,
-0.3525733054,
0.0801032186,
0.2087018937,
0.0533463843,
0.2673597336,
-0.4570311308,
0.039497003,
-0.3195517063,
-0.204874903,
-0.0434500612,
0.1353331506,
-0.0892048851,
0.2397127002,
0.1953058839,
0.1026562899,
0.2620455027,
-0.0591995418,
0.1804068983,
0.1563951224,
0.5779826641,
-0.2525074184,
-0.0220909454,
-0.0524836183,
-0.1771283448,
0.1634492576,
0.0495936684,
0.4309073389,
-0.315975368,
0.0107470937,
-0.1292082816,
-0.122249864,
0.3741809428,
-0.0066922279,
-0.2167482078,
-0.3019157052,
-0.3807262182,
0.3237061203,
-0.0806939304,
-0.2796666622,
0.2008900344,
-0.038002342,
-0.2599067688,
-0.2586503923,
0.1129403263,
-0.1048044115,
-0.1829636693,
0.0902454555,
-0.2681652308,
0.1017137915,
-0.057437662,
0.422822088,
-0.3233219981,
0.2451718003,
-0.0476450548,
0.2953232527,
0.3699170053,
-0.4035224915,
-0.4890655875,
-0.0134466812,
0.1843803674,
0.2378026545,
-0.0289863199,
0.2881444395,
0.2071385235,
-0.3476952016,
0.0604117215,
0.0866552219,
0.129802689,
0.3404926658,
-0.1914952099,
-0.4372930825,
0.4101483226,
0.1844230443,
-0.2632356286,
-0.5926272869,
-0.2096514106,
-0.2201001942,
0.0141151296,
-0.1993329376,
-0.2652761042,
-0.0197390094,
0.0547748953,
-0.3296551406,
-0.0481881946,
-0.21196419,
-0.0164221711,
0.4075004458,
-0.3809971809,
-0.0671112388,
0.4104774296,
-0.0325262696,
0.0160775408,
-0.1260008961,
-0.0352372341,
-0.3404730558,
0.0188441277,
-0.1246284693,
0.283957094,
0.1574339718,
-0.3725444674,
-0.2051101774,
0.2095500231,
-0.0340518691,
0.0869022757,
-0.1760457605,
0.2704809308,
-0.1397456229,
-0.1292766929,
0.2784686089,
-0.1737470925,
-0.0782441646,
-0.1303407699,
0.1673097759,
0.2626062334,
0.0895877481,
0.4840814173,
0.3674254715,
-0.0200815052,
-0.1213859171,
0.1851387769,
0.1798510253,
0.1355849653,
-0.233292073,
-0.017305512,
0.6673049927,
0.1584088951,
0.3197747469,
0.0334663689,
-0.3281885982,
-0.2300445139,
0.2341326475,
0.2983641624,
0.1582585871,
0.1329231262,
0.3186573684,
0.3469450474,
0.0745348707,
0.2600977421,
0.4380367398,
0.5110446811,
0.0919653028,
0.0119457003,
0.0950751603,
-0.3309468925,
-0.3392914832,
0.2181060314,
-0.1530642062,
0.6769608855,
0.005627133,
0.3541243672,
-0.027240511,
0.0668690652,
0.1295981407,
0.3824510574,
0.2104080319,
-0.402266711,
-0.1451213509,
0.0917437226,
-0.1520503163,
0.1312490404,
-0.1228130907,
-0.1388619095,
0.124456346,
-0.0374368355,
0.2275907248,
-0.0976808965,
0.3413504362,
0.1647022069,
-0.3163815737,
-0.1967754513,
-0.2952256203,
0.1427743137,
-0.041044835,
0.1379373074,
0.1017515287,
0.0208517611,
0.2121635079,
-0.2811616957,
-0.2880564332,
0.0261981785,
-0.0989260674,
-0.4190021157,
0.0419862568,
0.4446632266,
0.1643384397,
0.3288486004,
0.0893413126,
0.0642519966,
0.0240688287,
-0.1532047987,
0.0006648693,
-0.156073764,
0.1611468643,
-0.0476364382,
-0.3093942404,
-0.475726068,
-0.2025448382,
0.0364157148,
-0.2672003508,
-0.0642173216,
-0.22953704,
0.2396284342,
-0.0640904978,
0.1067320704,
-0.1079207659,
-0.1804133803,
0.064130418,
0.0804077834,
-0.140421629,
0.1453792453,
-0.5900021791,
0.0552156828,
0.3555608988,
-0.3177752197,
0.130136624,
-0.1445241272,
-0.4670262337,
0.2165988982,
-0.0551490113,
-0.1931109726,
-0.0490445346,
0.0232452154,
0.2404925823,
0.1181662381,
-0.197745055,
-0.0716170073,
-0.0266951676,
0.1901097149,
0.1161490083,
0.1605776101,
0.3275681734,
-0.132088244,
0.0220069736,
-0.1042535976,
0.0397160426,
0.0453551412,
-0.2836326957,
0.245972991,
-0.0552966818,
0.1432183087,
-0.1069722325,
0.8057384491,
0.2422969043,
-0.243199259,
-0.0703330562,
-0.0903488845,
-0.0702883303,
-0.1221859008,
-0.0307257622,
0.0003348477,
0.1754676104,
0.1227691248,
0.5753462315,
0.0227226429,
-0.2618603706,
-0.2323167771,
0.2291737199,
-0.0984153971,
-0.1660816818,
0.2451481968,
0.2188760787,
0.0562770441,
0.0476950631,
0.3273082972,
-0.4191040993,
-0.3323993981,
0.0446894988,
-0.1760572642,
-0.2805052102,
-0.2391747087,
-0.2002645731,
-0.3777950406,
-0.4251482785,
0.3432878554,
0.263582468,
0.5489297509,
0.1243813932,
-0.0968295634,
0.2191800922,
-0.1522559971,
0.0655358136,
-0.2366159111,
-0.2895757854,
0.1795132905,
-0.0333267413,
-0.3219074607,
0.0134369209,
-0.1971856654,
-0.0295847002,
0.3278867304,
-0.0153876953,
-0.1670004427,
-0.0544807874,
-0.2099220157,
0.0499692373,
0.0047977865,
0.0105058476,
-0.3588392437,
-0.1914760768,
0.3003989756,
0.1187165827,
-0.3400737345,
0.0776578337,
0.0090767257,
0.1560644209,
-0.2711518109,
0.0005420819,
0.3867948055,
0.5396572351,
0.2772964835,
-0.3942841887,
0.2889389396,
0.0025330568,
-0.1356239319,
0.2091218531,
0.1200827658,
0.1594481617,
0.3647194207,
0.0862558633,
-0.160955295,
0.1497532427,
0.1717621088,
0.0109473765,
0.0738904849,
-0.2870321572,
-0.0327258669,
-0.0553465039,
0.0564190932,
0.2335375547,
0.2024369538,
-0.3133953512,
-0.3247179389,
0.5959884524,
0.3803277016,
-0.0817591697,
0.0591214411,
-0.1594302058,
-0.212449342,
0.0031833276,
0.3320755363,
0.9616859555,
-0.0361312963,
0.2336976975,
0.2174377739,
-0.0309350342,
0.3985089958,
-0.3488453627,
-0.028482113,
-0.2169911563,
-0.2417106628,
-0.0772070736,
-0.1510021389,
0.2551008761,
0.1418094933,
-0.1065924615,
0.2904794812,
0.311155498,
0.0454585478,
-0.0738773271,
0.2952064872,
0.1969210058,
-0.0388145708,
-0.1757236421,
0.0622951388,
-0.042118188,
-0.2161155343,
0.1218046397,
-0.0506025851,
-0.5397320986,
0.1987794936,
-0.3056119978,
-0.2535443902,
-0.0059524253,
0.4464245737,
-0.0064621256,
-0.1360342056,
0.1693733037,
-0.1459731162,
0.3673000932,
-0.2042054683,
-0.1845325977,
0.0985409915,
0.1169229746,
0.3119470477,
0.4654178321,
-0.2429742515,
0.0305179749,
-0.1376770884,
-0.1448926628,
0.2899270356,
0.0243126526,
-0.4574977756,
-0.3292914331,
0.173350215,
0.3085518181,
-0.1377104521,
-0.1473699212,
0.1433728784,
-0.1439525932,
0.0311074257,
0.0259094052,
0.0657199174,
-0.1187403053,
0.8549550176,
-0.2891102731,
-0.2632471323,
-0.0914643183,
0.4993325174,
0.0986837596,
-0.4388749003,
0.2042729855,
-0.0517668873,
-0.2537936866,
-0.1439125836,
0.136764586,
0.0153956451,
-0.1277047694,
-0.0665010363,
-0.1762317568,
-0.1240037158,
0.3249715865,
0.1699851006,
0.2428447902,
-0.0837262869,
-0.1774937809,
0.2468435615,
-0.328785181,
0.0891060606,
0.2111871839,
0.3494646549,
-0.2351031899,
-0.1996964365,
-0.2085603923,
-0.1003243402,
-0.1379262507,
0.1160774752,
0.0410958752,
0.0242036022,
-0.2370049357,
-0.0795886815,
0.1163437963,
0.2017843723,
-0.1304215342,
0.2277388275,
-0.3855398297,
-0.0815223679,
0.0687747747,
0.1902276874,
-0.1212152094,
-0.4181317985,
-0.0599127151,
0.162943244,
-0.0284064617,
-0.2897573411,
-0.0809317827,
-0.4819075763,
-0.0330374055,
-0.1748723388,
0.5092719197,
-0.0284549184,
-0.0630908757,
0.3277778924,
-0.2139815986,
0.140726611,
0.193738386,
0.2686723471,
0.0791733265,
-0.2487016767,
0.335722208,
0.2605231106,
0.1490175426,
-0.0361261554,
0.151464656,
-0.1979042441,
-0.027911514,
0.2792246044,
0.1092879027,
0.8128836751,
-0.1213118508,
-0.595482111,
0.2628180385,
0.0550587848,
0.0709031075,
-0.2267871201,
0.4238923788,
0.1857878566,
0.0937290043,
0.2167641819,
0.424346149,
-0.0296486914,
0.0790285468,
-0.0533585064,
0.5252414942,
0.0559775457,
0.0300829671,
-0.164886564,
0.0284830481,
-0.1760410964,
-0.0447766781,
0.1498754621,
0.2445605397,
0.2723896801,
-0.3558056951,
0.6884695888,
0.3570596278,
0.1591427028,
-0.2962271869,
-0.0079386234,
0.2951878607,
0.5422818661,
0.1094152331,
0.2572138906,
0.2081541568,
0.3968317807,
-0.2889019251,
-0.0473376103,
-0.1906136572,
0.1498118192,
0.140030399,
-0.314050585,
0.2555695772,
-0.1108250618,
-0.0309339464,
0.0641813576,
-0.2243406922,
-0.1642582119,
0.2001769394,
0.0384884849,
0.0602519363,
-0.0766345859,
-0.0091734398,
0.0860625058,
0.2510479093,
0.0271172076,
0.1162602752,
0.3386169076,
0.0252005272,
0.3081497252,
0.274859488,
0.143607825,
-0.0221875235,
-0.6469066143,
-0.0227143541,
-0.2772479653,
0.2943778336,
0.0195345059,
0.1883044541,
-0.3704983592,
-0.0827373937,
0.2145933658,
0.0031120852,
0.0610564612,
-0.5787849426,
-0.0727087185,
-0.0034160558,
-0.2699395716,
-0.1305493116,
-0.0787746832,
0.1848436296,
0.0313293785,
0.1325550675,
-0.193487376,
-0.3762682378,
-0.0294353887,
-0.0934247524,
0.0093004517,
0.2366237342,
0.0202468373,
-0.1909414828,
0.7200249434,
0.0936730877,
-0.0714849085,
-0.13489829,
-0.4660582542,
-0.4094489515,
-0.0068381317,
-0.1019364223,
-0.248914212,
0.1159808412,
0.2929781675,
-0.1527726948,
0.3135236204,
0.3988291323,
0.6963893771,
-0.2271550298,
0.1801045686,
-0.5364995599,
0.1016409695,
-0.3078740537,
-0.2435194701,
-0.0656284168,
-0.1456209272,
0.1215389073,
-0.1681290567,
-0.1349670291,
-0.0808491707,
-0.0165274367,
0.0096098334,
0.1161242872,
0.4607245922,
0.3132439852,
0.3438240886,
0.2764531672,
-0.1613560915,
-0.1428871453,
0.1805913895,
-0.0459170714,
0.1733420938,
0.0033164062,
0.1465429813,
-0.1652984321,
0.1458809078,
0.0261310004,
0.7369500995,
0.1684113592,
-0.2778195143,
-0.1064316183,
-0.208754003,
0.1732776314,
0.1180634946,
-0.1529619396,
0.356854409,
0.1719871461,
0.5304066539,
-0.1909966469,
-0.061809849,
0.3194057047,
-0.4830408394,
-0.1014675722,
-0.2315473706,
0.0702892542,
0.0974431634,
0.0605835766,
-0.5089954734,
-0.3161847889,
0.3136835098,
0.1816223711,
0.0581691712,
0.2628868222,
-0.1175673306,
0.0733458847,
-0.2202982903,
-0.0775409862,
-0.3205028176,
0.0767266676,
-0.0646986067,
-0.0776448771
] |
https://github.com/huggingface/datasets/issues/1893 | wmt19 is broken | This was also mentioned in https://github.com/huggingface/datasets/issues/488
The bucket where is data was stored seems to be unavailable now. Maybe we can change the URL to the ones in https://conferences.unite.un.org/uncorpus/en/downloadoverview ? | 1. Check which lang pairs we have: `--dataset_name wmt19`:
Please pick one among the available configs: ['cs-en', 'de-en', 'fi-en', 'gu-en', 'kk-en', 'lt-en', 'ru-en', 'zh-en', 'fr-de']
2. OK, let's pick `ru-en`:
`--dataset_name wmt19 --dataset_config "ru-en"`
no cookies:
```
Traceback (most recent call last):
File "./run_seq2seq.py", line 661, in <module>
main()
File "./run_seq2seq.py", line 317, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 572, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 628, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt19/436092de5f3faaf0fc28bc84875475b384e90a5470fa6afaee11039ceddc5052/wmt_utils.py", line 755, in _split_generators
downloaded_files = dl_manager.download_and_extract(urls_to_download)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 276, in download_and_extract
return self.extract(self.download(url_or_urls))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 191, in download
downloaded_path_or_paths = map_nested(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 233, in map_nested
mapped = [
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 234, in <listcomp>
_single_map_nested((function, obj, types, None, True)) for obj in tqdm(iterable, disable=disable_tqdm)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 190, in _single_map_nested
mapped = [_single_map_nested((function, v, types, None, True)) for v in pbar]
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 190, in <listcomp>
mapped = [_single_map_nested((function, v, types, None, True)) for v in pbar]
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 172, in _single_map_nested
return function(data_struct)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 211, in _download
return cached_path(url_or_filename, download_config=download_config)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/file_utils.py", line 274, in cached_path
output_path = get_from_cache(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/file_utils.py", line 584, in get_from_cache
raise FileNotFoundError("Couldn't find file at {}".format(url))
FileNotFoundError: Couldn't find file at https://storage.googleapis.com/tfdataset-data/downloadataset/uncorpus/UNv1.0.en-ru.tar.gz
``` | 30 | wmt19 is broken
1. Check which lang pairs we have: `--dataset_name wmt19`:
Please pick one among the available configs: ['cs-en', 'de-en', 'fi-en', 'gu-en', 'kk-en', 'lt-en', 'ru-en', 'zh-en', 'fr-de']
2. OK, let's pick `ru-en`:
`--dataset_name wmt19 --dataset_config "ru-en"`
no cookies:
```
Traceback (most recent call last):
File "./run_seq2seq.py", line 661, in <module>
main()
File "./run_seq2seq.py", line 317, in main
datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/load.py", line 740, in load_dataset
builder_instance.download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 572, in download_and_prepare
self._download_and_prepare(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/builder.py", line 628, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
File "/home/stas/.cache/huggingface/modules/datasets_modules/datasets/wmt19/436092de5f3faaf0fc28bc84875475b384e90a5470fa6afaee11039ceddc5052/wmt_utils.py", line 755, in _split_generators
downloaded_files = dl_manager.download_and_extract(urls_to_download)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 276, in download_and_extract
return self.extract(self.download(url_or_urls))
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 191, in download
downloaded_path_or_paths = map_nested(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 233, in map_nested
mapped = [
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 234, in <listcomp>
_single_map_nested((function, obj, types, None, True)) for obj in tqdm(iterable, disable=disable_tqdm)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 190, in _single_map_nested
mapped = [_single_map_nested((function, v, types, None, True)) for v in pbar]
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 190, in <listcomp>
mapped = [_single_map_nested((function, v, types, None, True)) for v in pbar]
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/py_utils.py", line 172, in _single_map_nested
return function(data_struct)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/download_manager.py", line 211, in _download
return cached_path(url_or_filename, download_config=download_config)
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/file_utils.py", line 274, in cached_path
output_path = get_from_cache(
File "/mnt/nvme1/code/huggingface/datasets-master/src/datasets/utils/file_utils.py", line 584, in get_from_cache
raise FileNotFoundError("Couldn't find file at {}".format(url))
FileNotFoundError: Couldn't find file at https://storage.googleapis.com/tfdataset-data/downloadataset/uncorpus/UNv1.0.en-ru.tar.gz
```
This was also mentioned in https://github.com/huggingface/datasets/issues/488
The bucket where is data was stored seems to be unavailable now. Maybe we can change the URL to the ones in https://conferences.unite.un.org/uncorpus/en/downloadoverview ? | [
-0.3531682491,
-0.4596048594,
-0.0412609614,
0.3225631714,
0.1783248484,
-0.0362050459,
0.1937318742,
0.2335269153,
0.0660973191,
0.0606788397,
0.0147964992,
0.1090325415,
-0.1255614161,
0.5707480311,
0.1197271198,
-0.0096024955,
0.0067275241,
0.0419429056,
-0.4998803735,
0.0689935833,
-0.1902456582,
0.3333272338,
-0.2408049703,
0.0314800441,
-0.1947325468,
0.0247909892,
-0.1083270758,
0.1399193555,
-0.25319314,
-0.218459785,
0.1407963037,
0.0359902382,
-0.0891289115,
0.4259699881,
-0.0001029476,
-0.0148939863,
0.2780919373,
0.0244726166,
-0.2462791204,
-0.2544899583,
-0.3194820583,
-0.2060378641,
0.0056868084,
-0.0019714236,
-0.0526042059,
-0.2070801705,
-0.1086709797,
-0.1944905967,
0.5796177983,
0.2073671967,
0.349845469,
0.3110418022,
0.0414669067,
-0.3367853463,
0.1314580292,
0.1913622618,
-0.0752149224,
-0.052634351,
0.152730763,
0.0621747375,
0.0765528902,
0.3450376391,
0.0654331148,
-0.1494605094,
0.2028808594,
-0.1226102263,
0.0630016476,
-0.1020161659,
0.2366114855,
0.2687705755,
0.3145794272,
-0.1319088042,
-0.1572936177,
0.1063632816,
-0.2688269913,
-0.6988609433,
0.1989789456,
0.1747244298,
-0.0829618424,
0.1859146953,
-0.0768822059,
-0.1307183206,
-0.0139768347,
-0.1147556454,
-0.2573108673,
0.6457992792,
-0.1298798323,
0.0660454631,
0.0168190747,
-0.2135038972,
-0.1319765002,
0.3256303966,
-0.0784665942,
-0.2144068927,
-0.1820539385,
0.1821351647,
0.0829403996,
0.1146461889,
0.1386961192,
-0.0364184901,
-0.1732153594,
0.274096489,
0.1333289146,
0.0846898854,
-0.0378116369,
0.372231245,
0.1994920969,
-0.1968186349,
0.2560355067,
0.2929571569,
-0.002000235,
0.008912662,
0.0502563082,
-0.5564662814,
-0.095123902,
0.0572430827,
0.0931368172,
-0.0478079394,
-0.1790330708,
0.0198316984,
0.0519159734,
-0.1261248738,
0.0150335208,
0.4379253387,
0.0404033661,
-0.0652522147,
0.0184148252,
0.1667164266,
-0.1312855482,
-0.2226328999,
-0.2503520846,
-0.1263065636,
-0.2839660048,
0.1710804552,
0.161487937,
-0.1287925243,
0.309828639,
0.0590672754,
-0.2042984515,
-0.1414274424,
-0.1652148962,
0.0173628666,
0.1748153567,
0.2738702893,
-0.0161503926,
0.2083769441,
0.2654592693,
-0.2715435028,
-0.0453214124,
-0.08371865,
-0.1263061017,
-0.1878840625,
-0.0786033645,
0.3323677778,
0.0418703556,
0.1754484475,
-0.1218368113,
0.2003699541,
-0.008090619,
-0.064775452,
-0.0314941406,
-0.1032682583,
-0.4300822616,
-0.1006192341,
0.425793618,
0.4848940372,
-0.5752803683,
-0.2235398144,
-0.0691836476,
-0.1647975892,
0.0550693981,
0.2180525661,
-0.0357795432,
-0.0072340965,
-0.3086131215,
0.1856992245,
-0.2164033651,
-0.1183207706,
-0.0606828257,
-0.0117847249,
-0.1158542708,
-0.0222994685,
0.0701593459,
-0.2123700231,
-0.0756182149,
0.0206812248,
0.0376715064,
0.0554950163,
0.0734890476,
-0.1147502661,
-0.2230087817,
-0.4322019815,
0.5312094688,
-0.0340200216,
0.1467169672,
-0.3156555593,
0.1697200984,
-0.1503373384,
0.3136666715,
-0.0254013017,
0.0116740726,
0.0524742492,
0.2096418142,
0.0273614004,
0.0132031068,
-0.0901464447,
-0.5520674586,
0.2051402777,
-0.0901377648,
0.39275828,
-0.2127137035,
0.0717058033,
-0.2444404215,
0.0683689713,
-0.3136656284,
-0.3684315681,
0.2931484282,
0.1580294371,
-0.1629101932,
-0.1037699729,
-0.0295060501,
-0.0198190678,
0.0508931801,
0.1874932051,
-0.4635502398,
0.0206468552,
-0.2454173416,
-0.1044287011,
0.1565062106,
0.1352600604,
0.3007023633,
-0.0551720038,
-0.1239375249,
0.3911255002,
0.0701719746,
0.5023864508,
-0.1645807028,
-0.1782221198,
0.2495260984,
-0.6714121103,
-0.0299861878,
-0.0705697387,
-0.0209421366,
-0.1671120673,
-0.0189836808,
0.1479649842,
-0.1536424458,
0.2499715835,
0.2225047052,
0.1215655804,
0.2497499585,
-0.042422507,
-0.2755948901,
-0.3043380976,
0.1753040254,
-0.1138342619,
-0.0894694328,
0.0540229306,
0.0358071476,
0.0644631013,
0.3603799939,
0.1643046439,
0.1321124136,
-0.0437576137,
-0.2861378789,
-0.0584828332,
0.0737615973,
0.1758928001,
0.2659896016,
0.1356322765,
-0.0020134225,
0.2280425727,
0.1120294109,
-0.26754269,
0.3462763429,
-0.0811369121,
0.076207228,
0.2666006684,
-0.0454611219,
-0.1460238844,
-0.3090714812,
0.2241078913,
-0.1355581731,
0.0238547418,
-0.2537421584,
-0.0748047009,
-0.5700964928,
-0.1503540874,
-0.3576545715,
-0.2965596914,
-0.3699664176,
-0.2056929171,
-0.0746991485,
-0.2369470745,
-0.205753848,
0.3813062608,
0.2306604087,
-0.009750545,
-0.0720566884,
0.2905027866,
-0.197025314,
-0.0966324508,
-0.0782291144,
0.1418488622,
0.2689705789,
0.4118452072,
0.3864146471,
-0.3926119208,
-0.0617828108,
-0.0894833952,
-0.4258924425,
0.1954808086,
-0.1968224049,
0.5000597835,
0.259865582,
0.0435458012,
0.0032626837,
-0.3089445531,
0.5313379765,
-0.363528192,
0.0224789381,
-0.0262085907,
0.0161468983,
-0.168360576,
-0.2709484696,
-0.4824747443,
-0.3764452934,
-0.3136798739,
0.1400423497,
0.0832743272,
0.1126026139,
0.5596601367,
-0.039412342,
0.1088422239,
-0.2227113843,
0.2569582462,
-0.2365370691,
-0.1614937484,
0.1379234493,
-0.3226338625,
-0.315046221,
0.166568473,
0.1618140936,
0.5677850246,
-0.0527578779,
-0.3567388654,
0.191021651,
-0.0590971485,
0.0631304979,
0.1176171303,
0.1084914058,
0.297524631,
0.2475851178,
-0.1623529196,
-0.1300200224,
-0.4439030886,
0.1164975911,
-0.2083270997,
0.1733241826,
-0.1027669758,
0.2727059126,
-0.0579381362,
0.5740634799,
0.0900356174,
0.1567033231,
0.3807677031,
0.0893213004,
0.3447575867,
0.0577327795,
-0.4066469371,
0.1074613333,
-0.1301031411,
0.0447864607,
0.4078930616,
-0.0203961395,
0.2605622113,
-0.3096643686,
-0.056355726,
-0.3117496967,
-0.3440130651,
-0.1111938134,
0.3289455473,
0.26591748,
-0.0373778269,
0.0181176737,
0.0653880611,
0.0588203445,
0.1717752218,
0.2928850651,
-0.0539856888,
0.0369720832,
-0.2738021016,
-0.276831001,
-0.3234128058,
0.1019820496,
-0.01291015,
0.3148909211,
-0.1771433353,
-0.1706011295,
-0.1155195981,
-0.0268451199,
0.4513019621,
0.0661663637,
0.1779457033,
0.0517489016,
-0.0555670522,
-0.1506129503,
-0.0096933469,
-0.1581636667,
-0.2395306975,
0.0718365163,
0.5307102203,
-0.301789701,
-0.1330114007,
-0.0683955103,
0.3462325633,
-0.2519260347,
-0.1010982692,
-0.0519164465,
-0.325830102,
-0.2018244863,
-0.1577471942,
0.0122024193,
0.0089348406,
-0.2125342637,
0.0324788615,
0.0873219818,
-0.0878789276,
0.0805717707,
0.1848245859,
0.33298406,
-0.0717552081,
0.157333985,
0.1696557999,
0.3938363194,
0.3397708237,
0.4407485425,
-0.1332305968,
0.0009340569,
-0.0843151584,
-0.0430409834,
0.2639269829,
0.3094118834,
-0.2154772431,
0.2849522829,
0.1207695603,
0.4189075828,
-0.321680963,
-0.0813535452,
0.4164521992,
0.13627325,
-0.2010636926,
-0.132318005,
0.0803623125,
0.0103405491,
0.0315413475,
-0.1423969269,
0.129077822,
-0.3507482409,
0.0609089658,
0.0781516582,
0.827927351,
0.1066753417,
0.0505959839,
0.227273345,
0.0661897287,
0.3275760114,
0.138271153,
-0.1211389378,
-0.2004660666,
-0.331795007,
0.0248154178,
-0.0992224738,
0.1703276187,
-0.0614857338,
-0.5772719383,
0.3057086766,
0.0568099171,
0.2166910022,
-0.0032007787,
0.101964429,
-0.1573067456,
-0.2348062545,
-0.1303106993,
0.248047173,
-0.1064349264,
0.0918127373,
-0.176969409,
-0.0333319306,
0.2297402322,
-0.1552755088,
-0.2749798894,
0.0140249804,
-0.4671755433,
-0.0161019973,
-0.0833166093,
0.003231883,
0.4410224557,
0.0363414474,
0.0427440256,
0.4214237034,
-0.1386112273,
0.0637239739,
-0.0429339521,
0.1349080354,
-0.08417321,
-0.0047631599,
0.3453171849,
-0.1905311942,
-0.1987347007,
0.0011529848,
-0.0088487342,
0.2760787606,
-0.1280662715,
-0.1807636917,
0.0528135896,
-0.088657096,
0.0750105754,
-0.0534613803,
0.1075576469,
-0.3861638308,
0.2380412817,
-0.0373640582,
-0.2498445809,
0.2838938832,
0.0724155158,
-0.1974201947,
-0.218572095,
0.1967155784,
-0.0967932418,
0.0722097456,
0.4667074978,
0.6660532355,
-0.1134741008,
-0.3804568648,
-0.0415862128,
0.0729248375,
-0.4744260311,
0.0584275313,
0.2235014439,
0.0947007984,
0.0354970954,
0.524518609,
0.2763050497,
-0.1558059901,
0.0191006139,
-0.2525486946,
-0.2125606537,
0.2726360261,
0.1221406907,
0.1264694333,
0.0170027614,
0.0855213553,
-0.0519936495,
0.4380587339,
-0.4479951262,
0.0722152144,
-0.0592745245,
0.1075920165,
-0.0114654889,
-0.1288197637,
0.1122964174,
-0.0021175034,
0.1891263425,
0.1201663613,
-0.2339140922,
-0.3181345165,
-0.2554551363,
0.0638010353,
0.0222387537,
-0.3724784553,
0.1332535744,
0.0038005784,
0.1911074072,
-0.0588760376,
0.1395947635,
0.2633413672,
0.0707134008,
0.2352756858,
0.3863004446,
0.1705933511,
-0.035811808,
0.220204711,
0.2296402007,
0.1827302128,
-0.1206440926,
0.0040956168,
-0.1112101823,
-0.1543441862,
0.2066457421,
0.0368232727,
0.0823289752,
0.1238953993,
0.2084759325,
-0.2427890152,
-0.1854782254,
0.0495930091,
0.6940180063,
0.3251619041,
-0.003995344,
0.1033435166,
0.2818104327,
0.3300279081,
-0.4054026604,
0.2187286764,
0.0474138148,
-0.0549256355,
0.1595047116,
0.0900709555,
0.1510616839,
-0.2287930995,
0.0520993322,
0.1556001008,
-0.0261918176,
-0.1324228048,
0.2067632526,
0.3938089907,
-0.1332143545,
0.1219130084,
0.2359657437,
0.1771038324,
0.1712321937,
0.1749378145,
-0.1229385883,
0.2797957659,
-0.1340795457,
0.204788506,
0.2102100253,
-0.1498597264,
-0.0360555016,
0.2317605019,
-0.1721914709,
-0.2055543959,
-0.1639111936,
0.7347181439,
-0.1869656295,
-0.2501263618,
-0.1461639404,
0.4038398266,
-0.1279131025,
-0.2281079292,
-0.0500653461,
-0.160884738,
-0.4026356637,
-0.0049360618,
-0.0634755865,
-0.1126711145,
-0.1450129896,
0.2226956934,
-0.0667212158,
-0.2337393612,
-0.3430243134,
-0.0320747532,
-0.172200352,
-0.2374606431,
0.2469199896,
0.4927992523,
-0.0392520651,
-0.044383999,
0.2488894463,
0.3715668321,
0.2062989324,
-0.036203362,
0.1271965206,
0.0061101206,
-0.0001474544,
-0.2387240231,
0.2048412561,
0.0573934019,
-0.0297750812,
0.3240718544,
0.2761631012,
-0.3107549548,
-0.1467662156,
-0.0071533229,
0.0752681643,
-0.1834908724,
0.0840040147,
-0.2293252647,
-0.0188466683,
-0.2724255323,
-0.0399923548,
-0.4678648412,
0.2930606604,
0.4881743789,
-0.0980253145,
0.2136204392,
-0.3128413558,
0.1499490142,
-0.0095694968,
0.5411003232,
0.3447870016,
0.0321866758,
-0.3470340967,
-0.3034186363,
-0.7088704705,
0.1239585876,
0.1814082265,
0.3573846817,
-0.2041453123,
-0.1512263566,
0.1522309929,
0.0626369789,
0.0859172195,
0.230250448,
-0.008512456,
-0.1255458295,
-0.2412825227,
0.0564876907,
-0.0381071642,
-0.0573259145,
0.0210197419,
-0.0549896099,
0.2110700607,
-0.2144034505,
0.1125238612,
-0.3159066141,
0.0951720178,
-0.0801587105,
0.0805482268,
0.1205663234,
0.2367542088,
0.1540485919,
-0.2392391264,
-0.0573793054,
-0.1824192703,
-0.2834380269,
-0.1731166095,
0.3386115432,
-0.0379035175,
0.2053393126,
0.0080538094,
0.0012943675,
-0.3797042966,
0.156288445,
0.0118327737,
-0.1514140069,
-0.2223190218,
0.2561259568,
-0.0427352339,
-0.2844474316,
0.3420319557,
0.2845244408,
-0.006555967,
0.2018400729,
-0.2968759537,
-0.3999533653,
0.402936548,
-0.182915777,
-0.1783638448,
-0.2296224236,
0.0490625948,
0.1954255551,
-0.1067995727,
-0.1924479455,
0.0228689164,
0.1301561296,
0.0905378982,
-0.0717686489,
0.215465337,
-0.1068772972,
-0.1212995425,
-0.0754631385,
0.248825565,
0.5338648558,
-0.244704172,
-0.0007781237,
-0.0159924962
] |
https://github.com/huggingface/datasets/issues/1892 | request to mirror wmt datasets, as they are really slow to download | Yes that would be awesome. Not only the download speeds are awful, but also some files are missing.
We list all the URLs in the datasets/wmt19/wmt_utils.py so we can make a script to download them all and host on S3.
Also I think most of the materials are under the CC BY-NC-SA 3.0 license (must double check) so it should be possible to redistribute the data with no issues.
cc @patrickvonplaten who knows more about the wmt scripts | Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you! | 78 | request to mirror wmt datasets, as they are really slow to download
Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you!
Yes that would be awesome. Not only the download speeds are awful, but also some files are missing.
We list all the URLs in the datasets/wmt19/wmt_utils.py so we can make a script to download them all and host on S3.
Also I think most of the materials are under the CC BY-NC-SA 3.0 license (must double check) so it should be possible to redistribute the data with no issues.
cc @patrickvonplaten who knows more about the wmt scripts | [
-0.1942384243,
-0.3241636753,
0.0704936236,
0.2124050111,
0.0100602806,
0.0652932897,
0.1663088948,
0.4544949532,
0.0562023893,
-0.1142764091,
-0.2518415749,
-0.0738519654,
-0.06027551,
0.3107830882,
-0.0818378031,
-0.082689546,
-0.0564791411,
0.117088981,
-0.6240912676,
-0.0960812271,
0.049162589,
-0.1243581846,
-0.0437548459,
-0.0403343216,
-0.0081343129,
0.2965478301,
-0.2204904109,
0.0053675408,
-0.2378391922,
0.0302221179,
0.0825243741,
0.5002446771,
0.251403451,
0.3773463666,
-0.0001185067,
-0.126459077,
0.1999658495,
0.1051911861,
-0.2532367706,
0.0950694531,
-0.4915254712,
-0.368005842,
-0.2355643511,
-0.0161058605,
-0.1161405444,
-0.121174708,
-0.1031257287,
-0.0419019759,
0.2931005657,
-0.1038037539,
0.0933474377,
0.4699972272,
-0.489641577,
-0.0356688872,
0.530185163,
0.3358390033,
-0.2962785065,
0.322100997,
0.5346035361,
-0.1389747262,
0.3522563279,
0.4234989882,
-0.156877622,
0.3637657464,
0.608229816,
-0.1887510121,
0.0055284631,
-0.0134865753,
-0.2485119104,
0.6527907848,
0.720821023,
-0.1008874476,
-0.5376056433,
-0.0442305282,
-0.0979527682,
0.1767014563,
-0.3057172298,
0.1743835807,
0.0661839396,
0.2955293953,
-0.4052233398,
-0.3033836782,
-0.0579063445,
0.028833976,
-0.0891483873,
0.4804273844,
0.0012676418,
-0.0247035474,
-0.178448841,
-0.1355647594,
0.447549969,
-0.4463378191,
0.2125969082,
0.052524168,
0.0689760819,
-0.3961447477,
-0.4048392773,
-0.1257927716,
0.1217901856,
0.0788983181,
-0.0996375382,
0.0322677381,
0.0608738624,
-0.0059300326,
0.0533965342,
-0.0918022692,
-0.2332047522,
-0.2772551477,
0.276022613,
0.1725939363,
0.0821031928,
-0.0665009767,
-0.1882123649,
-0.1074699759,
-0.4842552543,
-0.0299398042,
0.0422553644,
-0.3843736947,
-0.1701831818,
-0.2568981946,
0.1291090846,
-0.1630942523,
-0.2392385453,
0.0911242962,
-0.2050315589,
0.2176316679,
0.3381383419,
0.136832267,
-0.1975699961,
-0.1293521225,
-0.0453771092,
-0.188868776,
0.1018537506,
0.4096264839,
0.1961643547,
0.1012697145,
0.0809570774,
0.0125230551,
0.2291606665,
0.3779651821,
0.2429534644,
0.2257289439,
0.001198329,
0.2455645502,
-0.2264271677,
0.5534644127,
-0.057104364,
0.4946643412,
-0.2405674309,
0.0189298447,
-0.3959585726,
-0.1411180943,
0.1163502857,
0.1434742212,
-0.2485121191,
-0.2411221564,
-0.3235341311,
-0.1215085685,
-0.35814026,
0.1436906159,
-0.0027514398,
-0.1603312492,
-0.1871884763,
0.0441946797,
0.1369812042,
0.5623064637,
-0.3566908836,
-0.0392404646,
0.131822899,
0.0986153483,
0.424931854,
0.4538915455,
-0.1545875221,
0.0098787751,
-0.2259877622,
-0.0569346398,
-0.089612633,
-0.219973892,
0.0895927846,
0.4570540488,
-0.2056547701,
-0.0597872883,
0.2275106907,
0.1248132437,
0.1572808921,
-0.3333367407,
-0.0803301334,
0.4827218056,
-0.0005444828,
-0.0532287396,
-0.2400193065,
-0.5139436722,
0.3213064671,
0.3085312843,
0.1036253944,
0.1355848759,
-0.1118638963,
0.0921422392,
0.4986329973,
-0.0057324041,
0.0308685079,
0.0384277143,
0.2461277843,
0.4436830878,
0.0216942094,
-0.0548782088,
0.1014436334,
0.096406959,
0.1379812956,
0.0048361532,
0.3148324788,
-0.2072589099,
0.0021637827,
-0.115176335,
0.0240760613,
0.2152529359,
0.0360905528,
-0.225575909,
-0.1383959502,
-0.3562446237,
-0.0284169167,
0.0065099206,
-0.0618622452,
0.0354508124,
0.3565734327,
0.390959084,
0.0855043307,
0.3153095841,
0.4274920225,
-0.1052078605,
0.2611541748,
0.0331895128,
-0.0113981124,
-0.2118982673,
-0.0324240625,
0.3801894486,
0.488373518,
0.4331640005,
0.295230478,
-0.5407387018,
0.589648664,
0.1838253736,
0.142156601,
-0.2557568848,
-0.1816962808,
0.2997313142,
-0.1090610176,
0.140494749,
0.0827983767,
-0.0776055753,
-0.0244505182,
-0.1666246802,
-0.1693378091,
0.0878171623,
0.2612861097,
-0.0619199686,
-0.279553771,
0.0631748587,
0.0156491995,
0.287946552,
0.1335858405,
-0.1323907226,
0.0646265298,
-0.0883880854,
-0.3523097038,
0.1846617162,
0.1971086562,
0.1618146747,
0.1227263883,
0.2624022961,
0.0814857632,
0.1348207295,
0.3998576999,
-0.2348160446,
0.0997166336,
-0.0268704183,
0.0908876956,
0.176591441,
-0.1573081911,
-0.0949690491,
-0.2645001709,
-0.1340387464,
-0.0339328125,
-0.1693637967,
-0.2057607025,
-0.2036469579,
-0.3950929642,
-0.1671647727,
-0.522464633,
-0.0712077394,
-0.1530240178,
-0.2038121521,
0.0044245683,
0.0359449759,
-0.0924447328,
-0.0787626877,
-0.1533592641,
0.1934636086,
-0.3043536842,
0.1676544994,
-0.2028640956,
0.0372742452,
0.0826230273,
-0.0348153338,
0.1339907199,
0.1073844284,
0.3420970738,
-0.1645021141,
0.0245060399,
-0.6451833844,
-0.1008390337,
-0.089119032,
0.2413384318,
-0.0052913204,
-0.0128092319,
0.4223403931,
0.3099901676,
0.1228702366,
0.1148652956,
-0.1887271404,
-0.0579358004,
0.0266150832,
-0.1791532934,
-0.1181853935,
-0.0117391348,
-0.1499120593,
-0.4867738187,
-0.1689321995,
-0.0910814703,
-0.2901758254,
-0.0047473814,
-0.0703092366,
-0.040264599,
-0.1997774243,
-0.229234755,
0.0202788487,
-0.1326332092,
-0.453529954,
0.4614348114,
0.0416385606,
-0.2373159379,
0.1984031945,
0.1946206391,
-0.296246171,
0.0509120934,
-0.4404748082,
-0.0259736814,
-0.0599780604,
-0.1563955396,
0.1378272772,
0.3616069853,
0.0748509914,
-0.0016158856,
-0.0182546601,
-0.1750897467,
-0.1703828424,
-0.3151820004,
0.0791269392,
0.3384955227,
0.0421885401,
0.1538708359,
0.11675594,
0.9036083221,
0.210612312,
0.0860703439,
0.169300884,
0.3139073849,
0.012604733,
0.3304952085,
-0.1501100659,
0.0602099895,
-0.1646617353,
0.1526874304,
0.3023821115,
0.1654992849,
-0.0322430693,
-0.1236023158,
-0.0410887562,
-0.072139211,
-0.1531145126,
-0.1666942686,
0.0026161554,
0.103264004,
0.2143985778,
-0.1322299242,
-0.0433785841,
-0.2789010108,
0.5671608448,
0.5380115509,
0.0727009252,
0.1175924093,
0.0948473513,
-0.2196178883,
-0.3919325471,
-0.0763271898,
-0.0541767366,
0.278693527,
0.0569843054,
0.1415186673,
-0.0046378374,
-0.0162835456,
0.5470104814,
-0.3564412296,
0.2949029803,
-0.1442958117,
-0.3694461584,
-0.1728059351,
-0.3832763433,
-0.1095412225,
-0.5588152409,
-0.0772591457,
0.3528341055,
0.0084958524,
-0.1516512632,
0.0020081364,
0.0590505898,
-0.3691159487,
-0.357565403,
0.2556676269,
-0.2719133496,
-0.0996837318,
-0.3776423633,
-0.0062341243,
-0.2807675898,
-0.252792567,
-0.0268178061,
0.0922122672,
0.4408901036,
0.3354247212,
-0.0946623981,
-0.0925572962,
0.3012339771,
0.3603006899,
0.0374417827,
0.239199087,
0.3606663346,
0.5046021938,
0.2717078626,
0.0542823672,
-0.4194882214,
-0.1345646828,
0.2902258337,
0.4311195016,
0.0573270358,
0.2111230344,
0.1722723097,
0.3007184863,
-0.5591009259,
-0.0802775621,
0.094420962,
-0.0634875447,
0.01727736,
-0.3364311755,
0.2843795419,
0.1423133612,
0.0836705267,
-0.2427265346,
0.0492842086,
0.1078375652,
-0.1778745055,
0.2658793926,
1.1352505684,
-0.3869093657,
0.376745373,
-0.2859295309,
-0.4389286637,
0.1930589378,
-0.2761821449,
-0.327101171,
-0.2529641986,
-0.3486737311,
-0.1005759537,
-0.1680057049,
-0.0770905167,
0.2319844067,
-0.0434465744,
0.3802528381,
0.2873780727,
0.1452065855,
0.0846100897,
0.2755839527,
-0.1763866246,
-0.1767055988,
0.0174673013,
0.0858034045,
-0.193377614,
0.3361292481,
-0.1472962499,
-0.2969938815,
0.0805654377,
0.1121013314,
-0.2119933963,
0.1694366783,
-0.0968711525,
-0.1284095049,
-0.3832775652,
0.0316642821,
0.5623490214,
-0.6436482668,
-0.0283583142,
0.4671005905,
-0.4409530461,
-0.0093006985,
-0.1883155555,
0.0606652386,
-0.20780164,
0.0433314256,
0.1382693201,
-0.1444745362,
-0.1458596289,
-0.2769359052,
0.2813338935,
0.0168469772,
-0.0551266074,
-0.4118455052,
0.1743116677,
0.1672119796,
-0.0446520522,
0.1962047815,
0.0632110983,
-0.027257368,
0.081934467,
0.2307800502,
-0.4344761968,
0.3629768491,
0.3060404956,
-0.0408435911,
-0.0279143602,
0.6274665594,
0.0486875065,
0.1014704406,
0.358007133,
0.5374691486,
-0.2415651232,
-0.1683885753,
0.03074947,
-0.0980750769,
-0.1062975079,
0.1364813149,
-0.0502545759,
0.0660018623,
-0.0690909773,
0.1345000267,
0.359329313,
-0.4689960778,
-0.2436364144,
-0.3942450285,
0.0242582783,
0.0535476953,
0.0327043012,
0.1865342557,
-0.2131732404,
0.0048585422,
0.0788605288,
-0.1300335079,
-0.2711831927,
0.2341392636,
0.1094053015,
0.1262316406,
-0.0035662409,
-0.0821690708,
-0.1568306983,
-0.0009768754,
-0.0895654336,
-0.1557525992,
-0.1373563856,
-0.1435672641,
-0.0431495905,
0.2149403989,
0.203358233,
-0.2076735795,
0.2152034938,
-0.4240145683,
-0.2052711546,
0.2071904093,
0.0006218031,
0.2079429626,
0.0590288006,
0.3071046472,
0.4449148774,
-0.4507369697,
-0.3996252418,
0.5348268747,
0.3253592849,
0.325548321,
-0.0671706125,
-0.2354675829,
0.2650062442,
-0.0123568773,
-0.0109518729,
-0.0831559971,
0.0932818726,
0.2349170446,
0.6755411029,
-0.2223101705,
0.2469662428,
0.3122732043,
0.5689751506,
0.1658486128,
0.0841730386,
-0.0623439997,
0.3127584159,
0.1107253581,
-0.265597105,
-0.2527963817,
0.0075159334,
0.1948584914,
0.0039294586,
0.031006027,
0.1690725237,
0.0471989512,
-0.3080033064,
-0.1037282944,
0.1783950031,
0.2707660794,
-0.2716915905,
0.1359701008,
0.3350217342,
0.0492314138,
-0.0688229427,
0.1073079407,
-0.0878665149,
-0.0548639335,
0.2023761272,
0.3228910863,
0.1374693066,
0.1845921725,
0.0004635602,
-0.2728708088,
-0.0769897327,
0.1307692528,
0.3602414131,
0.1649554521,
-0.1015259922,
-0.224044174,
0.3066638112,
-0.4101092815,
-0.1396854222,
0.4005136192,
0.1548840404,
-0.0920046866,
-0.0353095643,
-0.0261295214,
-0.0149947479,
0.1356625259,
-0.0866740867,
-0.1096813381,
0.2018982321,
0.2682109475,
0.3090908825,
-0.1002070829,
0.0417803898,
0.0281452537,
0.2311130762,
-0.1858752966,
-0.3661703765,
-0.1566651464,
0.1162602231,
0.2669991255,
0.3494160175,
0.0623408258,
0.0763204992,
-0.1407997161,
0.092416361,
-0.0587396175,
-0.0367527828,
0.1045640409,
0.1864999235,
-0.1557811797,
-0.2483769059,
0.1958195269,
0.0466491021,
-0.1271169186,
-0.142116189,
-0.3193196058,
-0.4644871652,
-0.2384033799,
0.143011421,
0.2091126442,
-0.0635755435,
-0.1490926296,
0.1420561224,
-0.2940700352,
0.0162313841,
0.6448604465,
-0.2574047148,
0.1037266329,
0.0241502821,
0.0357918739,
-0.3664198518,
0.4402468503,
0.5229848027,
0.3633785546,
-0.2721738815,
-0.047226727,
-0.2062494159,
-0.115587458,
-0.3925126195,
-0.0154037923,
0.1047282517,
0.0767886639,
0.2676980793,
0.0844926387,
0.1572559476,
0.4931558371,
-0.1959607601,
0.1494655311,
-0.113935329,
-0.0947786123,
-0.1895321906,
0.2923417985,
-0.1569336057,
-0.3024151623,
0.2051189244,
0.073002249,
-0.1011532769,
-0.0674512833,
0.1848723143,
0.2188967764,
0.2232159972,
0.0867183805,
-0.3278850913,
0.2595081925,
0.0074994564,
0.0889080837,
0.0261738766,
-0.0786308944,
-0.0229803585,
-0.0662233979,
-0.2500489354,
-0.1942214072,
0.0299949944,
0.0709098354,
-0.1323095709,
0.0677479655,
-0.1909640133,
0.0218406878,
-0.2647407651,
0.2825656235,
-0.5296421647,
0.1088238508,
0.08981511,
-0.0256011337,
-0.1168296561,
-0.2822607756,
-0.2311117649,
-0.1848533452,
0.5103884339,
-0.1827991903,
-0.1760084331,
-0.3596640825,
0.1117760837,
-0.3630666733,
0.3120628893,
-0.1571567208,
0.1808514595,
0.146075651,
-0.0923875272,
0.2985778451,
0.5052081347,
0.0931397378,
-0.0751510635,
-0.0936296284,
0.2878223658,
-0.0517689437,
0.0559037402,
-0.5135706663,
0.0270242691
] |
https://github.com/huggingface/datasets/issues/1892 | request to mirror wmt datasets, as they are really slow to download | Yeah, the scripts are pretty ugly! A big refactor would make sense here...and I also remember that the datasets were veeery slow to download | Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you! | 24 | request to mirror wmt datasets, as they are really slow to download
Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you!
