KoichiYasuoka commited on
Commit
0a15e28
1 Parent(s): ac9f541

initial release

Browse files
README.md CHANGED
@@ -1,3 +1,39 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - "de"
4
+ tags:
5
+ - "german"
6
+ - "token-classification"
7
+ - "pos"
8
+ - "dependency-parsing"
9
+ datasets:
10
+ - "universal_dependencies"
11
+ license: "mit"
12
+ pipeline_tag: "token-classification"
13
+ ---
14
+
15
+ # bert-base-german-upos
16
+
17
+ ## Model Description
18
+
19
+ This is a RoBERTa model pre-trained with [UD_German-HDT](https://github.com/UniversalDependencies/UD_German-HDT) for POS-tagging and dependency-parsing, derived from [bert-base-german-cased](https://huggingface.co/bert-base-german-cased). Every word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech).
20
+
21
+ ## How to Use
22
+
23
+ ```py
24
+ from transformers import AutoTokenizer,AutoModelForTokenClassification
25
+ tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/bert-base-german-upos")
26
+ model=AutoModelForTokenClassification.from_pretrained("KoichiYasuoka/bert-base-german-upos")
27
+ ```
28
+
29
+ or
30
+
31
+ ```py
32
+ import esupar
33
+ nlp=esupar.load("KoichiYasuoka/bert-base-german-upos")
34
+ ```
35
+
36
+ ## See Also
37
+
38
+ [esupar](https://github.com/KoichiYasuoka/esupar): Tokenizer POS-tagger and Dependency-parser with BERT/RoBERTa models
39
+
config.json ADDED
@@ -0,0 +1,355 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForTokenClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 768,
10
+ "id2label": {
11
+ "0": "ADJ",
12
+ "1": "ADJ+NOUN",
13
+ "2": "ADP",
14
+ "3": "ADP+DET",
15
+ "4": "ADV",
16
+ "5": "AUX",
17
+ "6": "B-ADJ",
18
+ "7": "B-ADP",
19
+ "8": "B-ADP+DET",
20
+ "9": "B-ADV",
21
+ "10": "B-AUX",
22
+ "11": "B-CCONJ",
23
+ "12": "B-DET",
24
+ "13": "B-DET+NOUN",
25
+ "14": "B-INTJ",
26
+ "15": "B-NOUN",
27
+ "16": "B-NOUN+PROPN",
28
+ "17": "B-NUM",
29
+ "18": "B-PRON",
30
+ "19": "B-PROPN",
31
+ "20": "B-PROPN+PROPN",
32
+ "21": "B-PUNCT",
33
+ "22": "B-SCONJ",
34
+ "23": "B-VERB",
35
+ "24": "B-X",
36
+ "25": "CCONJ",
37
+ "26": "CCONJ+NOUN",
38
+ "27": "DET",
39
+ "28": "DET+NOUN",
40
+ "29": "I-ADJ",
41
+ "30": "I-ADP",
42
+ "31": "I-ADP+DET",
43
+ "32": "I-ADV",
44
+ "33": "I-AUX",
45
+ "34": "I-CCONJ",
46
+ "35": "I-DET",
47
+ "36": "I-DET+NOUN",
48
+ "37": "I-INTJ",
49
+ "38": "I-NOUN",
50
+ "39": "I-NOUN+PROPN",
51
+ "40": "I-NUM",
52
+ "41": "I-PRON",
53
+ "42": "I-PROPN",
54
+ "43": "I-PROPN+PROPN",
55
+ "44": "I-PUNCT",
56
+ "45": "I-SCONJ",
57
+ "46": "I-VERB",
58
+ "47": "I-X",
59
+ "48": "INTJ",
60
+ "49": "NOUN",
61
+ "50": "NOUN+PROPN",
62
+ "51": "NOUN+X",
63
+ "52": "NUM",
64
+ "53": "PART",
65
+ "54": "PRON",
66
+ "55": "PROPN",
67
