Hugging Face's logo

ProMeText
/
aquilign_french_segmenter

Safetensors
bert
Model card Files
xet
Community
mattlev commited on
Commit
7e6c70e
·
verified ·
1 Parent(s): a4cab34

Upload 10 files

Browse files
Files changed (10) hide show
  1. config.json +36 -0
  2. eval.txt +6 -0
  3. metrics.json +1 -0
  4. model.safetensors +3 -0
  5. model_name +1 -0
  6. optimizer.pt +3 -0
  7. rng_state.pth +3 -0
  8. scheduler.pt +3 -0
  9. trainer_state.json +396 -0
  10. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "dbmdz/bert-base-french-europeana-cased",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "LABEL_0",
14
+ "1": "LABEL_1",
15
+ "2": "LABEL_2"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 3072,
19
+ "label2id": {
20
+ "LABEL_0": 0,
21
+ "LABEL_1": 1,
22
+ "LABEL_2": 2
23
+ },
24
+ "layer_norm_eps": 1e-12,
25
+ "max_position_embeddings": 512,
26
+ "model_type": "bert",
27
+ "num_attention_heads": 12,
28
+ "num_hidden_layers": 12,
29
+ "pad_token_id": 0,
30
+ "position_embedding_type": "absolute",
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.41.1",
33
+ "type_vocab_size": 2,
34
+ "use_cache": true,
35
+ "vocab_size": 32000
36
+ }
eval.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ | | Synt (None, Delim.) | Bert (None, Delim., Pad.) |
2
+ |-----------+-----------------------------------------------+-----------------------------------------------|
3
+ | Accuracy | 0.9547619047619048 | 0.9852173913043478 |
4
+ | Precision | [0.9417637271214643, 0.8188405797101449, 1.0] | [0.9921121822962313, 0.8743718592964824, 1.0] |
5
+ | Recall | [0.9783923941227312, 0.6174863387978142, 1.0] | [0.9783923941227312, 0.9508196721311475, 1.0] |
6
+ | F1-score | [0.9597286986011022, 0.7040498442367601, 1.0] | [0.9852045256744996, 0.9109947643979057, 1.0] |
metrics.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"loss": 0.0013, "grad_norm": 0.07273274660110474, "learning_rate": 2.25e-05, "epoch": 11.0, "step": 176, "eval_loss": 0.04350681230425835, "eval_accurracy": {"accuracy": 0.9921428571428571}, "eval_recall": [0.9926289926289926, 0.9473684210526315, 1.0], "eval_precision": [0.993849938499385, 0.9375, 1.0], "eval_f1": [0.9932390903503381, 0.9424083769633508, 1.0], "eval_runtime": 3.3327, "eval_samples_per_second": 15.003, "eval_steps_per_second": 0.3}
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b416d4cdd435138016707a2a2b3f110f0b0e6890a12420144e800a224ec2ee86
3
+ size 440139580
model_name ADDED
@@ -0,0 +1 @@
 
 
1
+ dbmdz/bert-base-french-europeana-cased
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f194d02d59651b88bd345151b100442d9977b1fcbf2710f83c3ddf77b85ef76
3
+ size 880398906
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9adf48fcdef69f723b8eb32c0a091cfd0124b0e8a46555654bb2e131ab620a35
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d12c31799628bb96494709d75be9daa503a77c472bf96805b97b6510cfee6e19