Yeah, the scripts are pretty ugly! A big refactor would make sense here...and I also remember that the datasets were veeery slow to download | [
-0.3805081248,
-0.1753541529,
0.0075874254,
0.1037698239,
-0.0104690157,
0.0599777699,
0.1670314968,
0.5077075958,
0.2716600299,
-0.030725494,
-0.2506195307,
-0.2216313183,
0.0339968055,
0.2432686985,
-0.1237287074,
-0.239030838,
-0.1749915481,
0.16517362,
-0.779135108,
-0.1799132079,
0.1061559394,
-0.1956076026,
-0.1282523721,
0.0072471853,
-0.0354573131,
0.1832385957,
-0.1280825734,
0.0092024757,
-0.2814995646,
0.1942980289,
0.0807315558,
0.4943602085,
0.1801086664,
0.2579833269,
-0.0001139865,
-0.1233247072,
0.3037296236,
0.1550534815,
-0.1294933856,
0.2597905993,
-0.3924543858,
-0.3952482939,
-0.1610326618,
-0.0658746883,
-0.0436392166,
-0.1273961961,
0.0423425958,
-0.1791129708,
0.2340648323,
-0.1549863219,
0.171645835,
0.3462651968,
-0.4791023731,
-0.0595959574,
0.3936363757,
0.286071986,
-0.3739833832,
0.364372313,
0.5096561313,
-0.1086650491,
0.0914192498,
0.3541179299,
0.0158587992,
0.2641913891,
0.4950575829,
-0.2407833636,
0.075598985,
0.0502777286,
-0.2511315644,
0.5559904575,
0.7872530222,
-0.0566436276,
-0.3280629516,
0.1233151257,
-0.1975927055,
0.2542442977,
-0.2248380929,
0.056847848,
0.024637863,
0.3577072918,
-0.4675729573,
-0.3055554032,
0.0396123081,
-0.0742741525,
-0.1059215739,
0.2843853235,
0.0716390312,
-0.0120374598,
-0.0921833813,
-0.1142880917,
0.2289707363,
-0.4278953075,
0.084317334,
0.0660024583,
0.0435409769,
-0.4074560404,
-0.3956960738,
0.0255871136,
0.2399977148,
0.1858922392,
0.0770378709,
0.1213210449,
-0.0796720162,
-0.0602354109,
0.1222131625,
0.2509381175,
-0.2132791579,
-0.3227464855,
0.3489449024,
0.041504506,
-0.0507074594,
0.0060409009,
-0.0399923138,
-0.1021284908,
-0.3949145675,
0.1237595454,
0.0564691722,
-0.3286578655,
-0.2215178311,
-0.2023864985,
0.0951995999,
-0.2182260007,
-0.1477466524,
0.1160877645,
-0.2152334899,
0.3245935738,
0.3515891433,
-0.0812582225,
-0.2032057941,
-0.1062378585,
-0.0876734182,
-0.0906695575,
-0.1020315588,
0.3286112547,
0.2291555554,
0.1653603911,
-0.0103696957,
-0.0122763775,
0.2687147856,
0.3333777785,
0.1240339875,
0.1441441476,
-0.0162178352,
0.1667737365,
-0.310200125,
0.4530284703,
-0.001628428,
0.5766243339,
-0.2522420287,
0.1281215996,
-0.4634129703,
-0.1512123793,
0.0933424607,
0.2227725983,
-0.0691979453,
-0.1474102139,
-0.337459445,
-0.1135534644,
-0.3930585384,
0.083895877,
-0.0484142676,
-0.0942336023,
-0.1931762397,
0.1001076922,
-0.0327635556,
0.5073940158,
-0.4661901295,
0.1300623864,
0.0478364043,
-0.0847914666,
0.5431499481,
0.5093154311,
-0.2762376666,
0.1490687728,
-0.3273759186,
-0.1876094043,
0.076281473,
-0.2453273088,
0.0465539619,
0.5941821933,
-0.2964410782,
-0.0330518,
0.229226619,
0.1116807386,
0.2743134499,
-0.2730725408,
-0.1289037168,
0.3666532636,
-0.0583093241,
-0.0247747488,
-0.2994121015,
-0.548987627,
0.2229861915,
0.265191257,
0.1280786693,
0.3034206629,
-0.0386642814,
0.0988603681,
0.6538225412,
-0.0089280922,
-0.0234395843,
-0.0003189817,
0.0472186655,
0.3054363132,
0.1835783422,
-0.0747922957,
0.0838960558,
0.032362245,
0.0159604624,
-0.0498866104,
0.5514829159,
-0.2607075572,
-0.137117818,
-0.1333979964,
0.1238406301,
0.2173396945,
0.0750877187,
-0.1869908869,
-0.3331971467,
-0.4164793491,
-0.0838468373,
0.3183396757,
-0.1707100719,
-0.0238183904,
0.3621007204,
0.3282919526,
0.0042817909,
0.4077358544,
0.2721312344,
-0.2106319219,
0.2551044226,
-0.0920155868,
-0.0419658087,
-0.1011897698,
-0.0889157057,
0.3310205042,
0.5127509832,
0.2911524773,
0.3215765953,
-0.4997346997,
0.5782854557,
0.2840393484,
0.0891665518,
-0.30639112,
-0.1912350506,
0.2503746152,
-0.1309269369,
0.0748780817,
0.1060548946,
-0.1318773329,
-0.0196297728,
-0.2404820323,
-0.1276594698,
0.0532319397,
0.4599596262,
0.123571977,
-0.1596833318,
0.1282086819,
-0.0428500772,
0.2617390752,
0.0907395184,
-0.131102547,
-0.0340097919,
-0.07171157,
-0.1973290294,
0.2812798023,
0.0401282609,
0.0109607726,
0.1533150375,
0.3378610015,
0.0244841389,
0.2259531617,
0.3778087497,
-0.2540163696,
0.0651435256,
-0.0915398449,
0.0669926256,
0.1442794502,
-0.0549814478,
-0.1041566953,
-0.4047584534,
-0.1812863797,
0.0463713929,
-0.097675927,
-0.1688510925,
-0.3446543217,
-0.4082674086,
-0.0992144421,
-0.5792313218,
0.0557991751,
-0.0046377042,
-0.2675743699,
0.1047199145,
0.0755683184,
-0.1571829021,
0.1201935709,
-0.1703462154,
0.2722744942,
-0.2807752192,
0.3172825277,
-0.288212508,
0.0609799996,
0.0805847198,
0.0599450544,
0.1933193207,
0.0593234189,
0.3335086107,
-0.1804133654,
0.0672978312,
-0.5876150131,
-0.1100950167,
-0.0758060142,
0.3115232885,
-0.0588632077,
-0.0066559911,
0.4242178798,
0.2608998418,
-0.0444028154,
-0.0744852349,
-0.1252696067,
-0.1272204518,
-0.1251818836,
-0.1961210519,
-0.1452279389,
-0.0458362661,
-0.2629902661,
-0.4468738139,
-0.1661348939,
-0.0215505026,
-0.2584235072,
0.0988363177,
0.0352509469,
-0.1489520371,
-0.3318729103,
-0.1897390187,
-0.0466474779,
-0.1492342204,
-0.3901193738,
0.457868129,
0.0261717141,
-0.2646771669,
0.1498560607,
0.2164312899,
-0.5094373226,
0.0981019586,
-0.2976003289,
-0.0140466578,
-0.1650476903,
-0.2264686823,
0.1445493996,
0.2360325456,
0.088688679,
-0.0517384522,
-0.0951495245,
-0.0671272054,
-0.0962918401,
-0.0946698785,
-0.0157100763,
0.3045806587,
0.0664639771,
0.1675761044,
-0.0576214381,
0.959654808,
0.021691002,
0.1754842103,
0.0679393783,
0.2717377245,
0.039869912,
0.2014281005,
-0.2668825686,
0.0835353807,
-0.1249187142,
0.0521823317,
0.3039143085,
0.1249791533,
-0.1708261669,
-0.0700298846,
-0.0757937729,
0.0712348446,
-0.1591446698,
-0.1414767057,
-0.1161919758,
-0.0078626908,
0.3021859527,
-0.0106130838,
-0.1702646017,
-0.2661761045,
0.4968168139,
0.5332591534,
-0.0456055067,
-0.0154970642,
0.0921605825,
-0.0783628225,
-0.3821451664,
-0.0482672155,
0.0627890527,
0.2886773646,
0.014483951,
0.1875742525,
0.046396181,
-0.0074964128,
0.5453279614,
-0.123771362,
0.3349528909,
-0.1271113455,
-0.276181519,
-0.0125231147,
-0.3755600452,
-0.0907004774,
-0.4050469995,
0.0913535953,
0.1981764138,
0.0271169692,
-0.1265957505,
0.3056604266,
0.0780079439,
-0.3180399835,
-0.4010605514,
0.2760355771,
-0.1486583352,
-0.0660197586,
-0.2761560678,
-0.0371491536,
-0.3086215556,
-0.2919816971,
0.0455661714,
0.0917200595,
0.3058345318,
0.3838375211,
-0.2576648295,
-0.005940564,
0.2473200262,
0.3797290325,
-0.0763752684,
0.4432213306,
0.1999128163,
0.4128923714,
0.2050314099,
0.1331305951,
-0.2609007955,
-0.1755682528,
0.144667536,
0.5312457681,
0.030027464,
0.0413483717,
0.1008012891,
0.2563354075,
-0.4412169755,
-0.0150057562,
0.0300204679,
-0.1436484009,
-0.1330214441,
-0.363894701,
0.4546467662,
0.2197043449,
0.0021267086,
0.1150175333,
0.0837421864,
0.1231453046,
-0.2407297641,
0.131878078,
1.2271447182,
-0.326397866,
0.164795965,
-0.3563348353,
-0.4868181348,
0.0334761888,
-0.4651016295,
-0.2087593377,
-0.3127734363,
-0.4907945096,
-0.1319436878,
-0.1764509678,
0.0160788223,
0.2103881836,
-0.0678400472,
0.3187937737,
0.2265431881,
0.1773726046,
0.0438163504,
0.4219675958,
-0.2083984166,
-0.1405155659,
0.2325100899,
0.1023765802,
-0.113053292,
0.2913127244,
-0.1807507277,
-0.2523024976,
0.0849612355,
0.088307485,
-0.3463484645,
0.1626555771,
-0.1756796241,
-0.0730064958,
-0.4221746922,
0.038206622,
0.454090625,
-0.597476542,
-0.0403134748,
0.4229363203,
-0.4424492121,
0.1386765093,
-0.2240237594,
0.1539199054,
-0.1801402122,
0.1034056097,
0.129721269,
-0.1762741953,
-0.2432311028,
-0.3747296333,
0.1003049612,
-0.1255984008,
-0.2650671899,
-0.3074171543,
0.3010383546,
0.2093643844,
-0.0423920378,
0.296833992,
0.0057013482,
-0.0042280853,
0.109611854,
0.1353616863,
-0.3067113757,
0.4592968225,
0.4877769351,
-0.1900091916,
-0.0723126903,
0.6181308031,
0.2643422782,
-0.0587379932,
0.3968242407,
0.3254401088,
-0.3243731856,
-0.2269504368,
-0.0082669035,
0.1197330952,
-0.165471226,
0.0481740646,
-0.2693099678,
0.0307193995,
0.0650306791,
0.2906698287,
0.2003008723,
-0.3191122711,
-0.3266789913,
-0.2999357581,
-0.1410444379,
0.0363922343,
-0.1130056977,
0.1646092236,
-0.0948688462,
0.1472031474,
0.0882496908,
-0.0622581765,
-0.3280103207,
0.1313580275,
0.192831859,
0.0267667919,
0.1010588259,
-0.0162748303,
-0.1377649605,
0.0935384184,
-0.0342330895,
-0.0437326245,
-0.2401140481,
-0.1801443547,
0.0280422494,
0.1960352957,
0.009221483,
-0.1458551139,
0.145878613,
-0.4362348914,
-0.2423063964,
0.2092885375,
-0.0175633468,
0.1895726472,
0.0336883925,
0.0602164306,
0.4817604125,
-0.4044142962,
-0.2955756783,
0.4526636899,
0.3606827855,
0.4359562397,
-0.1289250106,
-0.2172567248,
0.1653545499,
-0.0583173186,
-0.0379863232,
-0.0083535127,
0.253379494,
0.3885250688,
0.5414737463,
-0.0975565165,
0.2074244469,
0.332834214,
0.5465676188,
0.1835826039,
0.1167749465,
-0.2574369013,
0.1962171793,
0.1262338907,
-0.2941484153,
-0.1781191528,
0.2401687205,
0.1639473587,
-0.0225154012,
-0.0333559737,
0.1968993843,
-0.0072688162,
-0.0731638446,
-0.0811792091,
0.2850218415,
0.1132995412,
-0.3163086176,
0.1843848377,
0.2956699729,
0.0016436419,
0.0151050109,
0.2206532508,
0.0046650544,
0.2000064999,
0.1815549433,
0.3975971341,
0.2893585861,
0.2947401106,
-0.0303025022,
-0.2451178133,
0.0555298626,
0.2859324515,
0.2938245833,
0.1808363497,
0.1104945391,
-0.3314802647,
0.3333825767,
-0.3425589204,
-0.2218650281,
0.3521630764,
0.1152309477,
0.0076439194,
0.038280189,
-0.1178942174,
-0.0955641866,
0.0974495858,
-0.1117017865,
0.1432794333,
0.2680999339,
0.1403879374,
0.144807905,
-0.064090237,
0.0713653043,
-0.0701778829,
0.2951073349,
-0.2350680679,
-0.3420344889,
-0.2025131583,
0.191454187,
0.4404248297,
0.5600760579,
-0.1249939054,
0.2510113716,
0.0793599188,
-0.0655983165,
-0.1637024432,
-0.0240781642,
0.1137698293,
0.2061110437,
-0.1341823786,
-0.2268020213,
0.2327247411,
0.0812334418,
-0.1399888843,
-0.0647992939,
-0.2271707803,
-0.5099564791,
-0.2978397012,
0.1872097552,
0.2960660458,
-0.0391265675,
-0.1910185814,
-0.0217962228,
-0.1943466663,
-0.1120371744,
0.3948348165,
-0.2685812712,
0.0481977947,
0.0001684949,
0.0641376674,
-0.3175899088,
0.3527599871,
0.5571142435,
0.2938533723,
-0.2924823761,
-0.1421652585,
-0.3117377758,
-0.0294770002,
-0.3457057774,
0.0882663652,
0.0843097791,
0.0187881216,
0.3116801977,
0.0353749245,
0.3401125371,
0.3215562403,
-0.086168997,
0.0670942813,
-0.1379246116,
-0.0336063318,
-0.0977792591,
0.2399320751,
-0.1622872353,
-0.420037806,
0.3297254741,
0.1149038225,
-0.0178158544,
-0.0010441747,
0.02219145,
0.2019757926,
0.0660200119,
0.0297752395,
-0.2607840598,
0.3256411254,
-0.0511154011,
0.1036878228,
-0.0411340445,
-0.1182778478,
-0.0885495842,
-0.1841378063,
-0.0987275839,
-0.2638862729,
0.0237842277,
-0.0384650119,
-0.0475372486,
0.1023280919,
-0.2719503045,
0.1443004012,
-0.3689770401,
0.0997441933,
-0.4695649445,
0.0733556151,
0.0477810763,
0.011149887,
-0.026246652,
-0.1484720409,
-0.1620287448,
-0.1092316732,
0.4935833216,
0.1556566954,
-0.3027801514,
-0.3601826131,
0.1839683503,
-0.3446587324,
0.429477334,
-0.160238415,
0.0225388706,
0.1177979633,
-0.1433000118,
0.3758054674,
0.4795023203,
0.1060113609,
-0.1197684854,
-0.1107082665,
0.1266929805,
0.02238065,
0.0692921504,
-0.5538514256,
0.0501203537
] |
https://github.com/huggingface/datasets/issues/1892 | request to mirror wmt datasets, as they are really slow to download | I'm downloading them.
I'm starting with the ones hosted on http://data.statmt.org which are the slowest ones | Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you! | 16 | request to mirror wmt datasets, as they are really slow to download
Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you!
I'm downloading them.
I'm starting with the ones hosted on http://data.statmt.org which are the slowest ones | [
-0.4048578739,
-0.3140093684,
0.0369966589,
0.1695627272,
-0.0139839752,
0.1483639777,
0.0976713747,
0.4406712651,
0.2012498975,
-0.0260526091,
-0.346716702,
-0.2557104528,
0.089427039,
0.1072440669,
-0.1605698019,
-0.2548780441,
-0.0824237168,
0.1284086704,
-0.6802176833,
-0.1375782937,
0.091618076,
-0.2926720679,
-0.1066291854,
0.1005204096,
-0.1245002747,
0.1187914386,
-0.133661136,
-0.0403252169,
-0.2713429928,
0.2231260687,
0.0854253992,
0.3694525063,
0.1696946025,
0.2424312979,
-0.0001144641,
-0.1100507379,
0.1767381579,
0.2066613734,
-0.2290119082,
0.0147692934,
-0.4497108459,
-0.4104299545,
-0.2166116834,
0.0102257431,
-0.0480710827,
-0.2220498472,
-0.0168272518,
-0.1880831718,
0.2030467391,
-0.2183157206,
0.1400483996,
0.2997651994,
-0.4613547325,
-0.1171536893,
0.3193843365,
0.2256764472,
-0.2811762393,
0.3676024675,
0.3717102706,
0.0632949844,
0.1310900152,
0.2845406234,
-0.0112568922,
0.5359500051,
0.3685593009,
-0.1663260758,
0.1632972658,
0.1413365602,
-0.225880444,
0.6267237663,
0.8606652021,
-0.0362280495,
-0.310733974,
0.1644583046,
-0.1043654829,
0.308845073,
-0.1453814656,
0.2164488435,
0.0218990073,
0.3345196247,
-0.3599195778,
-0.262278527,
0.0219193697,
-0.0385219529,
-0.1772530377,
0.2454933375,
0.0386829115,
-0.0167624876,
-0.1030219346,
-0.1719417572,
0.2029069215,
-0.5088528991,
0.1482291818,
0.055997245,
0.047647737,
-0.3436841071,
-0.6021658182,
-0.0119135678,
0.2365086079,
0.1315685809,
-0.0645148084,
0.0113891624,
-0.1155206263,
-0.004034251,
0.1955769062,
0.0216698758,
-0.3318408728,
-0.3245334029,
0.2537921369,
-0.1227807477,
-0.120769538,
-0.0051749642,
-0.0560189299,
-0.0316803455,
-0.3978102803,
-0.120177947,
-0.0525078848,
-0.4321282804,
-0.2400132418,
-0.2409908026,
0.0043359399,
-0.2385877073,
-0.1764441431,
0.0435196981,
-0.215959385,
0.1803097874,
0.2785014808,
-0.055915039,
-0.1992188394,
-0.1912968457,
-0.1133067161,
-0.2688353658,
0.0762467533,
0.3992102444,
0.120294638,
0.0832928866,
-0.0349253714,
0.0040840283,
0.2557309866,
0.318690747,
0.2280759215,
0.0805882514,
-0.0507085025,
0.1632943749,
-0.1923896968,
0.5492449999,
0.0061652865,
0.5649520159,
-0.2161950767,
0.0496354848,
-0.4902124107,
-0.0830660462,
0.1664422452,
0.1738640666,
0.0148536861,
-0.1995099038,
-0.3133870065,
-0.0792063922,
-0.509786427,
0.1552451253,
-0.1281244159,
-0.0444270633,
-0.1475668252,
0.1225654855,
0.0442837477,
0.4127905071,
-0.2928538322,
0.0157919377,
0.0534082763,
-0.0462498218,
0.4894143939,
0.505053103,
-0.2413395941,
0.1250048727,
-0.2376469076,
-0.1175845712,
0.0515241027,
-0.1229909658,
-0.0325722434,
0.5273020267,
-0.3150414228,
-0.2354466915,
0.177016288,
0.2023500651,
0.2160565853,
-0.2299129069,
-0.073620297,
0.3995552063,
-0.0265601985,
-0.04338523,
-0.2944585681,
-0.486957103,
0.2586960793,
0.237485528,
0.0530504733,
0.1959837079,
-0.0057677813,
0.1948940903,
0.4590445757,
0.0916241556,
-0.0071795583,
-0.0429694243,
0.0914266109,
0.2943184376,
0.0716166943,
-0.0905446783,
0.1710484028,
-0.0536939166,
0.1256410778,
-0.0315475166,
0.5903394818,
-0.1503251493,
0.007481385,
-0.0631248429,
0.1456244439,
0.389059037,
0.0475445241,
-0.1497335136,
-0.253745079,
-0.290907979,
0.048539415,
0.3243512213,
-0.2179530561,
-0.0125326999,
0.5686087012,
0.4224883318,
0.0976048112,
0.4273555875,
0.3591607809,
-0.1608588845,
0.2127684355,
-0.0247064885,
-0.1188274771,
-0.1153676957,
-0.0895705372,
0.3114090562,
0.5767363906,
0.1853414923,
0.363009572,
-0.5933861732,
0.5415113568,
0.1586473435,
0.035349369,
-0.2379424572,
-0.1708574146,
0.2270597517,
-0.0726940259,
0.0169401169,
0.0890948027,
-0.0511586778,
0.0289125368,
-0.1342879534,
-0.0268505439,
0.0258781388,
0.4235849679,
0.0438393354,
-0.1089970022,
0.1734330058,
-0.0720521808,
0.1587712169,
0.0192032531,
-0.0871140212,
0.0364307016,
-0.0771666914,
-0.3994717598,
0.245752424,
0.1586455256,
0.0353570357,
0.1561482847,
0.309666872,
0.1416005194,
0.1711181849,
0.2627331316,
-0.1946191937,
0.0222405605,
-0.0450105816,
-0.0230080839,
0.1137784496,
-0.1154773235,
-0.142889291,
-0.2855249643,
-0.0421551391,
0.0820327625,
-0.0846407637,
-0.0614053197,
-0.3466532528,
-0.5324674845,
-0.1312477589,
-0.617423296,
-0.0382482186,
-0.1032449305,
-0.2420075387,
0.0315121375,
0.0466376245,
0.0030834973,
0.05545941,
-0.1959473193,
0.2636882961,
-0.2539595366,
0.2536250055,
-0.2014013529,
0.117072016,
0.0113310069,
0.0463703237,
0.1688545495,
0.2607700229,
0.3691004217,
-0.1120980605,
0.13260445,
-0.54004848,
0.0408737957,
-0.0130174104,
0.3866235912,
-0.0338756852,
-0.2162377238,
0.4594660997,
0.1832924485,
0.0527516268,
0.0272366703,
-0.1683707833,
-0.0866199136,
-0.0670037121,
-0.1649510711,
-0.0319065005,
0.0660421774,
-0.2619407177,
-0.5064026117,
-0.1169008762,
-0.0841071606,
-0.2575371861,
-0.0112296473,
-0.0393467396,
-0.1664271653,
-0.1902395487,
-0.22383371,
-0.0161361545,
-0.2360785007,
-0.4750449061,
0.4797660112,
0.07755889,
-0.2248680592,
0.1456605643,
0.2234229296,
-0.4729543328,
0.1071244478,
-0.3786622584,
0.0849280506,
-0.1833435148,
-0.1696144491,
0.1855351627,
0.2796225846,
-0.0561239123,
-0.1013823003,
-0.0752926618,
-0.0664280131,
-0.1520848423,
-0.1004991159,
0.0640999526,
0.3907660246,
0.087131843,
0.0543972328,
-0.0144240186,
0.9082938433,
0.2233340889,
0.0690707117,
0.0383335017,
0.2785711884,
-0.0878228843,
0.2137204409,
-0.2204850018,
0.0769745111,
-0.0739088506,
0.133490622,
0.4275665283,
0.2644350827,
-0.0535895899,
-0.0824765638,
-0.0538627282,
0.0244772658,
-0.1276632249,
-0.3043905199,
0.0564099178,
0.1423767507,
0.3426668644,
-0.1209505126,
-0.1897438169,
-0.3379051089,
0.4688176811,
0.6513609886,
-0.0609644651,
0.0805081055,
0.080833137,
-0.0314891189,
-0.2758722901,
-0.0102399215,
0.1072124317,
0.3037171364,
-0.0362745449,
0.242235437,
0.050625205,
0.0627504736,
0.2902598381,
-0.0846781209,
0.3149999976,
-0.2368749827,
-0.3210115433,
0.1225267574,
-0.2114393115,
-0.0314981677,
-0.4593895078,
0.1168688908,
0.2564662099,
0.0187604576,
-0.0877768248,
0.1811020821,
0.0854196697,
-0.3570490777,
-0.3601966798,
0.3576745093,
-0.1333043426,
0.1113695055,
-0.2920578718,
-0.1578163356,
-0.1651881337,
-0.1639720649,
0.0473702848,
0.0987864658,
0.2592986226,
0.4022814631,
-0.2731803656,
0.0022827722,
0.2368916571,
0.4151032567,
0.1408009827,
0.5216084719,
0.2833669186,
0.3856369853,
0.2477418184,
0.1359353811,
-0.3158936501,
-0.218878597,
0.1133714691,
0.3797289133,
0.002856385,
0.2040543705,
0.185921669,
0.2156557739,
-0.4926128089,
-0.0066149384,
-0.0361998491,
-0.0276057534,
-0.1115591601,
-0.296965152,
0.4388793409,
0.2685515583,
0.0987458229,
0.1024792343,
-0.0013481565,
0.2189722359,
-0.4021958709,
0.1367770135,
1.0715600252,
-0.3632162213,
0.109679997,
-0.3807943165,
-0.4657892883,
0.1052367166,
-0.4329254925,
-0.2255307734,
-0.2089775056,
-0.4920126796,
-0.1338612735,
-0.1546848416,
0.1261089146,
0.1819251925,
-0.0065343454,
0.1903233826,
0.3940102458,
0.2490643263,
0.0605359524,
0.368044883,
-0.1053002849,
-0.151558578,
0.1496789753,
0.096427463,
-0.1311602294,
0.2217652202,
-0.0750765651,
-0.2779451609,
0.08569704,
0.1454675496,
-0.2411843836,
0.0366278887,
-0.1807598919,
-0.0174848884,
-0.3960856795,
0.0377733558,
0.4888108969,
-0.6864606738,
0.0188972559,
0.3415190578,
-0.4918783307,
0.0173329562,
-0.2555530369,
0.0967731029,
-0.2075226605,
0.1782935858,
0.138438493,
-0.1854406297,
-0.1331969351,
-0.2244443446,
0.1818854064,
-0.0898345932,
-0.1923562586,
-0.4273498654,
0.2352770865,
0.1829025149,
0.0969033986,
0.2842638493,
0.0437415391,
-0.0054007396,
0.088881731,
0.1895173937,
-0.3294633627,
0.4207531512,
0.5548226237,
-0.1979377568,
-0.1259867847,
0.6357402802,
0.2047565579,
0.1227295622,
0.3180734813,
0.2730902433,
-0.309709698,
-0.1819448024,
0.0837130025,
-0.1224999428,
-0.088984549,
0.047645282,
-0.2326234281,
0.2021254152,
0.0015359074,
0.304435432,
0.2734974623,
-0.4514812231,
-0.350833714,
-0.3322855234,
-0.0971915498,
0.0723167211,
-0.25902161,
0.117458269,
-0.148048833,
0.2740130723,
0.0842932984,
-0.0004571471,
-0.2636135817,
0.1694959104,
0.2028913498,
-0.0421102196,
0.1046394706,
-0.0314336866,
-0.1653340459,
0.2372362316,
-0.0471887663,
-0.0556553043,
-0.197070539,
-0.1565997005,
0.0797372758,
0.2054679692,
0.0912317932,
-0.2244701982,
0.0881511271,
-0.4742830098,
-0.211529851,
0.2286707312,
-0.1130282953,
0.0611490309,
0.0590619408,
0.017581027,
0.3800585866,
-0.3338487446,
-0.402654469,
0.5654031634,
0.3676569462,
0.3162977695,
-0.1362123638,
-0.3233816922,
0.1768165827,
0.00321839,
-0.2345843613,
-0.0586540513,
0.1890222877,
0.1380799115,
0.5165478587,
-0.1515651494,
0.238173902,
0.4325606525,
0.5199491978,
0.0538493991,
0.1866321266,
-0.1702865362,
0.2538137138,
0.0755517483,
-0.259378016,
-0.194893837,
0.2842745185,
0.2228813171,
-0.0430592261,
-0.0183990151,
0.243673265,
0.0698681176,
-0.2601145208,
-0.1258621812,
0.1752543896,
0.1359933317,
-0.2857546508,
0.2617632151,
0.3761046231,
0.1796095967,
0.0130056906,
0.2156086564,
-0.0912376195,
0.1073672697,
0.2476565242,
0.3638362885,
0.2184050977,
0.1920842379,
0.0094170142,
-0.2894503474,
0.0534278937,
0.3052601218,
0.247010231,
0.0937667936,
0.0228028074,
-0.1456732303,
0.1766118854,
-0.4062262475,
-0.2408547252,
0.260761112,
0.0568460375,
-0.0434248596,
0.1096574217,
-0.0884290338,
-0.0807546377,
0.1009492129,
-0.0840277225,
0.0368064307,
0.2215910554,
0.1124881208,
0.2072283328,
-0.1745324731,
0.0337691456,
0.1124483794,
0.118323341,
-0.1860306561,
-0.3728147745,
-0.3661416769,
0.1247575879,
0.4770666063,
0.4669602811,
-0.2627885342,
0.0623644702,
-0.0314195864,
-0.1631103307,
-0.1262388527,
-0.0515968874,
0.154684484,
0.2232966125,
-0.1973984838,
-0.3956587911,
0.3107919693,
0.0714398697,
-0.0971719176,
-0.0413183272,
-0.2396619618,
-0.4381753802,
-0.4239857197,
0.2285811007,
0.229733929,
0.0753642619,
-0.1485551,
-0.1019668356,
-0.0453389324,
-0.1195877492,
0.3612488508,
-0.3295561969,
0.0384026654,
0.0436412543,
0.051850386,
-0.3699964285,
0.2957006693,
0.4970720708,
0.308634609,
-0.260517031,
-0.2158728689,
-0.255479604,
-0.0673460662,
-0.3787691593,
0.0402851924,
0.0965735018,
-0.0228542276,
0.3310456276,
0.1006317139,
0.2956649661,
0.4210749865,
-0.0227038227,
0.1257089376,
-0.1323170066,
-0.082852304,
-0.1275245249,
0.218967855,
-0.2469507456,
-0.3628220856,
0.3034453392,
0.1365979016,
-0.0504154861,
0.0687017217,
0.092595309,
0.2002923787,
0.0921233743,
0.033706788,
-0.2981727123,
0.2842431664,
0.0369537026,
0.1631712019,
0.075077787,
-0.0556000583,
-0.0350966826,
-0.2652784884,
-0.1652206331,
-0.200360924,
-0.0574179552,
0.068142876,
-0.0099162795,
0.0072869249,
-0.1741357893,
0.1290447563,
-0.3380171061,
0.1175005734,
-0.5509544611,
-0.0493197814,
0.1299544573,
-0.0076272888,
-0.0067693032,
-0.2790981531,
-0.0541579276,
0.0086200833,
0.3676298857,
-0.0132628046,
-0.121207431,
-0.4085980654,
0.0892245397,
-0.1533353925,
0.3862651289,
-0.0755103678,
-0.0240540281,
0.2824201584,
-0.1754868031,
0.4065972269,
0.4313980937,
0.1577371061,
-0.0796528459,
-0.1086574495,
0.1649681032,
-0.0654717982,
0.0759766027,
-0.4208133221,
0.1529711038
] |
https://github.com/huggingface/datasets/issues/1892 | request to mirror wmt datasets, as they are really slow to download | @lhoestq better to use our new git-based system than just raw S3, no? (that way we have built-in CDN etc.) | Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you! | 20 | request to mirror wmt datasets, as they are really slow to download
Would it be possible to mirror the wmt data files under hf? Some of them take hours to download and not because of the local speed. They are all quite small datasets, just extremely slow to download.
Thank you!
@lhoestq better to use our new git-based system than just raw S3, no? (that way we have built-in CDN etc.) | [
-0.2889100015,
-0.3306779265,
0.0120118596,
0.1093230769,
0.0525673293,
-0.0814259574,
0.1615227014,
0.4271322191,
0.0938708335,
-0.0190337226,
-0.3289594054,
-0.0868709683,
0.0092278495,
0.3145239353,
-0.2300495654,
-0.1343409121,
0.0110247098,
0.1879600286,
-0.566865325,
-0.1316291839,
0.1182781681,
-0.1686477512,
-0.0159089416,
-0.056058459,
0.0335923471,
0.2922866344,
-0.1292633116,
-0.0466059074,
-0.4010478556,
0.2060647607,
0.0686515272,
0.4430898428,
0.2044609636,
0.4233406782,
-0.0001147051,
-0.1628361642,
0.255114615,
0.1594581455,
-0.2941844165,
0.1020640731,
-0.5625472069,
-0.2940484285,
-0.1721083522,
0.0206452459,
-0.2282909006,
0.0565153658,
-0.0642556846,
-0.1407652646,
0.2692017853,
-0.2676654458,
0.1218053102,
0.3570900559,
-0.5695301294,
-0.0760532171,
0.4883638024,
0.3561577499,
-0.3749563694,
0.374475956,
0.4599543512,
0.0103054717,
0.2937454581,
0.3776448369,
0.0498917103,
0.4903296828,
0.5588127375,
-0.1442912668,
0.108404994,
0.101066485,
-0.2936746478,
0.6387221813,
0.8014045954,
-0.0934529006,
-0.4955382645,
0.0299212933,
-0.165800944,
0.1371796727,
-0.1666873693,
0.0896816105,
-0.0822559297,
0.2906906307,
-0.5258617997,
-0.4199447334,
-0.0245536417,
0.0971606076,
-0.1622618139,
0.3614452183,
-0.0499788187,
-0.0575058088,
-0.0215987191,
-0.0868923962,
0.2871053517,
-0.5339574218,
0.0277756639,
0.0092748106,
0.0158332884,
-0.5079793334,
-0.5281719565,
-0.1124781668,
0.1427507102,
0.0715579167,
-0.2179972231,
-0.061571721,
0.0635953322,
0.0075882529,
0.1665518433,
0.0226129405,
-0.3363719881,
-0.1470313668,
0.242810145,
-0.0826787651,
-0.062636435,
-0.1126048565,
-0.1358263493,
-0.058529254,
-0.4983780384,
-0.0591055378,
-0.0641232058,
-0.3968081772,
-0.0560138151,
-0.3704044223,
0.0905916765,
-0.2271957397,
-0.242527172,
-0.0228241943,
-0.1544203162,
0.1946365535,
0.2308921814,
0.0609243289,
-0.1811843812,
-0.0949188918,
-0.0341664739,
-0.353812784,
0.0227360278,
0.3785382509,
0.1024878472,
0.0440882668,
0.0467231348,
0.0125859454,
0.2723510563,
0.3997437358,
0.1198744103,
0.1355111301,
-0.0892270803,
0.1303316653,
-0.3001797795,
0.5513953567,
-0.0822591633,
0.5401504636,
-0.247490868,
0.0239215419,
-0.3828433752,
-0.1484459788,
-0.0855851918,
0.143490538,
-0.1461862624,
-0.2256683111,
-0.4768675566,
-0.1261457503,
-0.370220989,
0.0962531269,
-0.0732422397,
-0.0139552783,
-0.1799244285,
0.1466063261,
0.0096518397,
0.2471742183,
-0.2330055088,
-0.0535835326,
0.0196595527,
-0.0174877401,
0.545589149,
0.604721725,
-0.2435895503,
0.0085152946,
-0.1989865452,
-0.1130556464,
0.0107876286,
-0.2240660489,
-0.065666765,
0.5325779319,
-0.2225830406,
-0.0993729234,
0.1051367223,
0.1408649683,
0.1468099356,
-0.2809916437,
-0.2059144378,
0.5454109311,
-0.0261677243,
0.0190586224,
-0.2581398785,
-0.5230137706,
0.1062315702,
0.2587065697,
0.0255729519,
0.2686309814,
0.0157457069,
0.1611359715,
0.4051990807,
-0.0458651036,
-0.0070542283,
-0.006418216,
0.3837782145,
0.3038676679,
-0.0046904646,
-0.0034462884,
0.108939901,
0.06436342,
0.0773328319,
0.0847626776,
0.4987875819,
-0.1055862606,
0.0438250862,
-0.1553088278,
0.1709592789,
0.3255120814,
0.0487536415,
-0.1046047881,
-0.1730401218,
-0.3651241064,
-0.0260843039,
0.0542200804,
-0.0070964843,
-0.0475675315,
0.5461928844,
0.345230937,
0.0251750033,
0.3276381791,
0.2912020385,
-0.2134887874,
0.1773378998,
0.0235404931,
-0.0474238768,
-0.2182144076,
-0.1485567242,
0.4687214494,
0.5035500526,
0.4017014503,
0.3517150283,
-0.4376821518,
0.4014913738,
0.0752759129,
0.0569216199,
-0.2025917619,
-0.280410111,
0.156046018,
-0.0736739263,
0.0032423437,
-0.0218049586,
-0.187758863,
0.0269899517,
-0.1730398685,
-0.0699176192,
0.1036521345,
0.3284130394,
0.0244789161,
-0.2184793651,
0.203052178,
0.0279875323,
0.157550171,
0.2100996375,
-0.1309794486,
0.0774589106,
-0.0152056385,
-0.2841054499,
0.1486785561,
0.1963618994,
0.1138750762,
0.1733242273,
0.3031971455,
0.1898448914,
0.1547533274,
0.4193893075,
-0.2129914314,
0.061586991,
-0.0796333998,
0.2978202105,
0.2055938095,
-0.0357464068,
-0.1473862231,
-0.145506978,
0.0025364757,
0.0777264908,
-0.1255670339,
-0.0882336944,
-0.2426349223,
-0.3631208837,
-0.1498049498,
-0.4077998996,
-0.0824508667,
-0.0949651897,
-0.1468748152,
0.0854019523,
0.0987484232,
-0.0776861906,
-0.0030243061,
-0.018144168,
0.3205001056,
-0.2589110434,
0.0728622824,
-0.29400599,
0.0344226211,
0.0747305825,
0.0384314843,
0.1803444326,
-0.1175255477,
0.3160450757,
-0.1246630698,
0.0540365279,
-0.5954364538,
-0.0566792823,
0.0238986555,
0.3337823153,
0.1331884414,
-0.1336392611,
0.4270214438,
0.1770922244,
0.0450292826,
0.0378955528,
-0.1569054425,
-0.2003162503,
-0.1181854606,
-0.1908141971,
-0.0698715895,
-0.0480866134,
-0.1044222862,
-0.4379574656,
-0.1385015547,
-0.0337488204,
-0.3620451689,
-0.0011296198,
0.0098888502,
-0.0695374459,
-0.2653435767,
-0.2764092386,
-0.0574301369,
-0.2166294754,
-0.5555562973,
0.3413763046,
0.0805289596,
-0.1662790626,
0.0229266807,
0.3018040657,
-0.3650654554,
0.0230334736,
-0.4307639003,
-0.0529677793,
-0.0700541809,
-0.0424138308,
0.2023964822,
0.2894767821,
0.0451649241,
-0.1432103366,
-0.0733307973,
-0.1906095743,
-0.0912857577,
-0.2161954343,
0.1766274273,
0.3960529268,
0.0840988755,
0.2189152092,
0.123255372,
1.0079666376,
0.2507514954,
0.0240487941,
0.0811173171,
0.3469983041,
-0.0633272156,
0.1654547453,
-0.1328160763,
0.049280934,
-0.1263699532,
0.1854526252,
0.4246994853,
0.1636005938,
-0.3049124181,
-0.1610037088,
-0.0788752139,
-0.018702656,
-0.09023498,
-0.2591296434,
0.0133084711,
0.1059594005,
0.27032426,
-0.1721442491,
-0.0569165722,
-0.2881121337,
0.5134932995,
0.5014875531,
-0.0035768636,
0.0948698372,
0.1205561161,
-0.1066679806,
-0.4163665473,
-0.1466147006,
0.0493516698,
0.2472967803,
0.0047215298,
0.2376878262,
0.050456129,
-0.0322896577,
0.5050347447,
-0.2438181192,
0.2876740694,
-0.2740412951,
-0.4379414916,
0.0138901919,
-0.233836621,
-0.1396898478,
-0.4957697093,
0.1366994381,
0.3703024387,
-0.0324070677,
-0.1467791498,
0.0312685408,
-0.1002137512,
-0.239588812,
-0.430783689,
0.2779766917,
-0.112650454,
-0.0150615238,
-0.2945275307,
-0.0675636828,
-0.1654195189,
-0.1310253441,
0.079721272,
0.0226145554,
0.410975337,
0.2506613135,
-0.1289608777,
-0.1178852916,
0.2919683754,
0.3045598865,
0.0532694608,
0.3760148883,
0.2645437419,
0.5690960884,
0.340379715,
0.0349322297,
-0.2293999046,
-0.1585588604,
0.1091763079,
0.3894104362,
0.1780186892,
0.2084597945,
0.1500373483,
0.2509877384,
-0.5666326284,
-0.0764778331,
-0.0023037158,
-0.0262864176,
0.0252303071,
-0.121227704,
0.359023869,
0.1255482286,
0.0238725394,
-0.148084119,
0.1376605928,
0.1760544777,
-0.3433087468,
0.2689768076,
1.1076864004,
-0.3656045794,
0.2305892706,
-0.3309263587,
-0.6006499529,
0.1223599166,
-0.3119622767,
-0.2270286083,
-0.2131427824,
-0.3130867481,
-0.0980491936,
-0.149582088,
-0.0412906893,
0.1944206655,
-0.0749909878,
0.2023568302,
0.4440808296,
0.1437160373,
0.0984944403,
0.4232387245,
-0.0224347189,
-0.178252846,
0.1104896516,
0.1282366216,
-0.2176442593,
0.3711438775,
-0.1129857451,
-0.2857254744,
0.0327132791,
0.2146441042,
-0.2116669565,
0.1427697837,
0.0150580183,
-0.1655503511,
-0.3007995188,
-0.0127583295,
0.4979438186,
-0.6503031254,
0.0528462343,
0.3720951974,
-0.4815838039,
0.0828290582,
-0.1513361782,
0.1640636772,
-0.2222637832,
0.0016883425,
0.0279360097,
-0.2307056487,
-0.1817236245,
-0.2478945851,
0.2220162749,
-0.2231400907,
-0.0563503653,
-0.371516645,
0.1247260869,
0.2312970906,
0.0375287272,
0.2623558044,
0.0632218644,
0.023548089,
0.0910903886,
0.2097051144,
-0.3416233063,
0.3736175299,
0.4086954594,
-0.183056742,
-0.1204474419,
0.6932261586,
0.0599627383,
0.1400991231,
0.3065082729,
0.3477090597,
-0.296659112,
-0.1432988942,
0.0931541696,
-0.0949292108,
-0.2008300126,
0.0257395059,
-0.138359949,
-0.0713489503,
0.0436998941,
0.2620230913,
0.3745441437,
-0.4309702218,
-0.2084339261,
-0.257532835,
0.0982460231,
-0.01947283,
-0.269495666,
0.252156496,
-0.0434697457,
-0.0051667783,
0.0432021543,
-0.0195725299,
-0.2829039395,
0.1779471189,
0.0899411887,
0.1331411004,
-0.0084639378,
-0.1326051056,
-0.1195355952,
0.0352882259,
-0.0872822404,
-0.1757518798,
-0.1880128831,
-0.1848128438,
0.054184854,
0.2005055249,
0.178055793,
-0.212169677,
0.2198636681,
-0.355338335,
-0.2397075593,
0.2403866202,
-0.0003092289,
0.0835644305,
0.0511090681,
0.1653578132,
0.3959064484,
-0.469707936,
-0.3244508505,
0.5378161669,
0.3684097528,
0.2961510718,
-0.2040383071,
-0.2357988358,
0.3141791224,
-0.0028280318,
-0.0207211971,
-0.0554047525,
0.2595159113,
0.290294528,
0.5963609815,
-0.195512265,
0.2901700437,
0.4197299182,
0.5349559188,
-0.1353771091,
0.0373883061,
-0.1808125675,
0.3128831089,
0.1392180175,
-0.1906393617,
-0.336089313,
0.0202597752,
0.206243962,
0.0407680124,
0.0474688821,
0.2516184151,
0.1388721764,
-0.1557545215,
-0.1798748672,
0.1865205616,
0.2619704008,
-0.2746183574,
0.191422224,
0.2551442385,
0.0179770291,
-0.0672472343,
0.1334265172,
-0.1012008265,
0.0294645317,
0.2103697658,
0.1706474572,
0.1291403472,
0.3148682117,
0.0914533883,
-0.1671329141,
-0.0086859046,
0.1696700752,
0.3133715391,
0.1615932286,
0.009157991,
-0.4201601744,
0.289375037,
-0.4594334066,
-0.1767068505,
0.3615793586,
0.1225369126,
-0.1365914941,
-0.1155403256,
0.0052831471,
-0.000112772,
0.2047948241,
-0.2032440454,
0.0247589108,
0.1301854998,
0.2216450572,
0.2809832096,
-0.032511808,
0.0774458274,
0.1081179082,
0.2180998176,
-0.1721067727,
-0.1806696355,
-0.3341340423,
0.0832063407,
0.3756328821,
0.5546930432,
-0.1806424558,
0.1371493936,
-0.0031318441,
-0.0899247825,
-0.0146250837,
-0.0397590548,
0.0984541476,
0.1521315128,
-0.0985548869,
-0.2382950634,
0.2103475779,
0.066059269,
-0.1263801455,
-0.0434928052,
-0.3883407414,
-0.4532510638,
-0.2978821397,
0.1055049449,
0.2279826552,
0.0872059613,
-0.0487629771,
0.1077937409,
-0.184179455,
0.054880932,
0.3799108863,
-0.3713880777,
0.0885181949,
0.0712900311,
0.0576952547,
-0.380371213,
0.4389808178,
0.4249641001,
0.3350913525,
-0.2078576237,
-0.1208074093,
-0.0419808552,
-0.067568779,
-0.3867938221,
0.0545179024,
0.0605668426,
0.1189956814,
0.2939838171,
0.2040811926,
0.1446166933,
0.3913865983,
-0.151839152,
0.2199248523,
-0.1296269298,
-0.0603257529,
-0.0497742631,
0.2328107357,
-0.1578113735,
-0.355044961,
0.3176901937,
0.0918006524,
-0.0439622328,
0.0779463872,
0.1598031819,
0.2010507286,
0.1578119993,
0.0118869534,
-0.2411861867,
0.2156128138,
0.0102368221,
0.085598439,
0.1212219447,
-0.0877587646,
-0.1366538256,
-0.2215113789,
-0.2333173752,
-0.1392721534,
-0.041770272,
0.122054249,
-0.0421438366,
0.0073031634,
-0.1979624927,
0.1851389408,
-0.2294758558,
0.0798229799,
-0.5772966146,
0.0448306426,
0.1001191586,
-0.0192215052,
-0.0352137536,
-0.2984086871,
-0.2601256371,
0.000244081,
0.4037050009,
-0.1191802099,
-0.1109490097,
-0.3758730292,
0.0425347649,
-0.1837889552,
0.2944349647,
-0.1605953425,
0.0931733251,
0.1711372584,
-0.0880124047,
0.3685886264,
0.5453276038,
0.2308082283,
-0.0512092151,
-0.1287274659,
0.3854855299,
-0.0311309863,
0.0547294095,
-0.518409133,
0.0736174136
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | I started working on this. My idea is to first add the pyarrow Table wrappers InMemoryTable and MemoryMappedTable that both implement what's necessary regarding copy/pickle. Then have another wrapper that takes the concatenation of InMemoryTable/MemoryMappedTable objects.
