Hugging Face's logo

nlpso
/
m0_flat_ner_ocr_ptrn_cmbert_io

Token Classification
Transformers
PyTorch
French
camembert
Model card Files
xet
Community
1
nlpso commited on
Commit
20c292d
·
1 Parent(s): b10109a

Load model and tokenizer

Browse files
Files changed (11) hide show
  1. config.json +47 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scheduler.pt +3 -0
  6. sentencepiece.bpe.model +3 -0
  7. special_tokens_map.json +13 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +25 -0
  10. trainer_state.json +700 -0
  11. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "HueyNemud/das22-10-camembert_pretrained",
3
+ "architectures": [
4
+ "CamembertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 5,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 6,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "O",
16
+ "1": "I-PER",
17
+ "2": "I-TITRE",
18
+ "3": "I-ACT",
19
+ "4": "I-LOC",
20
+ "5": "I-CARDINAL",
21
+ "6": "I-FT"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 3072,
25
+ "label2id": {
26
+ "I-ACT": 3,
27
+ "I-CARDINAL": 5,
28
+ "I-FT": 6,
29
+ "I-LOC": 4,
30
+ "I-PER": 1,
31
+ "I-TITRE": 2,
32
+ "O": 0
33
+ },
34
+ "layer_norm_eps": 1e-05,
35
+ "max_position_embeddings": 514,
36
+ "model_type": "camembert",
37
+ "num_attention_heads": 12,
38
+ "num_hidden_layers": 12,
39
+ "output_past": true,
40
+ "pad_token_id": 1,
41
+ "position_embedding_type": "absolute",
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.25.1",
44
+ "type_vocab_size": 1,
45
+ "use_cache": true,
46
+ "vocab_size": 32005
47
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79a02b8e51543c87476b4eec6bde1416e69cc3a325c6b9a92157288b4911bcff
3
+ size 880409317
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86eb4f683a166991f0832e5c878c25c122e89e23ff6b8caafd3b628dd921af86
3
+ size 440218033
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af02fe94438ceddd21698804082934b6aa30ba09322dfac61528c7663af6ae34
3
+ size 14503
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3348d57e33a2df56ff570fa06068a10b38e92e5d9370fb65ef4dc6aceb702906
3
+ size 623
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:988bc5a00281c6d210a5d34bd143d0363741a432fefe741bf71e61b1869d4314
3
+ size 810912
special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "cls_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "mask_token": "<mask>",
10
+ "pad_token": "<pad>",
11
+ "sep_token": "</s>",
12
+ "unk_token": "<unk>"
13
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<s>NOTUSED",
4
+ "</s>NOTUSED"
5
+ ],
6
+ "bos_token": "<s>",
7
+ "cls_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "mask_token": {
10
+ "__type": "AddedToken",
11
+ "content": "<mask>",
12
+ "lstrip": true,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ "model_max_length": 512,
18
+ "name_or_path": "HueyNemud/das22-10-camembert_pretrained",
19
+ "pad_token": "<pad>",
20
+ "sep_token": "</s>",
21
+ "sp_model_kwargs": {},
22
+ "special_tokens_map_file": "/lrde/home2/jchazalo/.cache/huggingface/transformers/fe0e213c44079a9ee091098f81fff2941484006e9ba3001a9bf1ee9f87537599.cb3ec3a6c1200d181228d8825ae9767572abca54efa1bbb37fd83d721b2ef323",