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,396 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.03481420502066612,
3
+ "best_model_checkpoint": "results_bert-base-french-europeana-cased/epoch20_bs64/checkpoint-48",
4
+ "epoch": 11.0,
5
+ "eval_steps": 500,
6
+ "global_step": 176,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "grad_norm": 0.47372615337371826,
14
+ "learning_rate": 4.75e-05,
15
+ "loss": 0.3883,
16
+ "step": 16
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accurracy": {
21
+ "accuracy": 0.9764285714285714
22
+ },
23
+ "eval_f1": [
24
+ 0.9799147900182593,
25
+ 0.8114285714285714,
26
+ 1.0
27
+ ],
28
+ "eval_loss": 0.08146535605192184,
29
+ "eval_precision": [
30
+ 0.971049457177322,
31
+ 0.8875,
32
+ 1.0
33
+ ],
34
+ "eval_recall": [
35
+ 0.9889434889434889,
36
+ 0.7473684210526316,
37
+ 1.0
38
+ ],
39
+ "eval_runtime": 4.1114,
40
+ "eval_samples_per_second": 12.161,
41
+ "eval_steps_per_second": 0.243,
42
+ "step": 16
43
+ },
44
+ {
45
+ "epoch": 2.0,
46
+ "grad_norm": 0.4207443296909332,
47
+ "learning_rate": 4.5e-05,
48
+ "loss": 0.0566,
49
+ "step": 32
50
+ },
51
+ {
52
+ "epoch": 2.0,
53
+ "eval_accurracy": {
54
+ "accuracy": 0.9878571428571429
55
+ },
56
+ "eval_f1": [
57
+ 0.9895641497851443,
58
+ 0.91005291005291,
59
+ 1.0
60
+ ],
61
+ "eval_loss": 0.04085615277290344,
62
+ "eval_precision": [
63
+ 0.9889570552147239,
64
+ 0.9148936170212766,
65
+ 1.0
66
+ ],
67
+ "eval_recall": [
68
+ 0.9901719901719902,
69
+ 0.9052631578947369,
70
+ 1.0
71
+ ],
72
+ "eval_runtime": 3.9019,
73
+ "eval_samples_per_second": 12.814,
74
+ "eval_steps_per_second": 0.256,
75
+ "step": 32
76
+ },
77
+ {
78
+ "epoch": 3.0,
79
+ "grad_norm": 0.4004434049129486,
80
+ "learning_rate": 4.25e-05,
81
+ "loss": 0.0291,
82
+ "step": 48
83
+ },
84
+ {
85
+ "epoch": 3.0,
86
+ "eval_accurracy": {
87
+ "accuracy": 0.99
88
+ },
89
+ "eval_f1": [
90
+ 0.991389913899139,
91
+ 0.9270833333333334,
92
+ 1.0
93
+ ],
94
+ "eval_loss": 0.03481420502066612,
95
+ "eval_precision": [
96
+ 0.9926108374384236,
97
+ 0.9175257731958762,
98
+ 1.0
99
+ ],
100
+ "eval_recall": [
101
+ 0.9901719901719902,
102
+ 0.9368421052631579,
103
+ 1.0
104
+ ],
105
+ "eval_runtime": 3.5383,
106
+ "eval_samples_per_second": 14.131,
107
+ "eval_steps_per_second": 0.283,
108
+ "step": 48
109
+ },
110
+ {
111
+ "epoch": 4.0,
112
+ "grad_norm": 0.3703693747520447,
113
+ "learning_rate": 4e-05,
114
+ "loss": 0.0169,
115
+ "step": 64
116
+ },
117
+ {
118
+ "epoch": 4.0,
119
+ "eval_accurracy": {
120
+ "accuracy": 0.9885714285714285
121
+ },
122
+ "eval_f1": [
123
+ 0.9901719901719902,
124
+ 0.9157894736842105,
125
+ 1.0
126
+ ],
127
+ "eval_loss": 0.03665424883365631,
128
+ "eval_precision": [
129
+ 0.9901719901719902,
130
+ 0.9157894736842105,
131
+ 1.0
132
+ ],
133
+ "eval_recall": [
134
+ 0.9901719901719902,
135
+ 0.9157894736842105,
136
+ 1.0
137