What's important here is that concatenating two tables into one doesn't double the memory used (`total_allocated_bytes()` stays the same). | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 55 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
I started working on this. My idea is to first add the pyarrow Table wrappers InMemoryTable and MemoryMappedTable that both implement what's necessary regarding copy/pickle. Then have another wrapper that takes the concatenation of InMemoryTable/MemoryMappedTable objects.
What's important here is that concatenating two tables into one doesn't double the memory used (`total_allocated_bytes()` stays the same). | [
-0.3632808626,
0.171248883,
-0.0166257881,
0.1944021881,
0.099194482,
0.0310439691,
-0.2240344435,
0.2879285216,
-0.2779928446,
0.1697992235,
-0.0141434092,
0.6244734526,
-0.0076513197,
0.3612303734,
0.3155476749,
0.0208592787,
0.1660715491,
0.3219618499,
-0.4445088506,
0.2120196819,
-0.2109624445,
-0.2650684118,
-0.0310650542,
-0.3578846455,
-0.0414914936,
0.1385568529,
-0.3737603724,
0.0541285351,
-0.4198371172,
-0.4219734967,
0.036391966,
0.2514678836,
0.0210893042,
-0.0215109326,
-0.0001128412,
-0.0243959427,
-0.0898243636,
-0.1595998108,
-0.4942533672,
-0.0600704253,
-0.1329776496,
-0.3831797838,
0.0522672683,
-0.02179344,
0.2780460119,
-0.2891720533,
0.1262462586,
0.096600607,
-0.042054072,
-0.028358357,
0.1490675658,
0.2059525698,
0.3069912791,
-0.1393375546,
0.4071867764,
0.413789928,
-0.2051129639,
0.0767856836,
0.5211781859,
-0.2946155965,
0.1044792384,
-0.2324478328,
-0.1127732396,
-0.0687843636,
0.4034474492,
0.2039737999,
-0.2281583697,
-0.0668901801,
-0.0968059152,
0.2737795711,
0.4659917653,
-0.5767401457,
-0.3419586718,
-0.3257187009,
0.2530626953,
-0.3041680455,
0.083387211,
0.2824950814,
-0.170952484,
0.1324289143,
0.0093720593,
-0.31812644,
-0.2481419146,
0.2170401216,
0.316129446,
0.1951753497,
0.3829263151,
0.0876911208,
0.1879308224,
-0.0128816487,
0.1668946296,
-0.1282310039,
-0.2355228961,
0.1022402644,
-0.2337992787,
-0.0629914925,
0.1070390642,
-0.0313677117,
0.6003128886,
0.046158962,
0.3129304349,
0.0161991734,
-0.2231710404,
0.2855438292,
0.1177800372,
0.1156648174,
-0.3405876756,
0.0699795187,
-0.0161277521,
-0.1747213751,
0.2247428596,
0.0428508744,
-0.112451151,
0.0882591903,
0.1821429282,
-0.2519494593,
-0.0506822765,
0.1571141332,
0.0636107549,
-0.2553881109,
-0.072543852,
-0.1010020301,
-0.0092343539,
0.1719470769,
0.103839919,
-0.0815278441,
0.0429694578,
0.3130612969,
0.2300461829,
-0.1362038553,
0.0179161131,
0.0584183224,
-0.2758908868,
0.4048902988,
0.4093624353,
0.1198431253,
-0.1895425916,
0.2562189698,
0.0313366055,
0.3089536428,
0.0265611336,
-0.1521232873,
0.3627640307,
0.173697263,
0.0343496427,
-0.5147957802,
-0.1038980186,
-0.0499293692,
-0.3467099071,
0.582654655,
-0.0074535236,
-0.3105190694,
-0.2788462639,
0.1124337837,
-0.1399571151,
-0.1520404369,
-0.2362211943,
0.5187286139,
0.2234692723,
-0.3194288611,
0.0532799736,
0.1524676085,
-0.0710658133,
-0.487328887,
0.1770168841,
0.2461444288,
-0.6097145677,
0.0000971332,
0.1346833408,
-0.0355319977,
-0.0081759095,
0.1414320767,
-0.3650354743,
0.0534462482,
-0.3295628428,
0.3390139043,
0.1483227015,
0.0514437258,
-0.3597129583,
-0.1292902827,
-0.1312127709,
-0.0572624914,
0.4833110869,
0.4032013118,
0.2118996382,
0.1453686655,
0.2845697105,
0.353920728,
-0.2226087153,
-0.1017393321,
-0.0969371051,
-0.5097311735,
0.3827191591,
0.0436720848,
-0.1091199964,
0.05574264,
0.1622283608,
-0.5239160657,
0.2628687024,
-0.2847260237,
0.2222063988,
-0.1562276185,
0.4910983145,
-0.0552519597,
-0.1267485619,
-0.1807767004,
-0.3993272483,
0.1532473862,
-0.2859074473,
0.1690540165,
-0.2946664989,
-0.3228558898,
0.2363320738,
0.1322882175,
0.0230405107,
0.1099825799,
0.1279898286,
0.032816086,
0.0467835888,
-0.3443710804,
-0.1753305793,
0.0709775388,
0.1972886622,
0.0502543114,
-0.0657286048,
0.422205925,
-0.0130109712,
-0.2735291421,
0.0111991838,
0.138059631,
-0.0577249303,
0.1894704252,
0.1517650485,
0.4081635475,
-0.1975196749,
0.1372385621,
0.0834626853,
0.3099539876,
0.1968974322,
0.2757865787,
0.059356831,
-0.5478474498,
0.1720450372,
-0.1497484595,
-0.1026827022,
0.2931603789,
0.1232742667,
0.1167148575,
0.205556497,
-0.1688680351,
0.1156383976,
-0.0519310869,
0.0044766963,
-0.1561865658,
-0.0496025607,
0.2407089174,
0.0706176907,
0.3167554438,
-0.3992089629,
0.231384635,
0.2559007406,
-0.0257453211,
0.2315573692,
0.1496711671,
-0.1381302774,
-0.2459968925,
0.0611863285,
0.0652553886,
0.6371229887,
0.4014917612,
-0.0435562357,
0.0230939109,
-0.0723099038,
-0.1035804152,
0.1566608697,
0.0239404161,
0.1919995844,
0.2960339487,
0.514937222,
0.019223202,
-0.131281361,
0.006839864,
0.2437661588,
-0.0876246393,
-0.1007828638,
-0.233731389,
-0.3976405859,
-0.0855065808,
-0.2734544873,
-0.2590796947,
-0.149813354,
-0.0000417568,
0.261887908,
0.3025760949,
-0.1651453376,
0.1315742433,
0.1604928076,
0.502620995,
-0.4176717997,
-0.4203875661,
-0.0996634364,
-0.1015156209,
0.0813330859,
0.0966112316,
0.3859768212,
-0.0796461403,
0.4720605016,
0.4112361968,
-0.0381250307,
-0.3536837697,
-0.3720276952,
0.1493056715,
-0.0701526031,
0.133490935,
0.1294971406,
-0.0610316396,
0.122348249,
-0.450451076,
0.0899822265,
0.2273864448,
-0.3001633584,
0.2916742563,
-0.0307429414,
-0.158965528,
-0.3208266497,
-0.1860149503,
-0.2604091465,
-0.4231290817,
0.5149371624,
-0.0762138665,
0.1177769229,
0.0217033736,
0.2126740068,
-0.2492763698,
0.0497664027,
0.1925605834,
-0.0910726041,
-0.0414257199,
0.2398656756,
-0.120526731,
-0.0846001804,
-0.1582556963,
-0.188627094,
0.132732302,
0.4877554476,
-0.2162234783,
-0.1652865857,
-0.0899312049,
0.4463779628,
-0.0780020133,
0.062283434,
0.628362,
0.497443229,
-0.1046441048,
-0.0683522895,
-0.1453668624,
0.193040669,
0.2069642395,
-0.0469184518,
0.0595551096,
-0.1752708554,
0.0368187353,
0.3763083816,
-0.1202296987,
0.1142817289,
0.5036055446,
-0.0323612802,
0.3150960207,
-0.1933970451,
0.0417080373,
-0.1953703463,
-0.365473479,
-0.0698203668,
-0.0185966082,
-0.1062516123,
-0.2606144845,
-0.0054412708,
0.1965700537,
-0.2024163902,
-0.2465240657,
0.3875081241,
-0.0537444577,
0.2299596518,
0.0281375796,
-0.4425176382,
-0.3240292668,
0.0544876978,
-0.0609602965,
0.0671938881,
0.1160010099,
-0.1972330511,
-0.3786021173,
-0.4912510216,
-0.3914144635,
0.0640265271,
0.2047525048,
0.1848561764,
0.1703855693,
-0.416223228,
-0.0712062418,
-0.0527943075,
0.7359609008,
0.0948456302,
-0.4596620798,
0.0931819975,
0.0853066444,
0.1662998199,
0.0113303959,
0.1150951684,
-0.0030770972,
-0.2291936874,
0.1979465336,
-0.4122434258,
0.1610731035,
0.1100376099,
0.424136579,
0.0007144287,
-0.2770531774,
-0.0968424827,
-0.0321941376,
-0.4565472305,
-0.376570344,
0.0682293698,
0.1274804473,
0.1043494567,
-0.008660607,
-0.2238842696,
-0.0173048712,
0.046239946,
-0.0654449537,
0.3290602267,
-0.0729599893,
0.3960425854,
-0.3747378588,
-0.0979486257,
0.8761479855,
0.5169813633,
-0.2389470786,
-0.1651126444,
-0.0853902549,
-0.1969739497,
0.2817300856,
0.0562425517,
0.1388365626,
0.5503252745,
-0.0712953061,
-0.0063712336,
-0.4769187272,
-0.1635629982,
-0.1850142777,
-0.2612450719,
-0.2375257909,
-0.269775629,
0.4483608305,
0.250508666,
-0.1940822899,
0.3499962091,
0.0889330357,
-0.4363666773,
0.3113607466,
0.0886838511,
1.159001708,
0.0148964319,
0.3800123036,
0.1056020483,
-0.3035753667,
-0.0273734648,
-0.1147416905,
0.3279805183,
-0.2095257491,
-0.1091614589,
-0.053995464,
-0.2491600513,
-0.0816702172,
-0.0303637832,
-0.5228950381,
-0.1054130569,
-0.0092079639,
0.0421187952,
0.022153873,
0.1669180244,
-0.2668743134,
-0.3497604728,
-0.0076732766,
0.0527730845,
0.0839634612,
-0.4466321468,
-0.1137775183,
-0.1879320145,
-0.1947298646,
0.0875308216,
-0.2319943905,
-0.1305644214,
-0.3072327077,
0.1639656425,
-0.0106165791,
-0.1232781708,
-0.1258396506,
-0.0341870673,
-0.1937228888,
0.0677937269,
-0.1358164251,
-0.0937972516,
0.0887526646,
0.0209434256,
-0.1650477648,
-0.3840014935,
0.4343763888,
0.0698952898,
-0.1761591285,
0.0155024864,
-0.0875436738,
-0.3180527985,
-0.1743966788,
-0.0645117015,
0.2385872155,
-0.1678341031,
-0.1451399624,
0.3330943286,
0.0488706678,
-0.2631316185,
0.1011423469,
0.1563951671,
-0.0900354609,
0.1994000822,
-0.0967756659,
-0.0139270574,
-0.2291684002,
0.0532118902,
0.3081635535,
0.1532831937,
0.1787964404,
0.0110157281,
0.0155304521,
-0.1112136245,
0.0246140361,
0.1981815845,
-0.2538957596,
-0.0508716889,
0.1319767833,
-0.086240299,
-0.0479140542,
0.1837475598,
0.329377681,
0.3658778071,
-0.153763935,
-0.0920517445,
-0.457609117,
0.4486490786,
0.1664477289,
0.3751330674,
0.1310435981,
0.2656696439,
-0.0705384165,
0.3580205441,
-0.3201298118,
-0.1697636843,
-0.0794120282,
0.1343301237,
-0.0008912832,
0.1290604472,
-0.1456323415,
-0.2106569111,
0.0542132929,
0.135027051,
-0.0068690609,
-0.1512798071,
0.0569041967,
0.1238347068,
0.236327827,
0.2013335228,
-0.3025513887,
-0.2956585884,
0.0827313662,
-0.2491498291,
0.1192169487,
-0.0134782493,
-0.3082535267,
-0.028256312,
0.5552066565,
0.0543295145,
-0.1269166321,
0.3359415531,
0.0900056735,
0.1041657701,
-0.044035662,
0.2061875761,
-0.4008333385,
-0.011780329,
-0.0299180169,
0.3622280657,
0.2243475914,
-0.0661084652,
0.1021793485,
0.2828694284,
-0.3487551808,
-0.1331783533,
0.178449288,
0.261036396,
0.0997984111,
0.2106983066,
0.2864468694,
0.1986654103,
-0.3570128083,
-0.2535360456,
-0.2535092831,
-0.2603914142,
0.2006381899,
0.270041734,
0.1721265465,
0.1027308255,
-0.2891113758,
0.1050475389,
0.0649802983,
0.056005016,
-0.0073391274,
0.1163392663,
-0.0184896141,
-0.0767278373,
0.0518961102,
0.3210087121,
0.2086154968,
-0.0023517455,
-0.1749982387,
0.479883045,
0.1716332585,
-0.2350008488,
0.2683986425,
0.0498673394,
0.1248158962,
0.3608214259,
0.3734307289,
-0.0093327798,
0.2327897549,
0.4574439526,
0.1498829424,
-0.3676361442,
0.2930670083,
0.1533662677,
0.0267949551,
0.0329521224,
-0.0564512759,
-0.2492743284,
-0.2568693161,
-0.1258597374,
0.0693878904,
-0.2811411619,
0.4982887805,
-0.0051178299,
0.1935658157,
-0.2537724376,
0.382632941,
0.6115156412,
-0.0449693613,
-0.4045730233,
-0.1106405854,
0.3154495656,
-0.3348845243,
0.0536023155,
0.152083084,
0.4882125854,
0.2802979946,
0.0297360867,
0.1499828994,
-0.1370533705,
0.0802362636,
-0.1953177005,
-0.1656138599,
0.2606085241,
0.120690763,
0.1192939132,
0.1675412804,
-0.0253525116,
0.1888163537,
0.156555444,
-0.0386967324,
-0.406699419,
-0.1166393682,
0.4306778014,
-0.6115739346,
0.3541142642,
-0.0224138871,
-0.2569141388,
0.0828351602,
0.346321106,
-0.0350434519,
0.0142795816,
0.0515131019,
0.0750958994,
0.321184814,
0.2814216018,
0.2986443043,
0.1896282881,
-0.3491724432,
-0.2060579062,
-0.3519127965,
-0.0636400953,
-0.0258579571,
-0.3163740933,
-0.0514051653,
0.1494464576,
0.2918800414,
0.0563039519,
0.0356113315,
-0.0549108088,
-0.0545144863,
0.2851236761,
-0.2741578817,
-0.1244373173,
0.2524220347,
-0.0996355265,
0.0223419294,
-0.2017429322,
0.0379132293,
0.1373583823,
0.0389980003,
-0.3926255703,
0.0246426985,
0.2244709134,
0.2718216479,
0.5434370637,
0.0922375172,
0.304764241,
-0.2285117358,
0.0017849207,
0.0915166736,
-0.1626477838,
-0.374951303,
-0.0324522369,
-0.0192975327,
-0.0970245302,
-0.3878605664,
-0.0701936558,
-0.1041528434,
-0.0409862101,
-0.1244586557,
0.0511251166,
-0.2187105417,
0.0896853507,
0.1657112986,
-0.0279464889,
-0.0913174376,
0.1592526436,
0.1434433162,
0.0932312906,
-0.0916669369,
-0.2565833926,
0.0419634804,
0.2645943761,
-0.0430805199,
-0.0967156217,
-0.0352425501,
-0.1707375497,
0.1734446883,
-0.8052769899,
-0.2288546264,
0.2185245454,
0.0149885826,
-0.0133389905,
0.4682540298,
0.0607344061,
0.0070893541,
-0.3281798065,
0.362362951,
-0.1387653798,
-0.348947823,
-0.169995144,
-0.5305550098
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | Hi @lhoestq @albertvillanova,
I checked the linked issues and PR, this seems like a great idea. Would you mind elaborating on the in-memory and memory-mapped datasets?
Based on my understanding, it is something like this, please correct me if I am wrong:
1. For in-memory datasets, we don't have any dataset files so the entire dataset is pickled to the cache during loading, and then whenever required it is unpickled .
2. For on-disk/memory-mapped datasets, we have the data files provided, so they can be re-loaded from the paths, and only the file-paths are stored while pickling.
If this is correct, will the feature also handle pickling/unpickling of a concatenated dataset? Will this be cached?
This also leads me to ask whether datasets are chunked during pickling?
Thanks,
Gunjan | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 129 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
Hi @lhoestq @albertvillanova,
I checked the linked issues and PR, this seems like a great idea. Would you mind elaborating on the in-memory and memory-mapped datasets?
Based on my understanding, it is something like this, please correct me if I am wrong:
1. For in-memory datasets, we don't have any dataset files so the entire dataset is pickled to the cache during loading, and then whenever required it is unpickled .
2. For on-disk/memory-mapped datasets, we have the data files provided, so they can be re-loaded from the paths, and only the file-paths are stored while pickling.
If this is correct, will the feature also handle pickling/unpickling of a concatenated dataset? Will this be cached?
This also leads me to ask whether datasets are chunked during pickling?
Thanks,
Gunjan | [
-0.3151937127,
0.0239816308,
-0.0091473348,
0.3324129879,
-0.0929975286,
-0.0160012767,
-0.1415653974,
0.0513882749,
-0.1234132051,
-0.0343253613,
0.0126750283,
0.508906126,
0.015508946,
0.4306789339,
0.1224202663,
0.0571521707,
0.1958592981,
0.1693579555,
-0.4769207537,
0.1094918102,
-0.3694702983,
-0.2283259481,
-0.224198848,
-0.4162862897,
-0.2465882152,
0.0351606458,
-0.4585065544,
0.1797387749,
-0.5705539584,
-0.2730194032,
0.0891331732,
0.3404114246,
0.0331604779,
0.0554767139,
-0.0001229169,
-0.1215620041,
-0.1313506663,
-0.1438873112,
-0.4560646117,
-0.190041095,
-0.4508100152,
-0.4511182308,
0.0536839664,
0.0278963894,
0.4002718329,
-0.1607567817,
0.1130953431,
-0.1156750917,
-0.0251441151,
-0.0461408459,
0.0328180231,
0.1369231939,
0.2183957994,
-0.0649802685,
0.1313887089,
0.4564215541,
-0.2105990201,
-0.015487032,
0.4654990435,
-0.1170652062,
0.1297411919,
-0.2641113997,
-0.0361637212,
-0.2052613199,
0.4992967248,
0.0941745788,
-0.4199727178,
-0.2418173552,
-0.1013240963,
0.2512696683,
0.6391006708,
-0.3239814043,
-0.4200563431,
-0.3742828369,
0.1631270498,
-0.3263415098,
0.1764192283,
0.2312668562,
-0.0021590181,
0.1023017094,
-0.0673535392,
-0.4811718762,
-0.2130006999,
0.2608930469,
0.2544134557,
0.2590686977,
0.2200955153,
0.1209522337,
0.0647673309,
0.0149429385,
0.2294027358,
-0.3686331213,
-0.1657602936,
0.24533315,
-0.3499106467,
-0.0451852903,
0.1129572242,
-0.0423487462,
0.5332872868,
0.3057990372,
0.3247356713,
0.0516032018,
-0.526204288,
0.2660189271,
0.2431846857,
0.1561826319,
-0.096396409,
0.1706820577,
-0.0694976747,
-0.1371656209,
0.237611264,
0.0166340228,
-0.1367419064,
-0.0226722565,
0.0251741037,
-0.1259245872,
-0.0738965496,
0.1945526749,
0.1652773619,
-0.1814723313,
-0.0189236999,
-0.216298908,
0.0778373182,
-0.0743919834,
0.0621999726,
0.0568558052,
-0.013617374,
0.3514008224,
0.3560392857,
-0.1627320051,
0.0549260788,
-0.0814792961,
-0.2937661111,
0.3898350596,
0.3329312503,
0.0588357672,
-0.1926190853,
0.2399521619,
0.040908061,
0.202274859,
-0.1028450057,
-0.24468714,
0.4061220586,
0.0637180805,
0.0030737072,
-0.5130050778,
-0.0187864248,
-0.0329549834,
-0.3379788399,
0.5166099668,
-0.184938848,
-0.3788848221,
-0.3063002527,
-0.0302095786,
-0.1421592236,
-0.046722468,
-0.2177967578,
0.5460848212,
0.3333913088,
-0.2150405645,
0.114819698,
0.158397913,
-0.1153829917,
-0.4840550423,
0.1431282461,
0.4034635127,
-0.4882218242,
-0.0285891294,
0.0979507267,
0.0205814615,
0.0621083751,
0.1312433928,
-0.3371044993,
-0.025657177,
-0.3812057078,
0.298458606,
0.3309969008,
0.0592597798,
-0.1576828659,
-0.0075983331,
0.0237684548,
0.1462471187,
0.4264761508,
0.3102689385,
0.0304752141,
0.0982174277,
0.380361557,
0.3373634517,
-0.1726572067,
-0.3836883605,
-0.0291045681,
-0.4970317483,
0.443430841,
-0.1386879086,
-0.13001737,
0.2080399096,
0.2209340483,
-0.6818778515,
0.1491346061,
-0.1081966311,
0.2218991816,
-0.1341849118,
0.3370450735,
0.057609994,
-0.0608256422,
-0.1623553038,
-0.6230328083,
0.1108445972,
-0.3560384512,
0.0800078362,
-0.1722410321,
-0.2401235253,
0.2953733802,
0.0905296803,
-0.0372995213,
0.1557225287,
-0.0367937088,
0.058046069,
0.0346237868,
-0.4091597199,
-0.452103138,
0.2373279631,
0.0837839469,
0.0530398414,
-0.2692814171,
0.334718436,
0.1345412284,
-0.0918150619,
-0.1463080645,
0.2175553143,
0.0298359115,
0.1256483346,
0.0788571015,
0.4706088901,
-0.1577600539,
0.0567781478,
0.1230108589,
0.3961002231,
0.1758621484,
0.2842262089,
0.0698049366,
-0.7420749664,
0.2002218962,
-0.2985212207,
-0.0355481356,
0.3061637282,
-0.1743581891,
0.2200998962,
0.1943643838,
-0.1398901045,
0.0687407553,
-0.0497538634,
-0.2310174406,
-0.2290157676,
-0.0816832185,
0.2157758027,
0.062781319,
0.3361769915,
-0.525601387,
0.1454779953,
0.1831404716,
0.0464515239,
0.133184731,
0.1387511343,
-0.084919557,
-0.2781792879,
-0.0172249507,
0.0637633204,
0.7007014751,
0.2160706222,
0.0262392126,
-0.0180960465,
0.0740413666,
-0.1254708767,
0.2335385829,
-0.0067180023,
0.1687892675,
0.4369446635,
0.3996044099,
0.1107664704,
-0.0637715831,
0.0234743208,
0.261820823,
-0.2111874521,
-0.1233819872,
-0.2116715014,
-0.4366059303,
-0.0754052103,
-0.2309646755,
-0.3501525819,
-0.1998727918,
-0.0431916416,
0.0764247924,
0.2593501806,
0.0600145534,
0.1693040133,
0.1015221998,
0.7371382713,
-0.4011031389,
-0.364014715,
-0.0570498109,
0.0777585059,
0.1655645669,
-0.0318344831,
0.3715842366,
-0.0787614882,
0.3381141722,
0.2861677706,
-0.0335317403,
-0.2892335355,
-0.2818517685,
0.1511436105,
0.0167328846,
0.0454622135,
0.1730606109,
-0.099978976,
0.0313679576,
-0.4025484025,
0.048880212,
0.4723198414,
-0.2581705451,
0.1338030696,
0.0513441749,
-0.1683002412,
-0.2078545392,
-0.1897792816,
-0.2680079937,
-0.3550593853,
0.4333005846,
0.0087040365,
0.0841032416,
0.1416254044,
0.1919737458,
-0.3095082939,
0.2234873325,
0.185937494,
-0.2188667655,
-0.2029943764,
0.3382441103,
-0.1542314887,
-0.0722791851,
-0.177709043,
-0.1409834623,
0.1397571713,
0.4569011331,
-0.1309050918,
-0.0476003811,
0.0050767697,
0.4843652248,
-0.0939590335,
0.1180792227,
0.7127974629,
0.4727384746,
0.0355076119,
-0.1927273273,
-0.1481277943,
0.2134424895,
0.3343861699,
0.0561788566,
0.3356001973,
-0.2804924846,
-0.0089961141,
0.3415655792,
0.1196296588,
0.0275280904,
0.4800343812,
0.1667900085,
0.3584746122,
-0.1939474344,
-0.0502044037,
-0.2021127045,
-0.452494204,
0.0759774297,
-0.1718981266,
0.0070287101,
-0.1174065694,
-0.1100476012,
0.3091024458,
-0.1377838701,
-0.1023492962,
0.3258581758,
-0.3790818751,
0.4001636207,
0.0627168864,
-0.513910532,
-0.3369117975,
0.1134472489,
-0.1384173036,
0.1996137798,
0.3677075505,
-0.2280971855,
-0.4231455922,
-0.3757843077,
-0.2592119575,
0.0711830184,
0.2041252255,
0.2274398208,
0.1916835606,
-0.2750766575,
-0.051380191,
0.0553244017,
0.7124962807,
-0.0236195736,
-0.5619781613,
0.0524922013,
0.0781706944,
0.3489022255,
0.0581906885,
0.2769466341,
0.156277746,
-0.2433082014,
0.3843095303,
-0.3124508262,
0.1250409335,
0.0606208444,
0.400377959,
0.089004457,
-0.2711159587,
0.0492059514,
0.1524654478,
-0.3478346467,
-0.5306866765,
-0.028142266,
0.0754005015,
0.1031540334,
-0.0128378496,
-0.3287604153,
0.060139671,
0.0363635495,
-0.1553477347,
0.1983358413,
-0.0498029999,
0.3714056313,
-0.2416534573,
-0.0307151116,
0.7655329108,
0.6723150015,
-0.2359552383,
-0.239985615,
-0.1107771993,
-0.1825509667,
0.2655687928,
0.088351652,
0.120269008,
0.4290006757,
0.029699482,
-0.0531862788,
-0.4937258661,
-0.1347971112,
-0.0538790077,
-0.0830796137,
-0.3994486332,
-0.4151041806,
0.461601913,
0.1421403736,
-0.1688085347,
0.5554350019,
0.0548574179,
-0.4117570519,
0.1328788698,
0.05135094,
1.3276771307,
0.2869279385,
0.2386231273,
-0.0389147252,
-0.2739725709,
-0.0141868442,
-0.0437281355,
0.2706168592,
-0.0841444209,
-0.1475724578,
-0.2070042193,
-0.2532974184,
0.039106559,
0.0096354112,
-0.3656957448,
-0.1243710294,
-0.0491560921,
-0.0999155939,
0.0220853314,
0.0850073695,
-0.2020627707,
-0.354175806,
0.2298359126,
-0.0861268193,
0.1308623254,
-0.2966293991,
-0.0865639523,
-0.1409363598,
-0.1197370961,
0.0594424158,
-0.3053863049,
-0.1838283837,
-0.2568382621,
0.1923811734,
0.1816962063,
-0.1746620387,
0.079164505,
0.1379815936,
-0.1640586555,
-0.0747805983,
0.0173842143,
-0.1131509095,
0.2929673493,
-0.126858905,
-0.1666728407,
-0.2928452492,
0.5314505696,
0.1283070743,
-0.0943867192,
0.0102130212,
-0.1836510301,
-0.3001396656,
-0.2620707154,
-0.0297704265,
0.2410975546,
-0.2740917206,
-0.3464112282,
0.5255739093,
0.1118178591,
-0.1333804429,
-0.0171921924,
0.2821409404,
0.00417725,
0.2388520837,
-0.1153022796,
0.0447537154,
-0.172421962,
0.1028248146,
0.3626397252,
-0.0572488159,
0.1238453388,
-0.0124671906,
-0.0947056264,
-0.0052471161,
0.1010326371,
0.0888793841,
-0.1637594402,
-0.0873330384,
0.1111692339,
-0.0554685891,
-0.148193866,
0.2667477131,
0.2715302408,
0.5363761783,
-0.0959548801,
-0.2259960622,
-0.336122632,
0.3077060878,
0.0108234044,
0.3015134931,
-0.0119774267,
0.0568744093,
-0.0652879998,
0.4264637232,
-0.1813464463,
-0.1138805971,
0.0087324828,
0.0344174802,
0.1005262882,
0.1824614257,
-0.2513327003,
-0.199889496,
-0.0870506018,
0.0121297017,
-0.0571957342,
-0.0201099459,
0.0253753066,
0.2010985017,
0.3936628699,
0.0385297537,
-0.3091895282,
-0.4045880139,
-0.0642538965,
-0.3011092246,
0.0180379599,
0.006746456,
-0.2550587058,
-0.0453520194,
0.3187662959,
0.1587716043,
-0.2169714123,
0.4139186144,
0.2465448976,
0.179056555,
-0.0495938025,
0.1369212717,
-0.3104900122,
-0.0004517287,
-0.2512002289,
0.3576907516,
0.2786888778,
-0.0583897755,
0.2566399872,
0.1928195208,
-0.1270214915,
-0.2254947573,
0.093556352,
0.3229031265,
0.1165194064,
0.1785933673,
0.2229983211,
0.0470431224,
-0.3327701092,
-0.2509399951,
-0.2289907187,
-0.1038738862,
0.1372489333,
0.2295455933,
0.2040007114,
0.0853359997,
-0.2376066446,
0.0376167968,
0.0781702921,
-0.0080508497,
0.2104612887,
0.1018413901,
0.0391521156,
0.0759053305,
0.1027448699,
0.4434028864,
0.1278668493,
-0.0603461117,
-0.0536379106,
0.4059930146,
0.335498482,
-0.1564013362,
0.2840722203,
-0.1464850307,
0.1905841976,
0.4117324948,
0.2667860687,
0.1027267128,
0.1858246326,
0.3560812473,
-0.0903090537,
-0.3604804277,
0.3775742054,
0.1918887794,
0.1005820036,
0.0141636841,
-0.1952618062,
-0.3438720107,
-0.3334762752,
-0.124699384,
0.068407014,
-0.1662840247,
0.5352702141,
-0.1510174423,
0.1251707077,
-0.1867251843,
0.4833074212,
0.4300195575,
0.0550651327,
-0.3070124984,
-0.0052056527,
0.2363689244,
-0.2021863908,
0.1345390081,
0.1701175719,
0.4026608467,
0.3949686289,
0.1552559733,
0.1541090757,
-0.127664268,
0.0246708691,
-0.0342122205,
-0.036683619,
0.1181243062,
-0.0916445106,
0.1320811957,
0.0459390134,
0.0718483999,
0.4185603261,
0.121192351,
0.0658812523,
-0.4807811677,
-0.0521915033,
0.3336158991,
-0.5089946389,
0.4396709502,
0.0326909497,
-0.1054123268,
0.0539264642,
0.4718608856,
-0.0278708357,
0.0252904408,
0.0926609263,
0.0147378929,
0.363840133,
0.225744307,
0.4549017847,
0.1053725556,
-0.2259773016,
-0.2966222465,
-0.2613648176,
-0.1315376312,
0.312887013,
-0.379881233,
0.0001336299,
0.042566143,
0.3286976814,
0.075352706,
-0.0420679301,
-0.0879049003,
0.0292956531,
0.2934946418,
-0.224812299,
-0.1151870489,
0.2837563455,
0.1098928824,
-0.1114573926,
-0.2010267079,
0.0119722961,
0.3292584121,
-0.1495504975,
-0.2576147914,
-0.0294732749,
0.1874910742,
0.2973904908,
0.3829787076,
0.2143355608,
0.3231751621,
-0.2412548661,
0.056577757,
0.1228622645,
-0.103688769,
-0.2429078221,
0.2085928321,
0.0174815282,
-0.1900773644,
-0.541985631,
0.0445846319,
-0.0664343014,
-0.055745367,
-0.1266641021,
-0.0023190379,
-0.2712288499,
0.0344973952,
0.0491985381,
-0.1256289184,
0.1763835847,
0.3343781829,
0.1156436652,
0.0935374796,
0.053694576,
-0.0985950977,
-0.0166142732,
0.2888441384,
-0.0382957458,
-0.0881180465,
-0.1418968886,
-0.0524474978,
0.2187779695,
-0.8657222986,
-0.3556311727,
0.3075705171,
-0.1007840186,
0.1090639532,
0.3817705214,
0.0892792791,
-0.1302482337,
-0.3879982829,
0.3571622968,
-0.1040114015,
-0.3392214775,
-0.0915779397,
-0.4220615625
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | Hi ! Yes you're totally right about your two points :)
And in the case of a concatenated dataset, then we should reload each sub-table depending on whether it's in-memory or memory mapped. That means the dataset will be made of several blocks in order to keep track of what's from memory and what's memory mapped. This allows to pickle/unpickle concatenated datasets | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 62 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
Hi ! Yes you're totally right about your two points :)
And in the case of a concatenated dataset, then we should reload each sub-table depending on whether it's in-memory or memory mapped. That means the dataset will be made of several blocks in order to keep track of what's from memory and what's memory mapped. This allows to pickle/unpickle concatenated datasets | [
-0.3308896422,
0.080966264,
-0.0875296071,
0.1288807392,
0.1037874371,
0.0213833153,
-0.0953434408,
0.2642480731,
-0.0994139165,
0.1963125467,
-0.0066628084,
0.5743991137,
-0.0273892544,
0.4745151699,
0.1390600652,
0.0199647546,
0.1339085996,
0.2706728876,
-0.4599793255,
0.0995576456,
-0.2808739543,
-0.2470990419,
-0.0584457219,
-0.3346609473,
-0.0523942262,
-0.0421134606,
-0.3391370773,
0.1539797187,
-0.3864448667,
-0.4221207798,
0.0044406131,
0.2818989158,
0.1375906169,
-0.0432055183,
-0.0001011826,
-0.0590704307,
-0.0670984238,
-0.1675767899,
-0.5165484548,
-0.0967806131,
-0.3099039197,
-0.3545518517,
0.0518834293,
-0.1348393261,
0.2507438362,
-0.3207661808,
0.1727224588,
-0.0237122029,
0.0291028991,
-0.0417879075,
0.244840771,
0.201136753,
0.1695045233,
-0.1530704796,
0.1990845352,
0.3594321907,
-0.1890633553,
0.0671691969,
0.4142089486,
-0.1301231682,
0.0097993985,
-0.1467057467,
-0.0309296064,
-0.0901513919,
0.3890372217,
0.1517835259,
-0.1840357482,
-0.2165958583,
-0.1419533044,
0.2824299932,
0.5254633427,
-0.3413597643,
-0.4619013071,
-0.2848485112,
0.1841120273,
-0.3126704693,
0.1059342176,
0.2865485549,
-0.0368904732,
0.2204460651,
0.0238912068,
-0.2196093202,
-0.3139273822,
0.1313329041,
0.2414419949,
0.2709456682,
0.2284068763,
0.0479884744,
0.0606654137,
0.0512915887,
0.1425949633,
-0.1905995756,
-0.134885326,
0.123174116,
-0.1949243098,
-0.1008891687,
0.0449646786,
-0.0745253265,
0.5886336565,
0.2289019078,
0.3365896344,
0.1677314639,
-0.3017886877,
0.3579452634,
0.202316314,
0.0898765922,
-0.2345992178,
0.0051622242,
0.1279995739,
-0.2076892853,
0.1173693985,
-0.0112339621,
-0.1240905523,
0.1520698369,
0.0693647787,
-0.1491600871,
-0.0841784254,
0.1468492448,
0.0327504724,
-0.2261698842,
-0.0130925328,
-0.1123065501,
0.0061757453,
0.1253178269,
0.070429191,
-0.0281158946,
-0.021091979,
0.2245666981,
0.2642067373,
-0.0813479349,
0.0046338066,
-0.0850578398,
-0.2081481814,
0.3224072456,
0.3119049072,
0.0075855963,
-0.1021552011,
0.1868178844,
0.0192195103,
0.2130550742,
0.1430292428,
-0.1518513411,
0.4313819706,
0.1036881357,
0.0568900108,
-0.4961794913,
-0.1049025059,
-0.0678687543,
-0.3789528906,
0.4363247752,
-0.107694149,
-0.2815756202,
-0.3184338212,
0.2236292064,
-0.0530513413,
-0.1553883404,
-0.1535329223,
0.4602694511,
0.0986520797,
-0.2554015517,
0.0526077598,
0.1726688147,
-0.0268214196,
-0.4418765604,
0.1139300764,
0.1807295978,
-0.4583587646,
0.0623890385,
0.1553567648,
-0.128103286,
-0.0389841646,
0.1168559641,
-0.3931361437,
0.0349426493,
-0.396422863,
0.3418226242,
0.265509963,
0.0290630683,
-0.126969263,
-0.0477873869,
-0.0275524333,
0.0150118396,
0.4528052807,
0.3481982052,
0.2118476033,
0.0749008507,
0.2160174549,
0.3431639075,
-0.1340661794,
-0.1320325881,
-0.0794658661,
-0.4177036285,
0.3380464911,
-0.018900428,
-0.2303334475,
0.1269381493,
0.2450881898,
-0.5227223635,
0.2441468537,
-0.3239214122,
0.1686996222,
-0.1750643253,
0.4148916304,
0.0176655576,
-0.0822702572,
-0.1235152483,
-0.5488769412,
0.1035034359,
-0.31180197,
0.0981164724,
-0.1812511683,
-0.3896618485,
0.2531532645,
0.0668022931,
0.0256448444,
0.1171813831,
0.2458097041,
0.0483006015,
-0.0197435692,
-0.3945820332,
-0.3163654208,
0.0991428494,
0.0952568948,
-0.0008600876,
-0.1155862361,
0.2568220794,
-0.0641728044,
-0.1052488238,
0.0593799502,
0.1430230737,
-0.0712550133,
0.0776185095,
0.1616186202,
0.4332282245,
-0.090907529,
0.0953141004,
0.1977408677,
0.3503226638,
0.1969358474,
0.3186229765,
0.0536825806,
-0.4303309023,
0.1388094127,
-0.1494996995,
-0.2212159932,
0.4097175002,
0.0697908327,
0.1340531856,
0.2553194761,
-0.1307524741,
0.2639632225,
-0.0489562042,
-0.1614898145,
-0.2157243788,
-0.1472557187,
0.1894980222,
0.1447422653,
0.3336498141,
-0.4741409421,
0.2637765408,
0.249556601,
-0.011650227,
0.1791909039,
0.1175614372,
-0.1045675278,
-0.2616476715,
0.055511035,
0.1884202957,
0.531265974,
0.4590287805,
0.059164349,
-0.0612869263,
0.0004428178,
-0.125344649,
0.200202167,
0.0703895092,
0.1017497629,
0.307903558,
0.3970174789,
0.0371571071,
-0.1925428808,
-0.0835520253,
0.1319691092,
-0.1614087224,
-0.0441866703,
-0.2051666975,
-0.3770574331,
-0.0965348333,
-0.1838058084,
-0.2013024539,
-0.1883926988,
-0.1703037173,
0.2789219022,
0.2475674301,
-0.2998247743,
0.1375080347,
0.1295552701,
0.5335014462,
-0.3754704595,
-0.1796846688,
-0.108483687,
-0.089554362,
0.1045135558,
0.1845102608,
0.308083266,
-0.0592200793,
0.5809789896,
0.3431639969,
0.0040002242,
-0.3314963579,
-0.3079935312,
0.0987153202,
-0.049303811,
0.059814021,
0.1120493561,
-0.0000599511,
0.1649870574,
-0.4930721819,
0.0386007316,
0.2502208948,
-0.2884956002,
0.1441664398,
0.0897158682,
-0.171761632,
-0.3596301973,
-0.3185690641,
-0.3232240379,
-0.5316621661,
0.5087659359,
-0.012401633,
0.218919307,
0.1244734526,
0.0831148848,
-0.2596383095,
0.0898743421,
0.1883159578,
-0.2038052678,
-0.1830642074,
0.2191309631,
-0.2883408666,
-0.1422279775,
-0.1156650558,
-0.1629636139,
0.1477711201,
0.3719270229,
-0.1589663029,
-0.1688545942,
-0.0243643392,
0.3907032013,
-0.0688615292,
0.0487951264,
0.585606277,
0.5143857002,
-0.2438393682,
-0.1338955313,
0.0273833424,
0.1994644105,
0.2149877101,
0.012374144,
0.0917399824,
-0.2447971702,
-0.0339082628,
0.2672219574,
-0.0449007079,
-0.0083454065,
0.4802698791,
0.0296965782,
0.3378032446,
-0.1988240778,
0.0846577138,
-0.1542259455,
-0.4396434426,
-0.0395387001,
0.0810747817,
-0.0181270428,
-0.2163594514,
-0.0275656022,
0.2161122859,
-0.1812889278,
-0.2459590286,
0.3671939671,
-0.1107377559,
0.2527347505,
-0.0106518567,
-0.4284534752,
-0.2671548724,
0.1762674898,
-0.0477775857,
0.1248854995,
0.2901671231,
-0.2416697443,
-0.4778909683,
-0.3409611583,
-0.3137470484,
0.1387540549,
0.2407147884,
0.1306317598,
0.2050715387,
-0.4547717273,
-0.0807620436,
-0.0085478537,
0.6017178297,
-0.0057810247,
-0.4994776845,
0.0726115704,
0.0704668313,
0.2290144861,
0.0918517858,
0.0698180944,
0.0356100127,
-0.1553215683,
0.2879040241,
-0.3400049508,
0.0505486988,
0.0386341512,
0.3517750204,
-0.0046848431,
-0.2893883288,
-0.0051387697,
0.0565504432,
-0.486580193,
-0.4271866977,
-0.0341631994,
0.0121530928,
0.0056692101,
-0.1888548434,
