Hugging Face's logo

asahi417
/
tner-roberta-large-tweet-2020

Token Classification
Transformers
PyTorch
roberta
Inference Endpoints
Model card Files Community
1
asahi417 commited on
Commit
df1c50a
•
1 Parent(s): b3cb231

model update

Browse files
Files changed (8) hide show
  1. config.json +99 -99
  2. eval/metric.json +1 -0
  3. eval/prediction.2020.dev.json +0 -0
  4. eval/prediction.2020.test.json +0 -0
  5. eval/prediction.2021.test.json +0 -0
  6. pytorch_model.bin +2 -2
  7. tokenizer.json +0 -0
  8. tokenizer_config.json +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "./tner_output/search/roberta_large/twitter_ner/model_tpkhxk/epoch_10",
3
  "architectures": [
4
  "RobertaForTokenClassification"
5
  ],
@@ -12,38 +12,38 @@
12
  1.0,
13
  1.0,
14
  1.0,
15
- 0.0,
16
  1.0,
17
- 0.0,
 
 
18
  1.0,
19
  0.0,
20
  0.0,
21
- 1.0,
22
  0.0,
23
- 1.0,
24
  0.0,
25
- 1.0,
26
  0.0,
27
  0.0,
 
 
28
  1.0
29
  ],
30
  [
31
  1.0,
32
  1.0,
33
  1.0,
34
- 0.0,
35
  1.0,
36
- 0.0,
37
  1.0,
38
- 0.0,
39
  1.0,
40
  1.0,
41
  0.0,
42
  1.0,
43
  0.0,
44
- 1.0,
45
  0.0,
46
  0.0,
 
 
 
 
47
  1.0
48
  ],
49
  [
@@ -52,17 +52,17 @@
52
  1.0,
53
  1.0,
54
  1.0,
55
- 0.0,
56
  1.0,
57
  0.0,
58
  0.0,
59
  1.0,
60
  0.0,
61
- 1.0,
62
  0.0,
63
- 1.0,
64
  0.0,
65
  0.0,
 
 
66
  1.0
67
  ],
68
  [
@@ -71,54 +71,54 @@
71
  1.0,
72
  1.0,
73
  1.0,
74
- 0.0,
75
  1.0,
76
  0.0,
77
  0.0,
78
- 1.0,
79
  0.0,
80
  1.0,
81
  0.0,
82
- 1.0,
83
  0.0,
84
  0.0,
 
 
85
  1.0
86
  ],
87
  [
88
  1.0,
89
  1.0,
90
  1.0,
91
- 0.0,
92
  1.0,
93
  1.0,
94
  1.0,
95
  0.0,
96
  0.0,
97
- 1.0,
98
  0.0,
99
- 1.0,
100
  0.0,
101
  1.0,
102
  0.0,
103
  0.0,
 
 
104
  1.0
105
  ],
106
  [
107
  1.0,
108
  1.0,
109
  1.0,
110
- 0.0,
111
  1.0,
112
  1.0,
113
  1.0,
114
  0.0,
115
  0.0,
116
- 1.0,
117
  0.0,
118
- 1.0,
119
  0.0,
120
  1.0,
121
  0.0,
 
122
  0.0,
123
  1.0
124
  ],
@@ -126,132 +126,132 @@
126
  1.0,
127
  1.0,
128
  1.0,
129
- 0.0,
130
  1.0,
131
- 0.0,
132
  1.0,
133
  1.0,
134
- 0.0,
135
  1.0,
136
  0.0,
137
- 1.0,
138
  0.0,
139
- 1.0,
140
  0.0,
141
  0.0,
 
 
 
 
 
142
  1.0
143
  ],
144
  [
145
  1.0,
146
  1.0,
147
  1.0,
148
- 0.0,
149
  1.0,
150
- 0.0,
151
  1.0,
152
  1.0,
153
- 0.0,
154
  1.0,
155
- 0.0,
156
  1.0,
157
  0.0,
158
- 1.0,
159
  0.0,
160
  0.0,
 
 
 
 
 
161
  1.0
162
  ],
163
  [
164
  1.0,
165
  1.0,
166
  1.0,
167
- 0.0,
168
  1.0,
169
- 0.0,
170
  1.0,
171
- 0.0,
172
  1.0,
173
  1.0,
174
  0.0,
175
  1.0,
176
  0.0,
177
- 1.0,
178
  0.0,
179
  0.0,
 