+ ],
138
+ "eval_runtime": 3.4858,
139
+ "eval_samples_per_second": 14.344,
140
+ "eval_steps_per_second": 0.287,
141
+ "step": 64
142
+ },
143
+ {
144
+ "epoch": 5.0,
145
+ "grad_norm": 0.3292807936668396,
146
+ "learning_rate": 3.7500000000000003e-05,
147
+ "loss": 0.0109,
148
+ "step": 80
149
+ },
150
+ {
151
+ "epoch": 5.0,
152
+ "eval_accurracy": {
153
+ "accuracy": 0.9878571428571429
154
+ },
155
+ "eval_f1": [
156
+ 0.9894867037724181,
157
+ 0.9154228855721394,
158
+ 1.0
159
+ ],
160
+ "eval_loss": 0.03761892020702362,
161
+ "eval_precision": [
162
+ 0.9962640099626401,
163
+ 0.8679245283018868,
164
+ 1.0
165
+ ],
166
+ "eval_recall": [
167
+ 0.9828009828009828,
168
+ 0.968421052631579,
169
+ 1.0
170
+ ],
171
+ "eval_runtime": 3.6656,
172
+ "eval_samples_per_second": 13.64,
173
+ "eval_steps_per_second": 0.273,
174
+ "step": 80
175
+ },
176
+ {
177
+ "epoch": 6.0,
178
+ "grad_norm": 0.17665812373161316,
179
+ "learning_rate": 3.5e-05,
180
+ "loss": 0.0073,
181
+ "step": 96
182
+ },
183
+ {
184
+ "epoch": 6.0,
185
+ "eval_accurracy": {
186
+ "accuracy": 0.9878571428571429
187
+ },
188
+ "eval_f1": [
189
+ 0.9895897121861604,
190
+ 0.9081081081081082,
191
+ 1.0
192
+ ],
193
+ "eval_loss": 0.04045535624027252,
194
+ "eval_precision": [
195
+ 0.9865689865689866,
196
+ 0.9333333333333333,
197
+ 1.0
198
+ ],
199
+ "eval_recall": [
200
+ 0.9926289926289926,
201
+ 0.8842105263157894,
202
+ 1.0
203
+ ],
204
+ "eval_runtime": 3.1662,
205
+ "eval_samples_per_second": 15.792,
206
+ "eval_steps_per_second": 0.316,
207
+ "step": 96
208
+ },
209
+ {
210
+ "epoch": 7.0,
211
+ "grad_norm": 0.46033716201782227,
212
+ "learning_rate": 3.2500000000000004e-05,
213
+ "loss": 0.0043,
214
+ "step": 112
215
+ },
216
+ {
217
+ "epoch": 7.0,
218
+ "eval_accurracy": {
219
+ "accuracy": 0.9885714285714285
220
+ },
221
+ "eval_f1": [
222
+ 0.9901840490797545,
223
+ 0.9148936170212766,
224
+ 1.0
225
+ ],
226
+ "eval_loss": 0.0450633242726326,
227
+ "eval_precision": [
228
+ 0.9889705882352942,
229
+ 0.9247311827956989,
230
+ 1.0
231
+ ],
232
+ "eval_recall": [
233
+ 0.9914004914004914,
234
+ 0.9052631578947369,
235
+ 1.0
236
+ ],
237
+ "eval_runtime": 3.1686,
238
+ "eval_samples_per_second": 15.78,
239
+ "eval_steps_per_second": 0.316,
240
+ "step": 112
241
+ },
242
+ {
243
+ "epoch": 8.0,
244
+ "grad_norm": 0.7053983807563782,
245
+ "learning_rate": 3e-05,
246
+ "loss": 0.0029,
247
+ "step": 128
248
+ },
249
+ {
250
+ "epoch": 8.0,
251
+ "eval_accurracy": {
252
+ "accuracy": 0.9871428571428571
253
+ },
254
+ "eval_f1": [
255
+ 0.9889570552147239,
256
+ 0.9042553191489362,
257
+ 1.0
258
+ ],
259
+ "eval_loss": 0.055573731660842896,
260
+ "eval_precision": [
261
+ 0.9877450980392157,
262
+ 0.9139784946236559,
263
+ 1.0
264
+ ],
265
+ "eval_recall": [
266
+ 0.9901719901719902,
267
+ 0.8947368421052632,
268
+ 1.0
269
+ ],
270
+ "eval_runtime": 3.4353,
271
