momenalhamza commited on
Commit
e4cd8f2
·
verified ·
1 Parent(s): ba2e0d9

Upload fine-tuned DistilBERT intent classifier (6 intents, test F1 macro 0.94)

Browse files
config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "bos_token_id": null,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_dim": 3072,
13
+ "id2label": {
14
+ "0": "booking",
15
+ "1": "complaint",
16
+ "2": "farewell",
17
+ "3": "greeting",
18
+ "4": "inquiry",
19
+ "5": "other"
20
+ },
21
+ "initializer_range": 0.02,
22
+ "label2id": {
23
+ "booking": 0,
24
+ "complaint": 1,
25
+ "farewell": 2,
26
+ "greeting": 3,
27
+ "inquiry": 4,
28
+ "other": 5
29
+ },
30
+ "max_position_embeddings": 512,
31
+ "model_type": "distilbert",
32
+ "n_heads": 12,
33
+ "n_layers": 6,
34
+ "output_past": true,
35
+ "pad_token_id": 0,
36
+ "problem_type": "single_label_classification",
37
+ "qa_dropout": 0.1,
38
+ "seq_classif_dropout": 0.2,
39
+ "sinusoidal_pos_embds": false,
40
+ "tie_weights_": true,
41
+ "tie_word_embeddings": true,
42
+ "transformers_version": "5.8.0",
43
+ "use_cache": false,
44
+ "vocab_size": 119547
45
+ }
eval_results.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "distilbert-base-multilingual-cased",
3
+ "task": "intent",
4
+ "num_labels": 6,
5
+ "labels": {
6
+ "booking": 0,
7
+ "complaint": 1,
8
+ "farewell": 2,
9
+ "greeting": 3,
10
+ "inquiry": 4,
11
+ "other": 5
12
+ },
13
+ "source_checkpoint": "models/intent_classifier/runs/checkpoint-3336",
14
+ "test_metrics": {
15
+ "test_loss": 0.34578752517700195,
16
+ "test_model_preparation_time": 0.0051,
17
+ "test_accuracy": 0.9244604316546763,
18
+ "test_f1": 0.9245594281209359,
19
+ "test_f1_macro": 0.9371155856970953,
20
+ "test_precision": 0.9249189058091933,
21
+ "test_recall": 0.9244604316546763,
22
+ "test_runtime": 20.6374,
23
+ "test_samples_per_second": 107.766,
24
+ "test_steps_per_second": 6.735,
25
+ "epoch": 0.0
26
+ },
27
+ "classification_report": {
28
+ "booking": {
29
+ "precision": 0.8782961460446247,
30
+ "recall": 0.9173728813559322,
31
+ "f1-score": 0.8974093264248705,
32
+ "support": 472.0
33
+ },
34
+ "complaint": {
35
+ "precision": 0.9978401727861771,
36
+ "recall": 0.9978401727861771,
37
+ "f1-score": 0.9978401727861771,
38
+ "support": 463.0
39
+ },
40
+ "farewell": {
41
+ "precision": 1.0,
42
+ "recall": 0.9743589743589743,
43
+ "f1-score": 0.987012987012987,
44
+ "support": 156.0
45
+ },
46
+ "greeting": {
47
+ "precision": 0.9893048128342246,
48
+ "recall": 0.9685863874345549,
49
+ "f1-score": 0.9788359788359788,
50
+ "support": 191.0
51
+ },
52
+ "inquiry": {
53
+ "precision": 0.8790496760259179,
54
+ "recall": 0.8641188959660298,
55
+ "f1-score": 0.8715203426124197,
56
+ "support": 471.0
57
+ },
58
+ "other": {
59
+ "precision": 0.8948497854077253,
60
+ "recall": 0.8853503184713376,
61
+ "f1-score": 0.8900747065101388,
62
+ "support": 471.0
63
+ },
64
+ "accuracy": 0.9244604316546763,
65
+ "macro avg": {
66
+ "precision": 0.9398900988497783,
67
+ "recall": 0.9346046050621677,
68
+ "f1-score": 0.9371155856970953,
69
+ "support": 2224.0
70
+ },
71
+ "weighted avg": {
72
+ "precision": 0.9249189058091933,
73
+ "recall": 0.9244604316546763,
74
+ "f1-score": 0.9245594281209359,
75
+ "support": 2224.0
76
+ }
77
+ },
78
+ "per_language": {
79
+ "AR": {
80
+ "n": 736,
81
+ "accuracy": 0.8980978260869565,
82
+ "f1_weighted": 0.89834321233115,
83
+ "f1_macro": 0.9153021560624438
84
+ },
85
+ "EN": {
86
+ "n": 751,
87
+ "accuracy": 0.9494007989347537,
88
+ "f1_weighted": 0.9493381733107601,
89
+ "f1_macro": 0.958871593735452
90
+ },
91
+ "FR": {
92
+ "n": 737,
93
+ "accuracy": 0.9253731343283582,
94
+ "f1_weighted": 0.9252881695076642,
95
+ "f1_macro": 0.9364015453854347
96
+ }
97
+ },
98
+ "training": {
99
+ "epochs_completed": 3,
100
+ "epochs_planned": 5,
101
+ "note": "training stopped at end of epoch 3; checkpoint accepted as final (best eval_f1_macro=0.9402 at epoch 3, curve still improving but optimizer state was not saved due to save_only_model=True)."