 
 
 
180
  1.0
181
  ],
182
  [
183
  1.0,
184
  1.0,
185
  1.0,
186
- 0.0,
187
  1.0,
188
- 0.0,
189
- 1.0,
190
- 0.0,
191
- 0.0,
192
  1.0,
193
  1.0,
194
  1.0,
195
  0.0,
 
196
  1.0,
197
  0.0,
198
  0.0,
 
 
 
 
199
  1.0
200
  ],
201
  [
202
  1.0,
203
  1.0,
204
  1.0,
205
- 0.0,
206
  1.0,
207
- 0.0,
208
- 1.0,
209
- 0.0,
210
- 0.0,
211
  1.0,
212
  1.0,
213
  1.0,
214
  0.0,
 
 
215
  1.0,
216
  0.0,
217
  0.0,
 
 
 
218
  1.0
219
  ],
220
  [
221
  1.0,
222
  1.0,
223
  1.0,
224
- 0.0,
225
  1.0,
226
- 0.0,
 
227
  1.0,
228
  0.0,
229
  0.0,
230
- 1.0,
231
  0.0,
232
- 1.0,
233
- 1.0,
234
  1.0,
235
  0.0,
236
  0.0,
 
 
237
  1.0
238
  ],
239
  [
240
  1.0,
241
  1.0,
242
  1.0,
243
- 0.0,
244
  1.0,
245
- 0.0,
 
246
  1.0,
247
  0.0,
248
  0.0,
249
- 1.0,
250
  0.0,
251
- 1.0,
252
- 1.0,
253
  1.0,
254
  0.0,
 
255
  0.0,
256
  1.0
257
  ],
@@ -259,15 +259,15 @@
259
  1.0,
260
  1.0,
261
  1.0,
262
- 0.0,
263
  1.0,
264
- 0.0,
 
265
  1.0,
266
  0.0,
267
  0.0,
268
- 1.0,
269
  0.0,
270
- 1.0,
 
271
  0.0,
272
  1.0,
273
  1.0,
@@ -278,17 +278,17 @@
278
  1.0,
279
  1.0,
280
  1.0,
281
- 0.0,
282
  1.0,
283
- 0.0,
 
284
  1.0,
285
  0.0,
286
  0.0,
287
- 1.0,
288
  0.0,
289
- 1.0,
290
  0.0,
291
- 1.0,
 
 
292
  1.0,
293
  0.0,
294
  1.0
@@ -297,19 +297,19 @@
297
  1.0,
298
  1.0,
299
  1.0,
300
- 0.0,
301
  1.0,
302
- 0.0,
 
303
  1.0,
304
  0.0,
305
  0.0,
306
- 1.0,
307
  0.0,
308
- 1.0,
309
  0.0,
310
- 1.0,
311
  0.0,
312
  0.0,
 
 
 
313
  0.0
314
  ],
315
  [
@@ -629,40 +629,40 @@
629
  "hidden_dropout_prob": 0.1,
630
  "hidden_size": 1024,
631
  "id2label": {
632
- "0": "O",
633
- "1": "B-event",
634
- "2": "B-location",
635
- "3": "I-location",
636
- "4": "B-person",
637
- "5": "I-person",
638
- "6": "B-corporation",
639
  "7": "I-corporation",
640
- "8": "I-event",
641
- "9": "B-product",
642
- "10": "I-product",
643
- "11": "B-work of art",
644
- "12": "I-work of art",
645
- "13": "B-group",
646
- "14": "I-group"
647
  },
648
  "initializer_range": 0.02,
649
  "intermediate_size": 4096,
650
  "label2id": {
651
- "B-corporation": 6,
652
- "B-event": 1,
653
- "B-group": 13,
654
- "B-location": 2,
655
- "B-person": 4,
656
- "B-product": 9,
657
- "B-work of art": 11,
658
  "I-corporation": 7,
659
- "I-event": 8,
660
- "I-group": 14,
661
- "I-location": 3,
662
- "I-person": 5,
663
- "I-product": 10,
664
- "I-work of art": 12,
665
- "O": 0
666
  },
667
  "layer_norm_eps": 1e-05,
668
  "max_position_embeddings": 514,
@@ -672,7 +672,7 @@
672
  "pad_token_id": 1,
673
  "position_embedding_type": "absolute",
674
  "torch_dtype": "float32",
675
- "transformers_version": "4.12.5",
676
  "type_vocab_size": 1,
677
  "use_cache": true,
678
  "vocab_size": 50265
 