-0.2518097758,
-0.0151930787,
0.0212318897,
0.0047651008,
0.2259651273,
-0.0365258008,
0.3290245235,
-0.2873806059,
0.1535701752,
0.7977833748,
0.5349410772,
-0.3412919343,
-0.0700023323,
-0.0637581497,
-0.2462361008,
0.2890377939,
-0.0018978417,
0.1382668912,
0.4825796783,
-0.0300120581,
-0.0126199573,
-0.5641717315,
-0.1735908836,
-0.1112965196,
-0.1793779433,
-0.3051041961,
-0.377784133,
0.4603813589,
0.1201415658,
-0.2521438301,
0.3317559958,
-0.0150274672,
-0.4221540689,
0.3290471137,
0.0887540653,
1.1225886345,
0.1779589951,
0.3433840573,
0.0340346508,
-0.2753272951,
0.0926772058,
-0.112964794,
0.227816686,
-0.1867443919,
-0.123815909,
-0.1344875991,
-0.1599748433,
-0.0595984235,
0.029115973,
-0.521074295,
-0.1095250994,
0.0125245452,
-0.0728693306,
-0.1100047082,
0.2566877604,
-0.2936607003,
-0.3604158163,
-0.038054809,
0.1260077357,
0.1341445148,
-0.3623822033,
-0.0884262025,
-0.2095152438,
-0.1346686482,
0.0878353715,
-0.1085632667,
-0.1660429537,
-0.2215916663,
0.1632598639,
-0.0204982124,
-0.096108295,
-0.196172148,
-0.0393273532,
-0.1402608752,
0.0502506085,
-0.1339435875,
-0.1060588583,
0.0702750385,
0.0893012881,
-0.1681360453,
-0.2926284075,
0.5270133615,
0.0715628266,
-0.1033331305,
-0.0052108578,
-0.1177364886,
-0.3113012612,
-0.1454505622,
-0.0112221837,
0.111905545,
-0.2358522713,
-0.0929558277,
0.3725414872,
-0.1225112379,
-0.2287291437,
0.1796438694,
0.1737487167,
-0.0622074977,
0.2706394494,
0.0244436078,
0.0373506844,
-0.2022332847,
0.1054248363,
0.2561531961,
0.0977663696,
0.1157251298,
-0.0616131127,
-0.0979968607,
-0.2150929719,
0.1079406142,
0.125708431,
-0.0852990076,
-0.0342698358,
0.1577449143,
-0.1305824518,
-0.0722839758,
0.2747026086,
0.296936512,
0.3914659321,
-0.1860090792,
-0.1561175585,
-0.4850622118,
0.3692667782,
0.0839181542,
0.4723116755,
0.0426460654,
0.0909435898,
-0.0972829908,
0.3439356685,
-0.4360330999,
-0.2447358221,
-0.1786686778,
0.1886476278,
0.0621031001,
0.1072233766,
-0.1068105549,
-0.1632986665,
0.1243666261,
0.0660274774,
-0.0521616638,
-0.2705054879,
0.0005365312,
0.0886473879,
0.2666465044,
0.1510404199,
-0.2268210948,
-0.2388520837,
-0.0090216789,
-0.2859176695,
0.098426953,
-0.0472492017,
-0.2804763317,
0.0473793522,
0.4312939346,
0.1661451757,
-0.091177173,
0.2814756334,
0.1302238405,
0.1757021695,
-0.0284210239,
0.0774827674,
-0.5265858769,
-0.0447006896,
0.0518372208,
0.3191320598,
0.2118451148,
0.006655816,
0.2070165277,
0.3998594582,
-0.2470116168,
-0.1597464085,
0.2610473037,
0.2472994179,
0.0784808546,
0.1721437573,
0.2526984215,
0.320116967,
-0.3753231466,
-0.2055062205,
-0.2421288192,
-0.3017143011,
0.2398988754,
0.2074736059,
0.2079681754,
0.0633591115,
-0.2070697993,
-0.0001579486,
0.1082521826,
-0.0299706385,
0.0974072963,
0.0044742506,
-0.0765746683,
-0.0357469656,
0.0310704764,
0.2621297538,
0.2077346295,
0.0396182016,
-0.1450177133,
0.466124475,
0.3491326571,
-0.1293597519,
0.2394714653,
0.030702278,
0.217636779,
0.318343848,
0.3328392804,
0.0937068239,
0.2749660909,
0.2630878389,
-0.0392217785,
-0.3378152847,
0.3260612488,
0.1355717778,
0.058468435,
-0.0337923542,
-0.2106035799,
-0.2294226438,
-0.220093891,
-0.2160735577,
0.0881616175,
-0.2070173621,
0.3964640498,
0.0546579137,
0.2351970077,
-0.1943441927,
0.3281579614,
0.3741765916,
0.1182762831,
-0.307222724,
-0.0797846541,
0.3700534701,
-0.1761282682,
0.1467445791,
0.198632881,
0.3817966878,
0.3101078272,
0.0776181296,
0.168285206,
-0.1536345333,
0.0742750466,
-0.2141851038,
-0.1156187356,
0.1510536075,
-0.0136950165,
0.2354592979,
0.2452615947,
-0.0899586603,
0.303072691,
0.1031544358,
-0.0936659724,
-0.3618726134,
-0.047055237,
0.5336931944,
-0.558611095,
0.259732753,
-0.0543595515,
-0.2575384974,
0.015545141,
0.3132823706,
-0.0733167231,
0.1002572775,
0.0567995943,
0.1455907077,
0.2891752422,
0.4204459786,
0.3056247532,
0.1715647429,
-0.2166493088,
-0.1849777997,
-0.2979391813,
-0.0502009019,
0.1012847424,
-0.2598642707,
-0.0731846243,
0.0629110038,
0.284353286,
0.0627031997,
-0.0451716036,
-0.0960490853,
0.0298610441,
0.2053637207,
-0.3371419907,
-0.1072489768,
0.2484494597,
-0.0365945622,
0.0122252926,
-0.2142079473,
0.0440486744,
0.2075446844,
0.1077702492,
-0.228548333,
0.0429878831,
0.1554294527,
0.2202720195,
0.5512213707,
0.1182146296,
0.2511799037,
-0.2976634502,
0.0068752915,
0.0886665657,
-0.1955956221,
-0.3251900673,
0.0286134873,
0.0679930747,
-0.1258864254,
-0.2877642214,
-0.1205196828,
-0.077555865,
-0.0381677151,
-0.1025927737,
0.0462996736,
-0.2272877991,
0.1190912202,
0.1525844485,
-0.1543568224,
-0.0258812048,
0.2455317378,
0.1390735656,
-0.0061063841,
-0.0663333163,
-0.260060966,
0.0703027546,
0.2660230398,
-0.0264013037,
-0.0838247687,
0.0239694491,
-0.0870700181,
0.2527745962,
-0.7630252838,
-0.2411404699,
0.2996512055,
-0.1082538962,
-0.0248057991,
0.4120289087,
0.0739523321,
-0.0951399058,
-0.3477320373,
0.3045029938,
-0.0662198663,
-0.3062162101,
-0.1457573771,
-0.4974687099
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | Hi @lhoestq
Thanks, that sounds nice. Can you explain where the issue of the double memory may arise? Also, why is the existing `concatenate_datasets` not sufficient for this purpose? | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 29 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
Hi @lhoestq
Thanks, that sounds nice. Can you explain where the issue of the double memory may arise? Also, why is the existing `concatenate_datasets` not sufficient for this purpose? | [
-0.3921353221,
0.1095162034,
-0.0102484673,
0.369923979,
0.0255511552,
0.1648334861,
-0.2444255799,
0.199428305,
-0.1821031868,
0.136639297,
0.0495680943,
0.4666678905,
0.0200919919,
0.3617292047,
0.1092138886,
-0.0306007527,
0.1604874283,
0.2564076781,
-0.4017227292,
0.1629247218,
-0.3001019955,
-0.1728201509,
-0.1916833222,
-0.4789799452,
-0.1616440564,
0.1648574322,
-0.4099675715,
0.1326929331,
-0.5053353906,
-0.3240080476,
-0.0337191299,
0.2305969,
0.1441220194,
0.1116466299,
-0.0001183449,
-0.1393061131,
-0.0883798972,
-0.1393122077,
-0.4981407821,
-0.1376256943,
-0.3153512478,
-0.4476881325,
0.029385522,
0.0132386088,
0.3109205365,
-0.3076113462,
0.0527276322,
-0.1188959107,
-0.0023293048,
-0.0532881208,
0.1089301854,
0.0608279109,
0.2846044898,
-0.18300201,
0.1651659608,
0.4658091962,
-0.2200625986,
-0.0311096199,
0.4206275344,
-0.195269078,
0.2418723404,
-0.2360844761,
-0.1178748906,
-0.1633911729,
0.3987490535,
0.172732994,
-0.3037189841,
-0.131404683,
-0.0412973985,
0.2176211774,
0.5865793228,
-0.4311449826,
-0.334210515,
-0.2628728151,
0.2460076362,
-0.2954333127,
0.1332030296,
0.2357897162,
-0.1375349611,
0.1732290834,
0.0094013587,
-0.4855877161,
-0.1796580702,
0.2952821553,
0.2737633586,
0.2759508789,
0.3529568315,
0.1340425313,
0.2141886652,
0.0620447211,
0.2141939551,
-0.2129980773,
-0.254674077,
0.1283373684,
-0.3424056172,
-0.0140579119,
0.0599339977,
-0.1500601768,
0.5483287573,
0.1297709644,
0.2871188521,
-0.0448516533,
-0.4067798853,
0.3076330125,
0.0919862986,
0.2553530335,
-0.2252776921,
0.1636928171,
-0.0637871101,
-0.2130730599,
0.3011720181,
-0.0163452923,
-0.1694170833,
-0.073068276,
0.2098751664,
-0.2031954974,
-0.0720807463,
0.1567243338,
0.0766469464,
-0.2511427402,
-0.1469198763,
-0.1524337232,
0.0938945189,
-0.0379303358,
0.1397985816,
0.133142367,
0.1158499271,
0.4211546481,
0.3029701114,
-0.1989288032,
0.0078564435,
0.0276403539,
-0.2613210082,
0.3779643774,
0.3536388278,
0.1148885489,
-0.197372973,
0.3130427599,
-0.0070332121,
0.1946332604,
-0.0352145657,
-0.1502505839,
0.3124206364,
0.1093136519,
0.0563114136,
-0.4541545212,
-0.0355092138,
-0.0233206078,
-0.2534181476,
0.6008573771,
-0.1377721131,
-0.3231897354,
-0.5031118393,
0.0469380431,
-0.0669118017,
-0.023708662,
-0.2471251488,
0.5314825177,
0.3372123539,
-0.2383344918,
0.0611341745,
0.1914604604,
-0.175152868,
-0.4846290946,
0.221704796,
0.3638980985,
-0.4608434737,
0.0812962502,
0.0106249936,
0.0284838937,
0.0967996418,
0.1960921586,
-0.3160737157,
-0.0223340653,
-0.4253899455,
0.2173316181,
0.2514078319,
-0.0291146785,
-0.0864310563,
-0.0521699116,
-0.158220768,
0.0194213539,
0.5092132092,
0.4111897647,
0.0538949929,
0.1372525841,
0.431458801,
0.2517196238,
-0.1865966022,
-0.2832248509,
-0.0806563869,
-0.5113968253,
0.4175634682,
-0.1091269702,
-0.11199902,
0.1417609602,
0.1748512685,
-0.5082266331,
0.2905801535,
-0.1933504045,
0.2741512656,
-0.1444500834,
0.3921689689,
-0.0364354476,
-0.1589794457,
-0.158927083,
-0.5136504173,
0.1615584195,
-0.1798150539,
0.1487388909,
-0.1148836836,
-0.2364260852,
0.2844361961,
0.0667651817,
0.0482145622,
0.1300472468,
0.0389031619,
0.0784220695,
0.0149989054,
-0.3527720571,
-0.2117244005,
0.2260376513,
0.1711196899,
-0.0062984414,
-0.1863911152,
0.3505256772,
0.0912903473,
-0.0637104139,
-0.0793344229,
0.2012590617,
0.0118025644,
0.2366676629,
0.0625835955,
0.4659433961,
-0.197026968,
0.0314246714,
0.1769627482,
0.3687470555,
0.1420232356,
0.2543237805,
0.0953701586,
-0.558390677,
0.1644983739,
-0.2625583708,
-0.058306206,
0.2322971076,
0.0301948041,
0.1346608996,
0.0542905182,
-0.1179549247,
0.0882855132,
-0.0218890607,
-0.0929089785,
-0.2552784681,
0.0091868304,
0.2155606747,
-0.0128444061,
0.3394585252,
-0.5138896108,
0.1045173109,
0.1417917907,
0.0801381618,
0.1494961977,
0.1846862733,
-0.0975309983,
-0.2499240041,
0.0110410657,
0.1050736979,
0.7278345823,
0.3327361643,
-0.0270749331,
0.0240189079,
-0.0232417416,
-0.1357817799,
0.2460507751,
-0.0523551404,
0.2259605825,
0.3271228075,
0.517565906,
0.097870782,
-0.0231268033,
0.0357370898,
0.280864507,
-0.0925228447,
-0.1162447855,
-0.299050808,
-0.4070842564,
-0.0361857414,
-0.2963103652,
-0.2642365992,
-0.1862794459,
-0.0312163979,
0.1388067603,
0.3211250901,
-0.0032380968,
0.0747071654,
0.1011650115,
0.5987220407,
-0.3522721529,
-0.4385087192,
-0.0326090008,
0.0550651774,
0.0571301356,
0.0250622481,
0.4078397155,
-0.1361313462,
0.4487459362,
0.4095328152,
-0.1099671349,
-0.363945663,
-0.3495541215,
0.044286456,
0.0677514225,
0.0573994294,
0.0872166827,
-0.1416817755,
0.1894742846,
-0.4799759984,
0.0716042146,
0.4400593042,
-0.2252621055,
0.2626762986,
0.046937786,
-0.1753201485,
-0.3075056672,
-0.1543865949,
-0.1920005679,
-0.451446563,
0.4154673219,
-0.1236038059,
0.0698882192,
0.0775345117,
0.1950180829,
-0.3288706839,
0.1795967072,
0.1767162979,
-0.2035761476,
-0.1530010998,
0.3337665498,
-0.1142300963,
0.0010317191,
-0.1347980052,
-0.1396706998,
0.1908709407,
0.4543934464,
-0.1955450177,
-0.1592122614,
-0.0649241507,
0.3299048543,
-0.2098085135,
0.1074893996,
0.6054487824,
0.4714119732,
-0.059128359,
-0.0853747204,
-0.2051149309,
0.2540464699,
0.2081808895,
0.1479588449,
0.130411163,
-0.1937827766,
0.0110787004,
0.249563843,
0.0844727904,
0.1532489359,
0.4600156844,
0.0520455092,
0.3349920213,
-0.2104031742,
-0.1420198232,
-0.2575080395,
-0.3768160939,
-0.0843175352,
-0.1151892617,
0.0009575766,
-0.2153918743,
-0.1009140685,
0.3079531789,
-0.1786285192,
-0.212446928,
0.3111822009,
-0.2671620548,
0.3104089797,
0.0059288815,
-0.4796601534,
-0.3252388239,
0.0819748789,
-0.1227237582,
0.187094152,
0.1271076947,
-0.1675875485,
-0.3169614971,
-0.4128249288,
-0.3264816701,
0.0829284191,
0.1736969948,
0.2987654805,
0.1519920528,
-0.2542054355,
-0.0798255801,
0.0172684751,
0.8591120243,
0.0638371855,
-0.5751932263,
0.0738100708,
0.0964724571,
0.235324949,
0.0099885389,
0.2178255618,
0.104072541,
-0.2128965408,
0.249106884,
-0.3420882821,
0.1586685032,
0.1269886643,
0.4117609859,
0.0143964589,
-0.3251318634,
-0.0357559249,
0.1081977189,
-0.3782024384,
-0.5106706023,
0.1149501875,
0.2176655531,
0.0696090087,
-0.0149388984,
-0.2205721438,
-0.0238575749,
0.1370720416,
-0.1835567802,
0.3332089484,
0.0676085576,
0.4472171664,
-0.2835331559,
-0.1221531332,
0.8011989594,
0.6519252658,
-0.2067800909,
-0.2765524387,
-0.0635080636,
-0.1825511456,
0.3542740941,
0.0533777475,
0.0887250602,
0.4914829433,
-0.0025875531,
0.0044302288,
-0.4980610609,
-0.1264450848,
-0.0918322355,
-0.1753396243,
-0.4002380371,
-0.3667639196,
0.4565200508,
0.2283896059,
-0.1706125885,
0.5208794475,
0.0523539931,
-0.447489351,
0.1955935955,
0.0214553755,
1.2458560467,
0.1081588641,
0.301171273,
0.0293306969,
-0.3577950597,
0.0946928561,
-0.0979992673,
0.3106339574,
-0.1265075803,
-0.0006993376,
-0.0910703838,
-0.2204678357,
-0.0275887251,
-0.0188458506,
-0.5232801437,
-0.0967270508,
-0.0838883072,
-0.1347828358,
0.0285347626,
0.210478276,
-0.2344263941,
-0.3641815782,
0.2192933261,
-0.0080279671,
0.0928403065,
-0.4051950872,
-0.1335784644,
-0.1685973406,
-0.1312263012,
0.0406670645,
-0.4389451146,
-0.1573925912,
-0.2707947791,
0.2892298698,
-0.0143482964,
-0.1908157021,
0.0505685061,
0.0598166175,
-0.0464056693,
0.0665042847,
-0.0751335695,
-0.1513550878,
0.1128885895,
-0.0322070569,
-0.2315132022,
-0.4815511107,
0.4041182995,
0.131048277,
-0.2263628244,
-0.0356833786,
-0.0952384323,
-0.3578653336,
-0.2171874642,
-0.0105714388,
0.2572182715,
-0.148679167,
-0.3082505465,
0.4176058769,
0.15462102,
-0.2415662259,
0.0456598662,
0.3068847656,
-0.1014380157,
0.0988883972,
-0.163471207,
-0.0184570178,
-0.1666099578,
0.1040518731,
0.3685367405,
0.0014766976,
0.1319409311,
0.0362784341,
0.0489251837,
-0.0144582242,
0.0155188739,
0.1414502561,
-0.1418101788,
-0.0327541418,
0.1646203101,
-0.1074483395,
-0.1131367534,
0.1744047254,
0.3290686011,
0.4423856735,
0.0320086181,
-0.2129260451,
-0.400800854,
0.3655231297,
0.0545671843,
0.3841341734,
0.0187727064,
0.2157424986,
-0.1020369679,
0.4042104185,
-0.235641256,
-0.1241045818,
0.0039586946,
0.0871532857,
0.0489386916,
0.1895332336,
-0.1527878642,
-0.1534682512,
-0.0207304563,
0.0457590334,
-0.0048325602,
-0.0742739961,
0.0333283171,
0.1710639149,
0.257535249,
0.1527630985,
-0.2190362066,
-0.4636066556,
0.0118619204,
-0.2358220369,
0.0221935436,
0.1075059921,
-0.2809910178,
-0.0037314557,
0.3735971451,
0.0267493892,
-0.1658874303,
0.3852539361,
0.126620993,
0.1624552459,
-0.030215634,
0.2613482177,
-0.3178708255,
0.0073675066,
-0.244766444,
0.3067189753,
0.1848437041,
-0.0801893771,
0.0112201441,
0.2754110098,
-0.250664413,
-0.1051611304,
0.1765166223,
0.228017509,
0.0846745521,
0.2848002613,
0.2059249133,
0.1205370277,
-0.3507795334,
-0.235020265,
-0.3103943765,
-0.2013258636,
0.1200515851,
0.2938492298,
0.1429550648,
0.0684176832,
-0.2992197573,
0.0443542972,
0.0560020506,
0.0071745198,
0.1238171384,
-0.0145441741,
-0.0549443737,
-0.0123571977,
-0.0364686809,
0.2724399865,
0.1355479807,
0.0482495092,
-0.0523113199,
0.4605277479,
0.3122773468,
-0.1931197047,
0.1669676155,
-0.0077458695,
0.2087927759,
0.3916069269,
0.2493346334,
-0.0104126427,
0.2095511258,
0.3926912248,
-0.0135436356,
-0.2855476439,
0.2348790318,
0.1365749538,
0.0337785892,
0.0279447362,
0.0244909897,
-0.3163834214,
-0.2760236263,
-0.133628279,
0.1761217713,
-0.2296822518,
0.5138266087,
-0.0857446194,
0.0736973882,
-0.2091647387,
0.378633678,
0.653855741,
-0.0343053415,
-0.4184260964,
-0.0995004922,
0.3698891699,
-0.3635702729,
0.1004637033,
0.2781807482,
0.5315656066,
0.3679249287,
0.2171501517,
0.0727922022,
-0.1131785363,
0.055278793,
-0.1523784101,
-0.1277519017,
0.2600960732,
-0.0165480599,
0.0902415588,
0.1100359112,
0.0214056373,
0.2997893095,
0.126808539,
0.050965827,
-0.5596282482,
-0.0497351885,
0.3546371162,
-0.5050171018,
0.3200530112,
0.0705103874,
-0.2276431918,
-0.003228955,
0.3698562384,
-0.0468130559,
-0.0114679858,
0.0774336159,
0.0423710831,
0.3519232869,
0.1842330396,
0.3905679584,
0.1013740376,
-0.3784218132,
-0.3028890491,
-0.3700411916,
-0.0803644359,
0.0273217522,
-0.2456849217,
-0.0807178468,
0.1697348654,
0.2687173784,
0.0672255605,
0.0479568094,
-0.0700805485,
-0.041172348,
0.2574211657,
-0.2703320384,
-0.1060345769,
0.3794727325,
-0.0613818392,
0.0117306337,
-0.2595916092,
0.0768879354,
0.1721330583,
-0.0374331027,
-0.3302073181,
0.1232139394,
0.2427935749,
0.3225732446,
0.5413458943,
0.2077282816,
0.2530895472,
-0.230182007,
-0.0335408635,
0.0080999192,
-0.1351276338,
-0.3178328574,
0.103570126,
-0.0106339492,
-0.0577507466,
-0.6203402877,
-0.0652147382,
-0.0586420074,
-0.0344109647,
-0.1914422512,
0.0050821006,
-0.3712553084,
0.1478953212,
0.0911396146,
0.0203424431,
0.0294456407,
0.2493437231,
0.145234108,
0.1393391639,
-0.0469749719,
-0.2207227647,
-0.0059851706,
0.3501099348,
-0.0742329508,
-0.1108441725,
-0.0415952876,
-0.1768175066,
0.2330029458,
-0.8800833821,
-0.3228904307,
0.2580398321,
0.0318023451,
0.0130911544,
0.528670013,
-0.0296341274,
-0.1114501134,
-0.2562057376,
0.3106478453,
-0.1890329421,
-0.3711296618,
-0.0093599558,
-0.4192745388
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | Hi @lhoestq,
Will the `add_item` feature also help with lazy writing (or no caching) during `map`/`filter`? | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 16 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
Hi @lhoestq,
Will the `add_item` feature also help with lazy writing (or no caching) during `map`/`filter`? | [
-0.4765587151,
0.060836792,
-0.070722729,
0.0521825366,
0.0756109208,
0.0150532871,
-0.2226110101,
0.3435712159,
0.0733319372,
0.0877743587,
0.0297727436,
0.6137645841,
-0.0133163072,
0.4729108214,
0.1156109199,
0.0217178836,
0.1311561465,
0.2873377502,
-0.4156748354,
0.1589141637,
-0.2420462072,
-0.2265803814,
-0.0592830852,
-0.4308308959,
-0.004040096,
0.0221987497,
-0.2348328233,
0.046051316,
-0.4015109241,
-0.3713645637,
0.0051824525,
0.390306443,
0.0715250075,
0.0045820102,
-0.0001091117,
-0.0961321816,
-0.021398969,
-0.1660830677,
-0.4541788101,
-0.1732273549,
-0.3239760995,
-0.3942296803,
0.0978122875,
-0.0564115644,
0.2198664695,
-0.1939133555,
0.124648504,
-0.0921967626,
0.0510854498,
-0.0005945712,
0.1770360768,
0.1682466418,
0.1792140603,
-0.0644020215,
0.2513578534,
0.4772345722,
-0.2069152296,
-0.0169788934,
0.5437409878,
-0.2866383791,
0.0655304939,
-0.0498592183,
-0.1694200635,
-0.1345378757,
0.5407472253,
0.1566753387,
-0.0944055766,
-0.320982635,
-0.0573866703,
0.2175628543,
0.4746555388,
-0.4267604649,
-0.4392645657,
-0.2462814152,
0.1049217433,
-0.1887440383,
0.0964330435,
0.2361291051,
-0.1394098103,
0.2127307057,
-0.0762296394,
-0.3423057497,
-0.1845762879,
0.2520515621,
0.3293734789,
0.3210093975,
0.3376604021,
0.0152410371,
0.1598054767,
0.0632551759,
0.2139872909,
-0.3032206297,
-0.221729964,
0.1727043837,
-0.2810245454,
-0.0871395543,
0.0919325799,
0.0622821674,
0.488281399,
0.0839876905,
0.3856857121,
0.1145871133,
-0.2556547523,
0.399540484,
0.1065206602,
0.0492272004,
-0.186040014,
0.1075034291,
0.0387989953,
-0.2714676261,
0.1136783957,
-0.0038704826,
0.0327553451,
0.0663647354,
0.1317949444,
-0.2102600932,
-0.1473453343,
0.1261440217,
0.1737756282,
-0.2988485098,
-0.1008139625,
-0.1459760219,
0.0449212827,
0.0515248477,
0.1264668703,
0.0187105313,
0.0435858518,
0.2391198575,
0.3044168055,
-0.0500590056,
0.0550268479,
-0.1532794237,
-0.1885487735,
0.3325957358,
0.307651937,
-0.0385488272,
-0.1532030255,
0.2341606915,
-0.0660623536,
0.2304019332,
0.1649898589,
-0.1683607697,
0.4443360567,
0.2302445322,
0.0097290799,
-0.4723737836,
-0.0464850441,
-0.0752671957,
-0.3994234204,
0.4876354039,
-0.1724243611,
-0.3116793633,
-0.2956550419,
0.1612206399,
-0.0140458606,
-0.0826083869,
-0.1136310697,
0.5499398112,
0.1494812667,
-0.4128623903,
0.0774160177,
0.1391512305,
-0.2023866624,
-0.4482163191,
0.163980782,
0.233834818,
-0.5422616005,
-0.0117076039,
-0.015652854,
-0.0110325562,
0.0200102292,
0.1657969207,
-0.3622570932,
0.0428358838,
-0.3063445389,
0.2600966692,
0.3495649993,
-0.0053010285,
-0.2179080248,
0.087969169,
-0.1191325337,
0.041228205,
0.4386927485,
0.4519638419,
0.2561975121,
0.0754779875,
0.4115337729,
0.3871441782,
-0.25565961,
-0.1236730739,
-0.0714350492,
-0.3956528306,
0.359360069,
-0.0549359843,
-0.2022868991,
0.1958159208,
0.2372691929,
-0.6348540783,
0.1601855457,
-0.2363997102,
0.2605082393,
-0.1372979879,
0.5229121447,
-0.0187943354,
-0.1351220608,
-0.1498903632,
-0.476846844,
0.0531752929,
-0.2135917842,
0.0618376695,
-0.2344727069,
-0.3581312895,
0.245933339,
0.1108097136,
0.0520031974,
0.1000968218,
0.1329472065,
-0.0339877717,
0.0112844631,
-0.3725004196,
-0.2047202438,
0.1839381307,
0.1429639012,
-0.0163459294,
-0.0828609318,
0.3200977445,
-0.0363276079,
-0.132556051,
-0.0430019833,
0.1616273671,
-0.0701391846,
0.1692527235,
0.0912618488,
0.3741364479,
-0.203127861,
0.1750641763,
0.1850978285,
0.4629648328,
0.2235041857,
0.2856939435,
0.0593160018,
-0.4190937579,
0.1983950436,
-0.2558090091,
-0.2348175347,
0.3836273551,
0.0461269319,
0.1875650436,
0.185980469,
-0.059832301,
0.227047205,
-0.0241067782,
-0.1483355761,
-0.2353375852,
-0.0343405753,
0.1370094121,
0.0364805385,
0.363122642,
-0.31831792,
0.2373883873,
0.2140042335,
0.0234405324,
0.1837717146,
0.2534230351,
-0.0868164003,
-0.3094965816,
0.0938749388,
0.1604813784,
0.5992062688,
0.4242343307,
-0.0878362432,
-0.0488776788,
0.0394223221,
-0.0914465785,
0.2606402338,
0.0424241871,
0.1299593598,
0.3358278275,
0.508439064,
0.1565906405,
-0.1096974611,
-0.1188921258,
0.1821381897,
-0.1932060122,
-0.054929547,
-0.18518354,
-0.2611817718,
-0.0383877158,
-0.1816256493,
-0.1657209396,
-0.207585007,
-0.097721234,
0.2218883336,
0.3331530988,
-0.1846586913,
0.1372748315,
0.1484768391,
0.6496678591,
-0.276309669,
-0.4283194542,
-0.1805222332,
-0.1743448973,
0.1853332818,
0.061453253,
0.3481793404,
-0.045912344,
0.4973228276,
0.4200584292,
-0.0873127729,
-0.365924567,
-0.406437695,
0.0658802092,
0.0051847426,
0.1079394221,
0.0827778205,
-0.0497720577,
0.1289796084,
-0.5469329953,
0.0883529335,
0.2471524775,
-0.3253920078,
0.1705883145,
0.0624136627,
-0.1377744526,
-0.3601433635,
-0.1490685046,
-0.2621833682,
-0.5986359119,
0.4414025247,
-0.1484773904,
0.180031687,
0.0249651968,
0.1966512054,
-0.4081294239,
0.021418158,
0.2013934255,
-0.124220632,
-0.1794523001,
0.2523637116,
-0.1556439102,
-0.0787100196,
-0.1698462367,
-0.2005186081,
0.0577692986,
0.563129425,
-0.1712462902,
-0.2078758925,
-0.0578833669,
0.4450134039,
-0.0747941211,
0.014164323,
0.5879992843,
0.5294591188,
-0.1370871067,
-0.1355694532,
-0.1235171929,
0.1481077373,
0.2665315866,
-0.0376131088,
0.1700481325,
-0.0697884411,
-0.1092707664,
0.3424433172,
-0.0055069327,
0.0933607668,
0.4374441206,
0.1526982039,
0.3345213532,
-0.2667293251,
-0.0711629912,
-0.2311913669,
-0.4024606347,
-0.0510339774,
0.0386338681,
0.0045275725,
-0.3420060873,
-0.055498749,
0.2250030935,
-0.1947952062,
-0.2088431716,
0.3794308305,
-0.137475878,
0.380207777,
-0.0168419667,
-0.46713081,
-0.3867400289,
0.0967370719,
-0.0300534423,
0.1141521186,
0.2807099223,
-0.2234766483,
-0.4105625749,
-0.3334211409,
-0.4103843272,
0.0940466821,
0.2343838513,
0.1793727726,
0.0772436708,
-0.3287857771,
-0.1084488034,
0.0249275342,
0.7993979454,
-0.119070299,
-0.4681961536,
0.0222503468,
-0.027018059,
0.2350581288,
0.1083082855,
0.1514732838,
0.1005421132,
-0.047065597,
0.3932630122,
-0.222791627,
0.107361719,
-0.0357364751,
0.2548593581,
0.0443054959,
-0.2726072073,
0.0777117461,
0.0755955204,
-0.4649638236,
-0.478792429,
0.0730198845,
0.138793081,
0.156154722,
-0.1237869263,
-0.1623480469,
-0.0037985221,
0.0645300075,
-0.0983880311,
0.237167269,
0.0146074928,
0.340941906,
-0.2411068231,
-0.0266856849,
0.6412109733,
0.4557241797,
-0.3451504409,
-0.1704427898,
-0.1250293553,
-0.2065744996,
0.3252629638,
0.0364192016,
0.178078115,
0.5699096918,
0.04099565,
0.0123616345,
-0.5910300612,
-0.1712290794,
-0.1487498283,
-0.0998117104,
-0.3636201024,
-0.3279722035,
0.4662220478,
0.2714467347,
-0.238114357,
0.4522877336,
-0.1000938788,
-0.4977032542,
0.4367557764,
0.0994019136,
1.1413234472,
0.1355957687,
0.3166125119,
-0.1556840539,
-0.2941757739,
0.0338790752,
-0.1563635916,
0.3443220556,
-0.0829760432,
-0.160997659,
-0.1484384537,
-0.1936965436,
-0.0601994246,
-0.0071584322,
-0.5996420979,
-0.0995765775,
0.0189619064,
0.0549630485,
0.0502855778,
0.1950851083,
-0.2405221164,
-0.4536107779,
-0.0444340594,
0.043085549,
0.1018241346,
-0.3034267724,
-0.0788566172,
-0.2548259497,
-0.0428846367,
0.0972738117,
-0.2213404477,
-0.2444440722,
-0.2668028772,
0.2151145041,
0.0063613765,
-0.1317189634,
-0.0095417053,
-0.0237920545,
-0.1792589277,
0.0970389694,
-0.1027620807,
-0.1221287027,
0.2026908845,
0.0986105055,
-0.1388661563,
-0.3898613453,
0.4284770191,
0.1266144514,
-0.038052544,
-0.0174512789,
-0.1661408246,
-0.4061591327,
-0.1345536113,
-0.0337519646,
0.2091446817,
-0.1184433401,
-0.2625828981,
0.3084501922,
-0.0782310292,
-0.2305267155,
0.1201271117,
0.1115122885,
-0.0979481116,
0.2626575232,
-0.0174357295,
-0.0027792901,
-0.1698484421,
0.0950915515,
0.2190248519,
0.1056210995,
0.1143996492,
-0.0466690175,
-0.0904562548,
-0.1804322004,
-0.0006058142,
0.0840086341,
-0.0884604529,
0.0542774722,
0.1276235282,
-0.2729559839,
-0.0776808038,
0.2229893804,
0.3230029047,
0.4300009608,
-0.1942077279,
-0.2773550749,
-0.4286994934,
0.356061399,
0.0843130797,
0.3598293662,
0.0427449644,
0.1024754047,
-0.1521522403,
0.4299739003,
-0.3599180877,
-0.2498048842,
-0.0624087453,
0.1829364598,
0.0389305949,
0.1268908381,
-0.1238598526,
-0.1420402378,
0.0421582833,
0.1408631653,
-0.0729794204,
-0.1976671666,
0.0948440582,
0.1418450922,
0.1906208545,
0.1001479179,
-0.2256941348,
-0.3016276658,
-0.0575755164,
-0.3511756957,
0.1035140827,
-0.0092987269,
-0.3125888705,
-0.1604619175,
0.3754343092,
0.0410336182,
-0.1293419302,
0.3695025742,
0.088318646,
0.0355709195,
-0.0848019496,
0.1509501338,
-0.4421796799,
-0.0200396851,
-0.000544589,
0.3523771167,
0.1863223463,
-0.0164089408,
0.2933139801,
0.3991461396,
-0.2534356117,
0.0088067874,
0.2165990025,
0.1170319989,
0.0714501739,
0.2712442279,
0.2000563741,
0.2315376997,
-0.3683590293,
-0.3234625459,
-0.1228902638,
-0.2538639307,
0.289857626,
0.1271504611,
0.167693004,
0.0901787579,
-0.1946226656,
0.1269266307,
0.0460952669,
-0.0471629128,
0.1355333626,
0.0782917291,
0.1367094815,
-0.001041472,
0.090362601,
0.2777534425,
0.1658719182,
0.0078187343,
-0.0568726771,
0.4941706061,
0.3845893741,
-0.0948341191,
0.2010030746,
-0.0306094848,
0.2637597322,
0.2859853208,
0.2809198499,
0.1700624377,
0.2751067579,
0.1701166332,
-0.0839606076,
-0.3715267479,
0.291942209,
0.1412463039,
0.01876509,
-0.028315682,
-0.1185654476,
-0.2450383902,
-0.1537511945,
-0.1427407265,
0.1537353545,
-0.184352994,
0.5169634223,
0.0133688943,
0.3105065823,
-0.1622341722,
0.3076370955,
0.382340312,
0.1365793645,
-0.3617659807,
-0.0999964997,
0.1555229127,
-0.2377638072,
0.1416984648,
0.1683665663,
0.4571425915,
0.4111610651,
0.0801917464,
0.1701585352,
-0.1596922576,
-0.0086721033,
-0.1871577799,
-0.0559564121,
0.2579295635,
-0.0538846925,
0.1439668089,
0.1886512786,
-0.0323324427,
0.2839422226,
0.0560534969,
-0.0276474487,
-0.3535072505,
-0.0604579486,
0.5578277707,
-0.5780315399,
0.3011537492,
-0.0372193865,
-0.1526597142,
-0.0206110738,
0.4134004414,
-0.1001086831,
0.0743934214,
0.0755791515,
0.0957455933,
0.2858553231,
0.3127806485,
0.3436603248,
0.1175784916,
-0.3295322955,
-0.1820980608,
-0.255374819,
-0.1938429475,
0.0749248266,
-0.1611099541,
0.0014272965,
0.0788862929,
0.3104850352,
0.1100080088,
-0.0542922094,
-0.2407805771,
-0.0073100813,
0.1459168196,
-0.2963374555,
-0.0806995556,
0.228194356,
-0.0663476139,
0.0275354385,
-0.2774802446,
0.0943289995,
0.1722064614,
0.083730109,
-0.3141049445,
0.0166980773,
0.1884005219,
0.1818258315,
0.419016391,
0.1478976756,
0.2240952551,
-0.2689062655,
0.072251685,
0.1609318554,
-0.2032799125,
-0.3194251359,
0.0030531939,
0.0187072773,
-0.0467037596,
-0.5168741345,
-0.1386771947,
-0.1059471071,
0.0141492151,
-0.1697207689,
-0.0443771966,
-0.2834770679,
0.0758287832,
0.0713373497,
-0.074320212,
-0.0535411015,
0.2164404988,
0.1623055786,
0.0218602717,
-0.0475975424,
-0.2136846632,
0.0684084445,
0.1741073728,
-0.0640861019,
-0.0976743251,
-0.0648061112,
-0.1147765666,
0.3184700608,
-0.7344163656,
-0.3051046133,
0.2667625248,
-0.1206688806,
0.0448516458,
0.4788627923,
0.0444675162,
0.0331964344,
-0.3538217843,
0.2330825925,
-0.1090684459,
-0.3129501045,
-0.1438241452,
-0.5232961774
] |
https://github.com/huggingface/datasets/issues/1877 | Allow concatenation of both in-memory and on-disk datasets | > Can you explain where the issue of the double memory may arise?
We have to keep each block (in-memory vs memory mapped) separated in order to be able to reload them with pickle.
On the other hand we also need to have the full table from mixed in-memory and memory mapped data in order to iterate or extract data conveniently. That means that each block is accessible twice: once in the full table, and once in the separated blocks. But since pyarrow tables concatenation doesn't double the memory, then building the full table doesn't cost memory which is what we want :)
> Also, why is the existing concatenate_datasets not sufficient for this purpose?
The existing `concatenate_datasets` doesn't support having both in-memory and memory mapped data together (there's no fancy block separation logic). It works for datasets fully in-memory or fully memory mapped but not a mix of the two.
> Will the add_item feature also help with lazy writing (or no caching) during map/filter?
It will enable the implementation of the fast, masked filter from this discussion: https://github.com/huggingface/datasets/issues/1949
However I don't think this will affect map. | This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova | 188 | Allow concatenation of both in-memory and on-disk datasets
This is a prerequisite for the addition of the `add_item` feature (see #1870).
Currently there is one assumption that we would need to change: a dataset is either fully in memory (dataset._data_files is empty), or the dataset can be reloaded from disk (using the dataset._data_files).
This assumption is used for pickling for example:
- in-memory dataset can just be pickled/unpickled in-memory
- on-disk dataset can be unloaded to only keep the filepaths when pickling, and then reloaded from the disk when unpickling
Maybe let's have a design that allows a Dataset to have a Table that can be rebuilt from heterogenous sources like in-memory tables or on-disk tables ? This could also be further extended in the future
One idea would be to define a list of sources and each source implements a way to reload its corresponding pyarrow Table.
Then the dataset would be the concatenation of all these tables.
Depending on the source type, the serialization using pickle would be different. In-memory data would be copied while on-disk data would simply be replaced by the path to these data.
If you have some ideas you would like to share about the design/API feel free to do so :)
cc @albertvillanova
> Can you explain where the issue of the double memory may arise?
We have to keep each block (in-memory vs memory mapped) separated in order to be able to reload them with pickle.
On the other hand we also need to have the full table from mixed in-memory and memory mapped data in order to iterate or extract data conveniently. That means that each block is accessible twice: once in the full table, and once in the separated blocks. But since pyarrow tables concatenation doesn't double the memory, then building the full table doesn't cost memory which is what we want :)
> Also, why is the existing concatenate_datasets not sufficient for this purpose?
The existing `concatenate_datasets` doesn't support having both in-memory and memory mapped data together (there's no fancy block separation logic). It works for datasets fully in-memory or fully memory mapped but not a mix of the two.
> Will the add_item feature also help with lazy writing (or no caching) during map/filter?