23
+ "tokenizer_class": "CamembertTokenizer",
24
+ "unk_token": "<unk>"
25
+ }
trainer_state.json ADDED
@@ -0,0 +1,700 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9717367050948308,
3
+ "best_model_checkpoint": "/work/stual/res_ICDAR/method_0/tmp/42-flat-ner-pero-ocr-pretrained_camembert_ner/checkpoint-1400",
4
+ "epoch": 3.674540682414698,
5
+ "global_step": 1400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.26,
12
+ "eval_ACT": {
13
+ "f1": 0.9256820319849483,
14
+ "number": 519,
15
+ "precision": 0.9044117647058824,
16
+ "recall": 0.9479768786127167
17
+ },
18
+ "eval_CARDINAL": {
19
+ "f1": 0.9780701754385964,
20
+ "number": 678,
21
+ "precision": 0.9695652173913043,
22
+ "recall": 0.9867256637168141
23
+ },
24
+ "eval_FT": {
25
+ "f1": 0.0,
26
+ "number": 7,
27
+ "precision": 0.0,
28
+ "recall": 0.0
29
+ },
30
+ "eval_LOC": {
31
+ "f1": 0.9407166123778501,
32
+ "number": 761,
33
+ "precision": 0.9328165374677002,
34
+ "recall": 0.9487516425755584
35
+ },
36
+ "eval_PER": {
37
+ "f1": 0.9334308705193856,
38
+ "number": 679,
39
+ "precision": 0.9273255813953488,
40
+ "recall": 0.9396170839469808
41
+ },
42
+ "eval_TITRE": {
43
+ "f1": 0.44776119402985076,
44
+ "number": 41,
45
+ "precision": 0.5769230769230769,
46
+ "recall": 0.36585365853658536
47
+ },
48
+ "eval_accuracy": 0.9630473297899266,
49
+ "eval_f1": 0.9380432772332162,
50
+ "eval_loss": 0.22555255889892578,
51
+ "eval_precision": 0.9316678912564291,
52
+ "eval_recall": 0.9445065176908752,
53
+ "eval_runtime": 2.7451,
54
+ "eval_samples_per_second": 246.258,
55
+ "eval_steps_per_second": 15.664,
56
+ "step": 100
57
+ },
58
+ {
59
+ "epoch": 0.52,
60
+ "eval_ACT": {
61
+ "f1": 0.9429657794676807,
62
+ "number": 519,
63
+ "precision": 0.9305816135084428,
64
+ "recall": 0.9556840077071291
65
+ },
66
+ "eval_CARDINAL": {
67
+ "f1": 0.9809104258443465,
68
+ "number": 678,
69
+ "precision": 0.9766081871345029,
70
+ "recall": 0.9852507374631269
71
+ },
72
+ "eval_FT": {
73
+ "f1": 0.0,
74
+ "number": 7,
75
+ "precision": 0.0,
76
+ "recall": 0.0
77
+ },
78
+ "eval_LOC": {
79
+ "f1": 0.938961038961039,
80
+ "number": 761,
81
+ "precision": 0.9281129653401797,
82
+ "recall": 0.9500657030223391
83
+ },
84
+ "eval_PER": {
85
+ "f1": 0.9655172413793105,
86
+ "number": 679,
87
+ "precision": 0.9619883040935673,
88
+ "recall": 0.9690721649484536
89
+ },
90
+ "eval_TITRE": {
91
+ "f1": 0.575,
92
+ "number": 41,
93
+ "precision": 0.5897435897435898,
94
+ "recall": 0.5609756097560976
95
+ },
96
+ "eval_accuracy": 0.9681093394077449,
97
+ "eval_f1": 0.9504071058475203,
98
+ "eval_loss": 0.14738017320632935,
99
+ "eval_precision": 0.9444648767929386,
100
+ "eval_recall": 0.9564245810055866,
101
+ "eval_runtime": 2.0679,
102
+ "eval_samples_per_second": 326.898,
103
+ "eval_steps_per_second": 20.794,
104
+ "step": 200
105
+ },
106
+ {
107
+ "epoch": 0.79,
108
+ "eval_ACT": {
109
+ "f1": 0.9559386973180076,
110
+ "number": 519,
111
+ "precision": 0.9504761904761905,
112
+ "recall": 0.9614643545279383
113
+ },
114
+ "eval_CARDINAL": {
115
+ "f1": 0.9845474613686535,
116
+ "number": 678,
117