102
+ }
103
+ }
labels.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label_to_id": {
3
+ "booking": 0,
4
+ "complaint": 1,
5
+ "farewell": 2,
6
+ "greeting": 3,
7
+ "inquiry": 4,
8
+ "other": 5
9
+ },
10
+ "id_to_label": {
11
+ "0": "booking",
12
+ "1": "complaint",
13
+ "2": "farewell",
14
+ "3": "greeting",
15
+ "4": "inquiry",
16
+ "5": "other"
17
+ }
18
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a9d49e45733a5967a95156af811b29fb5fe5c6539ae5ee7e3cfcafbc04952ac
3
+ size 541329680
runs/checkpoint-3336/config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "bos_token_id": null,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_dim": 3072,
13
+ "id2label": {
14
+ "0": "booking",
15
+ "1": "complaint",
16
+ "2": "farewell",
17
+ "3": "greeting",
18
+ "4": "inquiry",
19
+ "5": "other"
20
+ },
21
+ "initializer_range": 0.02,
22
+ "label2id": {
23
+ "booking": 0,
24
+ "complaint": 1,
25
+ "farewell": 2,
26
+ "greeting": 3,
27
+ "inquiry": 4,
28
+ "other": 5
29
+ },
30
+ "max_position_embeddings": 512,
31
+ "model_type": "distilbert",
32
+ "n_heads": 12,
33
+ "n_layers": 6,
34
+ "output_past": true,
35
+ "pad_token_id": 0,
36
+ "problem_type": "single_label_classification",
37
+ "qa_dropout": 0.1,
38
+ "seq_classif_dropout": 0.2,
39
+ "sinusoidal_pos_embds": false,
40
+ "tie_weights_": true,
41
+ "tie_word_embeddings": true,
42
+ "transformers_version": "5.8.0",
43
+ "use_cache": false,
44
+ "vocab_size": 119547
45
+ }
runs/checkpoint-3336/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a9d49e45733a5967a95156af811b29fb5fe5c6539ae5ee7e3cfcafbc04952ac
3
+ size 541329680
runs/checkpoint-3336/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/checkpoint-3336/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "cls_token": "[CLS]",
4
+ "do_lower_case": false,
5
+ "is_local": false,
6
+ "local_files_only": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
runs/checkpoint-3336/trainer_state.json ADDED
@@ -0,0 +1,535 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 3336,
3
+ "best_metric": 0.9402247782964189,
4
+ "best_model_checkpoint": "/home/momen/Desktop/aispire/multilingual-intent-ner-chatbot/models/intent_classifier/runs/checkpoint-3336",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3336,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.044964028776978415,
14
+ "grad_norm": 5.444823265075684,
15
+ "learning_rate": 9.800000000000001e-06,
16
+ "loss": 3.48972900390625,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.08992805755395683,
21
+ "grad_norm": 5.337122917175293,
22
+ "learning_rate": 1.9600000000000002e-05,
23
+ "loss": 2.5280943298339844,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.13489208633093525,
28
+ "grad_norm": 9.257269859313965,
29
+ "learning_rate": 1.9824175824175825e-05,
30
+ "loss": 1.7847193908691406,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.17985611510791366,
35
+ "grad_norm": 10.680974960327148,
36
+ "learning_rate": 1.9644688644688644e-05,
37
+ "loss": 1.4386732482910156,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.22482014388489208,
42
+ "grad_norm": 14.049182891845703,
43
+ "learning_rate": 1.9461538461538462e-05,
44
+ "loss": 1.269312515258789,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.2697841726618705,
49
+ "grad_norm": 9.615478515625,
50
+ "learning_rate": 1.927838827838828e-05,
51