1
  {
2
+ "_name_or_path": "roberta-large",
3
  "architectures": [
4
  "RobertaForTokenClassification"
5
  ],
 
12
  1.0,
13
  1.0,
14
  1.0,
 
15
  1.0,
16
+ 1.0,
17
+ 1.0,
18
+ 1.0,
19
  1.0,
20
  0.0,
21
  0.0,
 
22
  0.0,
 
23
  0.0,
 
24
  0.0,
25
  0.0,
26
+ 1.0,
27
+ 0.0,
28
  1.0
29
  ],
30
  [
31
  1.0,
32
  1.0,
33
  1.0,
 
34
  1.0,
 
35
  1.0,
 
36
  1.0,
37
  1.0,
38
  0.0,
39
  1.0,
40
  0.0,
 
41
  0.0,
42
  0.0,
43
+ 0.0,
44
+ 0.0,
45
+ 1.0,
46
+ 0.0,
47
  1.0
48
  ],
49
  [
 
52
  1.0,
53
  1.0,
54
  1.0,
55
+ 1.0,
56
  1.0,
57
  0.0,
58
  0.0,
59
  1.0,
60
  0.0,
 
61
  0.0,
 
62
  0.0,
63
  0.0,
64
+ 1.0,
65
+ 0.0,
66
  1.0
67
  ],
68
  [
 
71
  1.0,
72
  1.0,
73
  1.0,
74
+ 1.0,
75
  1.0,
76
  0.0,
77
  0.0,
 
78
  0.0,
79
  1.0,
80
  0.0,
 
81
  0.0,
82
  0.0,
83
+ 1.0,
84
+ 0.0,
85
  1.0
86
  ],
87
  [
88
  1.0,
89
  1.0,
90
  1.0,
91
+ 1.0,
92
  1.0,
93
  1.0,
94
  1.0,
95
  0.0,
96
  0.0,
 
97
  0.0,
 
98
  0.0,
99
  1.0,
100
  0.0,
101
  0.0,
102
+ 1.0,
103
+ 0.0,
104
  1.0
105
  ],
106
  [
107
  1.0,
108
  1.0,
109
  1.0,
110
+ 1.0,
111
  1.0,
112
  1.0,
113
  1.0,
114
  0.0,
115
  0.0,
 
116
  0.0,
117
+ 0.0,
118
  0.0,
119
  1.0,
120
  0.0,
121
+ 1.0,
122
  0.0,
123
  1.0
124
  ],
 
126
  1.0,
127
  1.0,
128
  1.0,
 
129
  1.0,
 
130
  1.0,
131
  1.0,
 
132
  1.0,
133
  0.0,
 
134
  0.0,
 
135
  0.0,
136
  0.0,
137
+ 0.0,
138
+ 0.0,
139
+ 1.0,
140
+ 1.0,
141
+ 0.0,
142
  1.0
143
  ],
144
  [
145
  1.0,
146
  1.0,
147
  1.0,
 
148
  1.0,
 
149
  1.0,
150
  1.0,
 
151
  1.0,
 
152
  1.0,
153
  0.0,
 
154
  0.0,
155
  0.0,
156
+ 0.0,
157
+ 0.0,
158
+ 0.0,
159
+ 1.0,
160
+ 0.0,
161
  1.0
162
  ],
163
  [
164
  1.0,
165
  1.0,
166
  1.0,
 
167
  1.0,
 
168
  1.0,
 
169
  1.0,
170
  1.0,
171
  0.0,
172
  1.0,
173
  0.0,
 
174
  0.0,
175
  0.0,
176
+ 0.0,
177
+ 0.0,
178
+ 1.0,
179
+ 0.0,
180
  1.0
181
  ],
182
  [
183
  1.0,
184
  1.0,
185
  1.0,
 
186
  1.0,
 
 
 