+ "56": "PROPN+PROPN",
68
+ "57": "PROPN+X",
69
+ "58": "PUNCT",
70
+ "59": "SCONJ",
71
+ "60": "SYM",
72
+ "61": "VERB",
73
+ "62": "X"
74
+ },
75
+ "initializer_range": 0.02,
76
+ "intermediate_size": 3072,
77
+ "label2id": {
78
+ "ADJ": 0,
79
+ "ADJ+NOUN": 1,
80
+ "ADP": 2,
81
+ "ADP+DET": 3,
82
+ "ADV": 4,
83
+ "AUX": 5,
84
+ "B-ADJ": 6,
85
+ "B-ADP": 7,
86
+ "B-ADP+DET": 8,
87
+ "B-ADV": 9,
88
+ "B-AUX": 10,
89
+ "B-CCONJ": 11,
90
+ "B-DET": 12,
91
+ "B-DET+NOUN": 13,
92
+ "B-INTJ": 14,
93
+ "B-NOUN": 15,
94
+ "B-NOUN+PROPN": 16,
95
+ "B-NUM": 17,
96
+ "B-PRON": 18,
97
+ "B-PROPN": 19,
98
+ "B-PROPN+PROPN": 20,
99
+ "B-PUNCT": 21,
100
+ "B-SCONJ": 22,
101
+ "B-VERB": 23,
102
+ "B-X": 24,
103
+ "CCONJ": 25,
104
+ "CCONJ+NOUN": 26,
105
+ "DET": 27,
106
+ "DET+NOUN": 28,
107
+ "I-ADJ": 29,
108
+ "I-ADP": 30,
109
+ "I-ADP+DET": 31,
110
+ "I-ADV": 32,
111
+ "I-AUX": 33,
112
+ "I-CCONJ": 34,
113
+ "I-DET": 35,
114
+ "I-DET+NOUN": 36,
115
+ "I-INTJ": 37,
116
+ "I-NOUN": 38,
117
+ "I-NOUN+PROPN": 39,
118
+ "I-NUM": 40,
119
+ "I-PRON": 41,
120
+ "I-PROPN": 42,
121
+ "I-PROPN+PROPN": 43,
122
+ "I-PUNCT": 44,
123
+ "I-SCONJ": 45,
124
+ "I-VERB": 46,
125
+ "I-X": 47,
126
+ "INTJ": 48,
127
+ "NOUN": 49,
128
+ "NOUN+PROPN": 50,
129
+ "NOUN+X": 51,
130
+ "NUM": 52,
131
+ "PART": 53,
132
+ "PRON": 54,
133
+ "PROPN": 55,
134
+ "PROPN+PROPN": 56,
135
+ "PROPN+X": 57,
136
+ "PUNCT": 58,
137
+ "SCONJ": 59,
138
+ "SYM": 60,
139
+ "VERB": 61,
140
+ "X": 62
141
+ },
142
+ "layer_norm_eps": 1e-12,
143
+ "max_position_embeddings": 512,
144
+ "model_type": "bert",
145
+ "num_attention_heads": 12,
146
+ "num_hidden_layers": 12,
147
+ "pad_token_id": 0,
148
+ "position_embedding_type": "absolute",
149
+ "task_specific_params": {
150
+ "upos_multiword": {
151
+ "ADJ+NOUN": {
152
+ "selbst\u00e4ndigeDenken\u00b8": [
153
+ "selbst\u00e4ndige",
154
+ "Denken\u00b8"
155
+ ],
156
+ "\u00f6ffentlichenInternetcaf\u00e9s": [
157
+ "\u00f6ffentlichen",
158
+ "Internetcaf\u00e9s"
159
+ ]
160
+ },
161
+ "ADP+DET": {
162
+ "A.": [
163
+ "AN",
164
+ "DEM"
165
+ ],
166
+ "Am": [
167
+ "An",
168
+ "dem"
169
+ ],
170
+ "Ans": [
171
+ "An",
172
+ "das"
173
+ ],
174
+ "Beim": [
175
+ "Bei",
176
+ "dem"
177
+ ],
178
+ "F\u00fcrs": [
179
+ "F\u00fcr",
180
+ "das"
181
+ ],
182
+ "Im": [
183
+ "In",
184
+ "dem"
185
+ ],
186
+ "Ins": [
187
+ "In",
188
+ "das"
189
+ ],
190
+ "Ums": [
191
+ "Um",
192
+ "das"
193
+ ],
194
+ "Unterm": [
195
+ "Unter",
196
+ "dem"
197
+ ],
198
+ "Vom": [
199
+ "Von",
200
+ "dem"
201
+ ],
202
+ "Zum": [
203
+ "Zu",
204
+ "dem"
205
+ ],
206
+ "Zur": [
207
+ "Zu",
208
+ "der"
209
+ ],
210
+ "am": [
211
+ "an",
212
+ "dem"
213
+ ],
214
+ "ans": [
215
+ "an",
216
+ "das"
217
+ ],
218