+ "loss": 1.0742133331298829,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.3147482014388489,
56
+ "grad_norm": 13.242940902709961,
57
+ "learning_rate": 1.9095238095238098e-05,
58
+ "loss": 0.8703654479980468,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.3597122302158273,
63
+ "grad_norm": 18.3436279296875,
64
+ "learning_rate": 1.8912087912087913e-05,
65
+ "loss": 0.9773499298095704,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.40467625899280574,
70
+ "grad_norm": 14.933588981628418,
71
+ "learning_rate": 1.872893772893773e-05,
72
+ "loss": 0.7598931121826172,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.44964028776978415,
77
+ "grad_norm": 16.316469192504883,
78
+ "learning_rate": 1.854578754578755e-05,
79
+ "loss": 0.7906450653076171,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.49460431654676257,
84
+ "grad_norm": 27.295316696166992,
85
+ "learning_rate": 1.8362637362637363e-05,
86
+ "loss": 0.7821900177001954,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.539568345323741,
91
+ "grad_norm": 13.005910873413086,
92
+ "learning_rate": 1.817948717948718e-05,
93
+ "loss": 0.7028397369384766,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.5845323741007195,
98
+ "grad_norm": 15.019883155822754,
99
+ "learning_rate": 1.7996336996337e-05,
100
+ "loss": 0.6495774841308594,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.6294964028776978,
105
+ "grad_norm": 17.345195770263672,
106
+ "learning_rate": 1.7813186813186814e-05,
107
+ "loss": 0.6202023696899414,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.6744604316546763,
112
+ "grad_norm": 16.543298721313477,
113
+ "learning_rate": 1.7630036630036632e-05,
114
+ "loss": 0.6614398956298828,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.7194244604316546,
119
+ "grad_norm": 14.030974388122559,
120
+ "learning_rate": 1.7446886446886447e-05,
121
+ "loss": 0.6449069976806641,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.7643884892086331,
126
+ "grad_norm": 10.123924255371094,
127
+ "learning_rate": 1.7263736263736265e-05,
128
+ "loss": 0.6272428131103516,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.8093525179856115,
133
+ "grad_norm": 9.701709747314453,
134
+ "learning_rate": 1.708058608058608e-05,
135
+ "loss": 0.6356664276123047,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.85431654676259,
140
+ "grad_norm": 9.435210227966309,
141
+ "learning_rate": 1.6897435897435898e-05,
142
+ "loss": 0.6323908996582032,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.8992805755395683,
147
+ "grad_norm": 25.058605194091797,
148
+ "learning_rate": 1.6714285714285716e-05,
149
+ "loss": 0.6025145721435546,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.9442446043165468,
154
+ "grad_norm": 6.626457691192627,
155
+ "learning_rate": 1.653113553113553e-05,
156
+ "loss": 0.5902731323242187,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.9892086330935251,
161
+ "grad_norm": 20.808897018432617,
162
+ "learning_rate": 1.634798534798535e-05,
163
+ "loss": 0.6320363998413085,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 1.0,
168
+ "eval_accuracy": 0.9100314889788574,
169
+ "eval_f1": 0.9096123014459289,
170
+ "eval_f1_macro": 0.9233088628544217,
171
+ "eval_loss": 0.2721065282821655,
172
+ "eval_precision": 0.9129923578116932,
173
+ "eval_recall": 0.9100314889788574,
174
+ "eval_runtime": 20.1406,
175
+ "eval_samples_per_second": 110.374,
176
+ "eval_steps_per_second": 6.901,
