batterydata
commited on
Commit
•
65b25e3
1
Parent(s):
670833b
best model for abstract classification
Browse files- config.json +35 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- rng_state_4.pth +3 -0
- rng_state_5.pth +3 -0
- rng_state_6.pth +3 -0
- rng_state_7.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +247 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batterybert-uncased/",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
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": "battery",
|
14 |
+
"1": "non-battery"
|
15 |
+
},
|
16 |
+
"initializer_range": 0.02,
|
17 |
+
"intermediate_size": 3072,
|
18 |
+
"label2id": {
|
19 |
+
"battery": 0,
|
20 |
+
"non-battery": 1
|
21 |
+
},
|
22 |
+
"layer_norm_eps": 1e-12,
|
23 |
+
"max_position_embeddings": 512,
|
24 |
+
"model_type": "bert",
|
25 |
+
"num_attention_heads": 12,
|
26 |
+
"num_hidden_layers": 12,
|
27 |
+
"pad_token_id": 0,
|
28 |
+
"position_embedding_type": "absolute",
|
29 |
+
"problem_type": "single_label_classification",
|
30 |
+
"torch_dtype": "float32",
|
31 |
+
"transformers_version": "4.17.0.dev0",
|
32 |
+
"type_vocab_size": 2,
|
33 |
+
"use_cache": true,
|
34 |
+
"vocab_size": 30522
|
35 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2aca9796783520d6350d1cec97e5246b105ab3588d7835c210d45765f7c539bc
|
3 |
+
size 875996369
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:119c9a0e12dd1deaea9e31590c2e29db4c0438f7f4d7f6ab1fd2d25c58411019
|
3 |
+
size 438024457
|
rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3355f2122a347b7f14ae1fcd5d8f69fab8f46a76dc5c4d6adeaa2990aaa5d5d1
|
3 |
+
size 14657
|
rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfa7d5491a433f6b0c12dafb003eff8f28c257b3cb6471615196987da078944d
|
3 |
+
size 14657
|
rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:407760d7fe39c51317e3995d95087ca684bdec0e2f6e70fc26160335facde8e6
|
3 |
+
size 14657
|
rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1e395eca8e501734a010f0dc87fc30055b9a0020313fe65df5c55e733c799d2b
|
3 |
+
size 14657
|
rng_state_4.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa75f4c47025857ecc5d6bf64acda747bb658646c697832b4aff0bd5b93e0127
|
3 |
+
size 14657
|
rng_state_5.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ad42cece19fbad7e06cdddb9ed870c4980c82563a32b8a3947348d1ba3b541b9
|
3 |
+
size 14657
|
rng_state_6.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5950cb81fa19de929205741600c4ee1dd64ce35da909933927443374b0f3a50d
|
3 |
+
size 14657
|
rng_state_7.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:11e68014a36acd07f9270aa1af0cd59983e3128f7b10e748a44b795808a5e1b3
|
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, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/revision/bert/batterybert-uncased/", "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": 0.1561,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.98,
|
18 |
+
"learning_rate": 1.8694090760692135e-05,
|
19 |
+
"loss": 0.1369,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.0,
|
24 |
+
"eval_accuracy": 0.9633558392524719,
|
25 |
+
"eval_loss": 0.11954528838396072,
|
26 |
+
"eval_runtime": 3.6034,
|
27 |
+
"eval_samples_per_second": 2590.036,
|
28 |
+
"eval_steps_per_second": 81.034,
|
29 |
+
"step": 1021
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 1.47,
|
33 |
+
"learning_rate": 1.80411361410382e-05,
|
34 |
+
"loss": 0.0974,
|
35 |
+
"step": 1500
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 1.96,
|
39 |
+
"learning_rate": 1.7388181521384265e-05,
|
40 |
+
"loss": 0.1035,
|
41 |
+
"step": 2000
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 2.0,
|
45 |
+
"eval_accuracy": 0.9667845368385315,
|
46 |
+
"eval_loss": 0.12783248722553253,
|
47 |
+
"eval_runtime": 3.6648,
|
48 |
+
"eval_samples_per_second": 2546.636,
|
49 |
+
"eval_steps_per_second": 79.676,
|
50 |
+
"step": 2042
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.45,
|
54 |
+
"learning_rate": 1.673522690173033e-05,
|
55 |
+
"loss": 0.0744,
|
56 |
+
"step": 2500
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 2.94,
|
60 |
+
"learning_rate": 1.6082272282076395e-05,
|
61 |
+
"loss": 0.0734,
|
62 |
+
"step": 3000
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 3.0,
|
66 |
+
"eval_accuracy": 0.9661416411399841,
|
67 |
+
"eval_loss": 0.11725891381502151,
|
68 |
+
"eval_runtime": 3.8091,
|
69 |
+
"eval_samples_per_second": 2450.154,
|
70 |
+
"eval_steps_per_second": 76.658,
|
71 |
+
"step": 3063
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 3.43,
|
75 |
+
"learning_rate": 1.5429317662422463e-05,
|
76 |
+
"loss": 0.0586,
|
77 |
+
"step": 3500
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 3.92,
|
81 |
+
"learning_rate": 1.4776363042768528e-05,
|
82 |
+
"loss": 0.0448,
|
83 |
+