+ "aufs": [
219
+ "auf",
220
+ "das"
221
+ ],
222
+ "beim": [
223
+ "bei",
224
+ "dem"
225
+ ],
226
+ "durchs": [
227
+ "durch",
228
+ "das"
229
+ ],
230
+ "f\u00fcrs": [
231
+ "f\u00fcr",
232
+ "das"
233
+ ],
234
+ "hinterm": [
235
+ "hinter",
236
+ "dem"
237
+ ],
238
+ "hinters": [
239
+ "hinter",
240
+ "das"
241
+ ],
242
+ "im": [
243
+ "in",
244
+ "dem"
245
+ ],
246
+ "ins": [
247
+ "in",
248
+ "das"
249
+ ],
250
+ "ums": [
251
+ "um",
252
+ "das"
253
+ ],
254
+ "unterm": [
255
+ "unter",
256
+ "dem"
257
+ ],
258
+ "unters": [
259
+ "unter",
260
+ "das"
261
+ ],
262
+ "vom": [
263
+ "von",
264
+ "dem"
265
+ ],
266
+ "vorm": [
267
+ "vor",
268
+ "dem"
269
+ ],
270
+ "vors": [
271
+ "vor",
272
+ "das"
273
+ ],
274
+ "z.": [
275
+ "zu",
276
+ "dem"
277
+ ],
278
+ "zum": [
279
+ "zu",
280
+ "dem"
281
+ ],
282
+ "zur": [
283
+ "zu",
284
+ "der"
285
+ ],
286
+ "\u00dcbers": [
287
+ "\u00dcber",
288
+ "das"
289
+ ],
290
+ "\u00fcbers": [
291
+ "\u00fcber",
292
+ "das"
293
+ ]
294
+ },
295
+ "CCONJ+NOUN": {
296
+ "sowieInternetcaf\u00e9s": [
297
+ "sowie",
298
+ "Internetcaf\u00e9s"
299
+ ]
300
+ },
301
+ "DET+NOUN": {
302
+ "dieCr\u00e8me": [
303
+ "die",
304
+ "Cr\u00e8me"
305
+ ],
306
+ "dieFort\u00e9-Familie": [
307
+ "die",
308
+ "Fort\u00e9-Familie"
309
+ ]
310
+ },
311
+ "NOUN+PROPN": {
312
+ "Highend-ModellCli\u00e9": [
313
+ "Highend-Modell",
314
+ "Cli\u00e9"
315
+ ],
316
+ "McKinsey-BeraterRen\u00e9": [
317
+ "McKinsey-Berater",
318
+ "Ren\u00e9"
319
+ ],
320
+ "NachfolgemodellCli\u00e9": [
321
+ "Nachfolgemodell",
322
+ "Cli\u00e9"
323
+ ]
324
+ },
325
+ "PROPN+PROPN": {
326
+ "AlexanderArtop\u00e9": [
327
+ "Alexander",
328
+ "Artop\u00e9"
329
+ ],
330
+ "ClausS\u00f8rensen": [
331
+ "Claus",
332
+ "S\u00f8rensen"
333
+ ],
334
+ "Jean-LouisGass\u00e9e": [
335
+ "Jean-Louis",
336
+ "Gass\u00e9e"
337
+ ],
338
+ "JoelleRichardi\u00e9re": [
339
+ "Joelle",
340
+ "Richardi\u00e9re"
341
+ ],
342
+ "LeD\u00e9aut": [
343
+ "Le",
344
+ "D\u00e9aut"
345
+ ]
346
+ }
347
+ }
348
+ },
349
+ "tokenizer_class": "BertTokenizerFast",
350
+ "torch_dtype": "float32",
351
+ "transformers_version": "4.17.0",
352
+ "type_vocab_size": 2,
353
+ "use_cache": true,
354
+ "vocab_size": 30000
355
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22c187760763df4a0ba59d50560fd2464f373b3e771ffd63b4bf2a81430eae04
3
+ size 434244299
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
supar.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6767137a559c0cda3f40dacaa05190e965713588b2e3d5aea8a355f830e9cad
3
+ size 513838054
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "tokenizer_class": "BertTokenizerFast"}
vocab.txt ADDED
The diff for this file is too large to render. See raw diff