+ "precision": 0.9823788546255506,
118
+ "recall": 0.9867256637168141
119
+ },
120
+ "eval_FT": {
121
+ "f1": 0.4615384615384615,
122
+ "number": 7,
123
+ "precision": 0.5,
124
+ "recall": 0.42857142857142855
125
+ },
126
+ "eval_LOC": {
127
+ "f1": 0.9639344262295081,
128
+ "number": 761,
129
+ "precision": 0.9620418848167539,
130
+ "recall": 0.9658344283837057
131
+ },
132
+ "eval_PER": {
133
+ "f1": 0.9624724061810155,
134
+ "number": 679,
135
+ "precision": 0.961764705882353,
136
+ "recall": 0.9631811487481591
137
+ },
138
+ "eval_TITRE": {
139
+ "f1": 0.6933333333333332,
140
+ "number": 41,
141
+ "precision": 0.7647058823529411,
142
+ "recall": 0.6341463414634146
143
+ },
144
+ "eval_accuracy": 0.9758289040749177,
145
+ "eval_f1": 0.9622325581395349,
146
+ "eval_loss": 0.1064079999923706,
147
+ "eval_precision": 0.9613382899628253,
148
+ "eval_recall": 0.9631284916201117,
149
+ "eval_runtime": 2.071,
150
+ "eval_samples_per_second": 326.419,
151
+ "eval_steps_per_second": 20.763,
152
+ "step": 300
153
+ },
154
+ {
155
+ "epoch": 1.05,
156
+ "eval_ACT": {
157
+ "f1": 0.9364928909952606,
158
+ "number": 519,
159
+ "precision": 0.9216417910447762,
160
+ "recall": 0.9518304431599229
161
+ },
162
+ "eval_CARDINAL": {
163
+ "f1": 0.9845246868091377,
164
+ "number": 678,
165
+ "precision": 0.9837997054491899,
166
+ "recall": 0.9852507374631269
167
+ },
168
+ "eval_FT": {
169
+ "f1": 0.5454545454545454,
170
+ "number": 7,
171
+ "precision": 0.75,
172
+ "recall": 0.42857142857142855
173
+ },
174
+ "eval_LOC": {
175
+ "f1": 0.9541884816753926,
176
+ "number": 761,
177
+ "precision": 0.9504563233376793,
178
+ "recall": 0.9579500657030223
179
+ },
180
+ "eval_PER": {
181
+ "f1": 0.9823529411764705,
182
+ "number": 679,
183
+ "precision": 0.9809104258443465,
184
+ "recall": 0.9837997054491899
185
+ },
186
+ "eval_TITRE": {
187
+ "f1": 0.7654320987654322,
188
+ "number": 41,
189
+ "precision": 0.775,
190
+ "recall": 0.7560975609756098
191
+ },
192
+ "eval_accuracy": 0.9712730954188813,
193
+ "eval_f1": 0.9617952522255193,
194
+ "eval_loss": 0.1030876561999321,
195
+ "eval_precision": 0.9578869597340229,
196
+ "eval_recall": 0.9657355679702049,
197
+ "eval_runtime": 2.0342,
198
+ "eval_samples_per_second": 332.316,
199
+ "eval_steps_per_second": 21.138,
200
+ "step": 400
201
+ },
202
+ {
203
+ "epoch": 1.31,
204
+ "learning_rate": 9e-05,
205
+ "loss": 0.2317,
206
+ "step": 500
207
+ },
208
+ {
209
+ "epoch": 1.31,
210
+ "eval_ACT": {
211
+ "f1": 0.9608404966571156,
212
+ "number": 519,
213
+ "precision": 0.9526515151515151,
214
+ "recall": 0.9691714836223507
215
+ },
216
+ "eval_CARDINAL": {
217
+ "f1": 0.9859985261606484,
218
+ "number": 678,
219
+ "precision": 0.9852724594992637,
220
+ "recall": 0.9867256637168141
221
+ },
222
+ "eval_FT": {
223
+ "f1": 0.5454545454545454,
224
+ "number": 7,
225
+ "precision": 0.75,
226
+ "recall": 0.42857142857142855
227
+ },
228
+ "eval_LOC": {
229
+ "f1": 0.9454545454545454,
230
+ "number": 761,
231
+ "precision": 0.9345314505776636,
232
+ "recall": 0.9566360052562418
233
+ },
234
+ "eval_PER": {
235
+ "f1": 0.9685442574981712,
236
+ "number": 679,
237
+ "precision": 0.9622093023255814,