"step": 4000
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 4.0,
|
87 |
+
"eval_accuracy": 0.9683917164802551,
|
88 |
+
"eval_loss": 0.16057069599628448,
|
89 |
+
"eval_runtime": 3.7958,
|
90 |
+
"eval_samples_per_second": 2458.759,
|
91 |
+
"eval_steps_per_second": 76.927,
|
92 |
+
"step": 4084
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 4.41,
|
96 |
+
"learning_rate": 1.4123408423114595e-05,
|
97 |
+
"loss": 0.031,
|
98 |
+
"step": 4500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 4.9,
|
102 |
+
"learning_rate": 1.3470453803460662e-05,
|
103 |
+
"loss": 0.0321,
|
104 |
+
"step": 5000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 5.0,
|
108 |
+
"eval_accuracy": 0.9644272923469543,
|
109 |
+
"eval_loss": 0.16097016632556915,
|
110 |
+
"eval_runtime": 3.8317,
|
111 |
+
"eval_samples_per_second": 2435.729,
|
112 |
+
"eval_steps_per_second": 76.206,
|
113 |
+
"step": 5105
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 5.39,
|
117 |
+
"learning_rate": 1.2817499183806728e-05,
|
118 |
+
"loss": 0.0183,
|
119 |
+
"step": 5500
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 5.88,
|
123 |
+
"learning_rate": 1.2164544564152792e-05,
|
124 |
+
"loss": 0.0163,
|
125 |
+
"step": 6000
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 6.0,
|
129 |
+
"eval_accuracy": 0.9695703387260437,
|
130 |
+
"eval_loss": 0.21039049327373505,
|
131 |
+
"eval_runtime": 3.8392,
|
132 |
+
"eval_samples_per_second": 2430.966,
|
133 |
+
"eval_steps_per_second": 76.057,
|
134 |
+
"step": 6126
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 6.37,
|
138 |
+
"learning_rate": 1.1511589944498858e-05,
|
139 |
+
"loss": 0.0123,
|
140 |
+
"step": 6500
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 6.86,
|
144 |
+
"learning_rate": 1.0858635324844923e-05,
|
145 |
+
"loss": 0.0107,
|
146 |
+
"step": 7000
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 7.0,
|
150 |
+
"eval_accuracy": 0.9709632396697998,
|
151 |
+
"eval_loss": 0.2211793214082718,
|
152 |
+
"eval_runtime": 3.818,
|
153 |
+
"eval_samples_per_second": 2444.447,
|
154 |
+
"eval_steps_per_second": 76.479,
|
155 |
+
"step": 7147
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 7.35,
|
159 |
+
"learning_rate": 1.020568070519099e-05,
|
160 |
+
"loss": 0.009,
|
161 |
+
"step": 7500
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 7.84,
|
165 |
+
"learning_rate": 9.552726085537057e-06,
|
166 |
+
"loss": 0.005,
|
167 |
+
"step": 8000
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 8.0,
|
171 |
+
"eval_accuracy": 0.9671059846878052,
|
172 |
+
"eval_loss": 0.26634445786476135,
|
173 |
+
"eval_runtime": 3.8057,
|
174 |
+
"eval_samples_per_second": 2452.35,
|
175 |
+
"eval_steps_per_second": 76.726,
|
176 |
+
"step": 8168
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 8.33,
|
180 |
+
"learning_rate": 8.899771465883122e-06,
|
181 |
+
"loss": 0.0047,
|
182 |
+
"step": 8500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 8.81,
|
186 |
+
"learning_rate": 8.246816846229188e-06,
|
187 |
+
"loss": 0.0068,
|
188 |
+
"step": 9000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 9.0,
|
192 |
+
"eval_accuracy": 0.96924889087677,
|
193 |
+
"eval_loss": 0.2791275084018707,
|
194 |
+
"eval_runtime": 3.6479,
|
195 |
+
"eval_samples_per_second": 2558.477,
|
196 |
+
"eval_steps_per_second": 80.047,
|
197 |
+
"step": 9189
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 9.3,
|
201 |
+
"learning_rate": 7.593862226575253e-06,
|
202 |
+
"loss": 0.0041,
|
203 |
+
"step": 9500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 9.79,
|
207 |
+
"learning_rate": 6.94090760692132e-06,
|
208 |
+
"loss": 0.0042,
|
209 |
+
"step": 10000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 10.0,
|
213 |
+
"eval_accuracy": 0.9676417112350464,
|
214 |
+
"eval_loss": 0.2793237268924713,
|
215 |
+
"eval_runtime": 3.6466,
|
216 |
+
"eval_samples_per_second": 2559.346,
|
217 |
+
"eval_steps_per_second": 80.074,
|
218 |
+
"step": 10210
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 10.28,
|
222 |
+
"learning_rate": 6.287952987267385e-06,
|
223 |
+
"loss": 0.0028,
|
224 |
+
"step": 10500
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 10.77,
|
228 |
+
"learning_rate": 5.634998367613451e-06,
|
229 |
+
"loss": 0.0035,
|
230 |
+
"step": 11000
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 11.0,
|
234 |
+
"eval_accuracy": 0.9691417813301086,
|
235 |
+
"eval_loss": 0.28977808356285095,
|
236 |
+
"eval_runtime": 3.6274,
|
237 |
+
"eval_samples_per_second": 2572.927,
|
238 |
+
"eval_steps_per_second": 80.499,
|
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 sha256:71d575a046857c89a3e3ca140d1f7b5afd5e77d83311623b61c92ccff7b2913b
|
3 |
+
size 3183
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|