pkduongsu commited on
Commit
fbb67b8
1 Parent(s): 4751282

Model save

Browse files
Files changed (2) hide show
  1. README.md +4 -184
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -3,8 +3,6 @@ license: cc-by-4.0
3
  base_model: deepset/roberta-base-squad2
4
  tags:
5
  - generated_from_trainer
6
- datasets:
7
- - covid_qa_deepset
8
  model-index:
9
  - name: roberta-squad2-finetuned-covidQA
10
  results: []
@@ -15,9 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # roberta-squad2-finetuned-covidQA
17
 
18
- This model is a fine-tuned version of [deepset/roberta-base-squad2](https://huggingface.co/deepset/roberta-base-squad2) on the covid_qa_deepset dataset.
19
- It achieves the following results on the evaluation set:
20
- - Loss: 0.2338
21
 
22
  ## Model description
23
 
@@ -37,190 +33,14 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 2e-05
40
- - train_batch_size: 16
41
- - eval_batch_size: 16
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
- - lr_scheduler_warmup_steps: 250
46
  - num_epochs: 2
47
 
48
- ### Training results
49
-
50
- | Training Loss | Epoch | Step | Validation Loss |
51
- |:-------------:|:-----:|:----:|:---------------:|
52
- | 0.416 | 0.01 | 20 | 0.3362 |
53
- | 0.4503 | 0.02 | 40 | 0.3012 |
54
- | 0.1529 | 0.04 | 60 | 0.3735 |
55
- | 0.408 | 0.05 | 80 | 0.2852 |
56
- | 0.4379 | 0.06 | 100 | 0.2575 |
57
- | 0.2443 | 0.07 | 120 | 0.2880 |
58
- | 0.3613 | 0.08 | 140 | 0.3336 |
59
- | 0.3116 | 0.09 | 160 | 0.2532 |
60
- | 0.3361 | 0.11 | 180 | 0.2585 |
61
- | 0.3336 | 0.12 | 200 | 0.2854 |
62
- | 0.2891 | 0.13 | 220 | 0.2633 |
63
- | 0.3262 | 0.14 | 240 | 0.2311 |
64
- | 0.2053 | 0.15 | 260 | 0.4100 |
65
- | 0.2583 | 0.16 | 280 | 0.2908 |
66
- | 0.3646 | 0.18 | 300 | 0.2456 |
67
- | 0.2798 | 0.19 | 320 | 0.2468 |
68
- | 0.3079 | 0.2 | 340 | 0.2746 |
69
- | 0.4007 | 0.21 | 360 | 0.2521 |
70
- | 0.3548 | 0.22 | 380 | 0.2783 |
71
- | 0.3401 | 0.23 | 400 | 0.2667 |
72
- | 0.3405 | 0.25 | 420 | 0.2408 |
73
- | 0.3658 | 0.26 | 440 | 0.2376 |
74
- | 0.2781 | 0.27 | 460 | 0.2415 |
75
- | 0.1905 | 0.28 | 480 | 0.2597 |
76
- | 0.2666 | 0.29 | 500 | 0.2667 |
77
- | 0.2164 | 0.3 | 520 | 0.2394 |
78
- | 0.2155 | 0.32 | 540 | 0.2780 |
79
- | 0.2676 | 0.33 | 560 | 0.2831 |
80
- | 0.3552 | 0.34 | 580 | 0.2416 |
81
- | 0.2934 | 0.35 | 600 | 0.2362 |
82
- | 0.2138 | 0.36 | 620 | 0.2450 |
83
- | 0.1169 | 0.38 | 640 | 0.2686 |
84
- | 0.1815 | 0.39 | 660 | 0.2512 |
85
- | 0.3577 | 0.4 | 680 | 0.2632 |
86
- | 0.3298 | 0.41 | 700 | 0.2721 |
87
- | 0.2624 | 0.42 | 720 | 0.2667 |
88
- | 0.4011 | 0.43 | 740 | 0.2414 |
89
- | 0.4041 | 0.45 | 760 | 0.2264 |
90
- | 0.3107 | 0.46 | 780 | 0.2342 |
91
- | 0.3036 | 0.47 | 800 | 0.2202 |
92
- | 0.2474 | 0.48 | 820 | 0.2449 |
93
- | 0.2889 | 0.49 | 840 | 0.2601 |
94
- | 0.1131 | 0.5 | 860 | 0.3004 |
95
- | 0.2039 | 0.52 | 880 | 0.2730 |
96
- | 0.2916 | 0.53 | 900 | 0.2598 |
97
- | 0.2649 | 0.54 | 920 | 0.2425 |
98
- | 0.16 | 0.55 | 940 | 0.2319 |
99
- | 0.1761 | 0.56 | 960 | 0.2365 |