238
+ "recall": 0.9749631811487481
239
+ },
240
+ "eval_TITRE": {
241
+ "f1": 0.6,
242
+ "number": 41,
243
+ "precision": 0.6153846153846154,
244
+ "recall": 0.5853658536585366
245
+ },
246
+ "eval_accuracy": 0.9719058466211086,
247
+ "eval_f1": 0.9585338763420954,
248
+ "eval_loss": 0.10161029547452927,
249
+ "eval_precision": 0.9528892160471107,
250
+ "eval_recall": 0.964245810055866,
251
+ "eval_runtime": 2.1114,
252
+ "eval_samples_per_second": 320.168,
253
+ "eval_steps_per_second": 20.366,
254
+ "step": 500
255
+ },
256
+ {
257
+ "epoch": 1.57,
258
+ "eval_ACT": {
259
+ "f1": 0.9579349904397705,
260
+ "number": 519,
261
+ "precision": 0.9506641366223909,
262
+ "recall": 0.9653179190751445
263
+ },
264
+ "eval_CARDINAL": {
265
+ "f1": 0.9845474613686535,
266
+ "number": 678,
267
+ "precision": 0.9823788546255506,
268
+ "recall": 0.9867256637168141
269
+ },
270
+ "eval_FT": {
271
+ "f1": 0.36363636363636365,
272
+ "number": 7,
273
+ "precision": 0.5,
274
+ "recall": 0.2857142857142857
275
+ },
276
+ "eval_LOC": {
277
+ "f1": 0.9625738673670386,
278
+ "number": 761,
279
+ "precision": 0.9619422572178478,
280
+ "recall": 0.9632063074901446
281
+ },
282
+ "eval_PER": {
283
+ "f1": 0.9742836149889786,
284
+ "number": 679,
285
+ "precision": 0.9721407624633431,
286
+ "recall": 0.9764359351988218
287
+ },
288
+ "eval_TITRE": {
289
+ "f1": 0.7317073170731707,
290
+ "number": 41,
291
+ "precision": 0.7317073170731707,
292
+ "recall": 0.7317073170731707
293
+ },
294
+ "eval_accuracy": 0.9779802581624905,
295
+ "eval_f1": 0.9654403567447045,
296
+ "eval_loss": 0.08714007586240768,
297
+ "eval_precision": 0.9632925472747497,
298
+ "eval_recall": 0.9675977653631285,
299
+ "eval_runtime": 4.5111,
300
+ "eval_samples_per_second": 149.853,
301
+ "eval_steps_per_second": 9.532,
302
+ "step": 600
303
+ },
304
+ {
305
+ "epoch": 1.84,
306
+ "eval_ACT": {
307
+ "f1": 0.9476688867745005,
308
+ "number": 519,
309
+ "precision": 0.9360902255639098,
310
+ "recall": 0.9595375722543352
311
+ },
312
+ "eval_CARDINAL": {
313
+ "f1": 0.9867452135493373,
314
+ "number": 678,
315
+ "precision": 0.9852941176470589,
316
+ "recall": 0.9882005899705014
317
+ },
318
+ "eval_FT": {
319
+ "f1": 0.4,
320
+ "number": 7,
321
+ "precision": 0.6666666666666666,
322
+ "recall": 0.2857142857142857
323
+ },
324
+ "eval_LOC": {
325
+ "f1": 0.9587965990843689,
326
+ "number": 761,
327
+ "precision": 0.9544270833333334,
328
+ "recall": 0.9632063074901446
329
+ },
330
+ "eval_PER": {
331
+ "f1": 0.969208211143695,
332
+ "number": 679,
333
+ "precision": 0.964963503649635,
334
+ "recall": 0.9734904270986745
335
+ },
336
+ "eval_TITRE": {
337
+ "f1": 0.7804878048780488,
338
+ "number": 41,
339
+ "precision": 0.7804878048780488,
340
+ "recall": 0.7804878048780488
341
+ },
342
+ "eval_accuracy": 0.9765882055175905,
343
+ "eval_f1": 0.9625509825732296,
344
+ "eval_loss": 0.08122637122869492,
345
+ "eval_precision": 0.9582871908453304,
346
+ "eval_recall": 0.966852886405959,
347
+ "eval_runtime": 2.7384,
348
+ "eval_samples_per_second": 246.857,
349
+ "eval_steps_per_second": 15.702,
350
+ "step": 700
351
+ },
352
+ {