It will enable the implementation of the fast, masked filter from this discussion: https://github.com/huggingface/datasets/issues/1949
However I don't think this will affect map. | [
-0.4205481708,
0.1177791953,
-0.0135112107,
0.2829377949,
0.0721302778,
0.093980059,
-0.1494366378,
0.2320161015,
-0.1620395929,
0.1522656381,
0.0206327885,
0.5102052689,
-0.0248011276,
0.3577521741,
0.1537402868,
-0.0333432965,
0.1533056498,
0.2407621145,
-0.4325166345,
0.2023286223,
-0.2718375623,
-0.2322960049,
-0.1211220846,
-0.4407117069,
-0.1328162402,
0.105893001,
-0.3701848686,
0.085054189,
-0.461260885,
-0.4172993898,
0.0272536129,
0.2392501533,
0.1475352496,
0.020442266,
-0.0001143608,
-0.0304895937,
-0.0908041298,
-0.143208921,
-0.4931336641,
-0.1440989822,
-0.3029802144,
-0.4612714052,
-0.001470156,
-0.0359220281,
0.2839405537,
-0.3497852683,
0.0704568699,
0.0037822053,
-0.0004687868,
-0.0690575689,
0.1282016337,
0.0832637399,
0.2490190864,
-0.1535839885,
0.2830590308,
0.4180068374,
-0.2433201671,
0.0545795672,
0.483710289,
-0.229780674,
0.1604690105,
-0.1855497956,
-0.1114507467,
-0.1264530271,
0.3167589903,
0.2640161812,
-0.2564167678,
-0.1504637301,
-0.0628417283,
0.2610206902,
0.4842345417,
-0.4602854252,
-0.3712320328,
-0.3262186646,
0.16957362,
-0.29652071,
0.1417031139,
0.2897108495,
-0.1246988922,
0.1815573871,
-0.0053669102,
-0.3174613714,
-0.1614488363,
0.2725503743,
0.2950288653,
0.2683962286,
0.3663905263,
0.1452645063,
0.2541911602,
0.0573605448,
0.2632885873,
-0.1695788205,
-0.2025771588,
0.141052261,
-0.2625716031,
-0.0054383203,
0.0481507108,
-0.1600819528,
0.5918856859,
0.0995218605,
0.2872606218,
0.0227827691,
-0.2712805867,
0.323821038,
0.1234031618,
0.1842762232,
-0.2756659389,
0.1439843476,
-0.0581968203,
-0.1266409159,
0.2939160168,
0.015191501,
-0.1816809177,
-0.0323899798,
0.1392949969,
-0.1682103872,
-0.1076435745,
0.2026000917,
0.046782583,
-0.2848707736,
-0.1391759813,
-0.1701775789,
0.0513956323,
0.0725709051,
0.1154177934,
0.011062365,
0.0228985399,
0.3052881658,
0.3008717895,
-0.1711030304,
0.0266159736,
0.0167590454,
-0.3064672947,
0.3657281399,
0.3710740209,
0.0442398265,
-0.1348830909,
0.2717688084,
-0.0841673911,
0.3043231964,
-0.0053497925,
-0.1916877031,
0.3680021167,
0.1365815401,
0.0408081263,
-0.4994349778,
-0.0858417898,
-0.0360958129,
-0.2986931205,
0.6121867895,
-0.085869059,
-0.356138289,
-0.4303495288,
0.0681184083,
-0.0598042943,
-0.01372582,
-0.2639258504,
0.5533695817,
0.3309733272,
-0.2825149298,
0.0409157649,
0.1475727856,
-0.1118465513,
-0.5068644881,
0.2198457718,
0.2586517036,
-0.5528045893,
0.0585639328,
0.1289945245,
-0.0345777459,
0.111178726,
0.2120649964,
-0.3973548114,
0.045850236,
-0.3841031194,
0.2939403057,
0.1699151248,
0.0413727909,
-0.2576792836,
-0.1045227796,
-0.0269713774,
0.0448916033,
0.4616473317,
0.4100906551,
0.0950641781,
0.1630869806,
0.3194943964,
0.3140043616,
-0.2028829306,
-0.1930927336,
-0.0914327279,
-0.5326013565,
0.5056712031,
-0.0021548942,
-0.1473308653,
0.0495728888,
0.2052707821,
-0.521707952,
0.3012335002,
-0.2306773365,
0.2859787643,
-0.1803401709,
0.3845080435,
-0.0888176411,
-0.1506258547,
-0.1423925757,
-0.5311308503,
0.2147482038,
-0.2913578451,
0.1576972455,
-0.2815651894,
-0.2635594606,
0.2890248299,
0.0733938143,
-0.0002411995,
0.0676863566,
0.0810717717,
0.0158446487,
0.0758400559,
-0.3457560837,
-0.2225682735,
0.1293765157,
0.1590715647,
-0.0178134404,
-0.1510962099,
0.3943971097,
0.0196055882,
-0.1607126594,
-0.0763068199,
0.1959540695,
-0.0492700227,
0.1961109042,
0.119996801,
0.426999867,
-0.1109359413,
0.0082426704,
0.0712701976,
0.2727443576,
0.1453540921,
0.2931611836,
0.029671913,
-0.5536990166,
0.1499249041,
-0.1731538773,
-0.090544939,
0.2890940011,
0.1202576011,
0.1383825541,
0.1417950392,
-0.1665381789,
0.1052103266,
-0.0745430663,
0.0005405433,
-0.2177431136,
-0.0516587459,
0.2462766469,
0.1018996835,
0.313315928,
-0.4512613118,
0.2090352923,
0.2587775588,
0.0302926749,
0.1629490852,
0.1270131022,
-0.1180140376,
-0.2705498338,
0.0862551555,
0.1480767429,
0.6196273565,
0.3672585487,
-0.0011290591,
-0.0006008968,
-0.0432568602,
-0.099576503,
0.2032074481,
-0.0496496558,
0.2204274237,
0.311162591,
0.4858121276,
0.0462776199,
-0.0847223774,
0.0187399387,
0.242454946,
-0.023654623,
-0.0614497922,
-0.2882324457,
-0.4230682254,
-0.0768891126,
-0.2263133526,
-0.2073232085,
-0.1776858568,
-0.0387444496,
0.2044094205,
0.3083998263,
-0.1329578757,
0.1332467049,
0.1385301799,
0.4870513976,
-0.3491360247,
-0.3730885088,
-0.107597284,
-0.0817747861,
0.0371585861,
0.0833982751,
0.4131776094,
-0.1326896697,
0.443946749,
0.4218164384,
-0.1286579072,
-0.3170167208,
-0.3468725085,
0.125269562,
-0.0391195938,
0.0798843801,
0.0871992335,
-0.0926858112,
0.1265864223,
-0.4788064659,
0.1119152382,
0.3859834671,
-0.2844461799,
0.2566557527,
0.0273776762,
-0.1551005542,
-0.3382601738,
-0.2619176507,
-0.3027264178,
-0.4580032527,
0.4380856752,
-0.1251060367,
0.1152244657,
0.0953675359,
0.2327350825,
-0.2977922261,
0.0700942576,
0.1275105774,
-0.1762500852,
-0.1304357946,
0.3235553801,
-0.1357904971,
-0.0962291434,
-0.16813007,
-0.1917956173,
0.1267278194,
0.4574132562,
-0.21457991,
-0.0556683987,
-0.1127311811,
0.4080158174,
-0.1579322368,
0.0697520822,
0.6263286471,
0.5570223331,
-0.0811091959,
-0.0401121937,
-0.1109491289,
0.2213412821,
0.2556411922,
0.1116931215,
0.1188664958,
-0.1989122629,
0.0869155228,
0.372387737,
-0.0004664361,
0.0835758373,
0.4383435249,
-0.0265468359,
0.2976454496,
-0.2404049635,
-0.0903692842,
-0.2160645276,
-0.4245314002,
-0.0316983685,
-0.0570846722,
-0.0455009416,
-0.2517732978,
-0.0882366374,
0.231711939,
-0.1693065763,
-0.2507644892,
0.3463310897,
-0.119084537,
0.2554129362,
0.0230745822,
-0.4499954283,
-0.2813349366,
0.0694618523,
-0.0932987109,
0.0908188671,
0.155724138,
-0.1524041593,
-0.325556308,
-0.3939988613,
-0.3772606552,
0.1050976813,
0.3018628657,
0.1467110217,
0.1842961758,
-0.3467969,
-0.0701860189,
-0.0196162704,
0.8354031444,
0.0623186752,
-0.5227495432,
0.07037054,
0.1317286044,
0.2117172182,
0.0114878193,
0.1598885208,
0.0968084037,
-0.2083497494,
0.1697025299,
-0.4544526041,
0.1512105167,
0.0658321604,
0.4569221735,
0.0042451695,
-0.3004598618,
-0.068563506,
0.0551141575,
-0.4431815147,
-0.4144231975,
0.0910972953,
0.1811732501,
0.0586871654,
-0.0761779025,
-0.1391137242,
-0.0078496188,
0.0387606919,
-0.0608310401,
0.3464484811,
0.0200293809,
0.4484293461,
-0.2765004337,
-0.0650266409,
0.8608855605,
0.6849471331,
-0.2380896211,
-0.1312048584,
-0.0917248428,
-0.1726326495,
0.3094540238,
0.0511382483,
0.0947384536,
0.4905813336,
0.0457090102,
0.0370520987,
-0.4452129006,
-0.1305462867,
-0.090906918,
-0.2205589712,
-0.3609578907,
-0.3468238413,
0.3938201964,
0.17529805,
-0.1744602919,
0.4534806609,
0.0231147464,
-0.4244268537,
0.2573761344,
0.0153105073,
1.1907244921,
0.1039268672,
0.3567137718,
0.1535402536,
-0.2278426588,
0.0615460686,
-0.0887163356,
0.2979048491,
-0.1494075507,
-0.0145544112,
-0.0680170059,
-0.210012719,
-0.0067497771,
0.0345538184,
-0.4873149991,
-0.1321269274,
-0.0438017473,
-0.0675916225,
-0.0297098607,
0.1946684867,
-0.2373502403,
-0.339989543,
0.1423702836,
0.020314049,
0.0962793678,
-0.4308604598,
-0.0875315964,
-0.1844052076,
-0.194804877,
0.0438082963,
-0.2976351976,
-0.1206340492,
-0.3254966438,
0.3182812631,
-0.0280274712,
-0.1441755295,
-0.076507844,
0.0056235744,
-0.1649672091,
-0.0034902168,
-0.1046124995,
-0.120543547,
0.1330457032,
-0.0177099817,
-0.1660438925,
-0.433342129,
0.4511222839,
0.0919253677,
-0.2152075917,
-0.0488366894,
-0.1193017811,
-0.347933948,
-0.1846933216,
0.0115165636,
0.1461814642,
-0.2090688497,
-0.2435186207,
0.403971225,
0.1063709408,
-0.2542294562,
0.0761181936,
0.2311346382,
-0.0539383925,
0.1534841955,
-0.1724502593,
-0.0014168769,
-0.2240435779,
0.0823870599,
0.3592757285,
0.1005133986,
0.1333334297,
0.0378353,
-0.0127529055,
-0.038926512,
0.0005595386,
0.1835715771,
-0.1900925189,
-0.0854352862,
0.1291293055,
-0.0664900541,
-0.1210848764,
0.1926626861,
0.3050108254,
0.3546999693,
-0.0487309583,
-0.1518025845,
-0.4447177052,
0.4528211355,
0.1614198238,
0.3631178439,
0.0886075944,
0.3154398799,
-0.0791260898,
0.3983821869,
-0.2913196087,
-0.1211582795,
-0.0197573975,
0.1517444849,
-0.0519762039,
0.1548066735,
-0.0444977917,
-0.161250338,
0.0085243583,
0.0477796271,
-0.0568660088,
-0.1307120323,
-0.0081737116,
0.1341316402,
0.3104194999,
0.15466398,
-0.2584871352,
-0.4188140333,
0.0273017175,
-0.2525814474,
0.0860091373,
0.0694737285,
-0.2653083801,
0.0038917325,
0.469098419,
0.0464667082,
-0.1550284028,
0.3445218801,
0.0580565482,
0.1323040128,
0.0329740867,
0.2235347778,
-0.3659914136,
-0.0090111122,
-0.1355906129,
0.2857983708,
0.2216771394,
-0.0888827071,
0.1399483681,
0.2560238838,
-0.2851625681,
-0.224572897,
0.1921325177,
0.2152835727,
0.0877308697,
0.2562375963,
0.2321072221,
0.1659257859,
-0.3981328309,
-0.1963466853,
-0.283714205,
-0.252707392,
0.2140800059,
0.2647173703,
0.1927172542,
0.1320437342,
-0.2951048613,
0.0354305133,
0.1063152552,
-0.0030815694,
-0.0055839568,
0.0424553826,
-0.0371845812,
-0.0206306316,
-0.0071209706,
0.2337071896,
0.1869411767,
0.0185632054,
-0.1811138242,
0.4842148721,
0.2601225674,
-0.1952260733,
0.1806908697,
-0.024880223,
0.2080720663,
0.4031640291,
0.2998487651,
-0.0317452699,
0.1795700192,
0.4649271369,
0.0617627278,
-0.3081248701,
0.306102097,
0.1629643589,
0.0695118159,
0.0276518948,
-0.0616922602,
-0.2765994668,
-0.2785303593,
-0.1142995209,
0.1149339974,
-0.3045383692,
0.5424284339,
-0.0381481908,
0.1117519066,
-0.2194079161,
0.3140153289,
0.5822797418,
0.0182734076,
-0.3843542933,
-0.0587346777,
0.3683795333,
-0.380605042,
0.0478151254,
0.2346356064,
0.5222941041,
0.3223418891,
0.1351947337,
0.0563003421,
-0.0919187665,
0.0474995449,
-0.2079743445,
-0.0770852119,
0.2428857535,
0.0241843089,
0.1443094909,
0.1389950514,
-0.0102109481,
0.2100461125,
0.1940664053,
0.051765535,
-0.4397466183,
-0.152254492,
0.387455523,
-0.5414434671,
0.4035975933,
0.0093685724,
-0.2633354664,
0.0363723822,
0.3410356939,
-0.0467890054,
0.0430985391,
0.0376729704,
0.0665064082,
0.3350122869,
0.2623304129,
0.2968885303,
0.1776209772,
-0.3730182052,
-0.2074374557,
-0.3266671896,
-0.1075484753,
-0.0852781683,
-0.213211745,
-0.0943393111,
0.1443772614,
0.2860147655,
0.1380279511,
0.0169366077,
-0.0303847399,
-0.0389750376,
0.2556590438,
-0.2881510854,
-0.1359733194,
0.3313159049,
0.0026475023,
0.0365238637,
-0.2754091024,
0.1030919552,
0.1382572651,
-0.0082947388,
-0.3732227385,
0.0620639324,
0.3114856482,
0.3021299839,
0.6179404855,
0.1816812158,
0.2282119989,
-0.2298468798,
-0.0382053927,
0.0162862949,
-0.2088394314,
-0.2918353081,
0.0776325911,
0.0476222932,
-0.063734293,
-0.4909755886,
-0.1343867034,
-0.0859149694,
0.0522648543,
-0.1825255752,
0.1082528085,
-0.2580527961,
0.0688757673,
0.1717405468,
0.0054739788,
-0.0727895796,
0.1789419353,
0.1221987158,
0.0560631007,
-0.1198984236,
-0.2423340082,
0.0153740272,
0.2471013516,
-0.0438114777,
-0.0469027385,
-0.1139090806,
-0.0990877077,
0.2313877195,
-0.8545031548,
-0.2810122073,
0.2284172177,
0.0029593166,
-0.0042201504,
0.4445117712,
0.0436204635,
-0.0501917787,
-0.2877588868,
0.3475425243,
-0.1791467369,
-0.4259860516,
-0.1129207611,
-0.4514326453
] |
https://github.com/huggingface/datasets/issues/1876 | load_dataset("multi_woz_v22") NonMatchingChecksumError | Thanks for reporting !
This is due to the changes made in the data files in the multiwoz repo: https://github.com/budzianowski/multiwoz/pull/59
I'm opening a PR to update the checksums of the data files. | Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
| 32 | load_dataset("multi_woz_v22") NonMatchingChecksumError
Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
Thanks for reporting !
This is due to the changes made in the data files in the multiwoz repo: https://github.com/budzianowski/multiwoz/pull/59
I'm opening a PR to update the checksums of the data files. | [
-0.2244368196,
0.1764704883,
-0.0287194811,
0.1495723128,
0.1895723641,
0.0018984303,
0.365057677,
0.4820295274,
0.2514897585,
0.1666065454,
-0.0970347077,
0.1781595647,
-0.0822443962,
0.0992017984,
-0.184549123,
0.2732407451,
0.0765194818,
-0.0256413147,
-0.1682778746,
-0.0625797957,
-0.2430099845,
0.3292944431,
-0.15932706,
-0.0338404477,
-0.3486858606,
0.1303854585,
0.1739082336,
0.3362389207,
-0.0838697553,
-0.1523165703,
0.4145023823,
0.2369778305,
0.0439622179,
0.3686153591,
-0.0001172875,
0.1633386761,
0.2778438926,
-0.1433111429,
-0.1339642555,
-0.1817750186,
-0.5618034005,
-0.3552265167,
-0.1035683453,
-0.1154624894,
0.0706324726,
0.3033827245,
0.0572006106,
-0.0867422074,
0.2431904078,
0.1712264121,
0.1535670012,
0.3319172263,
0.2157701701,
0.0969050154,
0.0455293655,
-0.0774404109,
-0.1376800686,
0.0922585353,
0.2744396627,
-0.1569580734,
-0.2395651937,
0.2651385069,
-0.1804821938,
0.2693834603,
-0.0804270655,
-0.1510989368,
0.1479790807,
-0.1058239341,
0.2105876207,
0.5727722645,
0.2205697298,
-0.0554817878,
-0.1762657166,
-0.0319501609,
-0.2201737463,
0.0833649933,
0.3634192348,
0.1834742129,
-0.207865119,
0.0045872135,
-0.4487924278,
0.1443090737,
-0.0263799056,
0.2792591453,
0.2121901512,
0.3145954609,
0.2003568113,
0.0726619512,
0.227932483,
-0.2064420432,
0.1992098391,
-0.3010314107,
-0.1609570384,
0.1293215156,
-0.6801480651,
-0.0664965659,
0.0034522209,
0.3319551945,
0.44042629,
0.2289482802,
0.4523438513,
0.3736303449,
-0.1520799994,
0.1170620397,
0.3968932331,
0.1744448096,
0.0657050908,
-0.0545443483,
0.169239372,
0.2095876634,
-0.0953853205,
0.2373775989,
-0.030504927,
-0.3612298965,
0.3493167162,
0.1929920912,
0.2069613338,
-0.3434550762,
-0.1813287735,
0.0824042261,
-0.2454787642,
-0.1945313215,
0.271803081,
0.2862131596,
-0.0944883525,
0.5524675846,
-0.1653459072,
0.2156014144,
-0.1538059115,
-0.1847017407,
-0.257340163,
-0.1516558677,
0.073074013,
0.1531951427,
0.2943667769,
0.0656544417,
0.3595135212,
-0.0321133882,
0.3835367262,
-0.203120932,
0.3335245848,
0.0902097076,
-0.0699193627,
0.2554132044,
0.0134760961,
0.2010273635,
0.2036006153,
-0.0260661468,
-0.0622884706,
0.2220197171,
-0.3325723708,
-0.6167244911,
0.1093564481,
0.1607630253,
-0.3595764637,
-0.080491595,
0.0163278282,
-0.2801305056,
0.3050758839,
-0.4042750001,
-0.1545271724,
-0.3094390929,
-0.043020837,
-0.1179553494,
0.009863317,
0.0942920446,
-0.0934404433,
0.0649285913,
-0.0455776751,
-0.1832662672,
0.3468406796,
0.2667629123,
-0.2780412138,
-0.164542973,
-0.1394765228,
0.018730402,
0.2207068801,
-0.0911148638,
-0.3638111949,
0.2024420947,
-0.0716659203,
0.6232898235,
0.1856183112,
0.1097930819,
-0.3473640084,
-0.1108141243,
0.2366856933,
0.0823843926,
0.1350483447,
0.1764027029,
-0.2450099289,
-0.3122073114,
0.3476398885,
0.152895391,
0.0161079578,
-0.2277660668,
0.0021594316,
0.0893482119,
0.5888451338,
-0.2297042161,
-0.0971518904,
0.1738088727,
0.0959680974,
-0.2113370895,
0.0982793048,
0.0511621833,
-0.4866254926,
0.3225808442,
-0.156579718,
-0.0987990499,
0.3440407217,
-0.0598258227,
-0.0504282825,
-0.1761432588,
-0.3390327394,
0.0107524041,
0.080948852,
0.2748897076,
0.0834967047,
-0.2535200417,
-0.0927834511,
0.3571130335,
-0.3864996731,
-0.0857933909,
-0.4520959854,
0.1969178319,
-0.0879132226,
0.0851731449,
0.076490216,
0.1571477056,
0.1050595567,
-0.109420836,
-0.1870693415,
0.5129071474,
0.3213776052,
0.2384278327,
-0.0831714272,
0.3510249257,
-0.1519881785,
-0.049867142,
-0.1007989347,
0.5829588771,
-0.043825224,
-0.2062133551,
-0.1941186786,
0.4879204929,
-0.1729984134,
0.4535029829,
-0.0467058271,
0.1363263428,
0.3366039097,
-0.1314038932,
-0.2676326036,
-0.0946317986,
0.4545116723,
-0.1980917752,
-0.1669614613,
0.2530012131,
-0.2133248001,
0.0163291171,
0.3733876348,
-0.1440780461,
0.125990063,
-0.0151445484,
0.2458736002,
-0.2552569509,
-0.0198076405,
0.4922712445,
0.3632512689,
0.1011830121,
-0.0179459285,
-0.0273174345,
-0.229469195,
-0.1744434237,
0.119406268,
0.0973704606,
0.3296887279,
0.5370063782,
0.0413939059,
-0.3653717637,
-0.4267032146,
-0.0148495585,
0.0290298238,
0.1656201035,
-0.4414883256,
-0.2537639141,
-0.444193542,
0.228844732,
-0.3637366593,
-0.3436978161,
-0.4129475951,
-0.4648977816,
0.1210603118,
0.2281652242,
-0.0199533589,
0.3193201125,
-0.3111635447,
0.0683453232,
-0.1319281757,
0.060542874,
0.040054135,
-0.2777513266,
-0.0817122459,
-0.0046347044,
0.395745188,
0.0018831939,
-0.0578057617,
-0.307749927,
-0.0692628324,
-0.1458403319,
-0.1667574048,
-0.0308502838,
-0.110812515,
-0.0727641061,
0.2588977218,
0.302515626,
0.1404629052,
-0.3886935115,
0.2458182722,
-0.036740981,
-0.2834915817,
0.1877617091,
-0.1721365154,
-0.0189697687,
0.1081126034,
-0.3866265118,
-0.0947579294,
-0.3375903666,
-0.0101160184,
0.0664088875,
0.2717182934,
0.2532510161,
0.0835986882,
-0.0678554252,
0.1356462687,
0.2740627229,
-0.3200552762,
-0.8128378987,
0.4206299484,
-0.0225659702,
-0.2601570487,
-0.0509395339,
0.0214567855,
0.2018917948,
0.096438095,
-0.4472934008,
-0.2804217637,
-0.0651813447,
-0.3137296736,
0.1682510674,
-0.3037388027,
0.3359623253,
0.1188272834,
-0.0578323714,
-0.2449942231,
-0.2949228287,
0.3670243919,
0.0858880132,
0.515162468,
-0.2128700763,
0.2413334697,
0.0072190836,
0.5949299335,
0.2273383439,
0.2582990527,
0.0617829636,
0.1863198876,
0.1531589627,
-0.0817791671,
-0.3245138526,
0.1173865721,
-0.2651422322,
0.0044293851,
0.1789300889,
0.0948020071,
-0.2732841074,
-0.1016847715,
0.1297234297,
-0.1512692869,
-0.2311442196,
0.0542183816,
-0.0111722602,
0.0999286771,
0.2145665437,
0.0766037405,
-0.3172060251,
-0.3501292169,
0.1731935292,
0.2011539489,
-0.1653896868,
-0.1424285173,
-0.5950776935,
0.0641577691,
0.0246964507,
0.4123167992,
0.2499672323,
0.4031146765,
0.0229398012,
-0.3206758201,
0.0325472727,
-0.2579707503,
0.5845398307,
-0.3834360242,
0.3248412907,
0.0977155864,
-0.047182709,
-0.2630653381,
-0.1649410129,
-0.0380524062,
-0.0256653801,
0.3266569972,
0.3074329495,
-0.2269270271,
-0.2858473659,
0.1157698482,
0.3378463387,
-0.2878532708,
-0.168277964,
-0.3097969294,
-0.1108993888,
-0.3571150601,
-0.2201305926,
-0.3995673954,
0.347266078,
-0.1648674011,
-0.0812496543,
0.171175614,
-0.1453641355,
0.2767301798,
0.0399434417,
0.3021827936,
0.3644607365,
0.225110054,
0.0248756781,
0.3385648131,
0.2234017998,
0.4796162844,
-0.1350191534,
-0.0993255973,
-0.0563183874,
-0.2143765688,
0.0410138145,
0.2235025615,
0.0236764513,
-0.0814054161,
0.2965890169,
-0.1108628288,
-0.0433097072,
0.0397991277,
0.2845608294,
0.0299266726,
-0.3399110138,
-0.3018073142,
0.0595080778,
0.0007749796,
-0.1787183881,
0.2506116331,
-0.1772615314,
-0.0745184869,
0.0401595011,
0.3154737055,
0.8823342323,
0.0781966969,
0.088930279,
0.0817514583,
-0.1757193506,
-0.1717555374,
-0.4199718237,
0.1754764766,
-0.2713772655,
-0.2802042365,
-0.0789561421,
-0.2028965652,
0.1161566973,
0.3723819554,
-0.0140839145,
0.3294217587,
0.0507589579,
0.1718658209,
0.0152100958,
0.0928250104,
-0.1202959344,
-0.1144698635,
0.0429765694,
0.1203734577,
-0.1465318501,
0.1447480619,
-0.1230820715,
-0.0494484864,
0.2085238993,
-0.2759684324,
-0.1316171885,
0.1984800994,
-0.1262025535,
0.4007565677,
0.246336937,
-0.2069447041,
-0.2900714874,
0.4346533418,
0.093300581,
-0.0674181432,
-0.2564151287,
0.1446239054,
0.3509521186,
0.0936216116,
-0.0919712782,
-0.0624780692,
0.3511783481,
-0.1872356534,
-0.281373024,
0.0104380567,
-0.1517421603,
-0.6179360151,
-0.068176955,
-0.1264025867,
0.0984781235,
-0.1545959413,
-0.1535262614,
0.178663373,
0.0546398908,
-0.3420566618,
0.1221395358,
0.1330333203,
-0.3626444042,
0.3028177321,
0.129357338,
-0.2770990133,
-0.0944158137,
0.5548005104,
0.0386738293,
-0.0010046363,
0.7191019654,
0.0710802972,
-0.1204013824,
-0.2845616937,
0.3099562228,
0.2341083437,
-0.2122224867,
-0.0682359636,
-0.1181255132,
-0.0549927801,
0.0401895791,
0.5675612688,
0.2766494751,
0.3637249172,
-0.0064761043,
-0.6050378084,
-0.1222514361,
0.0505510792,
0.2656698525,
0.3840140402,
-0.3860803246,
-0.0716591924,
-0.1205711514,
0.2204844505,
-0.2977004647,
0.0136335753,
-0.3014143705,
-0.0014124531,
-0.3319446146,
-0.0595816895,
0.1033879668,
0.046647884,
0.1038635671,
-0.2260659337,
-0.2882828116,
-0.163638413,
-0.2158525586,
0.1587164402,
0.036347691,
-0.0632580519,
0.0698035583,
-0.093967393,
-0.1066645235,
-0.1600659639,
-0.0529935211,
0.208039254,
-0.0953486413,
0.1257883906,
0.2332841158,
-0.1949183792,
-0.1487827599,
-0.2617365718,
-0.2963094115,
-0.0960337296,
0.0908314586,
-0.1021257937,
0.0535944626,
-0.1696587205,
0.4369635582,
0.1701049805,
0.0245453175,
-0.0090561565,
0.2811066806,
-0.2891623378,
-0.0970127657,
0.2284563035,
0.3137276769,
0.3578689396,
-0.2104903609,
0.191317603,
0.0008259676,
0.223886162,
-0.2928494215,
-0.1559720039,
0.3667839468,
-0.1120519936,
0.0143715516,
0.0317224078,
-0.0716922656,
-0.3091430664,
0.5745686889,
0.1000952423,
-0.0016480461,
-0.3915390372,
0.1874296963,
0.3277112842,
-0.3724376857,
-0.0696391463,
-0.0045979545,
-0.2296411991,
0.2119151056,
-0.2145882547,
-0.1123066843,
0.4519321918,
0.0501494557,
0.250569135,
-0.1284567416,
-0.4928979874,
0.094471544,
0.1954618096,
-0.091243051,
0.0177288894,
0.0348487124,
0.4064949751,
-0.2336776853,
0.1827876419,
-0.702688992,
-0.0180880837,
0.0231122524,
-0.2515980303,
-0.348980099,
-0.378718853,
-0.1243418083,
0.162175402,
-0.0076933461,
0.1223286092,
0.3172166944,
0.027880678,
-0.0302278362,
-0.5480283499,
-0.1554401666,
-0.0721405745,
-0.1325325519,
-0.2546499968,
0.2110595703,
0.3817439973,
-0.116866909,
0.341118902,
0.2328886986,
0.3793443739,
0.3968304992,
0.0158853643,
-0.007507205,
-0.0547184721,
0.0275845379,
-0.1511167288,
0.7324590087,
0.1668840051,
-0.1373340935,
0.4767387211,
0.0521658659,
-0.1582672,
0.0700558126,
-0.0142775606,
-0.0927821547,
-0.0598388352,
0.7789543271,
0.038809415,
0.1453847885,
-0.0719825625,
0.233615756,
-0.308211267,
-0.3224420547,
0.0849670321,
0.0910015404,
0.2019015104,
-0.0258246455,
0.073482208,
-0.188868776,
0.6922151446,
-0.0983077735,
-0.2206100821,
-0.2996821105,
-0.2596407235,
-0.8195846081,
0.2501327693,
-0.1412822604,
0.3424732089,
-0.0689982027,
0.1562744379,
0.172787413,
0.2854845822,
0.2575878501,
-0.2129997313,
-0.1863705814,
-0.0521588176,
-0.3973253071,
0.0739046335,
0.4234282374,
0.0349838734,
-0.0069679096,
-0.2355775386,
0.18899858,
0.2631668448,
0.0424306244,
0.1342671663,
0.0111743882,
-0.0327079445,
-0.0246863049,
-0.0812195837,
0.1433466822,
0.5623226166,
-0.003130354,
-0.0114593133,
-0.1989396513,
-0.2911085188,
-0.0620974377,
0.1848046184,
-0.0502038412,
0.5143119693,
-0.1861061603,
-0.0308317896,
-0.2220528126,
0.200448364,
-0.0561482087,
-0.2992594242,
-0.128759712,
0.0325105563,
-0.0127721205,
0.0843318105,
0.1155449972,
0.0505927131,
0.0815806836,
0.2093013227,
-0.0561162457,
-0.0594240874,
0.7326384187,
-0.1133246943,
-0.1869965494,
-0.3649471402,
0.1383292973,
-0.3404733539,
-0.0648734123,
-0.4481022358,
0.0900492966,
0.3190637231,
0.1263612062,
-0.2260793746,
0.1400437802,
-0.257489711,
0.0272711366,
-0.0578921065,
0.0775433406,
0.2296062559,
-0.298792392,
-0.1100094616,
-0.1358180046
] |
https://github.com/huggingface/datasets/issues/1876 | load_dataset("multi_woz_v22") NonMatchingChecksumError | I just merged the fix. It will be available in the new release of `datasets` later today.
You'll be able to get the new version with
```
pip install --upgrade datasets
``` | Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
| 32 | load_dataset("multi_woz_v22") NonMatchingChecksumError
Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
I just merged the fix. It will be available in the new release of `datasets` later today.
You'll be able to get the new version with
```
pip install --upgrade datasets
``` | [
-0.2244368196,
0.1764704883,
-0.0287194811,
0.1495723128,
0.1895723641,
0.0018984303,
0.365057677,
0.4820295274,
0.2514897585,
0.1666065454,
-0.0970347077,
0.1781595647,
-0.0822443962,
0.0992017984,
-0.184549123,
0.2732407451,
0.0765194818,
-0.0256413147,
-0.1682778746,
-0.0625797957,
-0.2430099845,
0.3292944431,
-0.15932706,
-0.0338404477,
-0.3486858606,
0.1303854585,
0.1739082336,
0.3362389207,
-0.0838697553,
-0.1523165703,
0.4145023823,
0.2369778305,
0.0439622179,
0.3686153591,
-0.0001172875,
0.1633386761,
0.2778438926,
-0.1433111429,
-0.1339642555,
-0.1817750186,
-0.5618034005,
-0.3552265167,
-0.1035683453,
-0.1154624894,
0.0706324726,
0.3033827245,
0.0572006106,
-0.0867422074,
0.2431904078,
0.1712264121,
0.1535670012,
0.3319172263,
0.2157701701,
0.0969050154,
0.0455293655,
-0.0774404109,
-0.1376800686,
0.0922585353,
0.2744396627,
-0.1569580734,
-0.2395651937,
0.2651385069,
-0.1804821938,
0.2693834603,
-0.0804270655,
-0.1510989368,
0.1479790807,
-0.1058239341,
0.2105876207,
0.5727722645,
0.2205697298,
-0.0554817878,
-0.1762657166,
-0.0319501609,
-0.2201737463,
0.0833649933,
0.3634192348,
0.1834742129,
-0.207865119,
0.0045872135,
-0.4487924278,
0.1443090737,
-0.0263799056,
0.2792591453,
0.2121901512,
0.3145954609,
0.2003568113,
0.0726619512,
0.227932483,
-0.2064420432,
0.1992098391,
-0.3010314107,
-0.1609570384,
0.1293215156,
-0.6801480651,
-0.0664965659,
0.0034522209,
0.3319551945,
0.44042629,
0.2289482802,
0.4523438513,
0.3736303449,
-0.1520799994,
0.1170620397,
0.3968932331,
0.1744448096,
0.0657050908,
-0.0545443483,
0.169239372,
0.2095876634,
-0.0953853205,
0.2373775989,
-0.030504927,
-0.3612298965,
0.3493167162,
0.1929920912,
0.2069613338,
-0.3434550762,
-0.1813287735,
0.0824042261,
-0.2454787642,
-0.1945313215,
0.271803081,
0.2862131596,
-0.0944883525,
0.5524675846,
-0.1653459072,
0.2156014144,
-0.1538059115,
-0.1847017407,
-0.257340163,
-0.1516558677,
0.073074013,
0.1531951427,
0.2943667769,
0.0656544417,
0.3595135212,
-0.0321133882,
0.3835367262,
-0.203120932,
0.3335245848,
0.0902097076,
-0.0699193627,
0.2554132044,
0.0134760961,
0.2010273635,
0.2036006153,
-0.0260661468,
-0.0622884706,
0.2220197171,
-0.3325723708,
-0.6167244911,
0.1093564481,
0.1607630253,
-0.3595764637,
-0.080491595,
0.0163278282,
-0.2801305056,
0.3050758839,
-0.4042750001,
-0.1545271724,
-0.3094390929,
-0.043020837,
-0.1179553494,
0.009863317,
0.0942920446,
-0.0934404433,
0.0649285913,
-0.0455776751,
-0.1832662672,
0.3468406796,
0.2667629123,
-0.2780412138,
-0.164542973,
-0.1394765228,
0.018730402,
0.2207068801,
-0.0911148638,
-0.3638111949,
0.2024420947,
-0.0716659203,
0.6232898235,
0.1856183112,
0.1097930819,
-0.3473640084,
-0.1108141243,
0.2366856933,
0.0823843926,
0.1350483447,
0.1764027029,
-0.2450099289,
-0.3122073114,
0.3476398885,
0.152895391,
0.0161079578,
-0.2277660668,
0.0021594316,
0.0893482119,
0.5888451338,
-0.2297042161,
-0.0971518904,
0.1738088727,
0.0959680974,
-0.2113370895,
0.0982793048,
0.0511621833,
-0.4866254926,
0.3225808442,
-0.156579718,
-0.0987990499,
0.3440407217,
-0.0598258227,
-0.0504282825,
-0.1761432588,
-0.3390327394,
0.0107524041,
0.080948852,
0.2748897076,
0.0834967047,
-0.2535200417,
-0.0927834511,
0.3571130335,
-0.3864996731,
-0.0857933909,
-0.4520959854,
0.1969178319,
-0.0879132226,
0.0851731449,
0.076490216,
0.1571477056,
0.1050595567,
-0.109420836,
-0.1870693415,
0.5129071474,
0.3213776052,
0.2384278327,
-0.0831714272,
0.3510249257,
-0.1519881785,
-0.049867142,
-0.1007989347,
0.5829588771,
-0.043825224,
-0.2062133551,
-0.1941186786,
0.4879204929,
-0.1729984134,
0.4535029829,
-0.0467058271,
0.1363263428,
0.3366039097,
-0.1314038932,
-0.2676326036,
-0.0946317986,
0.4545116723,
-0.1980917752,
-0.1669614613,
0.2530012131,
-0.2133248001,
0.0163291171,
0.3733876348,
-0.1440780461,
0.125990063,
-0.0151445484,
0.2458736002,
-0.2552569509,
-0.0198076405,
0.4922712445,
0.3632512689,
0.1011830121,
-0.0179459285,
-0.0273174345,
-0.229469195,
-0.1744434237,
0.119406268,
0.0973704606,
0.3296887279,
0.5370063782,
0.0413939059,
-0.3653717637,
-0.4267032146,
-0.0148495585,
0.0290298238,
0.1656201035,
-0.4414883256,
-0.2537639141,
-0.444193542,
0.228844732,
-0.3637366593,
-0.3436978161,
-0.4129475951,
-0.4648977816,
0.1210603118,
0.2281652242,
-0.0199533589,
0.3193201125,
-0.3111635447,
0.0683453232,
-0.1319281757,
0.060542874,
0.040054135,
-0.2777513266,
-0.0817122459,
-0.0046347044,
0.395745188,
0.0018831939,
-0.0578057617,
-0.307749927,
-0.0692628324,
-0.1458403319,
-0.1667574048,
-0.0308502838,
-0.110812515,
-0.0727641061,
0.2588977218,
0.302515626,
0.1404629052,
-0.3886935115,
0.2458182722,
-0.036740981,
-0.2834915817,
0.1877617091,
-0.1721365154,
-0.0189697687,
0.1081126034,
-0.3866265118,
-0.0947579294,
-0.3375903666,
-0.0101160184,
0.0664088875,
0.2717182934,
0.2532510161,
0.0835986882,
-0.0678554252,
0.1356462687,
0.2740627229,
-0.3200552762,
-0.8128378987,
0.4206299484,
-0.0225659702,
-0.2601570487,
-0.0509395339,
0.0214567855,
0.2018917948,
0.096438095,
-0.4472934008,
-0.2804217637,
-0.0651813447,
-0.3137296736,
0.1682510674,
-0.3037388027,
0.3359623253,
0.1188272834,
-0.0578323714,
-0.2449942231,
-0.2949228287,
0.3670243919,
0.0858880132,
0.515162468,
-0.2128700763,
0.2413334697,
0.0072190836,
0.5949299335,
0.2273383439,
0.2582990527,
0.0617829636,
0.1863198876,
0.1531589627,
-0.0817791671,
-0.3245138526,
0.1173865721,
-0.2651422322,
0.0044293851,
0.1789300889,
0.0948020071,
-0.2732841074,
-0.1016847715,
0.1297234297,
-0.1512692869,
-0.2311442196,
0.0542183816,
-0.0111722602,
0.0999286771,
0.2145665437,
0.0766037405,
-0.3172060251,
-0.3501292169,
0.1731935292,
0.2011539489,
-0.1653896868,
-0.1424285173,
-0.5950776935,
0.0641577691,
0.0246964507,
0.4123167992,
0.2499672323,
0.4031146765,
0.0229398012,
-0.3206758201,
0.0325472727,
-0.2579707503,
0.5845398307,
-0.3834360242,
0.3248412907,
0.0977155864,
-0.047182709,
-0.2630653381,
-0.1649410129,
-0.0380524062,
-0.0256653801,
0.3266569972,
0.3074329495,
-0.2269270271,
-0.2858473659,
0.1157698482,
0.3378463387,
-0.2878532708,
-0.168277964,
-0.3097969294,
-0.1108993888,
-0.3571150601,
-0.2201305926,
-0.3995673954,
0.347266078,
-0.1648674011,
-0.0812496543,
0.171175614,
-0.1453641355,
0.2767301798,
0.0399434417,
0.3021827936,
0.3644607365,
0.225110054,
0.0248756781,
0.3385648131,
0.2234017998,
0.4796162844,
-0.1350191534,
-0.0993255973,
-0.0563183874,
-0.2143765688,
0.0410138145,
0.2235025615,
0.0236764513,
-0.0814054161,
0.2965890169,
-0.1108628288,
-0.0433097072,
0.0397991277,
0.2845608294,
0.0299266726,
-0.3399110138,
-0.3018073142,
0.0595080778,
0.0007749796,
-0.1787183881,
0.2506116331,
-0.1772615314,
-0.0745184869,
0.0401595011,
0.3154737055,
0.8823342323,
0.0781966969,
0.088930279,
0.0817514583,
-0.1757193506,
-0.1717555374,
-0.4199718237,
0.1754764766,
-0.2713772655,
-0.2802042365,
-0.0789561421,
-0.2028965652,
0.1161566973,
0.3723819554,
-0.0140839145,
0.3294217587,
0.0507589579,
0.1718658209,
0.0152100958,
0.0928250104,
-0.1202959344,
-0.1144698635,
0.0429765694,
0.1203734577,
-0.1465318501,
0.1447480619,
-0.1230820715,
-0.0494484864,
0.2085238993,
-0.2759684324,
-0.1316171885,
0.1984800994,
-0.1262025535,
0.4007565677,
0.246336937,
-0.2069447041,
-0.2900714874,
0.4346533418,
0.093300581,
-0.0674181432,
-0.2564151287,
0.1446239054,
0.3509521186,
0.0936216116,
-0.0919712782,
-0.0624780692,
0.3511783481,
-0.1872356534,
-0.281373024,
0.0104380567,
-0.1517421603,
-0.6179360151,
-0.068176955,
-0.1264025867,
0.0984781235,
-0.1545959413,
-0.1535262614,
0.178663373,
0.0546398908,
-0.3420566618,
0.1221395358,
0.1330333203,
-0.3626444042,
0.3028177321,
0.129357338,
-0.2770990133,
-0.0944158137,
0.5548005104,
0.0386738293,
-0.0010046363,
0.7191019654,
0.0710802972,
-0.1204013824,
-0.2845616937,
0.3099562228,
0.2341083437,
-0.2122224867,
-0.0682359636,
-0.1181255132,
-0.0549927801,
0.0401895791,
0.5675612688,
0.2766494751,
0.3637249172,
-0.0064761043,
-0.6050378084,
-0.1222514361,
0.0505510792,
0.2656698525,
0.3840140402,
-0.3860803246,
-0.0716591924,
-0.1205711514,
0.2204844505,
-0.2977004647,
0.0136335753,
-0.3014143705,
-0.0014124531,
-0.3319446146,
-0.0595816895,
0.1033879668,
0.046647884,
0.1038635671,
-0.2260659337,
-0.2882828116,
-0.163638413,
-0.2158525586,
0.1587164402,
0.036347691,
-0.0632580519,
0.0698035583,
-0.093967393,
-0.1066645235,
-0.1600659639,
-0.0529935211,
0.208039254,
-0.0953486413,
0.1257883906,
0.2332841158,
-0.1949183792,
-0.1487827599,
-0.2617365718,
-0.2963094115,
-0.0960337296,
0.0908314586,
-0.1021257937,
0.0535944626,
-0.1696587205,
0.4369635582,
0.1701049805,
0.0245453175,
-0.0090561565,
0.2811066806,
-0.2891623378,
-0.0970127657,
0.2284563035,
0.3137276769,
0.3578689396,
-0.2104903609,
0.191317603,
0.0008259676,
0.223886162,
-0.2928494215,
-0.1559720039,
0.3667839468,
-0.1120519936,
0.0143715516,
0.0317224078,
-0.0716922656,
-0.3091430664,
0.5745686889,
0.1000952423,
-0.0016480461,
-0.3915390372,
0.1874296963,
0.3277112842,
-0.3724376857,
-0.0696391463,
-0.0045979545,
-0.2296411991,
0.2119151056,
-0.2145882547,
-0.1123066843,
0.4519321918,
0.0501494557,
0.250569135,
-0.1284567416,
-0.4928979874,
0.094471544,
0.1954618096,
-0.091243051,
0.0177288894,
0.0348487124,
0.4064949751,
-0.2336776853,
0.1827876419,
-0.702688992,
-0.0180880837,
0.0231122524,
-0.2515980303,
-0.348980099,
-0.378718853,
-0.1243418083,
0.162175402,
-0.0076933461,
0.1223286092,
0.3172166944,
0.027880678,
-0.0302278362,
-0.5480283499,
-0.1554401666,
-0.0721405745,
-0.1325325519,
-0.2546499968,
0.2110595703,
0.3817439973,
-0.116866909,
0.341118902,
0.2328886986,
0.3793443739,
0.3968304992,
0.0158853643,
-0.007507205,
-0.0547184721,
0.0275845379,
-0.1511167288,
0.7324590087,
0.1668840051,
-0.1373340935,
0.4767387211,
0.0521658659,
-0.1582672,
0.0700558126,
-0.0142775606,
-0.0927821547,
-0.0598388352,
0.7789543271,
0.038809415,
0.1453847885,
-0.0719825625,
0.233615756,
-0.308211267,
-0.3224420547,
0.0849670321,
0.0910015404,
0.2019015104,
-0.0258246455,
0.073482208,
-0.188868776,
0.6922151446,
-0.0983077735,
-0.2206100821,
-0.2996821105,
-0.2596407235,
-0.8195846081,
0.2501327693,
-0.1412822604,
0.3424732089,
-0.0689982027,
0.1562744379,
0.172787413,
0.2854845822,
0.2575878501,
-0.2129997313,
-0.1863705814,
-0.0521588176,
-0.3973253071,
0.0739046335,
0.4234282374,
0.0349838734,
-0.0069679096,
-0.2355775386,
0.18899858,
0.2631668448,
0.0424306244,
0.1342671663,
0.0111743882,
-0.0327079445,
-0.0246863049,
-0.0812195837,
0.1433466822,
0.5623226166,
-0.003130354,
-0.0114593133,
-0.1989396513,
-0.2911085188,
-0.0620974377,
0.1848046184,
-0.0502038412,
0.5143119693,
-0.1861061603,
-0.0308317896,
-0.2220528126,
0.200448364,
-0.0561482087,
-0.2992594242,
-0.128759712,
0.0325105563,
-0.0127721205,
0.0843318105,
0.1155449972,
0.0505927131,
0.0815806836,
0.2093013227,
-0.0561162457,
-0.0594240874,
0.7326384187,
-0.1133246943,
-0.1869965494,
-0.3649471402,
0.1383292973,
-0.3404733539,
-0.0648734123,
-0.4481022358,
0.0900492966,
0.3190637231,
0.1263612062,
-0.2260793746,
0.1400437802,
-0.257489711,
0.0272711366,
-0.0578921065,
0.0775433406,
0.2296062559,
-0.298792392,
-0.1100094616,
-0.1358180046
] |
https://github.com/huggingface/datasets/issues/1876 | load_dataset("multi_woz_v22") NonMatchingChecksumError | Hi, I still meet the error when loading the datasets after upgradeing datasets.
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json'] | Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
| 27 | load_dataset("multi_woz_v22") NonMatchingChecksumError
Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
Hi, I still meet the error when loading the datasets after upgradeing datasets.