177
+ "step": 1112
178
+ },
179
+ {
180
+ "epoch": 1.0341726618705036,
181
+ "grad_norm": 8.032069206237793,
182
+ "learning_rate": 1.6164835164835167e-05,
183
+ "loss": 0.45827747344970704,
184
+ "step": 1150
185
+ },
186
+ {
187
+ "epoch": 1.079136690647482,
188
+ "grad_norm": 23.47137451171875,
189
+ "learning_rate": 1.5981684981684985e-05,
190
+ "loss": 0.4937872314453125,
191
+ "step": 1200
192
+ },
193
+ {
194
+ "epoch": 1.1241007194244603,
195
+ "grad_norm": 15.404353141784668,
196
+ "learning_rate": 1.5802197802197804e-05,
197
+ "loss": 0.39016223907470704,
198
+ "step": 1250
199
+ },
200
+ {
201
+ "epoch": 1.169064748201439,
202
+ "grad_norm": 19.353946685791016,
203
+ "learning_rate": 1.5619047619047622e-05,
204
+ "loss": 0.5053915405273437,
205
+ "step": 1300
206
+ },
207
+ {
208
+ "epoch": 1.2140287769784173,
209
+ "grad_norm": 14.485176086425781,
210
+ "learning_rate": 1.5435897435897436e-05,
211
+ "loss": 0.5772517013549805,
212
+ "step": 1350
213
+ },
214
+ {
215
+ "epoch": 1.2589928057553956,
216
+ "grad_norm": 24.316083908081055,
217
+ "learning_rate": 1.5252747252747253e-05,
218
+ "loss": 0.40472305297851563,
219
+ "step": 1400
220
+ },
221
+ {
222
+ "epoch": 1.3039568345323742,
223
+ "grad_norm": 16.71881675720215,
224
+ "learning_rate": 1.5069597069597071e-05,
225
+ "loss": 0.32287361145019533,
226
+ "step": 1450
227
+ },
228
+ {
229
+ "epoch": 1.3489208633093526,
230
+ "grad_norm": 28.510713577270508,
231
+ "learning_rate": 1.4886446886446887e-05,
232
+ "loss": 0.4578011703491211,
233
+ "step": 1500
234
+ },
235
+ {
236
+ "epoch": 1.393884892086331,
237
+ "grad_norm": 2.16450572013855,
238
+ "learning_rate": 1.4703296703296705e-05,
239
+ "loss": 0.4114567947387695,
240
+ "step": 1550
241
+ },
242
+ {
243
+ "epoch": 1.4388489208633093,
244
+ "grad_norm": 2.4565234184265137,
245
+ "learning_rate": 1.452014652014652e-05,
246
+ "loss": 0.4359485626220703,
247
+ "step": 1600
248
+ },
249
+ {
250
+ "epoch": 1.4838129496402876,
251
+ "grad_norm": 8.808877944946289,
252
+ "learning_rate": 1.4336996336996338e-05,
253
+ "loss": 0.3754317092895508,
254
+ "step": 1650
255
+ },
256
+ {
257
+ "epoch": 1.5287769784172662,
258
+ "grad_norm": 28.746694564819336,
259
+ "learning_rate": 1.4153846153846156e-05,
260
+ "loss": 0.5163775634765625,
261
+ "step": 1700
262
+ },
263
+ {
264
+ "epoch": 1.5737410071942446,
265
+ "grad_norm": 15.770833015441895,
266
+ "learning_rate": 1.3970695970695971e-05,
267
+ "loss": 0.3473609161376953,
268
+ "step": 1750
269
+ },
270
+ {
271
+ "epoch": 1.6187050359712232,
272
+ "grad_norm": 1.3721988201141357,
273
+ "learning_rate": 1.3787545787545789e-05,
274
+ "loss": 0.3472472381591797,
275
+ "step": 1800
276
+ },
277
+ {
278
+ "epoch": 1.6636690647482015,
279
+ "grad_norm": 3.2410874366760254,
280
+ "learning_rate": 1.3604395604395605e-05,
281
+ "loss": 0.3802450180053711,
282
+ "step": 1850
283
+ },
284
+ {
285
+ "epoch": 1.70863309352518,
286
+ "grad_norm": 1.0753153562545776,
287
+ "learning_rate": 1.3421245421245422e-05,
288
+ "loss": 0.4780339050292969,
289
+ "step": 1900
290
+ },
291
+ {
292
+ "epoch": 1.7535971223021583,
293
+ "grad_norm": 43.74039077758789,
294
+ "learning_rate": 1.3241758241758242e-05,
295
+ "loss": 0.4354381561279297,
296
+ "step": 1950
297
+ },
298
+ {
299
+ "epoch": 1.7985611510791366,
300
+ "grad_norm": 11.25211238861084,
301
+ "learning_rate": 1.305860805860806e-05,
