batterydata commited on
Commit
29434d8
1 Parent(s): 2491e21

best model for qa

Browse files
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batterybert-cased/",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
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
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.17.0.dev0",
23
+ "type_vocab_size": 2,
24
+ "use_cache": true,
25
+ "vocab_size": 28996
26
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6167d07179e12ee2f4c8fd825c6a7aac973f726329edca099ef07191cc0eb86
3
+ size 861894637
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1619a2cdf8533839b44ac017ead99e582e3b856ef8c9c6c23db4c20c6ba8ac7a
3
+ size 430973431
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff2d5ce636730f70bfe373ee1c3e846fd9a36e0e67b31556217369cc407dcae7
3
+ size 14657
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef3fc4a2d23f9ac92d4a26df5894d9c0551024856ac05f49b00c818c52982410
3
+ size 14657
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a25568c7ffa5fe9c9556ab19eadf561ce6b6e25806f863cfe0f47e8b7dfdf22c
3
+ size 14657
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e16be472446729ed097e98c56b1e4411d0b79f7c7aa63a0d9fa909fcac66dbc5
3
+ size 14657
rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7600ccdea87c2c34bc50cbfc117e1b87028fe99b24ea36f17018bbca0dcf98ad
3
+ size 14657
rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:504f99aa72a899d8fdbdd799e5fa64b624c9b0b85f2c3a28615f4d150c7f3e16
3
+ size 14657
rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c6309c06db648f512a8f8edda137fd2df67c22191e921d5d2cf60298d17f6e6
3
+ size 14657
rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f813994bf1fbba78c215010bb2cd8220372cfabaff81b08ca1ef417457bd0576
3
+ size 14657
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bd6a744238c609c30e3be91276089eebd5ddcaa94a27dce9e772fe8020be25c
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": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batterybert-cased/", "tokenizer_class": "BertTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,304 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "global_step": 22184,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.09,
12
+ "learning_rate": 1.9549224666426253e-05,
13
+ "loss": 2.2682,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.18,
18
+ "learning_rate": 1.909844933285251e-05,
19
+ "loss": 1.4753,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.27,
24
+ "learning_rate": 1.864767399927876e-05,
25
+ "loss": 1.3606,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.36,
30
+ "learning_rate": 1.8196898665705013e-05,
31
+ "loss": 1.2586,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.45,
36
+ "learning_rate": 1.7746123332131267e-05,
37
+ "loss": 1.2084,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.54,
42
+ "learning_rate": 1.729534799855752e-05,
43
+ "loss": 1.1679,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.63,
48
+ "learning_rate": 1.6844572664983772e-05,
49
+ "loss": 1.138,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.72,
54
+ "learning_rate": 1.6393797331410027e-05,
55
+ "loss": 1.136,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.81,
60
+ "learning_rate": 1.5943021997836278e-05,
61
+ "loss": 1.1001,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.9,
66
+ "learning_rate": 1.5492246664262532e-05,
67
+ "loss": 1.0728,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.99,
72
+ "learning_rate": 1.5041471330688787e-05,
73
+ "loss": 1.0669,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 1.0,
78
+ "eval_exact_match": 79.96215704824976,
79
+ "eval_f1": 87.68130943560381,
80
+ "step": 5546
81
+ },
82
+ {
83
+ "epoch": 1.08,
84
+ "learning_rate": 1.4590695997115038e-05,
85
+ "loss": 0.8342,
86
+ "step": 6000
87
+ },
88
+ {
89
+ "epoch": 1.17,
90
+ "learning_rate": 1.413992066354129e-05,
91
+ "loss": 0.8289,
92
+ "step": 6500
93
+ },
94
+ {
95
+ "epoch": 1.26,
96
+ "learning_rate": 1.3689145329967547e-05,
97
+ "loss": 0.8292,
98
+ "step": 7000
99
+ },
100
+ {
101
+ "epoch": 1.35,
102
+ "learning_rate": 1.3238369996393798e-05,
103
+ "loss": 0.8103,
104
+ "step": 7500
105
+ },
106
+ {
107
+ "epoch": 1.44,
108
+ "learning_rate": 1.278759466282005e-05,
109
+ "loss": 0.8194,
110
+ "step": 8000
111
+ },
112
+ {
113
+ "epoch": 1.53,
114
+ "learning_rate": 1.2336819329246306e-05,
115
+ "loss": 0.8103,
116
+ "step": 8500
117
+ },
118
+ {
119
+ "epoch": 1.62,
120
+ "learning_rate": 1.1886043995672558e-05,
121
+ "loss": 0.8075,
122
+ "step": 9000
123
+ },
124
+ {
125
+ "epoch": 1.71,
126
+ "learning_rate": 1.143526866209881e-05,
127
+ "loss": 0.7804,
128
+ "step": 9500
129
+ },
130
+ {
131
+ "epoch": 1.8,
132
+ "learning_rate": 1.0984493328525063e-05,
133
+ "loss": 0.8155,
134
+ "step": 10000
135
+ },
136
+ {
137
+ "epoch": 1.89,
138
+ "learning_rate": 1.0533717994951317e-05,
139
+ "loss": 0.795,
140
+ "step": 10500
141
+ },
142
+ {
143
+ "epoch": 1.98,
144
+ "learning_rate": 1.008294266137757e-05,
145
+ "loss": 0.7755,
146
+ "step": 11000
147
+ },
148
+ {
149
+ "epoch": 2.0,
150
+ "eval_exact_match": 81.84484389782403,
151
+ "eval_f1": 89.0471703841863,
152
+ "step": 11092
153
+ },
154
+ {
155
+ "epoch": 2.07,
156
+ "learning_rate": 9.632167327803823e-06,
157
+ "loss": 0.6198,
158
+ "step": 11500
159
+ },
160
+ {
161
+ "epoch": 2.16,
162
+ "learning_rate": 9.181391994230077e-06,
163
+ "loss": 0.6012,
164
+ "step": 12000
165
+ },
166
+ {
167
+ "epoch": 2.25,
168
+ "learning_rate": 8.73061666065633e-06,
169
+ "loss": 0.5688,
170
+ "step": 12500
171
+ },
172
+ {
173
+ "epoch": 2.34,
174
+ "learning_rate": 8.279841327082583e-06,
175
+ "loss": 0.574,
176
+ "step": 13000
177
+ },
178
+ {
179
+ "epoch": 2.43,
180
+ "learning_rate": 7.829065993508835e-06,
181
+ "loss": 0.5873,
182
+ "step": 13500
183
+ },
184
+ {
185
+ "epoch": 2.52,
186
+ "learning_rate": 7.378290659935089e-06,
187
+ "loss": 0.5984,
188
+ "step": 14000
189
+ },
190
+ {
191
+ "epoch": 2.61,
192
+ "learning_rate": 6.927515326361343e-06,
193
+ "loss": 0.5877,
194
+ "step": 14500
195
+ },
196
+ {
197
+ "epoch": 2.7,
198
+ "learning_rate": 6.476739992787595e-06,
199
+ "loss": 0.5892,
200
+ "step": 15000
201
+ },
202
+ {
203
+ "epoch": 2.79,
204
+ "learning_rate": 6.025964659213849e-06,
205
+ "loss": 0.5722,
206
+ "step": 15500
207
+ },
208
+ {
209
+ "epoch": 2.88,
210
+ "learning_rate": 5.575189325640102e-06,
211
+ "loss": 0.5848,
212
+ "step": 16000
213
+ },
214
+ {
215
+ "epoch": 2.98,
216
+ "learning_rate": 5.124413992066355e-06,
217
+ "loss": 0.568,
218
+ "step": 16500
219
+ },
220
+ {
221
+ "epoch": 3.0,
222
+ "eval_exact_match": 81.54210028382214,
223
+ "eval_f1": 88.9922400786148,
224
+ "step": 16638
225
+ },
226
+ {
227
+ "epoch": 3.07,
228
+ "learning_rate": 4.673638658492607e-06,
229
+ "loss": 0.4788,
230
+ "step": 17000
231
+ },
232
+ {
233
+ "epoch": 3.16,
234
+ "learning_rate": 4.222863324918861e-06,
235
+ "loss": 0.4398,
236
+ "step": 17500
237
+ },
238
+ {
239
+ "epoch": 3.25,
240
+ "learning_rate": 3.7720879913451143e-06,
241
+ "loss": 0.4524,
242
+ "step": 18000
243
+ },
244
+ {
245
+ "epoch": 3.34,
246
+ "learning_rate": 3.3213126577713674e-06,
247
+ "loss": 0.4265,
248
+ "step": 18500
249
+ },
250
+ {
251
+ "epoch": 3.43,
252
+ "learning_rate": 2.87053732419762e-06,
253
+ "loss": 0.4424,
254
+ "step": 19000
255
+ },
256
+ {
257
+ "epoch": 3.52,
258
+ "learning_rate": 2.4197619906238733e-06,
259
+ "loss": 0.4456,
260
+ "step": 19500
261
+ },
262
+ {
263
+ "epoch": 3.61,
264
+ "learning_rate": 1.9689866570501265e-06,
265
+ "loss": 0.4379,
266
+ "step": 20000
267
+ },
268
+ {
269
+ "epoch": 3.7,
270
+ "learning_rate": 1.5182113234763796e-06,
271
+ "loss": 0.4411,
272
+ "step": 20500
273
+ },
274
+ {
275
+ "epoch": 3.79,
276
+ "learning_rate": 1.0674359899026326e-06,
277
+ "loss": 0.4412,
278
+ "step": 21000
279
+ },
280
+ {
281
+ "epoch": 3.88,
282
+ "learning_rate": 6.166606563288857e-07,
283
+ "loss": 0.4424,
284
+ "step": 21500
285
+ },
286
+ {
287
+ "epoch": 3.97,
288
+ "learning_rate": 1.6588532275513888e-07,
289
+ "loss": 0.4241,
290
+ "step": 22000
291
+ },
292
+ {
293
+ "epoch": 4.0,
294
+ "eval_exact_match": 81.54210028382214,
295
+ "eval_f1": 89.15597314634459,
296
+ "step": 22184
297
+ }
298
+ ],
299
+ "max_steps": 22184,
300
+ "num_train_epochs": 4,
301
+ "total_flos": 6.955928551581286e+16,
302
+ "trial_name": null,
303
+ "trial_params": null
304
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a63234e1f0e4224cb9feef740c3db5ea28f6b24e417fcd54ef8cf1d66762f4c
3
+ size 3183
vocab.txt ADDED
The diff for this file is too large to render. See raw diff