 
187
  1.0,
188
  1.0,
189
  1.0,
190
  0.0,
191
+ 0.0,
192
  1.0,
193
  0.0,
194
  0.0,
195
+ 0.0,
196
+ 0.0,
197
+ 1.0,
198
+ 0.0,
199
  1.0
200
  ],
201
  [
202
  1.0,
203
  1.0,
204
  1.0,
 
205
  1.0,
 
 
 
 
206
  1.0,
207
  1.0,
208
  1.0,
209
  0.0,
210
+ 0.0,
211
+ 0.0,
212
  1.0,
213
  0.0,
214
  0.0,
215
+ 0.0,
216
+ 1.0,
217
+ 0.0,
218
  1.0
219
  ],
220
  [
221
  1.0,
222
  1.0,
223
  1.0,
 
224
  1.0,
225
+ 1.0,
226
+ 1.0,
227
  1.0,
228
  0.0,
229
  0.0,
 
230
  0.0,
231
+ 0.0,
 
232
  1.0,
233
  0.0,
234
  0.0,
235
+ 1.0,
236
+ 0.0,
237
  1.0
238
  ],
239
  [
240
  1.0,
241
  1.0,
242
  1.0,
 
243
  1.0,
244
+ 1.0,
245
+ 1.0,
246
  1.0,
247
  0.0,
248
  0.0,
 
249
  0.0,
250
+ 0.0,
251
+ 0.0,
252
  1.0,
253
  0.0,
254
+ 1.0,
255
  0.0,
256
  1.0
257
  ],
 
259
  1.0,
260
  1.0,
261
  1.0,
 
262
  1.0,
263
+ 1.0,
264
+ 1.0,
265
  1.0,
266
  0.0,
267
  0.0,
 
268
  0.0,
269
+ 0.0,
270
+ 0.0,
271
  0.0,
272
  1.0,
273
  1.0,
 
278
  1.0,
279
  1.0,
280
  1.0,
 
281
  1.0,
282
+ 1.0,
283
+ 1.0,
284
  1.0,
285
  0.0,
286
  0.0,
 
287
  0.0,
 
288
  0.0,
289
+ 0.0,
290
+ 0.0,
291
+ 0.0,
292
  1.0,
293
  0.0,
294
  1.0
 
297
  1.0,
298
  1.0,
299
  1.0,
 
300
  1.0,
301
+ 1.0,
302
+ 1.0,
303
  1.0,
304
  0.0,
305
  0.0,
 
306
  0.0,
 
307
  0.0,
 
308
  0.0,
309
  0.0,
310
+ 0.0,
311
+ 1.0,
312
+ 0.0,
313
  0.0
314
  ],
315
  [
 
629
  "hidden_dropout_prob": 0.1,
630
  "hidden_size": 1024,
631
  "id2label": {
632
+ "0": "B-corporation",
633
+ "1": "B-creative_work",
634
+ "2": "B-event",
635
+ "3": "B-group",
636
+ "4": "B-location",
637
+ "5": "B-person",
638
+ "6": "B-product",
639
  "7": "I-corporation",
640
+ "8": "I-creative_work",
641
+ "9": "I-event",
642
+ "10": "I-group",
643
+ "11": "I-location",
644
+ "12": "I-person",
645
+ "13": "I-product",
646
+ "14": "O"
647
  },
648
  "initializer_range": 0.02,
649
  "intermediate_size": 4096,
650
  "label2id": {
651
+ "B-corporation": 0,
652
+ "B-creative_work": 1,
653
+ "B-event": 2,
654
+ "B-group": 3,
655
+ "B-location": 4,
656
+ "B-person": 5,
657
+ "B-product": 6,
658
  "I-corporation": 7,
659
+ "I-creative_work": 8,
660
+ "I-event": 9,
661
+ "I-group": 10,
662
+ "I-location": 11,
663
+ "I-person": 12,
664
+ "I-product": 13,
665
+ "O": 14
666
  },
667
  "layer_norm_eps": 1e-05,
668
  "max_position_embeddings": 514,
 