302
+ "loss": 0.41097541809082033,
303
+ "step": 2000
304
+ },
305
+ {
306
+ "epoch": 1.843525179856115,
307
+ "grad_norm": 8.187655448913574,
308
+ "learning_rate": 1.2875457875457877e-05,
309
+ "loss": 0.48035263061523437,
310
+ "step": 2050
311
+ },
312
+ {
313
+ "epoch": 1.8884892086330936,
314
+ "grad_norm": 27.55592155456543,
315
+ "learning_rate": 1.2692307692307693e-05,
316
+ "loss": 0.4109255981445312,
317
+ "step": 2100
318
+ },
319
+ {
320
+ "epoch": 1.933453237410072,
321
+ "grad_norm": 35.064144134521484,
322
+ "learning_rate": 1.250915750915751e-05,
323
+ "loss": 0.4292829132080078,
324
+ "step": 2150
325
+ },
326
+ {
327
+ "epoch": 1.9784172661870505,
328
+ "grad_norm": 58.29966735839844,
329
+ "learning_rate": 1.2326007326007328e-05,
330
+ "loss": 0.38613071441650393,
331
+ "step": 2200
332
+ },
333
+ {
334
+ "epoch": 2.0,
335
+ "eval_accuracy": 0.9158794421952317,
336
+ "eval_f1": 0.9154080699867051,
337
+ "eval_f1_macro": 0.929822150568706,
338
+ "eval_loss": 0.280459463596344,
339
+ "eval_precision": 0.9161672054023756,
340
+ "eval_recall": 0.9158794421952317,
341
+ "eval_runtime": 19.8966,
342
+ "eval_samples_per_second": 111.728,
343
+ "eval_steps_per_second": 6.986,
344
+ "step": 2224
345
+ },
346
+ {
347
+ "epoch": 2.023381294964029,
348
+ "grad_norm": 10.369275093078613,
349
+ "learning_rate": 1.2142857142857142e-05,
350
+ "loss": 0.37480056762695313,
351
+ "step": 2250
352
+ },
353
+ {
354
+ "epoch": 2.068345323741007,
355
+ "grad_norm": 16.939577102661133,
356
+ "learning_rate": 1.195970695970696e-05,
357
+ "loss": 0.21281005859375,
358
+ "step": 2300
359
+ },
360
+ {
361
+ "epoch": 2.1133093525179856,
362
+ "grad_norm": 18.542627334594727,
363
+ "learning_rate": 1.1776556776556778e-05,
364
+ "loss": 0.2902416801452637,
365
+ "step": 2350
366
+ },
367
+ {
368
+ "epoch": 2.158273381294964,
369
+ "grad_norm": 22.95549774169922,
370
+ "learning_rate": 1.1593406593406595e-05,
371
+ "loss": 0.28962074279785155,
372
+ "step": 2400
373
+ },
374
+ {
375
+ "epoch": 2.2032374100719423,
376
+ "grad_norm": 0.19946469366550446,
377
+ "learning_rate": 1.1410256410256411e-05,
378
+ "loss": 0.32730098724365236,
379
+ "step": 2450
380
+ },
381
+ {
382
+ "epoch": 2.2482014388489207,
383
+ "grad_norm": 37.85325622558594,
384
+ "learning_rate": 1.1227106227106228e-05,
385
+ "loss": 0.22887514114379884,
386
+ "step": 2500
387
+ },
388
+ {
389
+ "epoch": 2.2931654676258995,
390
+ "grad_norm": 0.43131428956985474,
391
+ "learning_rate": 1.1043956043956046e-05,
392
+ "loss": 0.3389986801147461,
393
+ "step": 2550
394
+ },
395
+ {
396
+ "epoch": 2.338129496402878,
397
+ "grad_norm": 32.919307708740234,
398
+ "learning_rate": 1.086080586080586e-05,
399
+ "loss": 0.2869235992431641,
400
+ "step": 2600
401
+ },
402
+ {
403
+ "epoch": 2.383093525179856,
404
+ "grad_norm": 40.49617004394531,
405
+ "learning_rate": 1.0677655677655678e-05,
406
+ "loss": 0.23308815002441408,
407
+ "step": 2650
408
+ },
409
+ {
410
+ "epoch": 2.4280575539568345,
411
+ "grad_norm": 0.08940605819225311,
412
+ "learning_rate": 1.0494505494505496e-05,
413
+ "loss": 0.2910081100463867,
414
+ "step": 2700
415
+ },
416
+ {
417
+ "epoch": 2.473021582733813,
418
+ "grad_norm": 88.92381286621094,
419
+ "learning_rate": 1.0311355311355311e-05,
420
+ "loss": 0.3438394546508789,
421
+ "step": 2750
422
+ },
423
+ {
424
+ "epoch": 2.5179856115107913,
425
+ "grad_norm": 67.60303497314453,