353
+ "epoch": 2.1,
354
+ "eval_ACT": {
355
+ "f1": 0.9514747859181732,
356
+ "number": 519,
357
+ "precision": 0.9398496240601504,
358
+ "recall": 0.9633911368015414
359
+ },
360
+ "eval_CARDINAL": {
361
+ "f1": 0.9874908020603386,
362
+ "number": 678,
363
+ "precision": 0.9853157121879589,
364
+ "recall": 0.9896755162241888
365
+ },
366
+ "eval_FT": {
367
+ "f1": 0.4,
368
+ "number": 7,
369
+ "precision": 0.6666666666666666,
370
+ "recall": 0.2857142857142857
371
+ },
372
+ "eval_LOC": {
373
+ "f1": 0.9652914210870989,
374
+ "number": 761,
375
+ "precision": 0.9621409921671018,
376
+ "recall": 0.9684625492772667
377
+ },
378
+ "eval_PER": {
379
+ "f1": 0.9706314243759177,
380
+ "number": 679,
381
+ "precision": 0.9677891654465594,
382
+ "recall": 0.9734904270986745
383
+ },
384
+ "eval_TITRE": {
385
+ "f1": 0.8311688311688312,
386
+ "number": 41,
387
+ "precision": 0.8888888888888888,
388
+ "recall": 0.7804878048780488
389
+ },
390
+ "eval_accuracy": 0.9787395596051632,
391
+ "eval_f1": 0.9665800222799851,
392
+ "eval_loss": 0.07897698879241943,
393
+ "eval_precision": 0.9637171417993335,
394
+ "eval_recall": 0.9694599627560522,
395
+ "eval_runtime": 2.0295,
396
+ "eval_samples_per_second": 333.085,
397
+ "eval_steps_per_second": 21.187,
398
+ "step": 800
399
+ },
400
+ {
401
+ "epoch": 2.36,
402
+ "eval_ACT": {
403
+ "f1": 0.9570200573065903,
404
+ "number": 519,
405
+ "precision": 0.9488636363636364,
406
+ "recall": 0.9653179190751445
407
+ },
408
+ "eval_CARDINAL": {
409
+ "f1": 0.9859778597785978,
410
+ "number": 678,
411
+ "precision": 0.9867060561299852,
412
+ "recall": 0.9852507374631269
413
+ },
414
+ "eval_FT": {
415
+ "f1": 0.4,
416
+ "number": 7,
417
+ "precision": 0.6666666666666666,
418
+ "recall": 0.2857142857142857
419
+ },
420
+ "eval_LOC": {
421
+ "f1": 0.958688524590164,
422
+ "number": 761,
423
+ "precision": 0.9568062827225131,
424
+ "recall": 0.9605781865965834
425
+ },
426
+ "eval_PER": {
427
+ "f1": 0.9816041206769683,
428
+ "number": 679,
429
+ "precision": 0.9808823529411764,
430
+ "recall": 0.9823269513991163
431
+ },
432
+ "eval_TITRE": {
433
+ "f1": 0.8354430379746836,
434
+ "number": 41,
435
+ "precision": 0.868421052631579,
436
+ "recall": 0.8048780487804879
437
+ },
438
+ "eval_accuracy": 0.9784864591242723,
439
+ "eval_f1": 0.968186046511628,
440
+ "eval_loss": 0.08038043230772018,
441
+ "eval_precision": 0.9672862453531599,
442
+ "eval_recall": 0.9690875232774674,
443
+ "eval_runtime": 2.0985,
444
+ "eval_samples_per_second": 322.128,
445
+ "eval_steps_per_second": 20.49,
446
+ "step": 900
447
+ },
448
+ {
449
+ "epoch": 2.62,
450
+ "learning_rate": 8e-05,
451
+ "loss": 0.0696,
452
+ "step": 1000
453
+ },
454
+ {
455
+ "epoch": 2.62,
456
+ "eval_ACT": {
457
+ "f1": 0.970334928229665,
458
+ "number": 519,
459
+ "precision": 0.9638783269961977,
460
+ "recall": 0.976878612716763
461
+ },
462
+ "eval_CARDINAL": {
463
+ "f1": 0.9845474613686535,
464
+ "number": 678,
465
+ "precision": 0.9823788546255506,
466
+ "recall": 0.9867256637168141
467
+ },
468
+ "eval_FT": {
469
+ "f1": 0.36363636363636365,
470
+ "number": 7,
471
+ "precision": 0.5,
472
