Hugging Face's logo

ProMeText
/
aquilign-multilingual-segmenter

Safetensors
bert
Model card Files
xet
Community
2
mattlev commited on
Commit
83043cd
·
verified ·
1 Parent(s): 3098123

Upload 10 files

Browse files
Files changed (10) hide show
  1. config.json +41 -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 +240 -0
  10. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google-bert/bert-base-multilingual-cased",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "directionality": "bidi",
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
+ "pooler_fc_size": 768,
31
+ "pooler_num_attention_heads": 12,
32
+ "pooler_num_fc_layers": 3,
33
+ "pooler_size_per_head": 128,
34
+ "pooler_type": "first_token_transform",
35
+ "position_embedding_type": "absolute",
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.41.1",
38
+ "type_vocab_size": 2,
39
+ "use_cache": true,
40
+ "vocab_size": 119547
41
+ }
eval.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ | | Synt (None, Delim.) | Bert (None, Delim., Pad.) |
2
+ |-----------+------------------------------------------------+-----------------------------------------------|
3
+ | Accuracy | 0.976800588896207 | 0.9932784225057487 |
4
+ | Precision | [0.9004062759534017, 0.6344567522979024, 1.0] | [0.9817984308992155, 0.8651105984529787, 1.0] |
5
+ | Recall | [0.9627745109804392, 0.3776125683826624, 1.0] | [0.9761430457218289, 0.8942348155421518, 1.0] |
6
+ | F1-score | [0.9305465342859794, 0.47344354555047485, 1.0] | [0.9789625707064629, 0.8794316457442406, 1.0] |
metrics.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"loss": 0.0046, "grad_norm": 0.07699564099311829, "learning_rate": 4.4000000000000006e-05, "epoch": 6.0, "step": 360, "eval_loss": 0.010619055479764938, "eval_accurracy": {"accuracy": 0.9964112146950224}, "eval_recall": [0.9854655117681738, 0.9145922746781115, 1.0], "eval_precision": [0.9909254656968923, 0.8692637160921884, 1.0], "eval_f1": [0.9881879469310263, 0.8913520861654293, 1.0], "eval_runtime": 9.0321, "eval_samples_per_second": 77.612, "eval_steps_per_second": 0.886}
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fba7caab1fc24d54e4945648fb0607ca10634c0084ba0080accc1619b6d58055
3
+ size 709083980
model_name ADDED
@@ -0,0 +1 @@
 