426
+ "learning_rate": 1.012820512820513e-05,
427
+ "loss": 0.31167381286621093,
428
+ "step": 2800
429
+ },
430
+ {
431
+ "epoch": 2.56294964028777,
432
+ "grad_norm": 0.43290701508522034,
433
+ "learning_rate": 9.945054945054946e-06,
434
+ "loss": 0.3191693687438965,
435
+ "step": 2850
436
+ },
437
+ {
438
+ "epoch": 2.6079136690647484,
439
+ "grad_norm": 9.048511505126953,
440
+ "learning_rate": 9.761904761904762e-06,
441
+ "loss": 0.2612957000732422,
442
+ "step": 2900
443
+ },
444
+ {
445
+ "epoch": 2.652877697841727,
446
+ "grad_norm": 65.75765228271484,
447
+ "learning_rate": 9.57875457875458e-06,
448
+ "loss": 0.2293174934387207,
449
+ "step": 2950
450
+ },
451
+ {
452
+ "epoch": 2.697841726618705,
453
+ "grad_norm": 8.569268226623535,
454
+ "learning_rate": 9.395604395604396e-06,
455
+ "loss": 0.34392719268798827,
456
+ "step": 3000
457
+ },
458
+ {
459
+ "epoch": 2.7428057553956835,
460
+ "grad_norm": 6.776681423187256,
461
+ "learning_rate": 9.212454212454213e-06,
462
+ "loss": 0.24704719543457032,
463
+ "step": 3050
464
+ },
465
+ {
466
+ "epoch": 2.787769784172662,
467
+ "grad_norm": 58.88591003417969,
468
+ "learning_rate": 9.02930402930403e-06,
469
+ "loss": 0.26505115509033206,
470
+ "step": 3100
471
+ },
472
+ {
473
+ "epoch": 2.83273381294964,
474
+ "grad_norm": 18.955289840698242,
475
+ "learning_rate": 8.846153846153847e-06,
476
+ "loss": 0.24674781799316406,
477
+ "step": 3150
478
+ },
479
+ {
480
+ "epoch": 2.8776978417266186,
481
+ "grad_norm": 0.0676506906747818,
482
+ "learning_rate": 8.663003663003664e-06,
483
+ "loss": 0.34366771697998044,
484
+ "step": 3200
485
+ },
486
+ {
487
+ "epoch": 2.922661870503597,
488
+ "grad_norm": 19.662029266357422,
489
+ "learning_rate": 8.47985347985348e-06,
490
+ "loss": 0.24356376647949218,
491
+ "step": 3250
492
+ },
493
+ {
494
+ "epoch": 2.9676258992805753,
495
+ "grad_norm": 12.233282089233398,
496
+ "learning_rate": 8.296703296703298e-06,
497
+ "loss": 0.4039815521240234,
498
+ "step": 3300
499
+ },
500
+ {
501
+ "epoch": 3.0,
502
+ "eval_accuracy": 0.9248762932973459,
503
+ "eval_f1": 0.9248291093672378,
504
+ "eval_f1_macro": 0.9402247782964189,
505
+ "eval_loss": 0.314363569021225,
506
+ "eval_precision": 0.9249379287307709,
507
+ "eval_recall": 0.9248762932973459,
508
+ "eval_runtime": 19.833,
509
+ "eval_samples_per_second": 112.086,
510
+ "eval_steps_per_second": 7.009,
511
+ "step": 3336
512
+ }
513
+ ],
514
+ "logging_steps": 50,
515
+ "max_steps": 5560,
516
+ "num_input_tokens_seen": 0,
517
+ "num_train_epochs": 5,
518
+ "save_steps": 500,
519
+ "stateful_callbacks": {
520
+ "TrainerControl": {
521
+ "args": {
522
+ "should_epoch_stop": false,
523
+ "should_evaluate": false,
524
+ "should_log": false,
525
+ "should_save": true,
526
+ "should_training_stop": false
527
+ },
528
+ "attributes": {}
529
+ }
530
+ },
531
+ "total_flos": 342443337632496.0,
532
+ "train_batch_size": 8,
533
+ "trial_name": null,
534
+ "trial_params": null
535
+ }
runs/checkpoint-3336/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e26f59723a05b29e098d0fcd8c3d0c7efa64a2a616fc00fa0c461cc3edb23094
3
+ size 5329
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "cls_token": "[CLS]",
4
+ "do_lower_case": false,
5
+ "is_local": true,
6
+ "local_files_only": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:056725b756dd2cac913db51cce00c3ebdaf207677f62a1e5d14758fe7e974d5f
3
+ size 5329