+ "recall": 0.2857142857142857
473
+ },
474
+ "eval_LOC": {
475
+ "f1": 0.962671905697446,
476
+ "number": 761,
477
+ "precision": 0.9595300261096605,
478
+ "recall": 0.9658344283837057
479
+ },
480
+ "eval_PER": {
481
+ "f1": 0.9808823529411764,
482
+ "number": 679,
483
+ "precision": 0.9794419970631424,
484
+ "recall": 0.9823269513991163
485
+ },
486
+ "eval_TITRE": {
487
+ "f1": 0.7594936708860759,
488
+ "number": 41,
489
+ "precision": 0.7894736842105263,
490
+ "recall": 0.7317073170731707
491
+ },
492
+ "eval_accuracy": 0.982536066818527,
493
+ "eval_f1": 0.9700799107972496,
494
+ "eval_loss": 0.07389234751462936,
495
+ "eval_precision": 0.9681008902077152,
496
+ "eval_recall": 0.9720670391061452,
497
+ "eval_runtime": 3.1328,
498
+ "eval_samples_per_second": 215.784,
499
+ "eval_steps_per_second": 13.726,
500
+ "step": 1000
501
+ },
502
+ {
503
+ "epoch": 2.89,
504
+ "eval_ACT": {
505
+ "f1": 0.9571020019065777,
506
+ "number": 519,
507
+ "precision": 0.9471698113207547,
508
+ "recall": 0.9672447013487476
509
+ },
510
+ "eval_CARDINAL": {
511
+ "f1": 0.9889462048636699,
512
+ "number": 678,
513
+ "precision": 0.9882179675994109,
514
+ "recall": 0.9896755162241888
515
+ },
516
+ "eval_FT": {
517
+ "f1": 0.3333333333333333,
518
+ "number": 7,
519
+ "precision": 0.4,
520
+ "recall": 0.2857142857142857
521
+ },
522
+ "eval_LOC": {
523
+ "f1": 0.9625738673670386,
524
+ "number": 761,
525
+ "precision": 0.9619422572178478,
526
+ "recall": 0.9632063074901446
527
+ },
528
+ "eval_PER": {
529
+ "f1": 0.9823529411764705,
530
+ "number": 679,
531
+ "precision": 0.9809104258443465,
532
+ "recall": 0.9837997054491899
533
+ },
534
+ "eval_TITRE": {
535
+ "f1": 0.8433734939759037,
536
+ "number": 41,
537
+ "precision": 0.8333333333333334,
538
+ "recall": 0.8536585365853658
539
+ },
540
+ "eval_accuracy": 0.9808909136927361,
541
+ "eval_f1": 0.9699108469539376,
542
+ "eval_loss": 0.07423367351293564,
543
+ "eval_precision": 0.9673953316042979,
544
+ "eval_recall": 0.97243947858473,
545
+ "eval_runtime": 2.3205,
546
+ "eval_samples_per_second": 291.313,
547
+ "eval_steps_per_second": 18.53,
548
+ "step": 1100
549
+ },
550
+ {
551
+ "epoch": 3.15,
552
+ "eval_ACT": {
553
+ "f1": 0.9578544061302681,
554
+ "number": 519,
555
+ "precision": 0.9523809523809523,
556
+ "recall": 0.9633911368015414
557
+ },
558
+ "eval_CARDINAL": {
559
+ "f1": 0.9852724594992636,
560
+ "number": 678,
561
+ "precision": 0.9838235294117647,
562
+ "recall": 0.9867256637168141
563
+ },
564
+ "eval_FT": {
565
+ "f1": 0.3333333333333333,
566
+ "number": 7,
567
+ "precision": 0.4,
568
+ "recall": 0.2857142857142857
569
+ },
570
+ "eval_LOC": {
571
+ "f1": 0.9619921363040629,
572
+ "number": 761,
573
+ "precision": 0.9594771241830066,
574
+ "recall": 0.9645203679369251
575
+ },
576
+ "eval_PER": {
577
+ "f1": 0.9794117647058823,
578
+ "number": 679,
579
+ "precision": 0.9779735682819384,
580
+ "recall": 0.9808541973490427
581
+ },
582
+ "eval_TITRE": {
583
+ "f1": 0.8048780487804877,
584
+ "number": 41,
585
+ "precision": 0.8048780487804879,
586
+ "recall": 0.8048780487804879
587
+ },
588
+ "eval_accuracy": 0.9803847127309542,
589