100
- | 0.4593 | 0.57 | 980 | 0.2300 |
101
- | 0.3461 | 0.59 | 1000 | 0.2360 |
102
- | 0.2248 | 0.6 | 1020 | 0.2354 |
103
- | 0.3183 | 0.61 | 1040 | 0.2266 |
104
- | 0.179 | 0.62 | 1060 | 0.2332 |
105
- | 0.1995 | 0.63 | 1080 | 0.2321 |
106
- | 0.2084 | 0.65 | 1100 | 0.2222 |
107
- | 0.2419 | 0.66 | 1120 | 0.2307 |
108
- | 0.3359 | 0.67 | 1140 | 0.2212 |
109
- | 0.2263 | 0.68 | 1160 | 0.2300 |
110
- | 0.2362 | 0.69 | 1180 | 0.2326 |
111
- | 0.3108 | 0.7 | 1200 | 0.2410 |
112
- | 0.3218 | 0.72 | 1220 | 0.2906 |
113
- | 0.2954 | 0.73 | 1240 | 0.2518 |
114
- | 0.2026 | 0.74 | 1260 | 0.2348 |
115
- | 0.2149 | 0.75 | 1280 | 0.2338 |
116
- | 0.1686 | 0.76 | 1300 | 0.2362 |
117
- | 0.1928 | 0.77 | 1320 | 0.2308 |
118
- | 0.3103 | 0.79 | 1340 | 0.2183 |
119
- | 0.1686 | 0.8 | 1360 | 0.2521 |
120
- | 0.1691 | 0.81 | 1380 | 0.2509 |
121
- | 0.3721 | 0.82 | 1400 | 0.2239 |
122
- | 0.3334 | 0.83 | 1420 | 0.2304 |
123
- | 0.3117 | 0.84 | 1440 | 0.2185 |
124
- | 0.267 | 0.86 | 1460 | 0.2142 |
125
- | 0.2403 | 0.87 | 1480 | 0.2215 |
126
- | 0.3576 | 0.88 | 1500 | 0.2158 |
127
- | 0.2544 | 0.89 | 1520 | 0.2284 |
128
- | 0.2935 | 0.9 | 1540 | 0.2241 |
129
- | 0.2224 | 0.91 | 1560 | 0.2208 |
130
- | 0.2615 | 0.93 | 1580 | 0.2194 |
131
- | 0.1746 | 0.94 | 1600 | 0.2372 |
132
- | 0.2313 | 0.95 | 1620 | 0.2381 |
133
- | 0.1911 | 0.96 | 1640 | 0.2472 |
134
- | 0.2399 | 0.97 | 1660 | 0.2483 |
135
- | 0.2611 | 0.99 | 1680 | 0.2420 |
136
- | 0.313 | 1.0 | 1700 | 0.2234 |
137
- | 0.1456 | 1.01 | 1720 | 0.2327 |
138
- | 0.172 | 1.02 | 1740 | 0.2298 |
139
- | 0.2197 | 1.03 | 1760 | 0.2376 |
140
- | 0.1991 | 1.04 | 1780 | 0.2483 |
141
- | 0.1186 | 1.06 | 1800 | 0.2455 |
142
- | 0.1417 | 1.07 | 1820 | 0.2493 |
143
- | 0.2101 | 1.08 | 1840 | 0.2423 |
144
- | 0.1564 | 1.09 | 1860 | 0.2467 |
145
- | 0.1816 | 1.1 | 1880 | 0.2505 |
146
- | 0.2034 | 1.11 | 1900 | 0.3005 |
147
- | 0.2178 | 1.13 | 1920 | 0.2384 |
148
- | 0.2895 | 1.14 | 1940 | 0.2602 |
149
- | 0.1629 | 1.15 | 1960 | 0.2422 |
150
- | 0.2443 | 1.16 | 1980 | 0.2294 |
151
- | 0.1776 | 1.17 | 2000 | 0.2403 |
152
- | 0.181 | 1.18 | 2020 | 0.2302 |
153
- | 0.1757 | 1.2 | 2040 | 0.2273 |
154
- | 0.1523 | 1.21 | 2060 | 0.2272 |
155
- | 0.0763 | 1.22 | 2080 | 0.2422 |
156
- | 0.1534 | 1.23 | 2100 | 0.2445 |
157
- | 0.1994 | 1.24 | 2120 | 0.2487 |
158
- | 0.1826 | 1.26 | 2140 | 0.2569 |
159
- | 0.2475 | 1.27 | 2160 | 0.2389 |
160
- | 0.1977 | 1.28 | 2180 | 0.2290 |
161
- | 0.2891 | 1.29 | 2200 | 0.2395 |
162
- | 0.2049 | 1.3 | 2220 | 0.2292 |
163
- | 0.2526 | 1.31 | 2240 | 0.2410 |
164
- | 0.2927 | 1.33 | 2260 | 0.2270 |
165
- | 0.1325 | 1.34 | 2280 | 0.2566 |
166
- | 0.1331 | 1.35 | 2300 | 0.2400 |
167
- | 0.1198 | 1.36 | 2320 | 0.2416 |
168
- | 0.1766 | 1.37 | 2340 | 0.2407 |
169
- | 0.1698 | 1.38 | 2360 | 0.2398 |
170
- | 0.1545 | 1.4 | 2380 | 0.2437 |
171
- | 0.2406 | 1.41 | 2400 | 0.2587 |
172
- | 0.2583 | 1.42 | 2420 | 0.2292 |
173
- | 0.1562 | 1.43 | 2440 | 0.2374 |
174
- | 0.2528 | 1.44 | 2460 | 0.2326 |
175