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json'] | [
-0.2244368196,
0.1764704883,
-0.0287194811,
0.1495723128,
0.1895723641,
0.0018984303,
0.365057677,
0.4820295274,
0.2514897585,
0.1666065454,
-0.0970347077,
0.1781595647,
-0.0822443962,
0.0992017984,
-0.184549123,
0.2732407451,
0.0765194818,
-0.0256413147,
-0.1682778746,
-0.0625797957,
-0.2430099845,
0.3292944431,
-0.15932706,
-0.0338404477,
-0.3486858606,
0.1303854585,
0.1739082336,
0.3362389207,
-0.0838697553,
-0.1523165703,
0.4145023823,
0.2369778305,
0.0439622179,
0.3686153591,
-0.0001172875,
0.1633386761,
0.2778438926,
-0.1433111429,
-0.1339642555,
-0.1817750186,
-0.5618034005,
-0.3552265167,
-0.1035683453,
-0.1154624894,
0.0706324726,
0.3033827245,
0.0572006106,
-0.0867422074,
0.2431904078,
0.1712264121,
0.1535670012,
0.3319172263,
0.2157701701,
0.0969050154,
0.0455293655,
-0.0774404109,
-0.1376800686,
0.0922585353,
0.2744396627,
-0.1569580734,
-0.2395651937,
0.2651385069,
-0.1804821938,
0.2693834603,
-0.0804270655,
-0.1510989368,
0.1479790807,
-0.1058239341,
0.2105876207,
0.5727722645,
0.2205697298,
-0.0554817878,
-0.1762657166,
-0.0319501609,
-0.2201737463,
0.0833649933,
0.3634192348,
0.1834742129,
-0.207865119,
0.0045872135,
-0.4487924278,
0.1443090737,
-0.0263799056,
0.2792591453,
0.2121901512,
0.3145954609,
0.2003568113,
0.0726619512,
0.227932483,
-0.2064420432,
0.1992098391,
-0.3010314107,
-0.1609570384,
0.1293215156,
-0.6801480651,
-0.0664965659,
0.0034522209,
0.3319551945,
0.44042629,
0.2289482802,
0.4523438513,
0.3736303449,
-0.1520799994,
0.1170620397,
0.3968932331,
0.1744448096,
0.0657050908,
-0.0545443483,
0.169239372,
0.2095876634,
-0.0953853205,
0.2373775989,
-0.030504927,
-0.3612298965,
0.3493167162,
0.1929920912,
0.2069613338,
-0.3434550762,
-0.1813287735,
0.0824042261,
-0.2454787642,
-0.1945313215,
0.271803081,
0.2862131596,
-0.0944883525,
0.5524675846,
-0.1653459072,
0.2156014144,
-0.1538059115,
-0.1847017407,
-0.257340163,
-0.1516558677,
0.073074013,
0.1531951427,
0.2943667769,
0.0656544417,
0.3595135212,
-0.0321133882,
0.3835367262,
-0.203120932,
0.3335245848,
0.0902097076,
-0.0699193627,
0.2554132044,
0.0134760961,
0.2010273635,
0.2036006153,
-0.0260661468,
-0.0622884706,
0.2220197171,
-0.3325723708,
-0.6167244911,
0.1093564481,
0.1607630253,
-0.3595764637,
-0.080491595,
0.0163278282,
-0.2801305056,
0.3050758839,
-0.4042750001,
-0.1545271724,
-0.3094390929,
-0.043020837,
-0.1179553494,
0.009863317,
0.0942920446,
-0.0934404433,
0.0649285913,
-0.0455776751,
-0.1832662672,
0.3468406796,
0.2667629123,
-0.2780412138,
-0.164542973,
-0.1394765228,
0.018730402,
0.2207068801,
-0.0911148638,
-0.3638111949,
0.2024420947,
-0.0716659203,
0.6232898235,
0.1856183112,
0.1097930819,
-0.3473640084,
-0.1108141243,
0.2366856933,
0.0823843926,
0.1350483447,
0.1764027029,
-0.2450099289,
-0.3122073114,
0.3476398885,
0.152895391,
0.0161079578,
-0.2277660668,
0.0021594316,
0.0893482119,
0.5888451338,
-0.2297042161,
-0.0971518904,
0.1738088727,
0.0959680974,
-0.2113370895,
0.0982793048,
0.0511621833,
-0.4866254926,
0.3225808442,
-0.156579718,
-0.0987990499,
0.3440407217,
-0.0598258227,
-0.0504282825,
-0.1761432588,
-0.3390327394,
0.0107524041,
0.080948852,
0.2748897076,
0.0834967047,
-0.2535200417,
-0.0927834511,
0.3571130335,
-0.3864996731,
-0.0857933909,
-0.4520959854,
0.1969178319,
-0.0879132226,
0.0851731449,
0.076490216,
0.1571477056,
0.1050595567,
-0.109420836,
-0.1870693415,
0.5129071474,
0.3213776052,
0.2384278327,
-0.0831714272,
0.3510249257,
-0.1519881785,
-0.049867142,
-0.1007989347,
0.5829588771,
-0.043825224,
-0.2062133551,
-0.1941186786,
0.4879204929,
-0.1729984134,
0.4535029829,
-0.0467058271,
0.1363263428,
0.3366039097,
-0.1314038932,
-0.2676326036,
-0.0946317986,
0.4545116723,
-0.1980917752,
-0.1669614613,
0.2530012131,
-0.2133248001,
0.0163291171,
0.3733876348,
-0.1440780461,
0.125990063,
-0.0151445484,
0.2458736002,
-0.2552569509,
-0.0198076405,
0.4922712445,
0.3632512689,
0.1011830121,
-0.0179459285,
-0.0273174345,
-0.229469195,
-0.1744434237,
0.119406268,
0.0973704606,
0.3296887279,
0.5370063782,
0.0413939059,
-0.3653717637,
-0.4267032146,
-0.0148495585,
0.0290298238,
0.1656201035,
-0.4414883256,
-0.2537639141,
-0.444193542,
0.228844732,
-0.3637366593,
-0.3436978161,
-0.4129475951,
-0.4648977816,
0.1210603118,
0.2281652242,
-0.0199533589,
0.3193201125,
-0.3111635447,
0.0683453232,
-0.1319281757,
0.060542874,
0.040054135,
-0.2777513266,
-0.0817122459,
-0.0046347044,
0.395745188,
0.0018831939,
-0.0578057617,
-0.307749927,
-0.0692628324,
-0.1458403319,
-0.1667574048,
-0.0308502838,
-0.110812515,
-0.0727641061,
0.2588977218,
0.302515626,
0.1404629052,
-0.3886935115,
0.2458182722,
-0.036740981,
-0.2834915817,
0.1877617091,
-0.1721365154,
-0.0189697687,
0.1081126034,
-0.3866265118,
-0.0947579294,
-0.3375903666,
-0.0101160184,
0.0664088875,
0.2717182934,
0.2532510161,
0.0835986882,
-0.0678554252,
0.1356462687,
0.2740627229,
-0.3200552762,
-0.8128378987,
0.4206299484,
-0.0225659702,
-0.2601570487,
-0.0509395339,
0.0214567855,
0.2018917948,
0.096438095,
-0.4472934008,
-0.2804217637,
-0.0651813447,
-0.3137296736,
0.1682510674,
-0.3037388027,
0.3359623253,
0.1188272834,
-0.0578323714,
-0.2449942231,
-0.2949228287,
0.3670243919,
0.0858880132,
0.515162468,
-0.2128700763,
0.2413334697,
0.0072190836,
0.5949299335,
0.2273383439,
0.2582990527,
0.0617829636,
0.1863198876,
0.1531589627,
-0.0817791671,
-0.3245138526,
0.1173865721,
-0.2651422322,
0.0044293851,
0.1789300889,
0.0948020071,
-0.2732841074,
-0.1016847715,
0.1297234297,
-0.1512692869,
-0.2311442196,
0.0542183816,
-0.0111722602,
0.0999286771,
0.2145665437,
0.0766037405,
-0.3172060251,
-0.3501292169,
0.1731935292,
0.2011539489,
-0.1653896868,
-0.1424285173,
-0.5950776935,
0.0641577691,
0.0246964507,
0.4123167992,
0.2499672323,
0.4031146765,
0.0229398012,
-0.3206758201,
0.0325472727,
-0.2579707503,
0.5845398307,
-0.3834360242,
0.3248412907,
0.0977155864,
-0.047182709,
-0.2630653381,
-0.1649410129,
-0.0380524062,
-0.0256653801,
0.3266569972,
0.3074329495,
-0.2269270271,
-0.2858473659,
0.1157698482,
0.3378463387,
-0.2878532708,
-0.168277964,
-0.3097969294,
-0.1108993888,
-0.3571150601,
-0.2201305926,
-0.3995673954,
0.347266078,
-0.1648674011,
-0.0812496543,
0.171175614,
-0.1453641355,
0.2767301798,
0.0399434417,
0.3021827936,
0.3644607365,
0.225110054,
0.0248756781,
0.3385648131,
0.2234017998,
0.4796162844,
-0.1350191534,
-0.0993255973,
-0.0563183874,
-0.2143765688,
0.0410138145,
0.2235025615,
0.0236764513,
-0.0814054161,
0.2965890169,
-0.1108628288,
-0.0433097072,
0.0397991277,
0.2845608294,
0.0299266726,
-0.3399110138,
-0.3018073142,
0.0595080778,
0.0007749796,
-0.1787183881,
0.2506116331,
-0.1772615314,
-0.0745184869,
0.0401595011,
0.3154737055,
0.8823342323,
0.0781966969,
0.088930279,
0.0817514583,
-0.1757193506,
-0.1717555374,
-0.4199718237,
0.1754764766,
-0.2713772655,
-0.2802042365,
-0.0789561421,
-0.2028965652,
0.1161566973,
0.3723819554,
-0.0140839145,
0.3294217587,
0.0507589579,
0.1718658209,
0.0152100958,
0.0928250104,
-0.1202959344,
-0.1144698635,
0.0429765694,
0.1203734577,
-0.1465318501,
0.1447480619,
-0.1230820715,
-0.0494484864,
0.2085238993,
-0.2759684324,
-0.1316171885,
0.1984800994,
-0.1262025535,
0.4007565677,
0.246336937,
-0.2069447041,
-0.2900714874,
0.4346533418,
0.093300581,
-0.0674181432,
-0.2564151287,
0.1446239054,
0.3509521186,
0.0936216116,
-0.0919712782,
-0.0624780692,
0.3511783481,
-0.1872356534,
-0.281373024,
0.0104380567,
-0.1517421603,
-0.6179360151,
-0.068176955,
-0.1264025867,
0.0984781235,
-0.1545959413,
-0.1535262614,
0.178663373,
0.0546398908,
-0.3420566618,
0.1221395358,
0.1330333203,
-0.3626444042,
0.3028177321,
0.129357338,
-0.2770990133,
-0.0944158137,
0.5548005104,
0.0386738293,
-0.0010046363,
0.7191019654,
0.0710802972,
-0.1204013824,
-0.2845616937,
0.3099562228,
0.2341083437,
-0.2122224867,
-0.0682359636,
-0.1181255132,
-0.0549927801,
0.0401895791,
0.5675612688,
0.2766494751,
0.3637249172,
-0.0064761043,
-0.6050378084,
-0.1222514361,
0.0505510792,
0.2656698525,
0.3840140402,
-0.3860803246,
-0.0716591924,
-0.1205711514,
0.2204844505,
-0.2977004647,
0.0136335753,
-0.3014143705,
-0.0014124531,
-0.3319446146,
-0.0595816895,
0.1033879668,
0.046647884,
0.1038635671,
-0.2260659337,
-0.2882828116,
-0.163638413,
-0.2158525586,
0.1587164402,
0.036347691,
-0.0632580519,
0.0698035583,
-0.093967393,
-0.1066645235,
-0.1600659639,
-0.0529935211,
0.208039254,
-0.0953486413,
0.1257883906,
0.2332841158,
-0.1949183792,
-0.1487827599,
-0.2617365718,
-0.2963094115,
-0.0960337296,
0.0908314586,
-0.1021257937,
0.0535944626,
-0.1696587205,
0.4369635582,
0.1701049805,
0.0245453175,
-0.0090561565,
0.2811066806,
-0.2891623378,
-0.0970127657,
0.2284563035,
0.3137276769,
0.3578689396,
-0.2104903609,
0.191317603,
0.0008259676,
0.223886162,
-0.2928494215,
-0.1559720039,
0.3667839468,
-0.1120519936,
0.0143715516,
0.0317224078,
-0.0716922656,
-0.3091430664,
0.5745686889,
0.1000952423,
-0.0016480461,
-0.3915390372,
0.1874296963,
0.3277112842,
-0.3724376857,
-0.0696391463,
-0.0045979545,
-0.2296411991,
0.2119151056,
-0.2145882547,
-0.1123066843,
0.4519321918,
0.0501494557,
0.250569135,
-0.1284567416,
-0.4928979874,
0.094471544,
0.1954618096,
-0.091243051,
0.0177288894,
0.0348487124,
0.4064949751,
-0.2336776853,
0.1827876419,
-0.702688992,
-0.0180880837,
0.0231122524,
-0.2515980303,
-0.348980099,
-0.378718853,
-0.1243418083,
0.162175402,
-0.0076933461,
0.1223286092,
0.3172166944,
0.027880678,
-0.0302278362,
-0.5480283499,
-0.1554401666,
-0.0721405745,
-0.1325325519,
-0.2546499968,
0.2110595703,
0.3817439973,
-0.116866909,
0.341118902,
0.2328886986,
0.3793443739,
0.3968304992,
0.0158853643,
-0.007507205,
-0.0547184721,
0.0275845379,
-0.1511167288,
0.7324590087,
0.1668840051,
-0.1373340935,
0.4767387211,
0.0521658659,
-0.1582672,
0.0700558126,
-0.0142775606,
-0.0927821547,
-0.0598388352,
0.7789543271,
0.038809415,
0.1453847885,
-0.0719825625,
0.233615756,
-0.308211267,
-0.3224420547,
0.0849670321,
0.0910015404,
0.2019015104,
-0.0258246455,
0.073482208,
-0.188868776,
0.6922151446,
-0.0983077735,
-0.2206100821,
-0.2996821105,
-0.2596407235,
-0.8195846081,
0.2501327693,
-0.1412822604,
0.3424732089,
-0.0689982027,
0.1562744379,
0.172787413,
0.2854845822,
0.2575878501,
-0.2129997313,
-0.1863705814,
-0.0521588176,
-0.3973253071,
0.0739046335,
0.4234282374,
0.0349838734,
-0.0069679096,
-0.2355775386,
0.18899858,
0.2631668448,
0.0424306244,
0.1342671663,
0.0111743882,
-0.0327079445,
-0.0246863049,
-0.0812195837,
0.1433466822,
0.5623226166,
-0.003130354,
-0.0114593133,
-0.1989396513,
-0.2911085188,
-0.0620974377,
0.1848046184,
-0.0502038412,
0.5143119693,
-0.1861061603,
-0.0308317896,
-0.2220528126,
0.200448364,
-0.0561482087,
-0.2992594242,
-0.128759712,
0.0325105563,
-0.0127721205,
0.0843318105,
0.1155449972,
0.0505927131,
0.0815806836,
0.2093013227,
-0.0561162457,
-0.0594240874,
0.7326384187,
-0.1133246943,
-0.1869965494,
-0.3649471402,
0.1383292973,
-0.3404733539,
-0.0648734123,
-0.4481022358,
0.0900492966,
0.3190637231,
0.1263612062,
-0.2260793746,
0.1400437802,
-0.257489711,
0.0272711366,
-0.0578921065,
0.0775433406,
0.2296062559,
-0.298792392,
-0.1100094616,
-0.1358180046
] |
https://github.com/huggingface/datasets/issues/1876 | load_dataset("multi_woz_v22") NonMatchingChecksumError | This must be related to https://github.com/budzianowski/multiwoz/pull/72
Those files have changed, let me update the checksums for this dataset.
For now you can use `ignore_verifications=True` in `load_dataset` to skip the checksum verification. | Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
| 31 | load_dataset("multi_woz_v22") NonMatchingChecksumError
Hi, it seems that loading the multi_woz_v22 dataset gives a NonMatchingChecksumError.
To reproduce:
`dataset = load_dataset('multi_woz_v22','v2.2_active_only',split='train')`
This will give the following error:
```
raise NonMatchingChecksumError(error_msg + str(bad_urls))
datasets.utils.info_utils.NonMatchingChecksumError: Checksums didn't match for dataset source files:
['https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dialog_acts.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_003.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_004.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_005.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_006.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_007.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_008.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_009.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_010.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_012.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_013.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_014.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_015.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_016.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/train/dialogues_017.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/dev/dialogues_002.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_001.json', 'https://github.com/budzianowski/multiwoz/raw/master/data/MultiWOZ_2.2/test/dialogues_002.json']
```
This must be related to https://github.com/budzianowski/multiwoz/pull/72
Those files have changed, let me update the checksums for this dataset.
For now you can use `ignore_verifications=True` in `load_dataset` to skip the checksum verification. | [
-0.2244368196,
0.1764704883,
-0.0287194811,
0.1495723128,
0.1895723641,
0.0018984303,
0.365057677,
0.4820295274,
0.2514897585,
0.1666065454,
-0.0970347077,
0.1781595647,
-0.0822443962,
0.0992017984,
-0.184549123,
0.2732407451,
0.0765194818,
-0.0256413147,
-0.1682778746,
-0.0625797957,
-0.2430099845,
0.3292944431,
-0.15932706,
-0.0338404477,
-0.3486858606,
0.1303854585,
0.1739082336,
0.3362389207,
-0.0838697553,
-0.1523165703,
0.4145023823,
0.2369778305,
0.0439622179,
0.3686153591,
-0.0001172875,
0.1633386761,
0.2778438926,
-0.1433111429,
-0.1339642555,
-0.1817750186,
-0.5618034005,
-0.3552265167,
-0.1035683453,
-0.1154624894,
0.0706324726,
0.3033827245,
0.0572006106,
-0.0867422074,
0.2431904078,
0.1712264121,
0.1535670012,
0.3319172263,
0.2157701701,
0.0969050154,
0.0455293655,
-0.0774404109,
-0.1376800686,
0.0922585353,
0.2744396627,
-0.1569580734,
-0.2395651937,
0.2651385069,
-0.1804821938,
0.2693834603,
-0.0804270655,
-0.1510989368,
0.1479790807,
-0.1058239341,
0.2105876207,
0.5727722645,
0.2205697298,
-0.0554817878,
-0.1762657166,
-0.0319501609,
-0.2201737463,
0.0833649933,
0.3634192348,
0.1834742129,
-0.207865119,
0.0045872135,
-0.4487924278,
0.1443090737,
-0.0263799056,
0.2792591453,
0.2121901512,
0.3145954609,
0.2003568113,
0.0726619512,
0.227932483,
-0.2064420432,
0.1992098391,
-0.3010314107,
-0.1609570384,
0.1293215156,
-0.6801480651,
-0.0664965659,
0.0034522209,
0.3319551945,
0.44042629,
0.2289482802,
0.4523438513,
0.3736303449,
-0.1520799994,
0.1170620397,
0.3968932331,
0.1744448096,
0.0657050908,
-0.0545443483,
0.169239372,
0.2095876634,
-0.0953853205,
0.2373775989,
-0.030504927,
-0.3612298965,
0.3493167162,
0.1929920912,
0.2069613338,
-0.3434550762,
-0.1813287735,
0.0824042261,
-0.2454787642,
-0.1945313215,
0.271803081,
0.2862131596,
-0.0944883525,
0.5524675846,
-0.1653459072,
0.2156014144,
-0.1538059115,
-0.1847017407,
-0.257340163,
-0.1516558677,
0.073074013,
0.1531951427,
0.2943667769,
0.0656544417,
0.3595135212,
-0.0321133882,
0.3835367262,
-0.203120932,
0.3335245848,
0.0902097076,
-0.0699193627,
0.2554132044,
0.0134760961,
0.2010273635,
0.2036006153,
-0.0260661468,
-0.0622884706,
0.2220197171,
-0.3325723708,
-0.6167244911,
0.1093564481,
0.1607630253,
-0.3595764637,
-0.080491595,
0.0163278282,
-0.2801305056,
0.3050758839,
-0.4042750001,
-0.1545271724,
-0.3094390929,
-0.043020837,
-0.1179553494,
0.009863317,
0.0942920446,
-0.0934404433,
0.0649285913,
-0.0455776751,
-0.1832662672,
0.3468406796,
0.2667629123,
-0.2780412138,
-0.164542973,
-0.1394765228,
0.018730402,
0.2207068801,
-0.0911148638,
-0.3638111949,
0.2024420947,
-0.0716659203,
0.6232898235,
0.1856183112,
0.1097930819,
-0.3473640084,
-0.1108141243,
0.2366856933,
0.0823843926,
0.1350483447,
0.1764027029,
-0.2450099289,
-0.3122073114,
0.3476398885,
0.152895391,
0.0161079578,
-0.2277660668,
0.0021594316,
0.0893482119,
0.5888451338,
-0.2297042161,
-0.0971518904,
0.1738088727,
0.0959680974,
-0.2113370895,
0.0982793048,
0.0511621833,
-0.4866254926,
0.3225808442,
-0.156579718,
-0.0987990499,
0.3440407217,
-0.0598258227,
-0.0504282825,
-0.1761432588,
-0.3390327394,
0.0107524041,
0.080948852,
0.2748897076,
0.0834967047,
-0.2535200417,
-0.0927834511,
0.3571130335,
-0.3864996731,
-0.0857933909,
-0.4520959854,
0.1969178319,
-0.0879132226,
0.0851731449,
0.076490216,
0.1571477056,
0.1050595567,
-0.109420836,
-0.1870693415,
0.5129071474,
0.3213776052,
0.2384278327,
-0.0831714272,
0.3510249257,
-0.1519881785,
-0.049867142,
-0.1007989347,
0.5829588771,
-0.043825224,
-0.2062133551,
-0.1941186786,
0.4879204929,
-0.1729984134,
0.4535029829,
-0.0467058271,
0.1363263428,
0.3366039097,
-0.1314038932,
-0.2676326036,
-0.0946317986,
0.4545116723,
-0.1980917752,
-0.1669614613,
0.2530012131,
-0.2133248001,
0.0163291171,
0.3733876348,
-0.1440780461,
0.125990063,
-0.0151445484,
0.2458736002,
-0.2552569509,
-0.0198076405,
0.4922712445,
0.3632512689,
0.1011830121,
-0.0179459285,
-0.0273174345,
-0.229469195,
-0.1744434237,
0.119406268,
0.0973704606,
0.3296887279,
0.5370063782,
0.0413939059,
-0.3653717637,
-0.4267032146,
-0.0148495585,
0.0290298238,
0.1656201035,
-0.4414883256,
-0.2537639141,
-0.444193542,
0.228844732,
-0.3637366593,
-0.3436978161,
-0.4129475951,
-0.4648977816,
0.1210603118,
0.2281652242,
-0.0199533589,
0.3193201125,
-0.3111635447,
0.0683453232,
-0.1319281757,
0.060542874,
0.040054135,
-0.2777513266,
-0.0817122459,
-0.0046347044,
0.395745188,
0.0018831939,
-0.0578057617,
-0.307749927,
-0.0692628324,
-0.1458403319,
-0.1667574048,
-0.0308502838,
-0.110812515,
-0.0727641061,
0.2588977218,
0.302515626,
0.1404629052,
-0.3886935115,
0.2458182722,
-0.036740981,
-0.2834915817,
0.1877617091,
-0.1721365154,
-0.0189697687,
0.1081126034,
-0.3866265118,
-0.0947579294,
-0.3375903666,
-0.0101160184,
0.0664088875,
0.2717182934,
0.2532510161,
0.0835986882,
-0.0678554252,
0.1356462687,
0.2740627229,
-0.3200552762,
-0.8128378987,
0.4206299484,
-0.0225659702,
-0.2601570487,
-0.0509395339,
0.0214567855,
0.2018917948,
0.096438095,
-0.4472934008,
-0.2804217637,
-0.0651813447,
-0.3137296736,
0.1682510674,
-0.3037388027,
0.3359623253,
0.1188272834,
-0.0578323714,
-0.2449942231,
-0.2949228287,
0.3670243919,
0.0858880132,
0.515162468,
-0.2128700763,
0.2413334697,
0.0072190836,
0.5949299335,
0.2273383439,
0.2582990527,
0.0617829636,
0.1863198876,
0.1531589627,
-0.0817791671,
-0.3245138526,
0.1173865721,
-0.2651422322,
0.0044293851,
0.1789300889,
0.0948020071,
-0.2732841074,
-0.1016847715,
0.1297234297,
-0.1512692869,
-0.2311442196,
0.0542183816,
-0.0111722602,
0.0999286771,
0.2145665437,
0.0766037405,
-0.3172060251,
-0.3501292169,
0.1731935292,
0.2011539489,
-0.1653896868,
-0.1424285173,
-0.5950776935,
0.0641577691,
0.0246964507,
0.4123167992,
0.2499672323,
0.4031146765,
0.0229398012,
-0.3206758201,
0.0325472727,
-0.2579707503,
0.5845398307,
-0.3834360242,
0.3248412907,
0.0977155864,
-0.047182709,
-0.2630653381,
-0.1649410129,
-0.0380524062,
-0.0256653801,
0.3266569972,
0.3074329495,
-0.2269270271,
-0.2858473659,
0.1157698482,
0.3378463387,
-0.2878532708,
-0.168277964,
-0.3097969294,
-0.1108993888,
-0.3571150601,
-0.2201305926,
-0.3995673954,
0.347266078,
-0.1648674011,
-0.0812496543,
0.171175614,
-0.1453641355,
0.2767301798,
0.0399434417,
0.3021827936,
0.3644607365,
0.225110054,
0.0248756781,
0.3385648131,
0.2234017998,
0.4796162844,
-0.1350191534,
-0.0993255973,
-0.0563183874,
-0.2143765688,
0.0410138145,
0.2235025615,
0.0236764513,
-0.0814054161,
0.2965890169,
-0.1108628288,
-0.0433097072,
0.0397991277,
0.2845608294,
0.0299266726,
-0.3399110138,
-0.3018073142,
0.0595080778,
0.0007749796,
-0.1787183881,
0.2506116331,
-0.1772615314,
-0.0745184869,
0.0401595011,
0.3154737055,
0.8823342323,
0.0781966969,
0.088930279,
0.0817514583,
-0.1757193506,
-0.1717555374,
-0.4199718237,
0.1754764766,
-0.2713772655,
-0.2802042365,
-0.0789561421,
-0.2028965652,
0.1161566973,
0.3723819554,
-0.0140839145,
0.3294217587,
0.0507589579,
0.1718658209,
0.0152100958,
0.0928250104,
-0.1202959344,
-0.1144698635,
0.0429765694,
0.1203734577,
-0.1465318501,
0.1447480619,
-0.1230820715,
-0.0494484864,
0.2085238993,
-0.2759684324,
-0.1316171885,
0.1984800994,
-0.1262025535,
0.4007565677,
0.246336937,
-0.2069447041,
-0.2900714874,
0.4346533418,
0.093300581,
-0.0674181432,
-0.2564151287,
0.1446239054,
0.3509521186,
0.0936216116,
-0.0919712782,
-0.0624780692,
0.3511783481,
-0.1872356534,
-0.281373024,
0.0104380567,
-0.1517421603,
-0.6179360151,
-0.068176955,
-0.1264025867,
0.0984781235,
-0.1545959413,
-0.1535262614,
0.178663373,
0.0546398908,
-0.3420566618,
0.1221395358,
0.1330333203,
-0.3626444042,
0.3028177321,
0.129357338,
-0.2770990133,
-0.0944158137,
0.5548005104,
0.0386738293,
-0.0010046363,
0.7191019654,
0.0710802972,
-0.1204013824,
-0.2845616937,
0.3099562228,
0.2341083437,
-0.2122224867,
-0.0682359636,
-0.1181255132,
-0.0549927801,
0.0401895791,
0.5675612688,
0.2766494751,
0.3637249172,
-0.0064761043,
-0.6050378084,
-0.1222514361,
0.0505510792,
0.2656698525,
0.3840140402,
-0.3860803246,
-0.0716591924,
-0.1205711514,
0.2204844505,
-0.2977004647,
0.0136335753,
-0.3014143705,
-0.0014124531,
-0.3319446146,
-0.0595816895,
0.1033879668,
0.046647884,
0.1038635671,
-0.2260659337,
-0.2882828116,
-0.163638413,
-0.2158525586,
0.1587164402,
0.036347691,
-0.0632580519,
0.0698035583,
-0.093967393,
-0.1066645235,
-0.1600659639,
-0.0529935211,
0.208039254,
-0.0953486413,
0.1257883906,
0.2332841158,
-0.1949183792,
-0.1487827599,
-0.2617365718,
-0.2963094115,
-0.0960337296,
0.0908314586,
-0.1021257937,
0.0535944626,
-0.1696587205,
0.4369635582,
0.1701049805,
0.0245453175,
-0.0090561565,
0.2811066806,
-0.2891623378,
-0.0970127657,
0.2284563035,
0.3137276769,
0.3578689396,
-0.2104903609,
0.191317603,
0.0008259676,
0.223886162,
-0.2928494215,
-0.1559720039,
0.3667839468,
-0.1120519936,
0.0143715516,
0.0317224078,
-0.0716922656,
-0.3091430664,
0.5745686889,
0.1000952423,
-0.0016480461,
-0.3915390372,
0.1874296963,
0.3277112842,
-0.3724376857,
-0.0696391463,
-0.0045979545,
-0.2296411991,
0.2119151056,
-0.2145882547,
-0.1123066843,
0.4519321918,
0.0501494557,
0.250569135,
-0.1284567416,
-0.4928979874,
0.094471544,
0.1954618096,
-0.091243051,
0.0177288894,
0.0348487124,
0.4064949751,
-0.2336776853,
0.1827876419,
-0.702688992,
-0.0180880837,
0.0231122524,
-0.2515980303,
-0.348980099,
-0.378718853,
-0.1243418083,
0.162175402,
-0.0076933461,
0.1223286092,
0.3172166944,
0.027880678,
-0.0302278362,
-0.5480283499,
-0.1554401666,
-0.0721405745,
-0.1325325519,
-0.2546499968,
0.2110595703,
0.3817439973,
-0.116866909,
0.341118902,
0.2328886986,
0.3793443739,
0.3968304992,
0.0158853643,
-0.007507205,
-0.0547184721,
0.0275845379,
-0.1511167288,
0.7324590087,
0.1668840051,
-0.1373340935,
0.4767387211,
0.0521658659,
-0.1582672,
0.0700558126,
-0.0142775606,
-0.0927821547,
-0.0598388352,
0.7789543271,
0.038809415,
0.1453847885,
-0.0719825625,
0.233615756,
-0.308211267,
-0.3224420547,
0.0849670321,
0.0910015404,
0.2019015104,
-0.0258246455,
0.073482208,
-0.188868776,
0.6922151446,
-0.0983077735,
-0.2206100821,
-0.2996821105,
-0.2596407235,
-0.8195846081,
0.2501327693,
-0.1412822604,
0.3424732089,
-0.0689982027,
0.1562744379,
0.172787413,
0.2854845822,
0.2575878501,
-0.2129997313,
-0.1863705814,
-0.0521588176,
-0.3973253071,
0.0739046335,
0.4234282374,
0.0349838734,
-0.0069679096,
-0.2355775386,
0.18899858,
0.2631668448,
0.0424306244,
0.1342671663,
0.0111743882,
-0.0327079445,
-0.0246863049,
-0.0812195837,
0.1433466822,
0.5623226166,
-0.003130354,
-0.0114593133,
-0.1989396513,
-0.2911085188,
-0.0620974377,
0.1848046184,
-0.0502038412,
0.5143119693,
-0.1861061603,
-0.0308317896,
-0.2220528126,
0.200448364,
-0.0561482087,
-0.2992594242,
-0.128759712,
0.0325105563,
-0.0127721205,
0.0843318105,
0.1155449972,
0.0505927131,
0.0815806836,
0.2093013227,
-0.0561162457,
-0.0594240874,
0.7326384187,
-0.1133246943,
-0.1869965494,
-0.3649471402,
0.1383292973,
-0.3404733539,
-0.0648734123,
-0.4481022358,
0.0900492966,
0.3190637231,
0.1263612062,
-0.2260793746,
0.1400437802,
-0.257489711,
0.0272711366,
-0.0578921065,
0.0775433406,
0.2296062559,
-0.298792392,
-0.1100094616,
-0.1358180046
] |
https://github.com/huggingface/datasets/issues/1872 | Adding a new column to the dataset after set_format was called | Hi ! Indeed if you add a column to a formatted dataset, then the new dataset gets a new formatting in which:
```
new formatted columns = (all columns - previously unformatted columns)
```
Therefore the new column is going to be formatted using the `torch` formatting.
If you want your new column to be unformatted you can re-run this line:
```python
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
``` | Hi,
thanks for the nice library. I'm in the process of creating a custom dataset, which has a mix of tensors and lists of strings. I stumbled upon an error and want to know if its a problem on my side.
I load some lists of strings and integers, then call `data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)`. This converts the integer columns into tensors, but keeps the lists of strings as they are. I then call `map` to add a new column to my dataset, which is a **list of strings**. Once I iterate through my dataset, I get an error that the new column can't be converted into a tensor (which is probably caused by `set_format`).
Below some pseudo code:
```python
def augment_func(sample: Dict) -> Dict:
# do something
return {
"some_integer_column1" : augmented_data["some_integer_column1"], # <-- tensor
"some_integer_column2" : augmented_data["some_integer_column2"], # <-- tensor
"NEW_COLUMN": targets, # <-- list of strings
}
data = datasets.load_dataset(__file__, data_dir="...", split="train")
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
augmented_dataset = data.map(augment_func, batched=False)
for sample in augmented_dataset:
print(sample) # fails
```
and the exception:
```python
Traceback (most recent call last):
File "dataset.py", line 487, in <module>
main()
File "dataset.py", line 471, in main
for sample in augmented_dataset:
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 697, in __iter__
yield self._getitem(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1069, in _getitem
outputs = self._convert_outputs(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 890, in _convert_outputs
v = map_nested(command, v, **map_nested_kwargs)
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 851, in command
return torch.tensor(x, **format_kwargs)
TypeError: new(): invalid data type 'str'
```
Thanks!
| 67 | Adding a new column to the dataset after set_format was called
Hi,
thanks for the nice library. I'm in the process of creating a custom dataset, which has a mix of tensors and lists of strings. I stumbled upon an error and want to know if its a problem on my side.
I load some lists of strings and integers, then call `data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)`. This converts the integer columns into tensors, but keeps the lists of strings as they are. I then call `map` to add a new column to my dataset, which is a **list of strings**. Once I iterate through my dataset, I get an error that the new column can't be converted into a tensor (which is probably caused by `set_format`).
Below some pseudo code:
```python
def augment_func(sample: Dict) -> Dict:
# do something
return {
"some_integer_column1" : augmented_data["some_integer_column1"], # <-- tensor
"some_integer_column2" : augmented_data["some_integer_column2"], # <-- tensor
"NEW_COLUMN": targets, # <-- list of strings
}
data = datasets.load_dataset(__file__, data_dir="...", split="train")
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
augmented_dataset = data.map(augment_func, batched=False)
for sample in augmented_dataset:
print(sample) # fails
```
and the exception:
```python
Traceback (most recent call last):
File "dataset.py", line 487, in <module>
main()
File "dataset.py", line 471, in main
for sample in augmented_dataset:
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 697, in __iter__
yield self._getitem(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1069, in _getitem
outputs = self._convert_outputs(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 890, in _convert_outputs
v = map_nested(command, v, **map_nested_kwargs)
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 851, in command
return torch.tensor(x, **format_kwargs)
TypeError: new(): invalid data type 'str'
```
Thanks!
Hi ! Indeed if you add a column to a formatted dataset, then the new dataset gets a new formatting in which:
```
new formatted columns = (all columns - previously unformatted columns)
```
Therefore the new column is going to be formatted using the `torch` formatting.
If you want your new column to be unformatted you can re-run this line:
```python
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
``` | [
-0.1638377607,
-0.174875319,
-0.0401732959,
-0.0492340326,
0.472972095,
0.26253438,
0.6921061873,
0.4177581668,
0.1687510908,
-0.2805867493,
0.1421069652,
0.3377909958,
-0.2340399921,
-0.0259004869,
0.1978131086,
-0.1026415974,
0.1268424988,
0.0638930127,
0.2461462319,
0.058688134,
-0.3777549267,
0.1032360271,
-0.1540208608,
-0.2058054507,
-0.176567018,
-0.3008016348,
-0.0024786443,
-0.1966911852,
-0.1602552235,
-0.2089735717,
0.1871721447,
-0.2391695827,
-0.0005562305,
0.6491597891,
-0.0001178686,
0.1240977272,
0.1696621478,
0.0316144451,
-0.039999865,
-0.1928999871,
-0.2241787314,
-0.3875738978,
-0.1664158404,
-0.3292911947,
0.1402374804,
-0.0900860578,
-0.426571697,
-0.7763676643,
0.2612173557,
0.5016533732,
0.165812552,
0.1779141724,
-0.0243939087,
-0.0470924303,
0.2402332723,
0.2883872986,
-0.2155927122,
-0.0182734057,
-0.0107631646,
0.1259518564,
0.5092816949,
0.1936926693,
-0.2842832208,
0.1020677835,
0.1063983738,
0.2241103798,
-0.2291462421,
-0.042599123,
-0.029220432,
0.1993380785,
0.6996716261,
-0.2541805506,
-0.3642948866,
-0.2796713114,
0.0794098005,
-0.4402055144,
0.1190703809,
-0.0155838802,
-0.0778751224,
-0.1121090204,
0.0912997201,
-0.1839412451,
-0.0351649374,
0.3412308693,
-0.1856347471,
0.0189564358,
-0.0303622596,
0.1920932233,
-0.1479992568,
-0.0397578925,
-0.1474707425,
0.1262124479,
0.0943792015,
0.2419148088,
-0.4515654445,
-0.1876333356,
0.0336438641,
-0.5349577665,
-0.2338589877,
-0.1157704145,
0.1772545725,
0.0031546,
-0.1879830658,
0.1016872451,
0.1837741286,
-0.0243843719,
-0.2377841622,
0.2027950585,
-0.058970958,
-0.2029015422,
0.0786278397,
-0.1089506745,
0.2219045311,
-0.2733575404,
0.1720264554,
0.005271662,
0.415761888,
0.118195951,
-0.0770013109,
0.2277685106,
-0.2703931928,
-0.2729820609,
0.0384286977,
0.0916891247,
-0.0273759849,
0.1280098855,
0.1400444657,
0.2697373033,
-0.0306834467,
0.1996585429,
-0.0518040508,
0.0073223449,
-0.1836285889,
0.0456943512,
0.1025166661,
0.1123386472,
-0.1570650488,
0.2218845487,
-0.1367077082,
-0.0665036887,
0.0907950625,
-0.2595321238,
0.3443580568,
0.1951228827,
-0.1425126642,
0.3772318959,
0.1710850447,
-0.03731969,
-0.1732723415,
0.2277605236,
-0.2527370751,
-0.3213149309,
-0.4081804156,
0.0944306254,
-0.0362260938,
-0.109873414,
0.0837555677,
0.1692531705,
0.5626443624,
-0.1914576888,
0.1816347986,
-0.2311047614,
-0.3539710641,
-0.2420077622,
-0.0327600539,
0.1081715748,
-0.6691623926,
0.094157666,
0.0791117027,
0.3833470047,
0.0772274733,
0.1891555488,
-0.0684039593,
0.3359549344,
-0.2292345166,
0.0447730869,
0.6186285019,
0.0500695184,
-0.1436016858,
0.0653750226,
-0.1200989261,
0.1486836374,
-0.0731993616,
0.1894952059,
0.0845487714,
-0.0850282162,
0.3703550696,
0.0103161409,
-0.1337239444,
-0.1136996821,
-0.0130397901,
0.1656285822,
0.5390791893,
-0.2064687759,
0.0425461568,
0.1790867299,
-0.1250631511,
0.2282995582,
-0.0172854662,
-0.0337754712,
-0.007314302,
0.1034119576,
0.2751896381,
0.0653937757,
0.0355754569,
-0.2091536969,
-0.3796856999,
-0.0080607384,
0.4330010414,
0.0214434043,
-0.0080669336,
-0.1484539509,
-0.377635479,
0.007541582,
-0.011138333,
0.2967941165,
0.0175653063,
0.1665643901,
-0.1642279625,
-0.1772774458,
-0.338902086,
0.0949276015,
-0.1961667538,
-0.0513705686,
-0.2434314787,
0.1282279789,
0.1284629107,
-0.2729172111,
-0.1948997676,
0.3867042661,
0.1357641816,
-0.0230473336,
-0.1044371128,
0.2309469581,
0.0756758153,
-0.1251295507,
-0.4924200773,
0.2001787722,
0.1332603991,
0.1238414496,
-0.1664899141,
-0.1851829588,
-0.011389114,
-0.2167834044,
-0.0754594505,
0.0507363379,
0.0049536126,
0.2880735993,
-0.3090747893,
0.1987279803,
0.1029936895,
0.0727731138,
-0.2182666659,
-0.3290410638,
-0.0560584776,
0.0218477659,
0.044045642,
-0.2663611472,
-0.3668256998,
0.0163655207,
0.6044389009,
0.0032728687,
0.1298659146,
0.1421495378,
0.1310876012,
0.0832141638,
0.0108529236,
0.0890295655,
0.4790549278,
0.0814551264,
-0.0663492233,
0.0239848923,
-0.2592504323,
-0.0981753021,
0.0403770991,
0.2914122939,
0.2057982683,
0.0817340538,
0.2835639417,
0.0091480277,
0.0822450519,
0.0117118433,
0.2326989174,
0.1617717147,
-0.5018966794,
-0.0692813471,
-0.1272542626,
0.2658208311,
-0.2941195369,
-0.6862953901,
0.0144574381,
-0.3176653385,
-0.222138077,
0.0013731825,
0.0248834193,
0.0532939062,
-0.0381292328,
0.1013479307,
0.2265939415,
-0.4410065413,
-0.0932102203,
-0.2546208501,
-0.0083846515,
0.0140890852,
0.3546909094,
-0.3220355809,
0.4234780669,
0.1450336576,
-0.1792605668,
-0.1153638586,
-0.300004065,
-0.017930666,
-0.3884204328,
-0.0696966723,
0.257756561,
0.1516359448,
-0.1477589458,
-0.1888681352,
0.3712129295,
0.1980445385,
0.2000721842,
0.1980656087,
0.1245262027,
0.0569170415,
-0.0044003054,
-0.5688668489,
-0.1075389311,
-0.0420353785,
0.082547456,
-0.0878118053,
0.098474443,
0.0212100558,
0.2226236761,
0.1303483546,
0.2398897707,
-0.0894927979,
-0.3793281019,
-0.0635959059,
0.3975251317,
-0.3226931095,
-0.2073270977,
0.1397509128,
-0.027444683,
0.1810926944,
0.2993006706,
-0.3082209229,
-0.1508527696,
-0.1311354339,
0.173268348,
-0.2809302211,
-0.0994717032,
0.506487608,
0.0120290294,
-0.0551416352,
-0.1404470801,
-0.2117806077,
0.1956443936,
0.2700184584,
0.0633992478,
0.2628599703,
0.5470189452,
0.0884679705,
0.3008493781,
0.0786025301,
-0.497492224,
0.2550722659,
-0.300170362,
0.1966457367,
-0.2971630692,
-0.3937256634,
-0.0502178371,
-0.2342026085,
-0.0278097428,
-0.1490176618,
-0.1602449864,
-0.0322329178,
0.0147371814,
0.5779083371,
-0.1564668119,
-0.1584125161,
0.3528256416,
-0.4837838411,
0.3728367686,
0.0443865657,
0.151432842,
-0.5375988483,
-0.1796581,
0.1274441183,
-0.0810082853,
0.2100244761,
-0.1992973983,
-0.456898272,
-0.1856812239,
-0.4243350625,
0.299892664,
0.3504036665,
0.7206825018,
-0.142863974,
-0.0488888398,
-0.1048104018,
0.1505413055,
0.8693699837,
-0.5131116509,
-0.5122168064,
0.415135026,
0.2547288835,
-0.6216637492,
-0.005700022,
-0.1735691726,
0.0902562588,
0.0768062174,
0.6951244473,
-0.0104240477,
0.1401379406,
0.1530414224,
0.3333969116,
0.0032473132,
-0.1188210845,
-0.165908888,
-0.1762062907,
-0.154374212,
0.0938118994,
0.1468464881,
0.3185719252,
0.3098248541,
-0.278763473,
-0.0822798684,
-0.1979754865,
0.2580941916,
-0.1493848264,
0.2094888389,
0.4666827917,
0.1809091866,
0.0732227266,
-0.0207397193,
-0.0134860687,
0.2105143517,
-0.2058898658,
-0.042640619,
0.1515753865,
-0.3012673259,
0.3134800792,
0.1650595963,
-0.019084489,
0.1892260611,
-0.2329683155,
-0.0420835838,
-0.3310905397,
0.1621736586,
0.5876700878,
0.2107718289,
-0.3579181135,
-0.7363802195,
0.2013386488,
0.0528041124,
0.0794656128,
0.595702529,
-0.0100870654,
-0.1674927473,
0.4298977554,
0.1278736591,
0.8010363579,
-0.2321157902,
0.0219282564,
0.2473866791,
-0.1963775754,
0.4466971755,
-0.219720751,
0.2085923851,
-0.1105169654,
-0.0548964962,
-0.1673021913,
-0.2323620617,
0.3513605595,
0.0045413948,
-0.1806060821,
-0.1170174628,
-0.2128398269,
0.6637678742,
-0.022808779,
0.1052002162,
0.148750931,
-0.4528321624,
-0.3152984083,
0.1043531895,
-0.1289207488,
-0.2021535933,
-0.0943826362,
0.1820896566,
-0.1783923656,
-0.2084894776,
-0.3607930243,
-0.0663564056,
-0.0690188259,
-0.046882242,
0.0452603474,
-0.6082689762,
0.4602431059,
0.3592824042,
0.5291900635,
0.0879843086,
-0.1030568033,
-0.0565396175,
0.0255034342,
0.159540534,
0.2783370912,
-0.4173799753,
0.3079517782,
0.1974585652,
-0.1584799141,
0.3220858574,
0.0993193984,
-0.2854144275,
-0.1614234746,
0.0934313163,
0.5090248585,
-0.6628905535,
-0.3925189376,
0.058613155,
0.2449562252,
-0.2578345537,
0.0919436365,
-0.0922591388,
0.0138550401,
0.3762598038,
-0.326174438,
-0.0639197826,
0.0640055239,
0.2678511739,
0.0133869573,
-0.0737926662,
0.5867012739,
0.202155143,
-0.151566714,
-0.1159016714,
0.0242913961,
0.1093553379,
-0.2628644705,
0.0673721582,
-0.0161280129,
0.1062918454,
-0.1352066994,
0.0582708158,
-0.1605556309,
0.2047501206,
-0.0174371153,
-0.2758021653,
-0.1192292869,
-0.0141021535,
0.2751240134,
0.2663636506,
0.1359471381,
0.2398538142,
-0.1381956488,
0.0437088087,
-0.2351256311,
-0.1114267707,
0.1152759567,
0.2109807432,
0.0536023565,
0.0563840345,
0.1539652795,
-0.4075020552,
0.0870691538,
0.4069390595,
0.028960146,
-0.2118222266,
-0.144543767,
0.1017587632,
0.1174027473,
-0.2523896992,
0.1890086532,
-0.2974908352,
-0.1747688949,
-0.2785750031,
0.0193047076,
0.3081174493,
-0.2520096004,
-0.0051213428,
-0.1206485331,
0.1459066421,
-0.0088554695,
-0.0315668508,
-0.1970040351,
0.0310069323,
0.1981580406,
0.2192882001,
0.2022118568,
0.0440385044,
-0.3146222532,
0.2178010494,
0.1843612194,
0.0563189127,
0.2494205236,
-0.4320303202,
0.0137220323,
0.0009934306,
0.3522247672,
-0.0245945789,
-0.2163707912,
-0.0196517557,
0.2225754708,
0.1150810048,
-0.2363809496,
0.054697983,
0.3840653598,
0.1135784835,
0.1640509963,
0.2766043544,
0.3523633182,
0.1498499811,
-0.2969776988,
-0.028338097,
0.1019252166,
-0.3147269487,
0.3177814484,
0.6366593838,
-0.0096955895,
0.0432767756,
0.2121616602,
0.0385081917,
-0.0184340887,
0.2452430427,
0.1546956897,
0.132366538,
0.5537686944,
-0.0233238563,
-0.1268624663,
-0.3789405227,
0.3849284649,
0.0655263588,
0.1893934309,
0.3565951884,
0.217031464,
0.4540137649,
-0.1332536936,
-0.0467362255,
-0.0272153579,
0.002621308,
-0.1409090906,
0.0477858931,
-0.1536695063,
0.025579378,
0.061330393,
-0.2604689598,
-0.0698649883,
-0.2468578964,
0.2972399294,
-0.0647749677,
0.0898900181,
0.0116087832,
-0.0867137611,
-0.1450733393,
-0.1023247987,
-0.2088188231,
0.1428878605,
-0.0000562109,
-0.0023571718,
0.1685061455,
0.0797036067,
0.3124369979,
0.318662256,
0.0319545083,
0.3550900817,
0.1073564664,
-0.0189530291,
0.2020026892,
0.1107801944,
-0.1152406484,
0.0459669381,
0.2320190817,
0.1205592155,
-0.1050471887,
0.1272470653,
0.0860554874,
0.3713683486,
-0.174245283,
0.1659412831,
-0.0186980572,
-0.2119490206,
-0.2812005579,
-0.0256916992,
-0.0536648035,
-0.2551566362,
0.5833733082,
-0.0171414316,
0.0827369243,
0.1473909914,
0.0789523125,
0.0169139728,
0.1726587564,
0.1242831051,
-0.0562417507,
-0.2194080949,
-0.1105198264,
-0.4495844841,
0.0777269453,
-0.2444207519,
-0.1423134655,
0.1331400722,
0.1083741859,
0.3426305652,
-0.0715121031,
0.3546712101,
-0.4199869931,
-0.3278801143,
0.2354024202,
-0.0737501681,
-0.3474464118,
-0.1158577502,
-0.0218221471,
0.0036886185,
-0.1946222782,
0.1039045081,
0.3052024543,
-0.0131547488,
-0.288513869,
-0.0521493331,
-0.1672354341,
-0.0539736599,
0.5608683228,
0.0929078981,
0.4217212498,
-0.1400010288,
0.0661322996,
-0.0567795224,
0.0003703907,
-0.2170341611,
0.4287862778,
-0.1763378382,
0.4928042889,
-0.2431973219,
0.0879032314,
-0.359551549,
0.2333640009,
-0.0096809119,
0.0360780731,
-0.2453805655,
0.4265150428,
-0.1822774708,
-0.0511563756,
0.1605167091,
0.301143378,
0.0987176597,
0.6137599349,
-0.1845927685,
-0.2232501209,
0.4984791279,
-0.4976071417,
-0.4248705804,
0.0814749449,
-0.0286827311,
0.1198044568,
0.0639729947,
-0.5664727688,
-0.2958456278,
0.4186931849,
-0.099404335,
-0.0072001293,
0.0887091011,
-0.0215017721,
0.1368438154,
-0.1779534519,
0.192484796,
-0.1696066111,
-0.1580831259,
-0.0421707407,
-0.2822971344
] |
https://github.com/huggingface/datasets/issues/1872 | Adding a new column to the dataset after set_format was called | Ok cool :)
Also I just did a PR to mention this behavior in the documentation | Hi,
thanks for the nice library. I'm in the process of creating a custom dataset, which has a mix of tensors and lists of strings. I stumbled upon an error and want to know if its a problem on my side.
