Xenova HF staff commited on
Commit
8109bee
·
verified ·
1 Parent(s): 2eb430f

Upload folder using huggingface_hub

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<end-of-node>": 50266,
3
+ "[empty-title]": 50265
4
+ }
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "hf-internal-testing/tiny-random-MarkupLMModel",
4
+ "architectures": [
5
+ "MarkupLMModel"
6
+ ],
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "bos_token_id": 0,
9
+ "classifier_dropout": null,
10
+ "eos_token_id": 2,
11
+ "gradient_checkpointing": false,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 32,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 37,
17
+ "layer_norm_eps": 1e-12,
18
+ "max_depth": 10,
19
+ "max_position_embeddings": 512,
20
+ "max_xpath_subs_unit_embeddings": 30,
21
+ "max_xpath_tag_unit_embeddings": 20,
22
+ "model_type": "markuplm",
23
+ "num_attention_heads": 4,
24
+ "num_hidden_layers": 5,
25
+ "pad_token_id": 1,
26
+ "position_embedding_type": "absolute",
27
+ "subs_pad_id": 2,
28
+ "tag_pad_id": 2,
29
+ "transformers_version": "4.48.2",
30
+ "type_vocab_size": 16,
31
+ "use_cache": true,
32
+ "vocab_size": 50265,
33
+ "xpath_unit_hidden_size": 32
34
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dd6895f96237b7868ec844706aaad612ac1eda5a57396c16c7919f23efa3d09
3
+ size 6995415
preprocessor_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "feature_extractor_type": "MarkupLMFeatureExtractor",
3
+ "processor_class": "MarkupLMProcessor"
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,297 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "50265": {
45
+ "content": "[empty-title]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": false
51
+ },
52
+ "50266": {
53
+ "content": "<end-of-node>",
54
+ "lstrip": false,
55
+ "normalized": true,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": false
59
+ }
60
+ },
61
+ "bos_token": "<s>",
62
+ "clean_up_tokenization_spaces": false,
63
+ "cls_token": "<s>",
64
+ "eos_token": "</s>",
65
+ "errors": "replace",
66
+ "extra_special_tokens": {},
67
+ "mask_token": "<mask>",
68
+ "max_depth": 50,
69
+ "max_width": 1000,
70
+ "model_max_length": 512,
71
+ "only_label_first_subword": true,
72
+ "pad_token": "<pad>",
73
+ "pad_token_label": -100,
74
+ "pad_width": 1001,
75
+ "processor_class": "MarkupLMProcessor",
76
+ "sep_token": "</s>",
77
+ "tags_dict": {
78
+ "a": 0,
79
+ "abbr": 1,
80
+ "acronym": 2,
81
+ "address": 3,
82
+ "altGlyph": 4,
83
+ "altGlyphDef": 5,
84
+ "altGlyphItem": 6,
85
+ "animate": 7,
86
+ "animateColor": 8,
87
+ "animateMotion": 9,
88
+ "animateTransform": 10,
89
+ "applet": 11,
90
+ "area": 12,
91
+ "article": 13,
92
+ "aside": 14,
93
+ "audio": 15,
94
+ "b": 16,
95
+ "base": 17,
96
+ "basefont": 18,
97
+ "bdi": 19,
98
+ "bdo": 20,
99
+ "bgsound": 21,
100
+ "big": 22,
101
+ "blink": 23,
102
+ "blockquote": 24,
103
+ "body": 25,
104
+ "br": 26,
105
+ "button": 27,
106
+ "canvas": 28,
107
+ "caption": 29,
108
+ "center": 30,
109
+ "circle": 31,
110
+ "cite": 32,
111
+ "clipPath": 33,
112
+ "code": 34,
113
+ "col": 35,
114
+ "colgroup": 36,
115
+ "color-profile": 37,
116
+ "content": 38,
117
+ "cursor": 39,
118
+ "data": 40,
119
+ "datalist": 41,
120
+ "dd": 42,
121
+ "defs": 43,
122
+ "del": 44,
123
+ "desc": 45,
124
+ "details": 46,
125
+ "dfn": 47,
126
+ "dialog": 48,
127
+ "dir": 49,
128
+ "div": 50,
129
+ "dl": 51,
130
+ "dt": 52,
131
+ "ellipse": 53,
132
+ "em": 54,
133
+ "embed": 55,
134
+ "feBlend": 56,