- | 0.1665 | 1.45 | 2480 | 0.2366 |
176
- | 0.1893 | 1.47 | 2500 | 0.2323 |
177
- | 0.109 | 1.48 | 2520 | 0.2492 |
178
- | 0.1385 | 1.49 | 2540 | 0.2418 |
179
- | 0.1267 | 1.5 | 2560 | 0.2437 |
180
- | 0.2004 | 1.51 | 2580 | 0.2393 |
181
- | 0.1754 | 1.52 | 2600 | 0.2408 |
182
- | 0.2147 | 1.54 | 2620 | 0.2355 |
183
- | 0.1409 | 1.55 | 2640 | 0.2460 |
184
- | 0.1409 | 1.56 | 2660 | 0.2406 |
185
- | 0.1456 | 1.57 | 2680 | 0.2443 |
186
- | 0.1926 | 1.58 | 2700 | 0.2385 |
187
- | 0.1772 | 1.6 | 2720 | 0.2342 |
188
- | 0.2147 | 1.61 | 2740 | 0.2346 |
189
- | 0.2292 | 1.62 | 2760 | 0.2319 |
190
- | 0.2335 | 1.63 | 2780 | 0.2303 |
191
- | 0.1409 | 1.64 | 2800 | 0.2347 |
192
- | 0.1004 | 1.65 | 2820 | 0.2502 |
193
- | 0.281 | 1.67 | 2840 | 0.2296 |
194
- | 0.1071 | 1.68 | 2860 | 0.2360 |
195
- | 0.1152 | 1.69 | 2880 | 0.2402 |
196
- | 0.219 | 1.7 | 2900 | 0.2350 |
197
- | 0.1384 | 1.71 | 2920 | 0.2367 |
198
- | 0.1792 | 1.72 | 2940 | 0.2351 |
199
- | 0.1795 | 1.74 | 2960 | 0.2338 |
200
- | 0.1554 | 1.75 | 2980 | 0.2373 |
201
- | 0.1764 | 1.76 | 3000 | 0.2352 |
202
- | 0.2362 | 1.77 | 3020 | 0.2337 |
203
- | 0.1912 | 1.78 | 3040 | 0.2304 |
204
- | 0.1202 | 1.79 | 3060 | 0.2313 |
205
- | 0.146 | 1.81 | 3080 | 0.2327 |
206
- | 0.2677 | 1.82 | 3100 | 0.2305 |
207
- | 0.1919 | 1.83 | 3120 | 0.2331 |
208
- | 0.1535 | 1.84 | 3140 | 0.2317 |
209
- | 0.1032 | 1.85 | 3160 | 0.2341 |
210
- | 0.0792 | 1.87 | 3180 | 0.2341 |
211
- | 0.1419 | 1.88 | 3200 | 0.2355 |
212
- | 0.1179 | 1.89 | 3220 | 0.2369 |
213
- | 0.1948 | 1.9 | 3240 | 0.2363 |
214
- | 0.1651 | 1.91 | 3260 | 0.2362 |
215
- | 0.2293 | 1.92 | 3280 | 0.2351 |
216
- | 0.1542 | 1.94 | 3300 | 0.2358 |
217
- | 0.2852 | 1.95 | 3320 | 0.2347 |
218
- | 0.0927 | 1.96 | 3340 | 0.2350 |
219
- | 0.1746 | 1.97 | 3360 | 0.2337 |
220
- | 0.0902 | 1.98 | 3380 | 0.2341 |
221
- | 0.2275 | 1.99 | 3400 | 0.2338 |
222
-
223
-
224
  ### Framework versions
225
 
226
  - Transformers 4.34.1
 
3
  base_model: deepset/roberta-base-squad2
4
  tags:
5
  - generated_from_trainer
 
 
6
  model-index:
7
  - name: roberta-squad2-finetuned-covidQA
8
  results: []
 
13
 
14
  # roberta-squad2-finetuned-covidQA
15
 
16
+ This model is a fine-tuned version of [deepset/roberta-base-squad2](https://huggingface.co/deepset/roberta-base-squad2) on an unknown dataset.
 
 
17
 
18
  ## Model description
19
 
 
33
 
34
  The following hyperparameters were used during training:
35
  - learning_rate: 2e-05
36
+ - train_batch_size: 8
37
+ - eval_batch_size: 8
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
+ - lr_scheduler_warmup_steps: 500
42
  - num_epochs: 2
43
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44
  ### Framework versions
45
 
46
  - Transformers 4.34.1
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e567ab61e79ff464ec1f1f84018b73a5052429fa800f7a8d9b29efe160a5e5b7
3
  size 496295078
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00345b95dc1636e2e5ff25855f0d5ca5e6ba19ad0db9c6e63e30f9efbc094c6f
3
  size 496295078