 
1
+ google-bert/bert-base-multilingual-cased
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64fa743c60be9c4c25d0d06018797708fa5c5d8c13c40aa1803ee095622030cd
3
+ size 1418287674
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d92f8588ca4576196e1ff5806fd549535f1e969c714eae025e4c9f25d11bd5bb
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:005e6bda6cd88475018063d38f5944852738698cb0b5363e8ec1c9bf609c6c68
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.009304543025791645,
3
+ "best_model_checkpoint": "results_multilingual_model/epoch50_bs100/checkpoint-240",
4
+ "epoch": 6.0,
5
+ "eval_steps": 500,
6
+ "global_step": 360,
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": 0.0904545858502388,
14
+ "learning_rate": 4.9e-05,
15
+ "loss": 0.0833,
16
+ "step": 60
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accurracy": {
21
+ "accuracy": 0.9942178762266288
22
+ },
23
+ "eval_f1": [
24
+ 0.9811026821999458,
25
+ 0.812751677852349,
26
+ 1.0
27
+ ],
28
+ "eval_loss": 0.015753956511616707,
29
+ "eval_precision": [
30
+ 0.9769120093520974,
31
+ 0.8488317757009346,
32
+ 1.0
33
+ ],
34
+ "eval_recall": [
35
+ 0.9853294635163938,
36
+ 0.7796137339055794,
37
+ 1.0
38
+ ],
39
+ "eval_runtime": 9.1266,
40
+ "eval_samples_per_second": 76.809,
41
+ "eval_steps_per_second": 0.877,
42
+ "step": 60
43
+ },
44
+ {
45
+ "epoch": 2.0,
46
+ "grad_norm": 0.056562066078186035,
47
+ "learning_rate": 4.8e-05,
48
+ "loss": 0.0113,
49
+ "step": 120
50
+ },
51
+ {
52
+ "epoch": 2.0,
53
+ "eval_accurracy": {
54
+ "accuracy": 0.9961314345124399
55
+ },
56
+ "eval_f1": [
57
+ 0.9873171173619604,
58
+ 0.8784722222222222,
59
+ 1.0
60
+ ],
61
+ "eval_loss": 0.010700768791139126,
62
+ "eval_precision": [
63
+ 0.9861557254671312,
64
+ 0.8884986830553117,
65
+ 1.0
66
+ ],
67
+ "eval_recall": [
68
+ 0.988481248015963,
69
+ 0.8686695278969957,
70
+ 1.0
71
+ ],
72
+ "eval_runtime": 9.2093,
73
+ "eval_samples_per_second": 76.118,
74
+ "eval_steps_per_second": 0.869,
75
+ "step": 120
76
+ },
77
+ {
78
+ "epoch": 3.0,
79
+ "grad_norm": 0.046516694128513336,
80
+ "learning_rate": 4.7e-05,
81
+ "loss": 0.0082,
82
+ "step": 180
83
+ },
84
+ {
85
+ "epoch": 3.0,
86
+ "eval_accurracy": {
87
+ "accuracy": 0.9964975666032267
88
+ },
89
+ "eval_f1": [
90
+ 0.9885187617473222,
91
+ 0.8898544427547251,
92
+ 1.0
93
+ ],
94
+ "eval_loss": 0.009478472173213959,
95
+ "eval_precision": [
96
+ 0.987244436403112,
97
+ 0.9010118785745711,
98
+ 1.0
99
+ ],
100
+ "eval_recall": [
101
+ 0.9897963811165027,
102
+ 0.878969957081545,
103
+ 1.0
104
+ ],
105
+ "eval_runtime": 9.3442,
106
+ "eval_samples_per_second": 75.02,
107
+ "eval_steps_per_second": 0.856,
108
+ "step": 180
109
+ },
110
+ {
111
+ "epoch": 4.0,
112
+ "grad_norm": 0.06247434392571449,
113
+ "learning_rate": 4.600000000000001e-05,
114
+ "loss": 0.0066,
115
+ "step": 240
116
+ },
117
+ {
118
+ "epoch": 4.0,
119
+ "eval_accurracy": {
120
+ "accuracy": 0.9965321073665085
121
+ },
122
+ "eval_f1": [
123
+ 0.9886149728981924,
124
+ 0.8924823302634397,
125
+ 1.0
126
+ ],
127
+ "eval_loss": 0.009304543025791645,
128
+ "eval_precision": [
129
+ 0.988816804282733,
130
+ 0.890765284309534,
131
+ 1.0
132
+ ],
133
+ "eval_recall": [
134
+ 0.988413223890073,
135
+ 0.894206008583691,
136
+ 1.0
137
+ ],
138
+ "eval_runtime": 9.051,
139
+ "eval_samples_per_second": 77.45,
140
+ "eval_steps_per_second": 0.884,
141
+ "step": 240
142
+ },
143
+ {
144
+ "epoch": 5.0,
145
+ "grad_norm": 0.06775504350662231,
146
+ "learning_rate": 4.5e-05,
147
+ "loss": 0.0055,
148
+ "step": 300
149
+ },
150
+ {
151
+ "epoch": 5.0,
152
+ "eval_accurracy": {
153
+ "accuracy": 0.9966460918853385
154
+ },
155
+ "eval_f1": [
156
+ 0.9890169552873577,
157
+ 0.8934722984092156,
158
+ 1.0
159
+ ],
160
+ "eval_loss": 0.010034682229161263,
161
+ "eval_precision": [
162
+ 0.9867289593066558,
163
+ 0.9140291806958474,
164
+ 1.0
165
+ ],
166
+ "eval_recall": [
167
+ 0.9913155865947123,
168
+ 0.8738197424892704,
169
+ 1.0
170
+ ],
171
+ "eval_runtime": 9.0462,
172
+ "eval_samples_per_second": 77.491,
173
+ "eval_steps_per_second": 0.884,
174
+ "step": 300
175
+ },
176
+ {
177
+ "epoch": 6.0,
178
+ "grad_norm": 0.07699564099311829,
179
+ "learning_rate": 4.4000000000000006e-05,
180
+ "loss": 0.0046,
181
+ "step": 360
182
+ },
183
+ {
184
+ "epoch": 6.0,
185
+ "eval_accurracy": {
186
+ "accuracy": 0.9964112146950224
187
+ },
188
+ "eval_f1": [
189
+ 0.9881879469310263,
190
+ 0.8913520861654293,
191
+ 1.0
192
+ ],
193
+ "eval_loss": 0.010619055479764938,
194
+ "eval_precision": [
195
+ 0.9909254656968923,
196
+ 0.8692637160921884,
197
+ 1.0
198
+ ],
199
+ "eval_recall": [
200
+ 0.9854655117681738,
201
+ 0.9145922746781115,
202
+ 1.0
203
+ ],
204
+ "eval_runtime": 9.0321,
205
+ "eval_samples_per_second": 77.612,
206
+ "eval_steps_per_second": 0.886,
207
+ "step": 360
208
+ }
209
+ ],
210
+ "logging_steps": 500,
211
+ "max_steps": 3000,
212
+ "num_input_tokens_seen": 0,
213
+ "num_train_epochs": 50,
214
+ "save_steps": 500,
215
+ "stateful_callbacks": {
216
+ "EarlyStoppingCallback": {
217
+ "args": {
218
+ "early_stopping_patience": 10,
219
+ "early_stopping_threshold": 0.0
220
+ },
221
+ "attributes": {
222
+ "early_stopping_patience_counter": 0
223
+ }
224
+ },
225
+ "TrainerControl": {
226
+ "args": {
227
+ "should_epoch_stop": false,
228
+ "should_evaluate": false,
229
+ "should_log": false,
230
+ "should_save": true,
231
+ "should_training_stop": false
232
+ },
233
+ "attributes": {}
234
+ }
235
+ },
236
+ "total_flos": 8431673465625120.0,
237
+ "train_batch_size": 100,
238
+ "trial_name": null,
239
+ "trial_params": null
240
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56bcb5fa40f9f9fdb2343dd4f41005e8be349f6799361b66ccb28613cc60f2d4
3
+ size 5112