135
+ "feColorMatrix": 57,
136
+ "feComponentTransfer": 58,
137
+ "feComposite": 59,
138
+ "feConvolveMatrix": 60,
139
+ "feDiffuseLighting": 61,
140
+ "feDisplacementMap": 62,
141
+ "feDistantLight": 63,
142
+ "feFlood": 64,
143
+ "feFuncA": 65,
144
+ "feFuncB": 66,
145
+ "feFuncG": 67,
146
+ "feFuncR": 68,
147
+ "feGaussianBlur": 69,
148
+ "feImage": 70,
149
+ "feMerge": 71,
150
+ "feMergeNode": 72,
151
+ "feMorphology": 73,
152
+ "feOffset": 74,
153
+ "fePointLight": 75,
154
+ "feSpecularLighting": 76,
155
+ "feSpotLight": 77,
156
+ "feTile": 78,
157
+ "feTurbulence": 79,
158
+ "fieldset": 80,
159
+ "figcaption": 81,
160
+ "figure": 82,
161
+ "filter": 83,
162
+ "font": 89,
163
+ "font-face": 88,
164
+ "font-face-format": 84,
165
+ "font-face-name": 85,
166
+ "font-face-src": 86,
167
+ "font-face-uri": 87,
168
+ "footer": 90,
169
+ "foreignObject": 91,
170
+ "form": 92,
171
+ "frame": 93,
172
+ "frameset": 94,
173
+ "g": 95,
174
+ "glyph": 96,
175
+ "glyphRef": 97,
176
+ "h1": 98,
177
+ "h2": 99,
178
+ "h3": 100,
179
+ "h4": 101,
180
+ "h5": 102,
181
+ "h6": 103,
182
+ "head": 104,
183
+ "header": 105,
184
+ "hgroup": 106,
185
+ "hkern": 107,
186
+ "hr": 108,
187
+ "html": 109,
188
+ "i": 110,
189
+ "iframe": 111,
190
+ "image": 112,
191
+ "img": 113,
192
+ "input": 114,
193
+ "ins": 115,
194
+ "kbd": 116,
195
+ "keygen": 117,
196
+ "label": 118,
197
+ "legend": 119,
198
+ "li": 120,
199
+ "line": 121,
200
+ "linearGradient": 122,
201
+ "link": 123,
202
+ "main": 124,
203
+ "map": 125,
204
+ "mark": 126,
205
+ "marker": 127,
206
+ "marquee": 128,
207
+ "mask": 129,
208
+ "math": 130,
209
+ "menu": 131,
210
+ "menuitem": 132,
211
+ "meta": 133,
212
+ "metadata": 134,
213
+ "meter": 135,
214
+ "missing-glyph": 136,
215
+ "mpath": 137,
216
+ "nav": 138,
217
+ "nobr": 139,
218
+ "noembed": 140,
219
+ "noframes": 141,
220
+ "noscript": 142,
221
+ "object": 143,
222
+ "ol": 144,
223
+ "optgroup": 145,
224
+ "option": 146,
225
+ "output": 147,
226
+ "p": 148,
227
+ "param": 149,
228
+ "path": 150,
229
+ "pattern": 151,
230
+ "picture": 152,
231
+ "plaintext": 153,
232
+ "polygon": 154,
233
+ "polyline": 155,
234
+ "portal": 156,
235
+ "pre": 157,
236
+ "progress": 158,
237
+ "q": 159,
238
+ "radialGradient": 160,
239
+ "rb": 161,
240
+ "rect": 162,
241
+ "rp": 163,
242
+ "rt": 164,
243
+ "rtc": 165,
244
+ "ruby": 166,
245
+ "s": 167,
246
+ "samp": 168,
247
+ "script": 169,
248
+ "section": 170,
249
+ "select": 171,
250
+ "set": 172,
251
+ "shadow": 173,
252
+ "slot": 174,
253
+ "small": 175,
254
+ "source": 176,
255
+ "spacer": 177,
256
+ "span": 178,
257
+ "stop": 179,
258
+ "strike": 180,
259
+ "strong": 181,
260
+ "style": 182,
261
+ "sub": 183,
262
+ "summary": 184,
263
+ "sup": 185,
264
+ "svg": 186,
265
+ "switch": 187,
266
+ "symbol": 188,
267
+ "table": 189,
268
+ "tbody": 190,
269
+ "td": 191,
270
+ "template": 192,
271
+ "text": 193,
272
+ "textPath": 194,
273
+ "textarea": 195,
274
+ "tfoot": 196,
275
+ "th": 197,
276
+ "thead": 198,
277
+ "time": 199,
278
+ "title": 200,
279
+ "tr": 201,
280
+ "track": 202,
281
+ "tref": 203,
282
+ "tspan": 204,
283
+ "tt": 205,
284
+ "u": 206,
285
+ "ul": 207,
286
+ "use": 208,
287
+ "var": 209,
288
+ "video": 210,
289
+ "view": 211,
290
+ "vkern": 212,
291
+ "wbr": 213,
292
+ "xmp": 214
293
+ },
294
+ "tokenizer_class": "MarkupLMTokenizer",
295
+ "trim_offsets": false,
296
+ "unk_token": "<unk>"
297
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff