batterydata commited on
Commit
a104989
1 Parent(s): 6ed8c15

best model for abstract classification

Browse files
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryscibert-cased/",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "battery",
13
+ "1": "non-battery"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 3072,
17
+ "label2id": {
18
+ "battery": 0,
19
+ "non-battery": 1
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 12,
25
+ "num_hidden_layers": 12,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.17.0.dev0",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 31116
34
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f099df50b7e5c0d6834c510fea622b36a260eb4fa1866ae42066e0f4ea85614
3
+ size 879645905
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:106216c78429778fe24d6884bed0cea215e96c9b4063af0b415b1885e25a335c
3
+ size 439849225
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8db1aa659f6745b245fff5387dfce2dc42fcc110017b4172692c4401271585d
3
+ size 14657
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c1ed30faf525699b02f8be335b1365917ce2e31ed08d9d0dbee2b56d3d4fa96
3
+ size 14657
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c0277c941fc2c4f166b38546f86cafab4b7553ff0aeeedc40c8aaafc61c4530
3
+ size 14657
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c526bb1b261eb44a0653345ed064d1ec658edde2bd9d003463e374a4a6003c5
3
+ size 14657
rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbfc4a57c2e938b43bd5a41221aac22a5ffe7cd4a8a9fe14c7c53ea6c38f883b
3
+ size 14657
rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ab6e0570478854a463ebd591176fd38c5aa753afd7e6b43364dfca009bffbc2
3
+ size 14657
rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65627cea26b0e6a54b8bd945eebacd6d8756749c28c500baffde1408d655bc9d
3
+ size 14657
rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c893083dda2c09a972e6830ee348923abe59691ae3f170d80c9b7cc961dbb095
3
+ size 14657
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb2d6ea316aafd776d25c89bd0c36e3f70f448d4042f9bbb2dee342f9cb4d1aa
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batteryscibert-cased/", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,247 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 11.0,
5
+ "global_step": 11231,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.49,
12
+ "learning_rate": 1.9347045380346067e-05,
13
+ "loss": 0.1574,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.98,
18
+ "learning_rate": 1.8694090760692135e-05,
19
+ "loss": 0.1384,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 1.0,
24
+ "eval_accuracy": 0.9631415605545044,
25
+ "eval_loss": 0.12948766350746155,
26
+ "eval_runtime": 3.6006,
27
+ "eval_samples_per_second": 2592.076,
28
+ "eval_steps_per_second": 81.098,
29
+ "step": 1021
30
+ },
31
+ {
32
+ "epoch": 1.47,
33
+ "learning_rate": 1.80411361410382e-05,
34
+ "loss": 0.1024,
35
+ "step": 1500
36
+ },
37
+ {
38
+ "epoch": 1.96,
39
+ "learning_rate": 1.7388181521384265e-05,
40
+ "loss": 0.109,
41
+ "step": 2000
42
+ },
43
+ {
44
+ "epoch": 2.0,
45
+ "eval_accuracy": 0.9664630889892578,
46
+ "eval_loss": 0.11389727890491486,
47
+ "eval_runtime": 3.5512,
48
+ "eval_samples_per_second": 2628.103,
49
+ "eval_steps_per_second": 82.225,
50
+ "step": 2042
51
+ },
52
+ {
53
+ "epoch": 2.45,
54
+ "learning_rate": 1.673522690173033e-05,
55
+ "loss": 0.0816,
56
+ "step": 2500
57
+ },
58
+ {
59
+ "epoch": 2.94,
60
+ "learning_rate": 1.6082272282076395e-05,
61
+ "loss": 0.0764,
62
+ "step": 3000
63
+ },
64
+ {
65
+ "epoch": 3.0,
66
+ "eval_accuracy": 0.9662488102912903,
67
+ "eval_loss": 0.11298216134309769,
68
+ "eval_runtime": 3.5558,
69
+ "eval_samples_per_second": 2624.692,
70
+ "eval_steps_per_second": 82.118,
71
+ "step": 3063
72
+ },
73
+ {
74
+ "epoch": 3.43,
75
+ "learning_rate": 1.5429317662422463e-05,
76
+ "loss": 0.065,
77
+ "step": 3500
78
+ },
79
+ {
80
+ "epoch": 3.92,
81
+ "learning_rate": 1.4776363042768528e-05,
82
+ "loss": 0.0517,
83
+ "step": 4000
84
+ },
85
+ {
86
+ "epoch": 4.0,
87
+ "eval_accuracy": 0.9630343914031982,
88
+ "eval_loss": 0.1652730107307434,
89
+ "eval_runtime": 3.5703,
90
+ "eval_samples_per_second": 2614.08,
91
+ "eval_steps_per_second": 81.786,
92
+ "step": 4084
93
+ },
94
+ {
95
+ "epoch": 4.41,
96
+ "learning_rate": 1.4123408423114595e-05,
97
+ "loss": 0.0348,
98
+ "step": 4500
99
+ },
100
+ {
101
+ "epoch": 4.9,
102
+ "learning_rate": 1.3470453803460662e-05,
103
+ "loss": 0.0385,
104
+ "step": 5000
105
+ },
106
+ {
107
+ "epoch": 5.0,
108
+ "eval_accuracy": 0.9632486701011658,
109
+ "eval_loss": 0.18385031819343567,
110
+ "eval_runtime": 3.5894,
111
+ "eval_samples_per_second": 2600.184,
112
+ "eval_steps_per_second": 81.352,
113
+ "step": 5105
114
+ },
115
+ {
116
+ "epoch": 5.39,
117
+ "learning_rate": 1.2817499183806728e-05,
118
+ "loss": 0.0244,
119
+ "step": 5500
120
+ },
121
+ {
122
+ "epoch": 5.88,
123
+ "learning_rate": 1.2164544564152792e-05,
124
+ "loss": 0.0245,
125
+ "step": 6000
126
+ },
127
+ {
128
+ "epoch": 6.0,
129
+ "eval_accuracy": 0.9693560600280762,
130
+ "eval_loss": 0.1997281014919281,
131
+ "eval_runtime": 3.7802,
132
+ "eval_samples_per_second": 2468.931,
133
+ "eval_steps_per_second": 77.245,
134
+ "step": 6126
135
+ },
136
+ {
137
+ "epoch": 6.37,
138
+ "learning_rate": 1.1511589944498858e-05,
139
+ "loss": 0.0175,
140
+ "step": 6500
141
+ },
142
+ {
143
+ "epoch": 6.86,
144
+ "learning_rate": 1.0858635324844923e-05,
145
+ "loss": 0.0171,
146
+ "step": 7000
147
+ },
148
+ {
149
+ "epoch": 7.0,
150
+ "eval_accuracy": 0.9687131643295288,
151
+ "eval_loss": 0.1954147070646286,
152
+ "eval_runtime": 3.5659,
153
+ "eval_samples_per_second": 2617.281,
154
+ "eval_steps_per_second": 81.886,
155
+ "step": 7147
156
+ },
157
+ {
158
+ "epoch": 7.35,
159
+ "learning_rate": 1.020568070519099e-05,
160
+ "loss": 0.0127,
161
+ "step": 7500
162
+ },
163
+ {
164
+ "epoch": 7.84,
165
+ "learning_rate": 9.552726085537057e-06,
166
+ "loss": 0.0102,
167
+ "step": 8000
168
+ },
169
+ {
170
+ "epoch": 8.0,
171
+ "eval_accuracy": 0.9687131643295288,
172
+ "eval_loss": 0.23564255237579346,
173
+ "eval_runtime": 3.5772,
174
+ "eval_samples_per_second": 2609.01,
175
+ "eval_steps_per_second": 81.628,
176
+ "step": 8168
177
+ },
178
+ {
179
+ "epoch": 8.33,
180
+ "learning_rate": 8.899771465883122e-06,
181
+ "loss": 0.0092,
182
+ "step": 8500
183
+ },
184
+ {
185
+ "epoch": 8.81,
186
+ "learning_rate": 8.246816846229188e-06,
187
+ "loss": 0.0078,
188
+ "step": 9000
189
+ },
190
+ {
191
+ "epoch": 9.0,
192
+ "eval_accuracy": 0.966570258140564,
193
+ "eval_loss": 0.2585618495941162,
194
+ "eval_runtime": 3.6436,
195
+ "eval_samples_per_second": 2561.473,
196
+ "eval_steps_per_second": 80.14,
197
+ "step": 9189
198
+ },
199
+ {
200
+ "epoch": 9.3,
201
+ "learning_rate": 7.593862226575253e-06,
202
+ "loss": 0.0061,
203
+ "step": 9500
204
+ },
205
+ {
206
+ "epoch": 9.79,
207
+ "learning_rate": 6.94090760692132e-06,
208
+ "loss": 0.0034,
209
+ "step": 10000
210
+ },
211
+ {
212
+ "epoch": 10.0,
213
+ "eval_accuracy": 0.9699989557266235,
214
+ "eval_loss": 0.2755224406719208,
215
+ "eval_runtime": 3.5689,
216
+ "eval_samples_per_second": 2615.084,
217
+ "eval_steps_per_second": 81.818,
218
+ "step": 10210
219
+ },
220
+ {
221
+ "epoch": 10.28,
222
+ "learning_rate": 6.287952987267385e-06,
223
+ "loss": 0.0035,
224
+ "step": 10500
225
+ },
226
+ {
227
+ "epoch": 10.77,
228
+ "learning_rate": 5.634998367613451e-06,
229
+ "loss": 0.0029,
230
+ "step": 11000
231
+ },
232
+ {
233
+ "epoch": 11.0,
234
+ "eval_accuracy": 0.9706417918205261,
235
+ "eval_loss": 0.2912203371524811,
236
+ "eval_runtime": 3.5951,
237
+ "eval_samples_per_second": 2596.026,
238
+ "eval_steps_per_second": 81.221,
239
+ "step": 11231
240
+ }
241
+ ],
242
+ "max_steps": 15315,
243
+ "num_train_epochs": 15,
244
+ "total_flos": 2.363421365580595e+16,
245
+ "trial_name": null,
246
+ "trial_params": null
247
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3ae211534d287b71ae82210a04f784aca9fd4c31dbc0411f73c2d75fb814f1a
3
+ size 3183
vocab.txt ADDED
The diff for this file is too large to render. See raw diff