672
  "pad_token_id": 1,
673
  "position_embedding_type": "absolute",
674
  "torch_dtype": "float32",
675
+ "transformers_version": "4.17.0",
676
  "type_vocab_size": 1,
677
  "use_cache": true,
678
  "vocab_size": 50265
eval/metric.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"2020.dev": {"micro/f1": 0.6411716842961759, "micro/recall": 0.6175548589341693, "micro/precision": 0.6666666666666666, "macro/f1": 0.5804596425227457, "macro/recall": 0.5680055941903029, "macro/precision": 0.6008877546537066, "per_entity_metric": {"corporation": {"precision": 0.49710982658959535, "recall": 0.4236453201970443, "f1": 0.45744680851063824}, "creative_work": {"precision": 0.5, "recall": 0.5144230769230769, "f1": 0.5071090047393365}, "event": {"precision": 0.47368421052631576, "recall": 0.31640625, "f1": 0.3793911007025761}, "group": {"precision": 0.59, "recall": 0.5198237885462555, "f1": 0.5526932084309134}, "location": {"precision": 0.5799086757990868, "recall": 0.7016574585635359, "f1": 0.635}, "person": {"precision": 0.9099099099099099, "recall": 0.8444816053511706, "f1": 0.8759757155247182}, "product": {"precision": 0.6556016597510373, "recall": 0.6556016597510373, "f1": 0.6556016597510373}}}, "2021.test": {"micro/f1": 0.6435331230283912, "micro/recall": 0.6604995374653099, "micro/precision": 0.6274165202108963, "macro/f1": 0.5923174132013881, "macro/recall": 0.6162368271564922, "macro/precision": 0.5754366734391806, "per_entity_metric": {"corporation": {"precision": 0.47904191616766467, "recall": 0.5333333333333333, "f1": 0.504731861198738}, "creative_work": {"precision": 0.39166666666666666, "recall": 0.5143638850889193, "f1": 0.4447072738024837}, "event": {"precision": 0.5186915887850467, "recall": 0.40400363967242947, "f1": 0.4542199488491049}, "group": {"precision": 0.6072144288577155, "recall": 0.5988142292490118, "f1": 0.6029850746268657}, "location": {"precision": 0.6097852028639618, "recall": 0.7136871508379888, "f1": 0.6576576576576576}, "person": {"precision": 0.8180854821617803, "recall": 0.8539823008849557, "f1": 0.8356485657586145}, "product": {"precision": 0.6035714285714285, "recall": 0.6954732510288066, "f1": 0.6462715105162524}}}, "2020.test": {"micro/f1": 0.6537007023230686, "micro/recall": 0.6279190451478983, "micro/precision": 0.6816901408450704, "macro/f1": 0.6108426813075821, "macro/recall": 0.5938611241800162, "macro/precision": 0.6343275766766727, "per_entity_metric": {"corporation": {"precision": 0.5346534653465347, "recall": 0.5654450261780105, "f1": 0.549618320610687}, "creative_work": {"precision": 0.49, "recall": 0.547486033519553, "f1": 0.5171503957783642}, "event": {"precision": 0.5612244897959183, "recall": 0.41509433962264153, "f1": 0.47722342733188716}, "group": {"precision": 0.6419753086419753, "recall": 0.5016077170418006, "f1": 0.5631768953068591}, "location": {"precision": 0.675, "recall": 0.6545454545454545, "f1": 0.6646153846153846}, "person": {"precision": 0.8723021582733813, "recall": 0.8137583892617449, "f1": 0.842013888888889}, "product": {"precision": 0.6651376146788991, "recall": 0.6590909090909091, "f1": 0.6621004566210046}}}}
eval/prediction.2020.dev.json ADDED
The diff for this file is too large to render. See raw diff
 
eval/prediction.2020.test.json ADDED
The diff for this file is too large to render. See raw diff
 
eval/prediction.2021.test.json ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:937209089e9636154f1e3e1545a2486ec1564517fdcec93038271ac36cc77f05
3
- size 1417465457
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c15243d80be50209710c05aa9db3f081f04b0b6b2a4ab94efab20c2e24ced12
3
+ size 1417467377
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 512, "name_or_path": "./tner_output/search/roberta_large/twitter_ner/model_tpkhxk/epoch_10", "special_tokens_map_file": "./tner_output/search/roberta_large/twitter_ner/model_tpkhxk/epoch_10/special_tokens_map.json", "tokenizer_class": "RobertaTokenizer"}
 
1
+ {"errors": "replace", "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "add_prefix_space": false, "trim_offsets": true, "model_max_length": 512, "name_or_path": "roberta-large", "tokenizer_class": "RobertaTokenizer"}