+ "eval_f1": 0.967670011148272,
590
+ "eval_loss": 0.08098369836807251,
591
+ "eval_precision": 0.9655172413793104,
592
+ "eval_recall": 0.9698324022346368,
593
+ "eval_runtime": 2.0371,
594
+ "eval_samples_per_second": 331.847,
595
+ "eval_steps_per_second": 21.109,
596
+ "step": 1200
597
+ },
598
+ {
599
+ "epoch": 3.41,
600
+ "eval_ACT": {
601
+ "f1": 0.9578544061302681,
602
+ "number": 519,
603
+ "precision": 0.9523809523809523,
604
+ "recall": 0.9633911368015414
605
+ },
606
+ "eval_CARDINAL": {
607
+ "f1": 0.9830508474576272,
608
+ "number": 678,
609
+ "precision": 0.9823269513991163,
610
+ "recall": 0.9837758112094396
611
+ },
612
+ "eval_FT": {
613
+ "f1": 0.5,
614
+ "number": 7,
615
+ "precision": 0.4444444444444444,
616
+ "recall": 0.5714285714285714
617
+ },
618
+ "eval_LOC": {
619
+ "f1": 0.9607329842931938,
620
+ "number": 761,
621
+ "precision": 0.9569752281616688,
622
+ "recall": 0.9645203679369251
623
+ },
624
+ "eval_PER": {
625
+ "f1": 0.9786921381337251,
626
+ "number": 679,
627
+ "precision": 0.9765395894428153,
628
+ "recall": 0.9808541973490427
629
+ },
630
+ "eval_TITRE": {
631
+ "f1": 0.7317073170731707,
632
+ "number": 41,
633
+ "precision": 0.7317073170731707,
634
+ "recall": 0.7317073170731707
635
+ },
636
+ "eval_accuracy": 0.9816502151354087,
637
+ "eval_f1": 0.965478841870824,
638
+ "eval_loss": 0.06854782998561859,
639
+ "eval_precision": 0.9622641509433962,
640
+ "eval_recall": 0.9687150837988827,
641
+ "eval_runtime": 2.0277,
642
+ "eval_samples_per_second": 333.39,
643
+ "eval_steps_per_second": 21.207,
644
+ "step": 1300
645
+ },
646
+ {
647
+ "epoch": 3.67,
648
+ "eval_ACT": {
649
+ "f1": 0.9655172413793104,
650
+ "number": 519,
651
+ "precision": 0.96,
652
+ "recall": 0.9710982658959537
653
+ },
654
+ "eval_CARDINAL": {
655
+ "f1": 0.9845474613686535,
656
+ "number": 678,
657
+ "precision": 0.9823788546255506,
658
+ "recall": 0.9867256637168141
659
+ },
660
+ "eval_FT": {
661
+ "f1": 0.5454545454545454,
662
+ "number": 7,
663
+ "precision": 0.75,
664
+ "recall": 0.42857142857142855
665
+ },
666
+ "eval_LOC": {
667
+ "f1": 0.9665573770491803,
668
+ "number": 761,
669
+ "precision": 0.9646596858638743,
670
+ "recall": 0.9684625492772667
671
+ },
672
+ "eval_PER": {
673
+ "f1": 0.9808541973490427,
674
+ "number": 679,
675
+ "precision": 0.9808541973490427,
676
+ "recall": 0.9808541973490427
677
+ },
678
+ "eval_TITRE": {
679
+ "f1": 0.8395061728395061,
680
+ "number": 41,
681
+ "precision": 0.85,
682
+ "recall": 0.8292682926829268
683
+ },
684
+ "eval_accuracy": 0.9832953682611997,
685
+ "eval_f1": 0.9717367050948308,
686
+ "eval_loss": 0.06763775646686554,
687
+ "eval_precision": 0.9702933531377645,
688
+ "eval_recall": 0.9731843575418995,
689
+ "eval_runtime": 2.0487,
690
+ "eval_samples_per_second": 329.96,
691
+ "eval_steps_per_second": 20.989,
692
+ "step": 1400
693
+ }
694
+ ],
695
+ "max_steps": 5000,
696
+ "num_train_epochs": 14,
697
+ "total_flos": 566245460738232.0,
698
+ "trial_name": null,
699
+ "trial_params": null
700
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8a84955445123f1e446e55dc128a8508559a9b17b726e9b83d950e1dd479e87
3
+ size 3503