I load some lists of strings and integers, then call `data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)`. This converts the integer columns into tensors, but keeps the lists of strings as they are. I then call `map` to add a new column to my dataset, which is a **list of strings**. Once I iterate through my dataset, I get an error that the new column can't be converted into a tensor (which is probably caused by `set_format`).
Below some pseudo code:
```python
def augment_func(sample: Dict) -> Dict:
# do something
return {
"some_integer_column1" : augmented_data["some_integer_column1"], # <-- tensor
"some_integer_column2" : augmented_data["some_integer_column2"], # <-- tensor
"NEW_COLUMN": targets, # <-- list of strings
}
data = datasets.load_dataset(__file__, data_dir="...", split="train")
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
augmented_dataset = data.map(augment_func, batched=False)
for sample in augmented_dataset:
print(sample) # fails
```
and the exception:
```python
Traceback (most recent call last):
File "dataset.py", line 487, in <module>
main()
File "dataset.py", line 471, in main
for sample in augmented_dataset:
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 697, in __iter__
yield self._getitem(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1069, in _getitem
outputs = self._convert_outputs(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 890, in _convert_outputs
v = map_nested(command, v, **map_nested_kwargs)
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 851, in command
return torch.tensor(x, **format_kwargs)
TypeError: new(): invalid data type 'str'
```
Thanks!
| 16 | Adding a new column to the dataset after set_format was called
Hi,
thanks for the nice library. I'm in the process of creating a custom dataset, which has a mix of tensors and lists of strings. I stumbled upon an error and want to know if its a problem on my side.
I load some lists of strings and integers, then call `data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)`. This converts the integer columns into tensors, but keeps the lists of strings as they are. I then call `map` to add a new column to my dataset, which is a **list of strings**. Once I iterate through my dataset, I get an error that the new column can't be converted into a tensor (which is probably caused by `set_format`).
Below some pseudo code:
```python
def augment_func(sample: Dict) -> Dict:
# do something
return {
"some_integer_column1" : augmented_data["some_integer_column1"], # <-- tensor
"some_integer_column2" : augmented_data["some_integer_column2"], # <-- tensor
"NEW_COLUMN": targets, # <-- list of strings
}
data = datasets.load_dataset(__file__, data_dir="...", split="train")
data.set_format("torch", columns=["some_integer_column1", "some_integer_column2"], output_all_columns=True)
augmented_dataset = data.map(augment_func, batched=False)
for sample in augmented_dataset:
print(sample) # fails
```
and the exception:
```python
Traceback (most recent call last):
File "dataset.py", line 487, in <module>
main()
File "dataset.py", line 471, in main
for sample in augmented_dataset:
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 697, in __iter__
yield self._getitem(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 1069, in _getitem
outputs = self._convert_outputs(
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 890, in _convert_outputs
v = map_nested(command, v, **map_nested_kwargs)
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in command
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 850, in <listcomp>
return [map_nested(command, i, **map_nested_kwargs) for i in x]
File "lib/python3.8/site-packages/datasets/utils/py_utils.py", line 225, in map_nested
return function(data_struct)
File "lib/python3.8/site-packages/datasets/arrow_dataset.py", line 851, in command
return torch.tensor(x, **format_kwargs)
TypeError: new(): invalid data type 'str'
```
Thanks!
Ok cool :)
Also I just did a PR to mention this behavior in the documentation | [
-0.1638377607,
-0.174875319,
-0.0401732959,
-0.0492340326,
0.472972095,
0.26253438,
0.6921061873,
0.4177581668,
0.1687510908,
-0.2805867493,
0.1421069652,
0.3377909958,
-0.2340399921,
-0.0259004869,
0.1978131086,
-0.1026415974,
0.1268424988,
0.0638930127,
0.2461462319,
0.058688134,
-0.3777549267,
0.1032360271,
-0.1540208608,
-0.2058054507,
-0.176567018,
-0.3008016348,
-0.0024786443,
-0.1966911852,
-0.1602552235,
-0.2089735717,
0.1871721447,
-0.2391695827,
-0.0005562305,
0.6491597891,
-0.0001178686,
0.1240977272,
0.1696621478,
0.0316144451,
-0.039999865,
-0.1928999871,
-0.2241787314,
-0.3875738978,
-0.1664158404,
-0.3292911947,
0.1402374804,
-0.0900860578,
-0.426571697,
-0.7763676643,
0.2612173557,
0.5016533732,
0.165812552,
0.1779141724,
-0.0243939087,
-0.0470924303,
0.2402332723,
0.2883872986,
-0.2155927122,
-0.0182734057,
-0.0107631646,
0.1259518564,
0.5092816949,
0.1936926693,
-0.2842832208,
0.1020677835,
0.1063983738,
0.2241103798,
-0.2291462421,
-0.042599123,
-0.029220432,
0.1993380785,
0.6996716261,
-0.2541805506,
-0.3642948866,
-0.2796713114,
0.0794098005,
-0.4402055144,
0.1190703809,
-0.0155838802,
-0.0778751224,
-0.1121090204,
0.0912997201,
-0.1839412451,
-0.0351649374,
0.3412308693,
-0.1856347471,
0.0189564358,
-0.0303622596,
0.1920932233,
-0.1479992568,
-0.0397578925,
-0.1474707425,
0.1262124479,
0.0943792015,
0.2419148088,
-0.4515654445,
-0.1876333356,
0.0336438641,
-0.5349577665,
-0.2338589877,
-0.1157704145,
0.1772545725,
0.0031546,
-0.1879830658,
0.1016872451,
0.1837741286,
-0.0243843719,
-0.2377841622,
0.2027950585,
-0.058970958,
-0.2029015422,
0.0786278397,
-0.1089506745,
0.2219045311,
-0.2733575404,
0.1720264554,
0.005271662,
0.415761888,
0.118195951,
-0.0770013109,
0.2277685106,
-0.2703931928,
-0.2729820609,
0.0384286977,
0.0916891247,
-0.0273759849,
0.1280098855,
0.1400444657,
0.2697373033,
-0.0306834467,
0.1996585429,
-0.0518040508,
0.0073223449,
-0.1836285889,
0.0456943512,
0.1025166661,
0.1123386472,
-0.1570650488,
0.2218845487,
-0.1367077082,
-0.0665036887,
0.0907950625,
-0.2595321238,
0.3443580568,
0.1951228827,
-0.1425126642,
0.3772318959,
0.1710850447,
-0.03731969,
-0.1732723415,
0.2277605236,
-0.2527370751,
-0.3213149309,
-0.4081804156,
0.0944306254,
-0.0362260938,
-0.109873414,
0.0837555677,
0.1692531705,
0.5626443624,
-0.1914576888,
0.1816347986,
-0.2311047614,
-0.3539710641,
-0.2420077622,
-0.0327600539,
0.1081715748,
-0.6691623926,
0.094157666,
0.0791117027,
0.3833470047,
0.0772274733,
0.1891555488,
-0.0684039593,
0.3359549344,
-0.2292345166,
0.0447730869,
0.6186285019,
0.0500695184,
-0.1436016858,
0.0653750226,
-0.1200989261,
0.1486836374,
-0.0731993616,
0.1894952059,
0.0845487714,
-0.0850282162,
0.3703550696,
0.0103161409,
-0.1337239444,
-0.1136996821,
-0.0130397901,
0.1656285822,
0.5390791893,
-0.2064687759,
0.0425461568,
0.1790867299,
-0.1250631511,
0.2282995582,
-0.0172854662,
-0.0337754712,
-0.007314302,
0.1034119576,
0.2751896381,
0.0653937757,
0.0355754569,
-0.2091536969,
-0.3796856999,
-0.0080607384,
0.4330010414,
0.0214434043,
-0.0080669336,
-0.1484539509,
-0.377635479,
0.007541582,
-0.011138333,
0.2967941165,
0.0175653063,
0.1665643901,
-0.1642279625,
-0.1772774458,
-0.338902086,
0.0949276015,
-0.1961667538,
-0.0513705686,
-0.2434314787,
0.1282279789,
0.1284629107,
-0.2729172111,
-0.1948997676,
0.3867042661,
0.1357641816,
-0.0230473336,
-0.1044371128,
0.2309469581,
0.0756758153,
-0.1251295507,
-0.4924200773,
0.2001787722,
0.1332603991,
0.1238414496,
-0.1664899141,
-0.1851829588,
-0.011389114,
-0.2167834044,
-0.0754594505,
0.0507363379,
0.0049536126,
0.2880735993,
-0.3090747893,
0.1987279803,
0.1029936895,
0.0727731138,
-0.2182666659,
-0.3290410638,
-0.0560584776,
0.0218477659,
0.044045642,
-0.2663611472,
-0.3668256998,
0.0163655207,
0.6044389009,
0.0032728687,
0.1298659146,
0.1421495378,
0.1310876012,
0.0832141638,
0.0108529236,
0.0890295655,
0.4790549278,
0.0814551264,
-0.0663492233,
0.0239848923,
-0.2592504323,
-0.0981753021,
0.0403770991,
0.2914122939,
0.2057982683,
0.0817340538,
0.2835639417,
0.0091480277,
0.0822450519,
0.0117118433,
0.2326989174,
0.1617717147,
-0.5018966794,
-0.0692813471,
-0.1272542626,
0.2658208311,
-0.2941195369,
-0.6862953901,
0.0144574381,
-0.3176653385,
-0.222138077,
0.0013731825,
0.0248834193,
0.0532939062,
-0.0381292328,
0.1013479307,
0.2265939415,
-0.4410065413,
-0.0932102203,
-0.2546208501,
-0.0083846515,
0.0140890852,
0.3546909094,
-0.3220355809,
0.4234780669,
0.1450336576,
-0.1792605668,
-0.1153638586,
-0.300004065,
-0.017930666,
-0.3884204328,
-0.0696966723,
0.257756561,
0.1516359448,
-0.1477589458,
-0.1888681352,
0.3712129295,
0.1980445385,
0.2000721842,
0.1980656087,
0.1245262027,
0.0569170415,
-0.0044003054,
-0.5688668489,
-0.1075389311,
-0.0420353785,
0.082547456,
-0.0878118053,
0.098474443,
0.0212100558,
0.2226236761,
0.1303483546,
0.2398897707,
-0.0894927979,
-0.3793281019,
-0.0635959059,
0.3975251317,
-0.3226931095,
-0.2073270977,
0.1397509128,
-0.027444683,
0.1810926944,
0.2993006706,
-0.3082209229,
-0.1508527696,
-0.1311354339,
0.173268348,
-0.2809302211,
-0.0994717032,
0.506487608,
0.0120290294,
-0.0551416352,
-0.1404470801,
-0.2117806077,
0.1956443936,
0.2700184584,
0.0633992478,
0.2628599703,
0.5470189452,
0.0884679705,
0.3008493781,
0.0786025301,
-0.497492224,
0.2550722659,
-0.300170362,
0.1966457367,
-0.2971630692,
-0.3937256634,
-0.0502178371,
-0.2342026085,
-0.0278097428,
-0.1490176618,
-0.1602449864,
-0.0322329178,
0.0147371814,
0.5779083371,
-0.1564668119,
-0.1584125161,
0.3528256416,
-0.4837838411,
0.3728367686,
0.0443865657,
0.151432842,
-0.5375988483,
-0.1796581,
0.1274441183,
-0.0810082853,
0.2100244761,
-0.1992973983,
-0.456898272,
-0.1856812239,
-0.4243350625,
0.299892664,
0.3504036665,
0.7206825018,
-0.142863974,
-0.0488888398,
-0.1048104018,
0.1505413055,
0.8693699837,
-0.5131116509,
-0.5122168064,
0.415135026,
0.2547288835,
-0.6216637492,
-0.005700022,
-0.1735691726,
0.0902562588,
0.0768062174,
0.6951244473,
-0.0104240477,
0.1401379406,
0.1530414224,
0.3333969116,
0.0032473132,
-0.1188210845,
-0.165908888,
-0.1762062907,
-0.154374212,
0.0938118994,
0.1468464881,
0.3185719252,
0.3098248541,
-0.278763473,
-0.0822798684,
-0.1979754865,
0.2580941916,
-0.1493848264,
0.2094888389,
0.4666827917,
0.1809091866,
0.0732227266,
-0.0207397193,
-0.0134860687,
0.2105143517,
-0.2058898658,
-0.042640619,
0.1515753865,
-0.3012673259,
0.3134800792,
0.1650595963,
-0.019084489,
0.1892260611,
-0.2329683155,
-0.0420835838,
-0.3310905397,
0.1621736586,
0.5876700878,
0.2107718289,
-0.3579181135,
-0.7363802195,
0.2013386488,
0.0528041124,
0.0794656128,
0.595702529,
-0.0100870654,
-0.1674927473,
0.4298977554,
0.1278736591,
0.8010363579,
-0.2321157902,
0.0219282564,
0.2473866791,
-0.1963775754,
0.4466971755,
-0.219720751,
0.2085923851,
-0.1105169654,
-0.0548964962,
-0.1673021913,
-0.2323620617,
0.3513605595,
0.0045413948,
-0.1806060821,
-0.1170174628,
-0.2128398269,
0.6637678742,
-0.022808779,
0.1052002162,
0.148750931,
-0.4528321624,
-0.3152984083,
0.1043531895,
-0.1289207488,
-0.2021535933,
-0.0943826362,
0.1820896566,
-0.1783923656,
-0.2084894776,
-0.3607930243,
-0.0663564056,
-0.0690188259,
-0.046882242,
0.0452603474,
-0.6082689762,
0.4602431059,
0.3592824042,
0.5291900635,
0.0879843086,
-0.1030568033,
-0.0565396175,
0.0255034342,
0.159540534,
0.2783370912,
-0.4173799753,
0.3079517782,
0.1974585652,
-0.1584799141,
0.3220858574,
0.0993193984,
-0.2854144275,
-0.1614234746,
0.0934313163,
0.5090248585,
-0.6628905535,
-0.3925189376,
0.058613155,
0.2449562252,
-0.2578345537,
0.0919436365,
-0.0922591388,
0.0138550401,
0.3762598038,
-0.326174438,
-0.0639197826,
0.0640055239,
0.2678511739,
0.0133869573,
-0.0737926662,
0.5867012739,
0.202155143,
-0.151566714,
-0.1159016714,
0.0242913961,
0.1093553379,
-0.2628644705,
0.0673721582,
-0.0161280129,
0.1062918454,
-0.1352066994,
0.0582708158,
-0.1605556309,
0.2047501206,
-0.0174371153,
-0.2758021653,
-0.1192292869,
-0.0141021535,
0.2751240134,
0.2663636506,
0.1359471381,
0.2398538142,
-0.1381956488,
0.0437088087,
-0.2351256311,
-0.1114267707,
0.1152759567,
0.2109807432,
0.0536023565,
0.0563840345,
0.1539652795,
-0.4075020552,
0.0870691538,
0.4069390595,
0.028960146,
-0.2118222266,
-0.144543767,
0.1017587632,
0.1174027473,
-0.2523896992,
0.1890086532,
-0.2974908352,
-0.1747688949,
-0.2785750031,
0.0193047076,
0.3081174493,
-0.2520096004,
-0.0051213428,
-0.1206485331,
0.1459066421,
-0.0088554695,
-0.0315668508,
-0.1970040351,
0.0310069323,
0.1981580406,
0.2192882001,
0.2022118568,
0.0440385044,
-0.3146222532,
0.2178010494,
0.1843612194,
0.0563189127,
0.2494205236,
-0.4320303202,
0.0137220323,
0.0009934306,
0.3522247672,
-0.0245945789,
-0.2163707912,
-0.0196517557,
0.2225754708,
0.1150810048,
-0.2363809496,
0.054697983,
0.3840653598,
0.1135784835,
0.1640509963,
0.2766043544,
0.3523633182,
0.1498499811,
-0.2969776988,
-0.028338097,
0.1019252166,
-0.3147269487,
0.3177814484,
0.6366593838,
-0.0096955895,
0.0432767756,
0.2121616602,
0.0385081917,
-0.0184340887,
0.2452430427,
0.1546956897,
0.132366538,
0.5537686944,
-0.0233238563,
-0.1268624663,
-0.3789405227,
0.3849284649,
0.0655263588,
0.1893934309,
0.3565951884,
0.217031464,
0.4540137649,
-0.1332536936,
-0.0467362255,
-0.0272153579,
0.002621308,
-0.1409090906,
0.0477858931,
-0.1536695063,
0.025579378,
0.061330393,
-0.2604689598,
-0.0698649883,
-0.2468578964,
0.2972399294,
-0.0647749677,
0.0898900181,
0.0116087832,
-0.0867137611,
-0.1450733393,
-0.1023247987,
-0.2088188231,
0.1428878605,
-0.0000562109,
-0.0023571718,
0.1685061455,
0.0797036067,
0.3124369979,
0.318662256,
0.0319545083,
0.3550900817,
0.1073564664,
-0.0189530291,
0.2020026892,
0.1107801944,
-0.1152406484,
0.0459669381,
0.2320190817,
0.1205592155,
-0.1050471887,
0.1272470653,
0.0860554874,
0.3713683486,
-0.174245283,
0.1659412831,
-0.0186980572,
-0.2119490206,
-0.2812005579,
-0.0256916992,
-0.0536648035,
-0.2551566362,
0.5833733082,
-0.0171414316,
0.0827369243,
0.1473909914,
0.0789523125,
0.0169139728,
0.1726587564,
0.1242831051,
-0.0562417507,
-0.2194080949,
-0.1105198264,
-0.4495844841,
0.0777269453,
-0.2444207519,
-0.1423134655,
0.1331400722,
0.1083741859,
0.3426305652,
-0.0715121031,
0.3546712101,
-0.4199869931,
-0.3278801143,
0.2354024202,
-0.0737501681,
-0.3474464118,
-0.1158577502,
-0.0218221471,
0.0036886185,
-0.1946222782,
0.1039045081,
0.3052024543,
-0.0131547488,
-0.288513869,
-0.0521493331,
-0.1672354341,
-0.0539736599,
0.5608683228,
0.0929078981,
0.4217212498,
-0.1400010288,
0.0661322996,
-0.0567795224,
0.0003703907,
-0.2170341611,
0.4287862778,
-0.1763378382,
0.4928042889,
-0.2431973219,
0.0879032314,
-0.359551549,
0.2333640009,
-0.0096809119,
0.0360780731,
-0.2453805655,
0.4265150428,
-0.1822774708,
-0.0511563756,
0.1605167091,
0.301143378,
0.0987176597,
0.6137599349,
-0.1845927685,
-0.2232501209,
0.4984791279,
-0.4976071417,
-0.4248705804,
0.0814749449,
-0.0286827311,
0.1198044568,
0.0639729947,
-0.5664727688,
-0.2958456278,
0.4186931849,
-0.099404335,
-0.0072001293,
0.0887091011,
-0.0215017721,
0.1368438154,
-0.1779534519,
0.192484796,
-0.1696066111,
-0.1580831259,
-0.0421707407,
-0.2822971344
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | Hi @alejandrocros it looks like an incompatibility with the current Trainer @sgugger
Indeed currently the Trainer of `transformers` doesn't support a dataset with a transform
It looks like it comes from this line: https://github.com/huggingface/transformers/blob/f51188cbe74195c14c5b3e2e8f10c2f435f9751a/src/transformers/trainer.py#L442
This line sets the format to not return certain unused columns. But this has two issues:
1. it forgets to also set the format_kwargs (this causes the error you got):
```python
dataset.set_format(type=dataset.format["type"], columns=columns, format_kwargs=dataset.format["format_kwargs"])
```
2. the Trainer wants to keep only the fields that are used as input for a model. However for a dataset with a transform, the output fields are often different from the columns fields. For example from a column "text" in the dataset, the strings can be transformed on-the-fly into "input_ids". If you want your dataset to only output certain fields and not other you must change your transform function.
| Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 139 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
Hi @alejandrocros it looks like an incompatibility with the current Trainer @sgugger
Indeed currently the Trainer of `transformers` doesn't support a dataset with a transform
It looks like it comes from this line: https://github.com/huggingface/transformers/blob/f51188cbe74195c14c5b3e2e8f10c2f435f9751a/src/transformers/trainer.py#L442
This line sets the format to not return certain unused columns. But this has two issues:
1. it forgets to also set the format_kwargs (this causes the error you got):
```python
dataset.set_format(type=dataset.format["type"], columns=columns, format_kwargs=dataset.format["format_kwargs"])
```
2. the Trainer wants to keep only the fields that are used as input for a model. However for a dataset with a transform, the output fields are often different from the columns fields. For example from a column "text" in the dataset, the strings can be transformed on-the-fly into "input_ids". If you want your dataset to only output certain fields and not other you must change your transform function.
| [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | FYI that option can be removed with `remove_unused_columns = False` in your `TrainingArguments`, so there is a workaround @alexvaca0 while the fix in `Trainer` is underway.
@lhoestq I think I will just use the line you suggested and if someone is using the columns that are removed in their transform they will need to change `remove_unused_columns` to `False`. We might switch the default of that argument in the next version if that proves too bug-proof. | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 75 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
FYI that option can be removed with `remove_unused_columns = False` in your `TrainingArguments`, so there is a workaround @alexvaca0 while the fix in `Trainer` is underway.
@lhoestq I think I will just use the line you suggested and if someone is using the columns that are removed in their transform they will need to change `remove_unused_columns` to `False`. We might switch the default of that argument in the next version if that proves too bug-proof. | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | I've tried your solutions @sgugger @lhoestq and the good news is that it throws no error. However, TPU training is taking forever, in 1 hour it has only trained 1 batch of 8192 elements, which doesn't make much sense... Is it possible that "on the fly" tokenization of batches is slowing down TPU training to that extent? | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 57 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
I've tried your solutions @sgugger @lhoestq and the good news is that it throws no error. However, TPU training is taking forever, in 1 hour it has only trained 1 batch of 8192 elements, which doesn't make much sense... Is it possible that "on the fly" tokenization of batches is slowing down TPU training to that extent? | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | I don't know what the value of `padding` is in your lines of code pasted above so I can't say for sure. The first batch will be very slow on TPU since it compiles everything, so that's normal (1 hour is long but 8192 elements is also large). Then if your batches are not of the same lengths, it will recompile everything at each step instead of using the same graph, which will be very slow, so you should double check you are using padding to make everything the exact same shape. | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 92 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
I don't know what the value of `padding` is in your lines of code pasted above so I can't say for sure. The first batch will be very slow on TPU since it compiles everything, so that's normal (1 hour is long but 8192 elements is also large). Then if your batches are not of the same lengths, it will recompile everything at each step instead of using the same graph, which will be very slow, so you should double check you are using padding to make everything the exact same shape. | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | I have tried now on a GPU and it goes smooth! Amazing feature .set_transform() instead of .map()! Now I can pre-train my model without the hard disk limitation. Thanks for your work all HuggingFace team!! :clap: | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 36 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
I have tried now on a GPU and it goes smooth! Amazing feature .set_transform() instead of .map()! Now I can pre-train my model without the hard disk limitation. Thanks for your work all HuggingFace team!! :clap: | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | In the end, to make it work I turned to A-100 gpus instead of TPUS, among other changes. Set_transform doesn't work as expected and slows down training very much even in GPUs, and applying map destroys the disk, as it multiplies by 100 the size of the data passed to it (due to inefficient implementation converting strings to int64 floats I guess). For that reason, I chose to use datasets to load the data as text, and then edit the Collator from Transformers to tokenize every batch it receives before processing it. That way, I'm being able to train fast, without memory breaks, without the disk being unnecessarily filled, while making use of GPUs almost all the time I'm paying for them (the map function over the whole dataset took ~15hrs, in which you're not training at all). I hope this info helps others that are looking for training a language model from scratch cheaply, I'm going to close the issue as the optimal solution I found after many experiments to the problem posted in it is explained above. | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 179 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
In the end, to make it work I turned to A-100 gpus instead of TPUS, among other changes. Set_transform doesn't work as expected and slows down training very much even in GPUs, and applying map destroys the disk, as it multiplies by 100 the size of the data passed to it (due to inefficient implementation converting strings to int64 floats I guess). For that reason, I chose to use datasets to load the data as text, and then edit the Collator from Transformers to tokenize every batch it receives before processing it. That way, I'm being able to train fast, without memory breaks, without the disk being unnecessarily filled, while making use of GPUs almost all the time I'm paying for them (the map function over the whole dataset took ~15hrs, in which you're not training at all). I hope this info helps others that are looking for training a language model from scratch cheaply, I'm going to close the issue as the optimal solution I found after many experiments to the problem posted in it is explained above. | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1867 | ERROR WHEN USING SET_TRANSFORM() | Great comment @alexvaca0 . I think that we could re-open the issue as a reformulation of why it takes so much space to save the arrow. Saving a 1% of oscar corpus takes more thank 600 GB (it breaks when it pass 600GB because it is the free memory that I have at this moment) when the full dataset is 1,3 TB. I have a 1TB M.2 NVMe disk that I can not train on because the saved .arrow files goes crazily big. If you can share your Collator I will be grateful. | Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
| 93 | ERROR WHEN USING SET_TRANSFORM()
Hi, I'm trying to use dataset.set_transform(encode) as @lhoestq told me in this issue: https://github.com/huggingface/datasets/issues/1825#issuecomment-774202797
However, when I try to use Trainer from transformers with such dataset, it throws an error:
```
TypeError: __init__() missing 1 required positional argument: 'transform'
[INFO|trainer.py:357] 2021-02-12 10:18:09,893 >> The following columns in the training set don't have a corresponding argument in `AlbertForMaskedLM.forward` and have been ignored: text.
Exception in device=TPU:0: __init__() missing 1 required positional argument: 'transform'
Traceback (most recent call last):
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 330, in _mp_start_fn
_start_fn(index, pf_cfg, fn, args)
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/torch_xla/distributed/xla_multiprocessing.py", line 324, in _start_fn
fn(gindex, *args)
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 368, in _mp_fn
main()
File "/home/alejandro_vaca/transformers/examples/language-modeling/run_mlm_wwm.py", line 332, in main
data_collator=data_collator,
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 286, in __init__
self._remove_unused_columns(self.train_dataset, description="training")
File "/anaconda3/envs/torch-xla-1.7/lib/python3.6/site-packages/transformers/trainer.py", line 359, in _remove_unused_columns
dataset.set_format(type=dataset.format["type"], columns=columns)
File "/home/alejandro_vaca/datasets/src/datasets/fingerprint.py", line 312, in wrapper
out = func(self, *args, **kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/arrow_dataset.py", line 818, in set_format
_ = get_formatter(type, **format_kwargs)
File "/home/alejandro_vaca/datasets/src/datasets/formatting/__init__.py", line 112, in get_formatter
return _FORMAT_TYPES[format_type](**format_kwargs)
TypeError: __init__() missing 1 required positional argument: 'transform'
```
The code I'm using:
```{python}
def tokenize_function(examples):
# Remove empty lines
examples["text"] = [line for line in examples["text"] if len(line) > 0 and not line.isspace()]
return tokenizer(examples["text"], padding=padding, truncation=True, max_length=data_args.max_seq_length)
datasets.set_transform(tokenize_function)
data_collator = DataCollatorForWholeWordMask(tokenizer=tokenizer, mlm_probability=data_args.mlm_probability)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=datasets["train"] if training_args.do_train else None,
eval_dataset=datasets["val"] if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
```
I've installed from source, master branch.
Great comment @alexvaca0 . I think that we could re-open the issue as a reformulation of why it takes so much space to save the arrow. Saving a 1% of oscar corpus takes more thank 600 GB (it breaks when it pass 600GB because it is the free memory that I have at this moment) when the full dataset is 1,3 TB. I have a 1TB M.2 NVMe disk that I can not train on because the saved .arrow files goes crazily big. If you can share your Collator I will be grateful. | [
-0.1826190799,
-0.2732515335,
0.139303118,
0.1847401708,
0.7292687297,
0.1197056025,
0.6009584069,
0.2287763655,
-0.3227849305,
0.0544776842,
0.2355117202,
0.033326745,
-0.0883406848,
-0.0234048478,
0.1892214566,
-0.5358200073,
-0.0138221234,
0.1288612187,
-0.4634345174,
-0.1831161976,
-0.276696831,
0.3134230077,
-0.3415371776,
0.1193405986,
-0.3186004162,
-0.0096086636,
-0.0476956666,
0.0164713264,
0.0844154507,
-0.1718305945,
0.2334781885,
-0.2800756395,
0.123195827,
0.3974504173,
-0.0001255372,
0.1195577979,
0.0606355742,
-0.2890720367,
-0.236786887,
-0.3003558517,
-0.1437312514,
-0.1552793086,
-0.0154345073,
-0.2408747971,
-0.4479369521,
-0.1890556216,
0.0066992268,
-0.2687550187,
0.5653690696,
0.7264814973,
0.1215874627,
0.2663603425,
0.0996967629,
-0.2382418662,
-0.1261350513,
0.1903081685,
-0.1291117519,
-0.022498332,
0.0317757204,
0.2165265381,
0.133624211,
0.161550492,
0.2006310821,
-0.1752623618,
0.4064541161,
-0.0619788989,
-0.1365287602,
0.0435065553,
0.0006193062,
0.0104682148,
0.2298872173,
-0.3454309404,
-0.4208543003,
-0.2735255957,
0.165444687,
-0.2141008526,
0.2700581551,
0.150387466,
-0.0663979799,
0.2279438376,
-0.2600677013,
-0.1098377854,
-0.058866486,
0.0313305967,
-0.1728425324,
0.162386477,
-0.006507393,
0.1303313971,
0.093178913,
0.1813862324,
0.2739948332,
0.1153551936,
0.1876619756,
0.3649874032,
-0.2653560042,
-0.0547861755,
-0.143478632,
0.0108284783,
-0.2889723778,
-0.2492739558,
-0.1723645329,
-0.2085099518,
0.0230884254,
-0.0149521548,
0.107146956,
0.6613351703,
-0.2454049289,
0.4977245033,
0.0151178241,
-0.112073496,
0.2516208589,
-0.0635673925,
-0.0569486432,
-0.3780822158,
-0.0096243173,
0.4175061285,
0.0937966555,
-0.0060419217,
-0.2874678969,
-0.1932771206,
-0.2089737654,
-0.0578819402,
0.1327710748,
0.217986092,
0.0588068366,
0.2713344395,
0.5032312274,
0.3485565484,
-0.1059521288,
-0.2050583065,
-0.1715609133,
0.1350133121,
-0.3857187629,
-0.1006991267,
0.1464199722,
-0.1412292868,
0.3286193609,
0.0532121845,
0.0755416751,
0.3815330565,
0.1520246267,
-0.1911822408,
0.2819839716,
0.1436704695,
-0.0143269077,
0.2014977187,
0.4693850875,
0.1014425009,
-0.186176762,
0.2664016783,
-0.2484503984,
-0.4411481917,
0.0609497875,
0.0914024562,
-0.0530005321,
-0.0377963297,
-0.1594835818,
0.1583796144,
0.420529604,
-0.0675123632,
0.0763092935,
-0.4352329671,
0.0418042317,
0.1099923402,
0.0900249705,
0.0778380707,
-0.2328577191,
-0.3426038027,
0.3258579969,
0.1530644447,
-0.1433434039,
0.2205608785,
-0.4285852015,
0.2562271357,
-0.0625739545,
-0.1584209055,
0.2098878324,
-0.3840964735,
-0.0710870028,
-0.099427551,
-0.1054711267,
-0.2191637754,
0.0696845651,
-0.1317370385,
-0.0530191436,
-0.1544255912,
-0.2207259089,
0.5086042881,
-0.0608971938,
0.2324943841,
0.0340902358,
-0.022244487,
0.1885961741,
0.1376517266,
0.0248881485,
0.1576398015,
-0.3205076456,
0.5120713711,
0.0987584144,
-0.2441779077,
0.2444195896,
0.4102390707,
-0.2440974563,
-0.1919647902,
0.0468019992,
0.2764774561,
-0.0697960556,
-0.1489025503,
0.1784001291,
0.196351558,
-0.0604719222,
0.0484522581,
-0.2369386256,
-0.0641726255,
-0.213375181,
-0.1585211158,
-0.0149896629,
0.1101618856,
0.0207734182,
-0.0464727804,
-0.2484814525,
0.2723910213,
-0.0954265073,
0.1349617839,
-0.5259901881,
0.2280209959,
-0.2145735323,
-0.0914451331,
-0.0678842664,
0.2231203914,
0.2404942662,
-0.230519712,
-0.3247488737,
0.3036766052,
-0.2122561038,
0.0371920615,
-0.3943193555,
0.245739758,
0.3090949953,
-0.5342990756,
0.0032726359,
0.263441503,
0.1639745384,
-0.127455771,
-0.3830228448,
0.1136110127,
0.075772807,
0.331451565,
0.1057743579,
0.3340391815,
0.0536120757,
-0.0884055272,
-0.2020369768,
-0.0776446611,
-0.0154919419,
-0.0838232711,
0.1649817824,
0.0597496852,
-0.2288422137,
-0.198387742,
0.4673169553,
0.1488420963,
0.021630194,
0.1293900311,
-0.1397811174,
0.0306706131,
0.1618633717,
-0.0950354263,
0.5624227524,
0.0324374251,
-0.2305445224,
0.1875145584,
0.069772251,
-0.1226136982,
0.4002858102,
-0.0731958747,
0.0252315477,
-0.1065489352,
0.0615076907,
0.0952067524,
-0.1649063826,
-0.2894913256,
0.2562519908,
0.2521975935,
-0.4736780524,
0.2136496603,
0.0259738415,
0.1696037203,
-0.5800776482,
-0.3827399909,
0.0345896259,
-0.2872624993,
-0.10989663,
-0.0208653025,
0.1248782724,
0.4439494312,
0.1025985628,
0.0722723082,
0.2448717207,
-0.1580226123,
-0.0618828833,
0.0459854193,
-0.1911006272,
0.0143577084,
-0.1871763319,
-0.2879668176,
-0.0504698977,
-0.3237555623,
-0.0552907735,
-0.103540495,
-0.1793443859,
0.1855939776,
-0.2955769002,
0.0113008711,
0.3231824636,
0.1871946305,
-0.1992334276,
-0.3188308179,
0.4316529632,
0.0009159446,
0.0959324315,
-0.0127254203,
-0.0703297779,
-0.2425716221,
-0.0997436419,
-0.4264296889,
-0.0446454957,
-0.1689537615,
0.0303075016,
-0.2177822143,
-0.1313050091,
0.4203009009,
0.2371351123,
0.0843031406,
0.2997520566,
0.1740505993,
-0.019832084,
-0.2111929953,
0.4776296914,
-0.2067955285,
-0.3576398194,
-0.127142489,
-0.0129220597,
0.2218755186,
0.2340478301,
-0.1764792502,
-0.178255707,
-0.2040158808,
-0.1126634032,
-0.3014697433,
0.4359829426,
0.4326147437,
0.0603175908,
0.168608129,
-0.050866209,
-0.5156285763,
0.3171767294,
0.1424026191,
0.4169047475,
-0.0929956511,
0.4798955321,
0.0866111964,
0.7751234174,
0.0111009851,
-0.3219003379,
0.260661304,
-0.1630744338,
0.2260282636,
-0.1403172016,
-0.4625391066,
0.2768844664,
0.2438236326,
0.031610623,
-0.0728214681,
-0.0910807401,
-0.0698522329,
-0.1445437223,
0.2413929403,
-0.105251506,
-0.1522357762,
0.1416440457,
-0.1808566004,
0.5583440661,
0.0370700955,
0.3895748556,
-0.2105326951,
0.0093454383,
0.0729867369,
-0.1095193475,
0.0688645095,
0.1238155365,
-0.1383382678,
-0.5897535086,
-0.1902555525,
0.1065890193,
-0.0117326379,
0.3856188953,
-0.1194273829,
-0.0214354992,
0.1513936818,
0.1890705526,
0.7310062051,
0.0729367658,
-0.32788077,
-0.0309845544,
-0.2211036235,
-0.6239109635,
-0.0709217936,
0.0394207835,
-0.165397644,
-0.1467244625,
0.1922558099,
-0.1503566056,
-0.1719056964,
0.3554315567,
0.0347531699,
-0.2583992481,
0.0776253715,
-0.0642112046,
-0.5231647491,
-0.375294894,
0.0817700922,
0.2867944539,
0.1763401031,
0.2706167698,
0.370555371,
-0.0252024289,
-0.0146309137,
0.1825231314,
0.0229771212,
0.2532840073,
0.120098263,
-0.0711412877,
0.2466864288,
0.2663378119,
0.0747499987,
0.3784387112,
0.0218266249,
-0.1386083066,
-0.0059656389,
-0.0273004398,
0.3917845488,
0.315844357,
-0.1038671136,
-0.2018497884,
0.000328552,
0.0675853863,
-0.0518517867,
0.2731792927,
0.4600664377,
0.0707788318,
-0.1523293853,
-0.5686460733,
0.1935224235,
0.1283531934,
0.1165831685,
0.0412618369,
0.1834687889,
-0.4985372424,
0.3000657856,
0.2441838086,
0.9627008438,
0.0045671933,
0.4122145176,
0.4066114426,
-0.1718393564,
-0.0448784269,
-0.1133906618,
0.1925233006,
-0.2689212263,
-0.374070704,
-0.126744166,
-0.1576801836,
0.1537522674,
0.1447482705,
-0.4471734762,
0.257401973,
-0.1016520262,
0.132446304,
-0.0517068133,
-0.0063032955,
-0.119157508,
-0.3786029518,
-0.4207050502,
0.0605766214,
0.0085387621,
0.2239294499,
-0.0211464986,
-0.0518367924,
-0.0725241452,
-0.3276564479,
-0.1989794225,
0.1438138932,
-0.342553407,
0.116275534,
0.0215493292,
-0.3508207202,
0.5728161335,
0.476177901,
0.5377931595,
0.0480026267,
-0.0448533632,
0.2743346691,
-0.2238796353,
0.2336568087,
0.0026794486,
0.1144547313,
0.4311789572,
-0.0262234248,
0.0106902868,
0.2838857472,
-0.1292874664,
-0.0491684601,
-0.1847701669,
-0.270021677,
0.0675759465,
-0.3728923798,
-0.21563375,
0.1479791701,
-0.0138649493,
-0.1213468537,
0.0629487038,
-0.1453509778,
-0.2010598481,
0.1748766899,
-0.1926722229,
-0.3175872564,
-0.0471323915,
0.5520837307,
-0.0154432608,
-0.0259579904,
0.9083887339,
0.1442696452,
0.0290533006,
-0.0506253056,
0.002759479,
0.5083622336,
-0.7060850859,
0.190207392,
0.044279214,
-0.1163030788,
-0.2586609423,
0.2225266695,
0.0910592377,
-0.1223972365,
0.106648311,
0.0268161297,
-0.3815153539,
0.043569807,
0.1826021522,
0.1099414527,
-0.0063494071,
0.4157491922,
0.3691781163,
-0.1609195173,
-0.2136099339,
0.085434325,
0.0152924135,
0.2436428517,
0.1348816156,
-0.1812809557,
0.1046393663,
-0.21582748,
0.0344827399,
-0.011733301,
0.0288340151,
-0.1036320776,
-0.3030481339,
0.0960220173,
0.1902741492,
-0.1844202429,
-0.0305876564,
-0.3210197687,
-0.2273690403,
-0.2802014947,
0.1485086679,
0.2875710726,
-0.0231883079,
0.4670306146,
-0.0447170474,
-0.1073347777,
-0.1710857153,
-0.3624246418,
-0.0558780432,
-0.2588468194,
0.2030041367,
0.1647433639,
0.0901971385,
-0.0337182581,
-0.5971802473,
0.2267416418,
0.248593837,
0.2303081602,
0.0678241402,
-0.2015693486,
0.1524336785,
0.1655962169,
0.2326399982,
-0.0341042653,
-0.2902601063,
0.020633243,
0.1774319261,
0.0617807582,
-0.285402149,
0.3496540189,
-0.0410755202,
0.0037133172,
0.0104911849,
0.2545005679,
0.0580313504,
-0.0115493834,
0.0494864509,
0.0794291943,
0.2100776732,
-0.0605450496,
0.3345695138,
0.77545017,
-0.2804123759,
0.2012108266,
0.2750224769,
0.145867154,
0.2543020546,
0.3205239177,
-0.1166608334,
0.3449279964,
-0.018806614,
-0.0199499018,
-0.0652347058,
-0.1228848994,
-0.310480684,
-0.1817119718,
0.0402110703,
0.0242763497,
0.0616915077,
0.5748894215,
-0.2628113329,
-0.3483904302,
-0.2398827225,
0.27559039,
-0.1278466731,
-0.0154439062,
-0.0172689334,
0.0287495852,
-0.3195880651,
-0.1499737203,
-0.212563023,
-0.1377665102,
0.6361115575,
0.0198254418,
-0.0023894086,
-0.1892207563,
-0.0454765595,
0.2628239095,
0.0819953158,
-0.1269654185,
0.241137132,
0.1011123508,
-0.1883557737,
0.0976071656,
0.4976043999,
0.2201209664,
0.2194080353,
-0.0463455468,
-0.0102585554,
-0.0782286674,
-0.0712296888,
0.1536552608,
0.2663583457,
-0.1075949371,
0.3812071979,
0.2045421749,
0.0881866962,
-0.0691199899,
0.1516285688,
-0.2430787385,
0.3113427758,
-0.2424249202,
0.5272915959,
-0.0208063424,
-0.1454467177,
-0.1479762644,
0.0642576814,
-0.1720504761,
0.1480784863,
0.3423068523,
0.0077573732,
0.1103841066,
-0.1328212023,
0.0115131028,
0.2430802286,
0.2766731977,
0.520721972,
-0.2535750568,
-0.3213944435,
-0.2156716585,
-0.3150945306,
-0.0384825096,
0.1294474602,
0.017416127,
-0.0245441925,
-0.176449731,
0.3041369021,
-0.0779176503,
0.0809711814,
-0.106515795,
-0.1017207205,
0.2112428546,
-0.0689959899,
-0.3980530798,
-0.1088813692,
0.379093498,
-0.0227049664,
-0.0859458745,
-0.1128537953,
0.4846833348,
-0.0293583721,
-0.3693024814,
0.0441123694,
-0.1199800968,
-0.1286894828,
0.2373431325,
0.0764207393,
0.2086418569,
0.0588338003,
0.2778176963,
0.0176055059,
-0.2506175041,
-0.3266899586,
0.0686158389,
-0.0857319683,
0.4292595983,
-0.2401909828,
-0.1764760315,
-0.4258552492,
0.0071113259,
0.1781209409,
0.0464420021,
-0.4590158165,
0.4230371118,
-0.3671389222,
0.0823737979,
0.241683796,
0.3583959937,
0.0034973957,
0.1476844996,
-0.3863492012,
-0.424439013,
0.6188311577,
-0.1814698726,
-0.4834170341,
0.1206564903,
-0.1362354755,
-0.0846379772,
-0.1597630382,
-0.5339646339,
0.0789653957,
0.1070880815,
-0.2078461051,
-0.3984490633,
0.2419901639,
0.1348472834,
0.0488823503,
-0.0512378328,
0.4404886067,
-0.0221920311,
-0.2021323442,
-0.1810315847,
-0.0351454914
] |
https://github.com/huggingface/datasets/issues/1859 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU) | Hi @corticalstack ! Thanks for reporting. Indeed in the recent versions of Faiss we must use `getDevice` to check if the index in on GPU.
I'm opening a PR | Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
| 29 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU)
Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
Hi @corticalstack ! Thanks for reporting. Indeed in the recent versions of Faiss we must use `getDevice` to check if the index in on GPU.