+ "eval_samples_per_second": 14.555,
272
+ "eval_steps_per_second": 0.291,
273
+ "step": 128
274
+ },
275
+ {
276
+ "epoch": 9.0,
277
+ "grad_norm": 0.16288042068481445,
278
+ "learning_rate": 2.7500000000000004e-05,
279
+ "loss": 0.0023,
280
+ "step": 144
281
+ },
282
+ {
283
+ "epoch": 9.0,
284
+ "eval_accurracy": {
285
+ "accuracy": 0.9892857142857143
286
+ },
287
+ "eval_f1": [
288
+ 0.9908144519289651,
289
+ 0.918918918918919,
290
+ 1.0
291
+ ],
292
+ "eval_loss": 0.05289832502603531,
293
+ "eval_precision": [
294
+ 0.9877899877899878,
295
+ 0.9444444444444444,
296
+ 1.0
297
+ ],
298
+ "eval_recall": [
299
+ 0.9938574938574939,
300
+ 0.8947368421052632,
301
+ 1.0
302
+ ],
303
+ "eval_runtime": 3.8415,
304
+ "eval_samples_per_second": 13.016,
305
+ "eval_steps_per_second": 0.26,
306
+ "step": 144
307
+ },
308
+ {
309
+ "epoch": 10.0,
310
+ "grad_norm": 0.19770997762680054,
311
+ "learning_rate": 2.5e-05,
312
+ "loss": 0.0021,
313
+ "step": 160
314
+ },
315
+ {
316
+ "epoch": 10.0,
317
+ "eval_accurracy": {
318
+ "accuracy": 0.9892857142857143
319
+ },
320
+ "eval_f1": [
321
+ 0.990791896869245,
322
+ 0.9206349206349206,
323
+ 1.0
324
+ ],
325
+ "eval_loss": 0.04967246204614639,
326
+ "eval_precision": [
327
+ 0.9901840490797545,
328
+ 0.925531914893617,
329
+ 1.0
330
+ ],
331
+ "eval_recall": [
332
+ 0.9914004914004914,
333
+ 0.9157894736842105,
334
+ 1.0
335
+ ],
336
+ "eval_runtime": 3.3345,
337
+ "eval_samples_per_second": 14.995,
338
+ "eval_steps_per_second": 0.3,
339
+ "step": 160
340
+ },
341
+ {
342
+ "epoch": 11.0,
343
+ "grad_norm": 0.07273274660110474,
344
+ "learning_rate": 2.25e-05,
345
+ "loss": 0.0013,
346
+ "step": 176
347
+ },
348
+ {
349
+ "epoch": 11.0,
350
+ "eval_accurracy": {
351
+ "accuracy": 0.9921428571428571
352
+ },
353
+ "eval_f1": [
354
+ 0.9932390903503381,
355
+ 0.9424083769633508,
356
+ 1.0
357
+ ],
358
+ "eval_loss": 0.04350681230425835,
359
+ "eval_precision": [
360
+ 0.993849938499385,
361
+ 0.9375,
362
+ 1.0
363
+ ],
364
+ "eval_recall": [
365
+ 0.9926289926289926,
366
+ 0.9473684210526315,
367
+ 1.0
368
+ ],
369
+ "eval_runtime": 3.3327,
370
+ "eval_samples_per_second": 15.003,
371
+ "eval_steps_per_second": 0.3,
372
+ "step": 176
373
+ }
374
+ ],
375
+ "logging_steps": 500,
376
+ "max_steps": 320,
377
+ "num_input_tokens_seen": 0,
378
+ "num_train_epochs": 20,
379
+ "save_steps": 500,
380
+ "stateful_callbacks": {
381
+ "TrainerControl": {
382
+ "args": {
383
+ "should_epoch_stop": false,
384
+ "should_evaluate": false,
385
+ "should_log": false,
386
+ "should_save": true,
387
+ "should_training_stop": false
388
+ },
389
+ "attributes": {}
390
+ }
391
+ },
392
+ "total_flos": 168415447860000.0,
393
+ "train_batch_size": 64,
394
+ "trial_name": null,
395
+ "trial_params": null
396
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:565d807bdec83fad94e5b5f7622f7e6d9e89088170c0eae871ad5962098a37a1
3
+ size 5176