I'm opening a PR | [
-0.071238637,
-0.3793424368,
-0.0126396045,
0.1242899075,
0.325030148,
0.1440669,
0.3014211953,
0.5358148217,
0.477586031,
0.3618175983,
0.068253383,
-0.0320641324,
0.1739505231,
-0.0848502219,
-0.1260115355,
0.1587207317,
0.4708420038,
0.2794695497,
0.4284846783,
-0.0465699695,
-0.4897719026,
0.0811467543,
0.0083704442,
0.0710057169,
0.0153227076,
-0.036025852,
-0.0760060325,
-0.0318542533,
-0.1387519836,
-0.4813568294,
-0.128130585,
-0.2600699365,
0.4309388995,
0.1854506284,
-0.0001175451,
0.0909123048,
0.2596595585,
-0.1979711056,
-0.0848164633,
-0.0664211959,
0.3481679857,
0.0831487328,
0.1509593129,
-0.3121952415,
-0.2275110483,
-0.3267766237,
-0.0085449554,
-0.1260352731,
-0.1798978746,
0.2844126821,
0.1657339185,
-0.1552809775,
0.460234642,
-0.0657819957,
-0.0353392363,
0.08556775,
-0.264105618,
-0.1356429458,
0.0376554914,
0.5125662088,
0.2819298804,
0.3212585449,
0.0330726504,
-0.2802489996,
-0.180924207,
-0.0009557866,
0.6283460259,
-0.345906496,
0.1626063436,
-0.1556596458,
0.0611041337,
-0.1890731007,
-0.345608145,
0.289567709,
0.2570061684,
-0.4825875461,
-0.063222833,
0.044311963,
-0.2356058955,
0.1721523851,
0.6439071894,
-0.2015717328,
0.0719454139,
-0.1651589274,
0.203163594,
-0.052415967,
-0.0817488655,
0.0932568759,
0.0007189042,
-0.132093519,
-0.1938660294,
0.054487519,
0.1594011933,
0.1064693034,
-0.3842538595,
-0.1285609752,
0.0794625729,
-0.4333072305,
-0.1163277626,
0.1132656932,
-0.4117618203,
0.013839582,
0.1884584725,
0.0297547746,
-0.2788293064,
-0.0231556259,
-0.0122408457,
-0.202277422,
-0.0987562686,
0.0644985586,
0.04379116,
-0.3351604342,
-0.162530601,
0.0169741213,
-0.3138407469,
0.1340077966,
0.264089644,
-0.3600541949,
-0.5404112339,
0.1515233815,
-0.2051827013,
0.2541143,
0.1186715961,
0.3394497335,
0.3243596852,
0.1193252057,
0.2600006163,
0.294056505,
-0.2567107677,
0.0723284483,
-0.1546052694,
-0.1204362437,
0.3157217503,
0.1273837984,
-0.0999877229,
-0.4395748973,
-0.0559441373,
0.0610692576,
-0.2190775424,
0.0148684308,
0.039002791,
-0.3672311902,
0.1371367425,
0.4929430187,
0.0687846392,
0.1333897859,
0.3695408702,
-0.1329856813,
-0.1448983848,
0.3369979262,
0.0195288602,
-0.2087259591,
-0.4237850606,
0.1341566741,
0.2424477935,
-0.0107824393,
0.3436017931,
0.0485057794,
0.0278703067,
-0.0147275329,
0.0952939689,
0.1335859299,
0.0630661249,
-0.2642408013,
0.4410451651,
0.0570826642,
-0.0643291175,
-0.0238724649,
0.1438175887,
-0.1553905308,
0.2431761026,
0.4800499082,
0.4059357643,
0.236102134,
-0.3073672056,
0.297696352,
0.4374848306,
-0.292992413,
-0.0848620832,
0.0937045962,
-0.1128478199,
-0.4517081976,
0.3368116617,
0.0265277736,
0.1492989361,
-0.0534780696,
0.2210409045,
-0.14187482,
-0.0379883796,
-0.1587761343,
-0.3545044363,
-0.3207325339,
0.3053148091,
-0.0515019,
-0.0426947847,
0.1113009155,
-0.1175485328,
-0.5868123174,
0.1098725796,
-0.0735581517,
-0.0875956416,
-0.0569468252,
0.932297647,
0.027881667,
0.2512435317,
-0.0496050268,
0.2162653208,
-0.0534442663,
0.0089389905,
-0.1493975222,
-0.3996338844,
-0.0829539672,
-0.0047224909,
-0.0199987032,
0.3082150221,
0.1228877306,
0.0533142537,
-0.1235306114,
-0.1038724333,
0.2014549524,
-0.2821292281,
-0.1687681526,
-0.1973064244,
0.0413984694,
0.0628547072,
0.1378223598,
-0.337028116,
-0.4472597539,
-0.1559963524,
0.2468760312,
0.3647454381,
-0.2399561852,
-0.0033570006,
0.3587985933,
0.1815852225,
-0.0744803399,
0.2085340917,
0.0854676664,
-0.0874268636,
-0.0320983008,
-0.0945049301,
0.473426342,
0.4213306904,
-0.0320823267,
0.1105893701,
0.3010747731,
0.2076683939,
0.0950360745,
-0.1479112506,
0.1323514283,
0.4306557477,
0.0921608955,
0.0992562473,
-0.1312623024,
-0.097616531,
0.3995726705,
0.1208031699,
-0.2026576698,
-0.610147655,
0.1275204122,
-0.0361715928,
-0.0758983344,
0.2764010429,
-0.0513892584,
-0.1054838151,
0.0300884023,
0.0499134585,
-0.8405377865,
0.1827555448,
0.0439396724,
-0.2287988961,
-0.1709544212,
-0.0537618548,
-0.1136792451,
0.3696150184,
-0.154349789,
-0.1558541209,
0.1671735197,
-0.1639796793,
-0.0550497472,
-0.0069373921,
0.0262609795,
-0.1074864939,
0.0510733053,
-0.3698428273,
0.0085147917,
0.0789711103,
0.2077314109,
-0.293381989,
-0.1897428632,
-0.0669157952,
0.0192803741,
0.1494901478,
0.2111280859,
-0.026634682,
0.2719706595,
-0.2457416505,
0.2465219498,
0.2640500069,
0.0451858267,
-0.1872446984,
-0.0682792291,
-0.0503306501,
-0.0573592484,
0.0380740985,
0.0472575091,
0.1168287396,
-0.1141841784,
0.0822906047,
-0.2946703136,
0.0565170348,
0.0235372093,
-0.2850289643,
-0.1330880672,
-0.1121564358,
0.2326911986,
0.089738667,
0.1159243286,
0.1796199381,
-0.0158458501,
-0.2538018823,
0.1694007367,
-0.2895035148,
0.0810748786,
-0.2157639861,
-0.0930475742,
-0.015356807,
-0.2302936167,
0.057571888,
-0.0329137854,
0.1174359471,
-0.1404097676,
0.2576201558,
0.2002504468,
0.3634290397,
-0.2162123919,
-0.1813195944,
-0.1422647834,
0.1777140349,
0.0998023748,
-0.4149799347,
0.3305245638,
0.1879828274,
-0.0797433108,
0.0312865861,
0.0676987469,
0.1965958774,
0.1595302373,
0.1310822815,
-0.3357593119,
0.1694400012,
0.4078320861,
0.0772914067,
-0.1620265841,
-0.0135010779,
-0.0939563662,
0.0710932612,
0.0425382927,
0.2906357646,
-0.192415148,
0.3714407384,
-0.1209216118,
0.5271897912,
-0.0674376041,
-0.4186007977,
0.203013733,
-0.095128715,
0.1291924119,
-0.0828326494,
-0.1675728858,
0.4965704978,
0.366768986,
-0.3319827616,
0.0488785505,
-0.4631190598,
0.0197122414,
0.1105109155,
0.2581324279,
-0.0941583663,
-0.0665520057,
0.0787322596,
-0.0464893989,
0.2993481457,
-0.5147417188,
0.3564273417,
-0.0554457344,
0.211345613,
0.2320540249,
0.453791678,
0.2656051815,
-0.1263950616,
0.2204960436,
-0.2902863622,
-0.3792003393,
0.4615425169,
-0.0284015741,
0.3594942093,
-0.0985930562,
-0.1411705166,
0.286729604,
0.0377817936,
0.5474436283,
0.2116204351,
-0.3290473819,
0.4518079162,
0.1415421367,
-0.3843214512,
-0.1611050367,
-0.1790841669,
0.2378774732,
0.0109718144,
0.4116705358,
-0.1174165756,
-0.3329497278,
-0.3428001404,
-0.0415189974,
-0.2220502496,
-0.2910756469,
-0.4210226536,
-0.3375882804,
-0.1586616039,
0.4674515128,
0.0171479061,
0.3032742143,
-0.1070556641,
0.1508087367,
-0.4915827513,
-0.1076555252,
-0.0780056044,
-0.176077947,
0.2934149802,
-0.1969273686,
-0.09105023,
0.0964751095,
0.2985574305,
0.4689680338,
-0.1375472546,
0.3453900218,
0.1153776348,
0.2162031829,
0.0484415442,
0.0482279137,
0.1535749286,
-0.0105876252,
0.167019546,
-0.1051010936,
0.1631276757,
-0.5299461484,
-0.1543182135,
0.2289402187,
-0.0426615551,
-0.2358105928,
0.1510491371,
0.3084451556,
-0.1525322795,
0.0727143586,
0.0318391807,
0.2313716561,
-0.1411704719,
0.7170222402,
0.4281052947,
0.9727648497,
-0.0457558744,
-0.1917943656,
0.157251358,
-0.4654709697,
0.2642843425,
-0.1838361472,
0.4179236889,
-0.5206820965,
-0.0384493619,
-0.0848184079,
0.0788293183,
0.2849553823,
-0.1983232498,
-0.4596635699,
0.0784824863,
-0.4503542781,
-0.1716122031,
-0.1771043837,
0.0814976022,
0.0242339466,
-0.2860444784,
-0.1874103844,
0.0645479038,
0.0486764535,
0.2211969793,
0.1126721129,
-0.067312941,
-0.1354816556,
0.0555883646,
-0.2241008729,
0.032032989,
-0.1085604802,
0.3049996793,
-0.1301366389,
-0.2326380908,
-0.1463972628,
-0.3277445436,
0.4981687665,
-0.3548032939,
-0.2450336069,
-0.166322872,
0.0345947444,
-0.1157076508,
-0.1562321186,
-0.1925413907,
0.2612093389,
0.0618070439,
-0.1635158956,
0.0063225813,
0.0223619789,
0.0268576369,
0.0025244206,
-0.1410479248,
-0.3572548032,
-0.313978672,
-0.4244966507,
-0.0923325941,
0.2318002284,
0.0967235342,
0.0967093483,
0.3062936962,
-0.2717125714,
0.1072379723,
-0.2875133157,
-0.1340826601,
0.0060450854,
0.335435003,
0.0758269653,
0.020216763,
0.3766196072,
-0.306086421,
-0.1705391705,
-0.1801344156,
-0.2744092643,
0.1611407995,
-0.0199052468,
0.0499290824,
-0.0482470021,
-0.2829882503,
-0.0480852872,
0.1993075907,
-0.0310296156,
-0.1004889607,
-0.6292184591,
-0.0166367516,
-0.4327394962,
0.0409637839,
-0.2383728325,
0.2573253512,
-0.182534039,
0.3144521117,
-0.0718003437,
-0.3056448102,
-0.2587477863,
0.0695552006,
-0.4101382196,
0.2657018304,
-0.4145555496,
-0.1361317337,
0.0721303523,
-0.0040341355,
0.1892806292,
0.0023425985,
-0.044790484,
-0.1383941323,
-0.1356261671,
0.1331773549,
0.0876084864,
0.2126124203,
-0.1529172212,
-0.0640686452,
-0.1222837567,
-0.0091961678,
-0.0924093723,
0.2961058319,
-0.0596885979,
0.1375634372,
0.1278683543,
0.2047434896,
-0.1311390698,
-0.052943062,
0.236430794,
0.3273603022,
-0.1198692024,
0.1237517893,
-0.398075223,
0.1246480942,
-0.3058547676,
0.136051476,
0.3669175506,
0.103776589,
0.0543199368,
0.0582996383,
-0.1199311912,
0.111245811,
-0.1013381109,
0.2528006136,
-0.1698139012,
-0.0065536909,
0.1894347668,
0.2006903738,
-0.0020724684,
0.1720244139,
-0.0062770378,
-0.285020411,
-0.1459519416,
0.1465513855,
0.3709390759,
0.0773442388,
-0.1627675891,
-0.1122218221,
0.1676340252,
0.1602595747,
-0.1290874034,
-0.225076884,
0.2475019246,
0.221237123,
0.168211922,
0.004840767,
0.4208073318,
0.4026451409,
0.0048013441,
0.1288182586,
0.5096387863,
0.1529027522,
-0.0295124389,
-0.0068310723,
0.2534530759,
-0.1015482917,
-0.005008515,
-0.1515846252,
-0.0859878957,
0.0324030444,
-0.4543078244,
-0.3302356005,
0.0706348494,
0.2346271574,
-0.4260692596,
-0.0737658218,
0.1905767918,
0.2589165568,
-0.0605847239,
-0.308482796,
0.2100087404,
0.1119435281,
-0.0233669542,
0.1212686449,
-0.1889851093,
0.323212862,
-0.4511888027,
0.502024889,
0.3923590183,
-0.0609439611,
0.0253052544,
-0.0984659642,
-0.1560996473,
0.0754405707,
0.350864023,
0.5285478234,
0.4314031303,
-0.3667829335,
0.293722868,
0.2441829145,
-0.0594789833,
0.1149276793,
-0.2624688148,
-0.0268696547,
0.2984541059,
0.1137852371,
0.0836742669,
0.0034766197,
-0.3272369504,
-0.092096284,
0.3319784999,
0.0844721347,
0.181508258,
-0.382091403,
-0.089977257,
-0.2460792363,
0.0526117943,
-0.2649682164,
0.0140134208,
0.0373250432,
-0.2223137617,
0.0476384051,
0.2294640988,
0.0397456959,
-0.1147419512,
0.039275296,
0.0888323039,
0.0258671902,
-0.3434678912,
-0.0146148056,
0.005543679,
0.2804378271,
0.2627859116,
0.0692642927,
0.2042124569,
0.5844689608,
-0.0433290154,
-0.2042547166,
0.0936053768,
-0.0386311784,
0.398542732,
0.4201739132,
0.0109179467,
0.1280183196,
-0.508053124,
-0.269748956,
-0.0381363034,
0.0823268592,
0.254004389,
0.1065944508,
0.1364921182,
0.0152402036,
-0.0982608795,
-0.0470897332,
0.0751610994,
0.6740571856,
0.0045943856,
-0.0781617612,
-0.0839429721,
-0.2627613246,
0.1842109412,
0.0143032894,
-0.3596513569,
0.078445822,
-0.0378655046,
-0.0361996405,
-0.0588832833,
-0.2341378033,
-0.2605667114,
0.7164089084,
-0.0687382072,
-0.3527013361,
-0.1285430342,
0.2969986796,
-0.3471459746,
-0.2056798339,
0.3894694448,
0.0512244962,
0.0723173693,
0.4429088235,
-0.2615902424,
-0.3091919422,
0.3434889317,
0.0125177354,
0.0193108991,
-0.0837197006,
0.382817328,
0.2854710519,
0.5122026205,
-0.8527294397,
-0.0467747152,
0.1794885099,
0.0681954771,
-0.0979240164,
-0.2164348066,
0.0820887461,
-0.0615526028,
-0.1021637097,
-0.0230057165,
0.0768950433,
0.4291135371,
-0.058200866,
-0.1954108477
] |
https://github.com/huggingface/datasets/issues/1859 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU) | I fixed this issue. It should work fine now.
Feel free to try it out by installing `datasets` from source.
Otherwise you can wait for the next release of `datasets` (in a few days) | Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
| 34 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU)
Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
I fixed this issue. It should work fine now.
Feel free to try it out by installing `datasets` from source.
Otherwise you can wait for the next release of `datasets` (in a few days) | [
-0.2384790629,
-0.2405905426,
-0.0386405401,
0.1195870563,
0.3427582085,
0.1524923295,
0.3444894552,
0.5221390128,
0.4608198106,
0.2513042092,
-0.0791419297,
0.1412859261,
0.121612519,
-0.0570593365,
-0.139593631,
0.1362419724,
0.4380638003,
0.1934216321,
0.3012019396,
-0.0583266057,
-0.4008606374,
0.0912524611,
0.0359446108,
0.0129265916,
-0.1873641163,
-0.0472336039,
-0.0613627657,
-0.0423773006,
-0.1474944651,
-0.4240971208,
-0.05774194,
-0.154525578,
0.4029409885,
0.3433765769,
-0.0001137439,
0.0171600059,
0.2634913921,
-0.1986214817,
-0.1538048089,
-0.1396833807,
0.2965039909,
0.0760406181,
0.1519308984,
-0.3508681655,
-0.2409551293,
-0.2891053557,
-0.0485754386,
-0.2439467907,
-0.1022811979,
0.2978605926,
0.2040631026,
-0.0874998942,
0.4052406549,
-0.1325296909,
0.0500700325,
0.1100266129,
-0.2817079425,
-0.1140929684,
0.0078178961,
0.4504533708,
0.2506673038,
0.2985078692,
-0.0069531873,
-0.259696871,
-0.091654852,
-0.0183353946,
0.3743927479,
-0.3835611343,
0.2296167314,
-0.121256128,
0.1396842301,
-0.3347600698,
-0.3617610633,
0.2229542136,
0.1872789562,
-0.4804654419,
-0.0755152106,
0.0607193485,
-0.2271498144,
0.1341764927,
0.5808407068,
-0.1446944326,
0.0683105364,
-0.1654361933,
0.1633942723,
0.0585160851,
-0.1198726669,
0.1065059453,
0.0583302267,
-0.1843252033,
-0.0953071639,
0.0642275065,
0.1779869199,
0.1287583709,
-0.5028724074,
-0.0867341012,
0.0247370936,
-0.4174592793,
-0.0460306481,
0.0724531412,
-0.3467906415,
0.1453313977,
0.1766993999,
0.0592346191,
-0.2870435715,
0.0337666571,
-0.0274503455,
-0.1457376182,
0.006659145,
0.0383950397,
0.0344574079,
-0.3303355575,
-0.1660042405,
-0.0315667838,
-0.3789484203,
0.0761541501,
0.1963503808,
-0.3188080788,
-0.5184015632,
0.1290773451,
-0.0136289001,
0.2290710956,
0.1416978538,
0.4266785383,
0.3175202012,
0.035345342,
0.264555037,
0.2988888025,
-0.2037707418,
0.066984117,
-0.1860589385,
-0.1827612966,
0.2399476171,
0.164498657,
0.0138556361,
-0.4622968435,
0.0648644418,
0.0778760016,
-0.1108914837,
0.1539482772,
0.0939458087,
-0.2833530307,
0.0944205076,
0.4206938148,
0.0270253047,
0.2014475763,
0.310233891,
-0.071125105,
-0.1758989692,
0.3338375986,
-0.0073357988,
-0.2411532104,
-0.5131024122,
0.1490999907,
0.2137339264,
-0.0529841557,
0.2680778205,
0.0859571546,
0.1033838391,
-0.0308675952,
0.0688060448,
0.0914392844,
0.0358480662,
-0.3160327971,
0.4951671362,
0.0128610358,
-0.1686312854,
-0.0478119031,
0.1192982644,
-0.1202772111,
0.1759349555,
0.5109656453,
0.331297189,
0.1718975902,
-0.270388931,
0.300778985,
0.4135612845,
-0.330041796,
-0.1987008452,
0.0647557825,
-0.0192647576,
-0.3254493177,
0.1825144589,
0.0729462802,
0.2850209475,
-0.0070305355,
0.2279456556,
-0.0312300585,
0.0020126142,
-0.1611646116,
-0.378584832,
-0.3233619928,
0.4342169762,
0.0403838158,
-0.0712840185,
0.0773932636,
-0.1179750711,
-0.6022037268,
0.1653823555,
-0.1568249166,
-0.042019099,
-0.0337345079,
0.9180985093,
0.0664085746,
0.2482101917,
-0.0480837114,
0.0745989382,
0.0211199,
0.1085083783,
-0.1187089235,
-0.4229266644,
-0.1303249747,
0.0156873576,
-0.0758008137,
0.2126974165,
0.0030955207,
0.0767876506,
-0.1611405015,
-0.1121303663,
0.1742816567,
-0.3099658191,
-0.0134584568,
-0.1755194366,
0.0850169063,
-0.0379408598,
0.1815438271,
-0.3314420879,
-0.4518474042,
-0.1660559773,
0.2707014084,
0.3499481678,
-0.2712793052,
-0.0366941318,
0.3937571347,
0.093265757,
-0.0825445801,
0.1525272429,
0.0257431716,
0.0043488108,
-0.0181567445,
0.0768708065,
0.4259207845,
0.321875602,
-0.0369860306,
0.0271980017,
0.3717030585,
0.2110739499,
0.054855004,
-0.1873321533,
0.1108267456,
0.4963284135,
0.1157272682,
0.0822517872,
-0.1317608953,
-0.0508648269,
0.2979767323,
0.1687547117,
-0.2600298524,
-0.6055716276,
0.250757575,
0.0219843946,
-0.0076651871,
0.2330779135,
-0.0285571776,
-0.1289516687,
0.0609240085,
0.1440111697,
-0.704290688,
0.1790255755,
0.0768936947,
-0.1703106761,
-0.0954486728,
-0.1197596788,
-0.1216631532,
0.3036921918,
-0.2471763641,
-0.1520450413,
0.1871306449,
-0.1039697081,
-0.0682562292,
-0.0682849661,
-0.0093313381,
-0.1152831018,
0.0440149754,
-0.3456336856,
0.0185123757,
0.0293474048,
0.2506074905,
-0.2507231236,
-0.2679326236,
-0.0913451612,
-0.0071540214,
0.1249449849,
0.2417453974,
-0.0073972717,
0.3755146563,
-0.199212566,
0.2434767187,
0.2827227712,
0.0155905392,
-0.1905138791,
-0.1179161146,
0.0238808114,
-0.0235804301,
0.0435440131,
0.0319895633,
0.1436031163,
-0.0923393667,
0.1365602612,
-0.3200149834,
-0.04510317,
-0.0297242031,
-0.3586836159,
-0.0365154818,
-0.0903932005,
0.1915639341,
0.0737379193,
0.0106565766,
0.1889193654,
-0.0424713492,
-0.1706619561,
0.1441256553,
-0.2356060147,
0.0102607831,
-0.2453028709,
-0.1029475704,
-0.0885345116,
-0.3024117947,
0.0760999024,
0.0420685783,
0.1523383856,
-0.2104491889,
0.2786489129,
0.1540669501,
0.3944187164,
-0.2255446613,
-0.2214525342,
-0.2535200119,
0.3299394846,
-0.1207607538,
-0.4436139464,
0.3380655348,
0.2211084813,
0.0755024999,
0.0768548548,
-0.0015969053,
0.3228557408,
0.1641943753,
0.2143815011,
-0.2751165926,
0.2191841006,
0.337636143,
0.0541074686,
-0.1252538711,
-0.0836513788,
-0.0990651026,
0.0410785377,
-0.0058503575,
0.3779467344,
-0.2007444799,
0.4935575128,
-0.0686326101,
0.5453832746,
0.0227309689,
-0.4466870129,
0.2302496135,
-0.0553987026,
0.1944939792,
-0.1707658172,
-0.243855834,
0.5023387671,
0.2270677388,
-0.1800375581,
-0.0540658198,
-0.4180532694,
0.0261120033,
0.0925812274,
0.1768015027,
-0.0753672421,
-0.1083803922,
0.1398135424,
-0.0579094626,
0.2643488348,
-0.4492537379,
0.3536350429,
-0.1488877833,
0.1971689612,
0.234036386,
0.3468433619,
0.211570397,
-0.1303754449,
0.1627739668,
-0.1963834465,
-0.5711723566,
0.4759764671,
-0.1632493436,
0.3026890755,
-0.0298791006,
-0.1258693188,
0.2209756672,
0.0263588168,
0.6625295877,
0.1776608825,
-0.3577190936,
0.4814566374,
0.1116920114,
-0.5040153861,
-0.1274018586,
-0.1381823719,
0.3122321963,
0.0321445242,
0.412440747,
-0.1865828782,
-0.341437757,
-0.3256059289,
-0.0043906793,
-0.2450977266,
-0.2345227599,
-0.3622906506,
-0.2913391888,
-0.2380203903,
0.3276259601,
-0.0114639364,
0.3348122835,
-0.1287940443,
0.0263397992,
-0.4946047962,
-0.116637893,
-0.0544774793,
-0.0450535938,
0.3728666902,
-0.1497162282,
0.0129615963,
0.079767026,
0.3228400648,
0.4381985962,
-0.0120700765,
0.2011637986,
0.1447053701,
0.1515205353,
-0.0496983901,
0.0190382712,
0.1335831285,
-0.0207424127,
0.1335238516,
-0.2026962489,
0.1749547124,
-0.450062871,
-0.1119611114,
0.2253981829,
0.0185721256,
-0.3241938055,
0.0502264723,
0.3169850409,
-0.1579329222,
0.0697198361,
0.1620564759,
0.1921929866,
-0.1824703813,
0.7100051641,
0.3957364559,
0.9556046128,
-0.0562951826,
-0.0954113454,
0.1049594805,
-0.4326786399,
0.2646965981,
-0.1638126373,
0.3374712765,
-0.5618674755,
-0.0877073035,
-0.0307712182,
0.0842925161,
0.2589507401,
-0.1019880921,
-0.4649597406,
0.0948773623,
-0.4682741463,
-0.1472352594,
-0.2312412858,
0.0857870877,
0.0226012375,
-0.3416079283,
-0.1620430052,
0.1439853907,
0.1054653376,
0.1562861949,
0.0815875977,
-0.0745508671,
-0.1078783497,
0.0366035625,
-0.1978428662,
-0.001073271,
-0.1509224325,
0.3344519138,
-0.1510010511,
-0.3497728109,
-0.080325976,
-0.2889745831,
0.5595018864,
-0.2791687846,
-0.217313692,
-0.1567485929,
-0.003770547,
-0.1009950191,
-0.1617239267,
-0.2079298496,
0.2510701418,
0.0504245609,
-0.2247954905,
0.0285596699,
0.0856408477,
-0.0402890146,
0.0778369009,
-0.1377172172,
-0.3344850838,
-0.3021645546,
-0.461712122,
-0.0396451615,
0.2221158892,
-0.0165138617,
0.1410472244,
0.3233229518,
-0.2969527543,
0.2007444948,
-0.3336688876,
-0.0935505256,
-0.0683641434,
0.4073937535,
0.0004037907,
0.0350848064,
0.3433610201,
-0.1498123407,
-0.2154189646,
-0.1667423844,
-0.2526398897,
0.2233088464,
-0.0360391662,
0.0633848757,
-0.167549789,
-0.2924132347,
-0.0721022561,
0.1874260306,
-0.0347482376,
-0.0919861421,
-0.5462042093,
-0.0152213126,
-0.4797926843,
0.0193543397,
-0.1716103554,
0.3138586879,
-0.2071560919,
0.214840427,
-0.043570593,
-0.243059516,
-0.3138627708,
0.1492474377,
-0.3146575689,
0.2671284378,
-0.3454401493,
-0.1002849936,
0.1255984008,
0.0150477029,
0.1757669002,
0.0711518079,
-0.0695873499,
-0.1896983832,
-0.1923273653,
0.1164016351,
0.1130907312,
0.1149256155,
-0.1560168415,
-0.1207176894,
-0.2156418711,
-0.0848517939,
0.0179862753,
0.3290677071,
-0.0386980884,
0.1870075911,
0.1404235363,
0.1338133514,
0.0028005391,
-0.0495499931,
0.3240999281,
0.315885812,
-0.0934159309,
0.0927626565,
-0.3825969994,
0.0782682598,
-0.2551831305,
0.0975088775,
0.3156036735,
0.0965934843,
0.0957465023,
-0.0200309418,
-0.1255622357,
0.21827887,
0.0375533551,
0.3016256094,
-0.185734272,
0.0153936297,
0.2364436537,
0.2276838124,
-0.0599946454,
0.0629109442,
0.048512537,
-0.2579557896,
-0.0974105597,
0.1441824138,
0.3739403486,
0.0622025281,
-0.0634850264,
-0.0815522224,
0.2122593373,
0.0998334438,
0.0115571581,
-0.2535571754,
0.191652745,
0.1791356504,
0.2260338813,
0.0954770371,
0.4428194165,
0.5543107986,
-0.0909989923,
0.2256236523,
0.5728284717,
0.0962592065,
0.0338157974,
-0.0322622582,
0.2113484144,
-0.1478865743,
-0.0385957994,
-0.1483107209,
-0.0765078068,
0.0906022042,
-0.4630674422,
-0.4397492707,
0.0545201078,
0.248170346,
-0.4364410937,
-0.0726334974,
0.1797018647,
0.1748315096,
-0.0361980349,
-0.2937140167,
0.1201874912,
0.0101768123,
0.029214045,
0.0758816451,
-0.1556368619,
0.1881145984,
-0.2491923273,
0.4484765828,
0.3650005162,
0.0280820206,
0.0773644745,
-0.0085191093,
-0.1375134885,
0.0097258352,
0.2962389588,
0.533088088,
0.4502039552,
-0.3733898103,
0.3845909834,
0.2183308899,
-0.0557166673,
0.1354076117,
-0.216899693,
-0.0477176718,
0.1838822514,
0.1873998344,
0.1257197559,
-0.0600480437,
-0.351975441,
-0.02423089,
0.402941227,
0.0626424402,
0.0988752991,
-0.3894559741,
-0.0816496909,
-0.2280412465,
-0.0090599135,
-0.3260864913,
0.0363085158,
0.1406640112,
-0.2598182857,
0.0531366728,
0.1167773753,
0.0698551983,
-0.1794430017,
0.1243244261,
-0.0248716809,
0.0554885492,
-0.3125728071,
-0.1278847605,
-0.1540424526,
0.2752578855,
0.2603756785,
0.1083726659,
0.1266847551,
0.5640342236,
-0.0361204818,
-0.2216991782,
0.2161679417,
-0.1263414323,
0.4234744012,
0.3542955816,
-0.0601987243,
0.1872883141,
-0.5837618113,
-0.2151584625,
0.0077526495,
0.0600414872,
0.2648928463,
0.0928937271,
0.1579352766,
-0.0195696726,
-0.0677332804,
-0.0965491235,
0.0310988277,
0.7332782745,
-0.0058207586,
-0.0975355506,
-0.0607820675,
-0.2373835742,
0.1060013771,
-0.016163189,
-0.2784540653,
0.0840942189,
0.0688182637,
0.07030496,
-0.0650185198,
-0.254414171,
-0.3064649999,
0.6995818615,
-0.011248678,
-0.3947992921,
-0.1250038594,
0.3383896947,
-0.3028246164,
-0.1947580576,
0.3656844497,
0.1907239556,
0.0350593701,
0.4770620167,
-0.2215889394,
-0.3079065681,
0.427732408,
0.0150557496,
-0.1008342505,
-0.0568255559,
0.2920415401,
0.3486465812,
0.3909555078,
-0.8917965293,
-0.0410389304,
0.1061533913,
0.0174170807,
-0.1242295504,
-0.0979674757,
0.0490944572,
-0.0471265167,
-0.157671243,
0.0149086192,
0.1378750652,
0.3350590169,
-0.0517827906,
-0.1984498948
] |
https://github.com/huggingface/datasets/issues/1859 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU) | Thanks for such a quick fix and merge to master, pip installed git master, tested all OK | Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
| 17 | Error "in void don't know how to serialize this type of index" when saving index to disk when device=0 (GPU)
Error serializing faiss index. Error as follows:
`Error in void faiss::write_index(const faiss::Index*, faiss::IOWriter*) at /home/conda/feedstock_root/build_artifacts/faiss-split_1612472484670/work/faiss/impl/index_write.cpp:453: don't know how to serialize this type of index`
Note:
`torch.cuda.is_available()` reports:
```
Cuda is available
cuda:0
```
Adding index, device=0 for GPU.
`dataset.add_faiss_index(column='embeddings', index_name='idx_embeddings', device=0)`
However, during a quick debug, self.faiss_index has no attr "device" when checked in` search.py, method save`, so fails to transform gpu index to cpu index. If I add index without device, index is saved OK.
```
def save(self, file: str):
"""Serialize the FaissIndex on disk"""
import faiss # noqa: F811
if (
hasattr(self.faiss_index, "device")
and self.faiss_index.device is not None
and self.faiss_index.device > -1
):
index = faiss.index_gpu_to_cpu(self.faiss_index)
else:
index = self.faiss_index
faiss.write_index(index, file)
```
Thanks for such a quick fix and merge to master, pip installed git master, tested all OK | [
-0.1146762371,
-0.3390220404,
-0.0211907737,
0.1105611846,
0.3648001552,
0.1579401493,
0.2933632731,
0.5287510157,
0.5046114922,
0.3331741989,
0.0193730742,
0.0713497102,
0.1117652133,
-0.0630020797,
-0.1396041811,
0.1634270251,
0.4838933051,
0.2276034355,
0.3357007504,
-0.0381597579,
-0.4010097384,
0.1175064892,
0.0855752528,
0.0203528591,
-0.0746074691,
0.0274525266,
-0.027267538,
-0.0571346134,
-0.1761831194,
-0.4307223558,
-0.0974788144,
-0.2497539818,
0.3530039191,
0.298730284,
-0.0001167819,
0.0765006691,
0.2537882328,
-0.2185034752,
-0.1212786958,
-0.1918152124,
0.3125520945,
0.0238259733,
0.2112212479,
-0.3889564574,
-0.2219995856,
-0.229141295,
-0.0148388641,
-0.0824066848,
-0.0860947743,
0.2563897371,
0.1728009284,
-0.1268796921,
0.4269101024,
-0.0993107408,
0.0486690328,
0.1324173212,
-0.2699908912,
-0.0813281536,
0.0310124122,
0.4924758077,
0.3509413004,
0.3124846816,
-0.0470330343,
-0.3270620406,
-0.1348572373,
-0.0353981853,
0.6270087361,
-0.4113227129,
0.1714872122,
-0.1755582541,
0.0252775177,
-0.2577891648,
-0.3364471793,
0.2810266316,
0.1917248517,
-0.4463644326,
-0.1149211377,
0.0594533756,
-0.2262238413,
0.1179194301,
0.605045855,
-0.2198891193,
0.0826856494,
-0.1929790974,
0.2028477043,
0.0529565513,
-0.107448481,
0.0764328018,
0.0183133855,
-0.1368498653,
-0.2525523305,
0.0414269641,
0.1735973507,
0.1504005492,
-0.4110239148,
-0.1585876495,
0.0316705368,
-0.4751076996,
-0.1077525318,
0.1082185656,
-0.3547214568,
0.053616561,
0.2421880513,
0.0392323434,
-0.302978158,
0.0532990135,
-0.0838187486,
-0.1678363234,
-0.0528732985,
0.0116876177,
0.0138861537,
-0.369096458,
-0.1901766062,
0.0614839643,
-0.339402616,
0.1539419591,
0.2605769336,
-0.3318856955,
-0.5334683657,
0.1121125221,
-0.0934317186,
0.2243352383,
0.1077441052,
0.3150539994,
0.3123529851,
0.0913959742,
0.2175911963,
0.3172948956,
-0.2520931661,
0.1195912063,
-0.1601219028,
-0.1714918613,
0.2880522609,
0.0833574757,
-0.0394233279,
-0.4399603903,
0.0343989655,
0.1377116442,
-0.1891627312,
0.037174508,
0.0833465606,
-0.3589979112,
0.0460222699,
0.4789455235,
0.0850206316,
0.1684922576,
0.3972040415,
-0.1274075359,
-0.1334165186,
0.3661161363,
-0.0772540122,
-0.1925713271,
-0.5117433667,
0.141272366,
0.212868154,
-0.0281150546,
0.1784330308,
0.0310088024,
0.1252805293,
-0.0399938598,
0.0840366334,
0.1249943748,
0.1171111465,
-0.2871429026,
0.4271498621,
-0.014120847,
-0.0355851166,
-0.0640767962,
0.1413881034,
-0.1511021852,
0.2421637326,
0.5180214643,
0.4163946509,
0.2536925673,
-0.2635124624,
0.285296768,
0.4269221425,
-0.331413269,
-0.1441106349,
0.095061712,
-0.0796565413,
-0.3813626766,
0.2808742225,
0.0293667689,
0.2075393945,
-0.1095142886,
0.189942807,
-0.1288817823,
-0.0122312736,
-0.14324525,
-0.3757480681,
-0.3430811167,
0.3114046454,
-0.066041708,
-0.0519272462,
0.1272797585,
-0.1354819685,
-0.5923422575,
0.1196644828,
-0.1409868449,
-0.0515022203,
-0.0383367874,
0.9742364883,
0.0753186941,
0.2244440019,
-0.0890065804,
0.1671756208,
-0.0822715685,
0.019758597,
-0.1304288805,
-0.4603840709,
-0.0925273523,
0.0079037733,
-0.1039909869,
0.2397684753,
0.0991541147,
0.0758389235,
-0.1709727347,
-0.1432719678,
0.1668156385,
-0.2967721224,
-0.1551456302,
-0.129832387,
0.1190659702,
0.0699759349,
0.1621151567,
-0.3236828744,
-0.489902854,
-0.2084936351,
0.2800620198,
0.3638555706,
-0.2614918351,
-0.0434343815,
0.3414681256,
0.030494649,
-0.0277126208,
0.1776824743,
0.0977630913,
-0.0440906137,
-0.0421459526,
-0.0168213043,
0.5092594028,
0.4029054642,
-0.0575101189,
0.0993843898,
0.2719666362,
0.1814071387,
0.096909672,
-0.1989656836,
0.0933969915,
0.4091674685,
0.0790103823,
0.0928937942,
-0.1172558963,
0.0038423731,
0.350977093,
0.1438661814,
-0.2183191925,
-0.5096515417,
0.1285247654,
-0.0430182368,
-0.0327101909,
0.264970541,
0.00115706,
-0.1233770847,
0.0679902732,
0.0005133078,
-0.7569185495,
0.1782088578,
0.0353667699,
-0.2462669909,
-0.1474381536,
-0.0495022796,
-0.1168399006,
0.3600617349,
-0.1883469522,
-0.1926209033,
0.2656211257,
-0.0701293871,
-0.0270441696,
0.0180689022,
0.0745555833,
-0.1314168125,
0.0339826867,
-0.32213974,
0.0470150486,
0.1045816839,
0.2126230448,
-0.2320212126,
-0.2859652042,
-0.088802509,
-0.0023956746,
0.1029872745,
0.2418350428,
-0.0353086181,
0.2771550715,
-0.1721948087,
0.1879621744,
0.2446773648,
0.0301039349,
-0.2019223124,
-0.1163960919,
0.0249053575,
-0.0574816689,
0.0562346056,
-0.0197933894,
0.100903824,
-0.1188318133,
0.0695455074,
-0.2965685725,
-0.0026124045,
-0.0111069437,
-0.366713196,
-0.0308047067,
-0.1012417972,
0.2145469487,
0.0998618379,
0.0570709184,
0.2294717431,
0.0136416331,
-0.2189857215,
0.1472661644,
-0.2851096094,
0.0995736197,
-0.2638254166,
-0.0314238742,
-0.0933675915,
-0.2771866918,
0.0463003442,
-0.0066173524,
0.1611281931,
-0.1750765741,
0.2500218153,
0.109631516,
0.3826162517,
-0.19946675,
-0.1604937911,
-0.2403208911,
0.2455212921,
0.0141069815,
-0.4077499509,
0.3803669214,
0.2327859849,
0.0825454593,
-0.016832931,
0.0531430393,
0.2793094814,
0.1768025905,
0.1367385387,
-0.3660622537,
0.2124080062,
0.3130117655,
0.0957228765,
-0.114920944,
-0.053412579,
-0.0269128121,
0.0467293411,
0.0268323906,
0.3343471587,
-0.1983145773,
0.4334014058,
-0.117956534,
0.6137671471,
-0.029454492,
-0.33391729,
0.2581991553,
-0.0253976882,
0.1039872319,
-0.1415957659,
-0.2537580132,
0.4257482588,
0.3052313924,
-0.2542216778,
-0.0381647609,
-0.4093990624,
0.0523650274,
0.1372279823,
0.2153997719,
-0.0554610193,
-0.0866813958,
0.1114594117,
-0.0794080347,
0.2802916169,
-0.5015255213,
0.285998106,
-0.1337615699,
0.1897230148,
0.2340592742,
0.4296455383,
0.2158711851,
-0.1023618951,
0.1824276447,
-0.3492434621,
-0.5001864433,
0.4314859509,
-0.1153530777,
0.2361045182,
-0.057021223,
-0.099565886,
0.2641095519,
0.0280010514,
0.5748894811,
0.250570029,
-0.3606672287,
0.4638264775,
0.1107284874,
-0.4079249501,
-0.1878313422,
-0.1110823825,
0.3629106283,
0.0571910553,
0.4573306739,
-0.2006698549,
-0.3740855753,
-0.3333070278,
-0.0210818537,
-0.2284049541,
-0.2749688029,
-0.4151601195,
-0.3075050116,
-0.194904983,
0.3872946501,
0.0377009585,
0.3573036492,
-0.1309339851,
0.122271955,
-0.4832381308,
-0.0675466806,
-0.08159291,
-0.2251809388,
0.3386276066,
-0.2394340932,
-0.0769013762,
-0.0219503939,
0.285957396,
0.4177362621,
-0.04913041,
0.3201038837,
0.186142534,
0.1888229251,
0.0004614927,
0.0391884446,
0.1098860502,
0.0540728197,
0.1841255426,
-0.1428145766,
0.2230831683,
-0.4668781757,
-0.1193053871,
0.2227191329,
-0.0280993842,
-0.2630175948,
0.1655524522,
0.3173097074,
-0.1281888634,
0.1277456582,
0.1194430292,
0.1563390493,
-0.1931025088,
0.6817430258,
0.3997468352,
0.9798758626,
0.0252919178,
-0.0914611295,
0.0858019888,
-0.4167231321,
0.2120492458,
-0.1581187546,
0.3874967992,
-0.5338545442,
-0.0146613028,
-0.0704165697,
0.0637784004,
0.227236867,
-0.2236260325,
-0.4985606968,
0.1207726151,
-0.4409579635,
-0.1280458122,
-0.1978638321,
0.018963106,
-0.0129704615,
-0.2784304321,
-0.1050782651,
0.079292804,
0.0551650636,
0.2143960297,
0.0763616115,
-0.1182078645,
-0.1765974909,
0.0644144118,
-0.2304027379,
-0.0265106447,
-0.1730614007,
0.3446364701,
-0.0993859023,
-0.3877473772,
-0.1328043193,
-0.3671299219,
0.4722645283,
-0.2943335474,
-0.2061363906,
-0.1880918443,
0.0174348336,
-0.0513621196,
-0.1856401563,
-0.1617178321,
0.2138828784,
0.0408921763,
-0.1677395105,
0.0168895535,
0.0026909336,
-0.0189809278,
0.0521011055,
-0.1252660751,
-0.3506317735,
-0.3085308671,
-0.4875867665,
-0.037599992,
0.241218403,
0.0331750512,
0.1144584715,
0.3455032408,
-0.3167863488,
0.1765204221,
-0.3581044972,
-0.1160879135,
-0.0079591507,
0.3641250432,
0.0735052824,
0.016895555,
0.4057068527,
-0.2747247517,
-0.1524932683,
-0.1529944837,
-0.2313353866,
0.1319375783,
-0.0770362914,
0.1316166073,
-0.1380921602,
-0.3396428227,
-0.0646574497,
0.2435810268,
0.0010780331,
-0.0346701592,
-0.6063768268,
0.0421925932,
-0.4536465108,
0.0480269827,
-0.1868365109,
0.2728899419,
-0.2377520651,
0.2444856018,
-0.0362340137,
-0.3157742023,
-0.2671248019,
0.1573024392,
-0.3896793723,
0.2999818623,
-0.3915309608,
-0.1758680493,
0.0934170932,
0.0215158686,
0.1660154313,
0.0189266503,
-0.0389737748,
-0.1631835699,
-0.1572704315,
0.1302211881,
0.0838365406,
0.1632905602,
-0.1456516236,
0.0065036491,
-0.1009744853,
-0.0033867881,
-0.0315148309,
0.2484488785,
-0.020478338,
0.15043208,
0.0951674953,
0.2308659852,
-0.0633923784,
-0.0546428449,
0.2786683738,
0.2993780971,
-0.1352015883,
0.0992387608,
-0.3457975686,
0.1191848665,
-0.3002995253,
0.0524718091,
0.3520542681,
0.108747296,
0.0988349989,
-0.0381997153,
-0.1049233377,
0.2200043499,
-0.0865345746,
0.2196075618,
-0.1954568475,
0.0227088705,
0.1914826632,
0.2191763967,
0.0411764085,
0.1304050833,
0.0471723489,
-0.2500387728,
-0.1444495618,
0.1082359478,
0.37427634,
0.0224192739,
-0.1011443585,
-0.0451783836,
0.2204566598,
0.1478140503,
-0.0020098016,
-0.2099750638,
0.2430303991,
0.2296788841,
0.1241615936,
0.0390941799,
0.3866071999,
0.4460995793,
0.016255755,
0.1503821313,
0.5617634654,
0.0576332621,
0.0258301049,
0.0887889117,
0.118650198,
-0.149478361,
-0.0015851222,
-0.1095034257,
-0.0852056742,
0.0175592937,
-0.4877848327,
-0.4000404477,
0.1134632826,
0.2814671993,
-0.4551028013,
-0.0638952553,
0.1260293275,
0.2360827476,
-0.0055894628,
-0.3023850918,
0.1427739114,
0.1391484737,
-0.0389773883,
0.1191958636,
-0.1901587546,
0.3054991364,
-0.3720205426,
0.5323476791,
0.4078014195,
-0.0345838256,
0.0908470526,
0.0159076266,
-0.1674891263,
-0.0154256374,
0.4096111655,
0.5584619641,
0.4596505165,
-0.4080402851,
0.3030124903,
0.2462729067,
-0.0842655748,
0.1901927292,
-0.2424070239,
-0.0100642182,
0.2692171335,
0.047301013,
0.0936595947,
-0.0415211804,
-0.2905817032,
-0.0935948864,
0.2913462222,
0.0783217996,
0.185355112,
-0.3981408179,
-0.1571914405,
-0.1800521016,
0.0423385091,
-0.3538242579,
0.0492832586,
0.0853732899,
-0.1960120648,
0.035372179,
0.1670020521,
0.0526830703,
-0.1624988317,
0.0931416005,
0.0648179278,
0.0099919727,
-0.3806732893,
-0.0841451809,
0.0213940442,
0.234654516,
0.2559272051,
0.1386157721,
0.1663260311,
0.5535817742,
-0.088434279,
-0.2022337019,
0.1693255901,
-0.1345929205,
0.411342591,
0.3900166452,
-0.0215613171,
0.1709012389,
-0.4755472243,
-0.2536108196,
0.0076708794,
0.1050749123,
0.2578586638,
0.0936069191,
0.1616384685,
0.0079905428,
-0.0863536,
-0.0575325117,
0.1221468002,
0.6473210454,
0.0037712324,
-0.131472975,
-0.0434659161,
-0.3159878254,
0.2457978725,
0.024757728,
-0.355896771,
0.0329872742,
-0.029260572,
0.0657525361,
-0.0683965683,
-0.2140353769,
-0.3025998175,
0.6512045264,
-0.0724926814,
-0.3786832094,
-0.1211365312,
0.2815716863,
-0.2921560705,
-0.22173962,
0.3673094511,
0.1116153896,
0.0860566646,
0.5192093253,
-0.3003047705,
-0.2952008247,
0.358333379,
0.0133009255,
-0.0779230595,
-0.1047351211,
0.3780348599,
0.3131188452,
0.4897957444,
-0.8266088367,
-0.050542362,
0.1462623477,
0.0598301664,
-0.1069388464,
-0.0695539564,
0.1198460162,
-0.0517985038,
-0.0920975283,
-0.0096074305,
0.1192866564,
0.3867179751,
-0.0559398793,
-0.2090595216
] |
Subsets and Splits