EhsanAghazadeh
commited on
Commit
•
b23bbeb
1
Parent(s):
d5e0035
Add model and its dependencies
Browse files- all_results.json +16 -0
- config.json +27 -0
- eval_results.json +11 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +195 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.8578431372549019,
|
4 |
+
"eval_combined_score": 0.8790933899332929,
|
5 |
+
"eval_f1": 0.9003436426116839,
|
6 |
+
"eval_loss": 0.48073676228523254,
|
7 |
+
"eval_runtime": 1.1509,
|
8 |
+
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 354.51,
|
10 |
+
"eval_steps_per_second": 44.314,
|
11 |
+
"train_loss": 0.17106628703034443,
|
12 |
+
"train_runtime": 327.4394,
|
13 |
+
"train_samples": 3668,
|
14 |
+
"train_samples_per_second": 112.021,
|
15 |
+
"train_steps_per_second": 3.512
|
16 |
+
}
|
config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "bert-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"finetuning_task": "mrpc",
|
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 |
+
"problem_type": "single_label_classification",
|
22 |
+
"torch_dtype": "float32",
|
23 |
+
"transformers_version": "4.9.2",
|
24 |
+
"type_vocab_size": 2,
|
25 |
+
"use_cache": true,
|
26 |
+
"vocab_size": 30522
|
27 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.8578431372549019,
|
4 |
+
"eval_combined_score": 0.8790933899332929,
|
5 |
+
"eval_f1": 0.9003436426116839,
|
6 |
+
"eval_loss": 0.48073676228523254,
|
7 |
+
"eval_runtime": 1.1509,
|
8 |
+
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 354.51,
|
10 |
+
"eval_steps_per_second": 44.314
|
11 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a6bfeb9dc85b3366879f7c6e3f65b6bd9e714d49094131e06112c8eed58908cf
|
3 |
+
size 438019245
|
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": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"train_loss": 0.17106628703034443,
|
4 |
+
"train_runtime": 327.4394,
|
5 |
+
"train_samples": 3668,
|
6 |
+
"train_samples_per_second": 112.021,
|
7 |
+
"train_steps_per_second": 3.512
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8578431372549019,
|
3 |
+
"best_model_checkpoint": "./bert-base-uncased/fine_tuned_models/checkpoint-460",
|
4 |
+
"epoch": 10.0,
|
5 |
+
"global_step": 1150,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 1.0,
|
12 |
+
"learning_rate": 1e-05,
|
13 |
+
"loss": 0.6308,
|
14 |
+
"step": 115
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.0,
|
18 |
+
"eval_accuracy": 0.7058823529411765,
|
19 |
+
"eval_combined_score": 0.7639204049572351,
|
20 |
+
"eval_f1": 0.8219584569732937,
|
21 |
+
"eval_loss": 0.5484543442726135,
|
22 |
+
"eval_runtime": 1.1176,
|
23 |
+
"eval_samples_per_second": 365.063,
|
24 |
+
"eval_steps_per_second": 45.633,
|
25 |
+
"step": 115
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 2.0,
|
29 |
+
"learning_rate": 2e-05,
|
30 |
+
"loss": 0.488,
|
31 |
+
"step": 230
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 2.0,
|
35 |
+
"eval_accuracy": 0.7990196078431373,
|
36 |
+
"eval_combined_score": 0.8344304388422037,
|
37 |
+
"eval_f1": 0.86984126984127,
|
38 |
+
"eval_loss": 0.44723159074783325,
|
39 |
+
"eval_runtime": 1.121,
|
40 |
+
"eval_samples_per_second": 363.966,
|
41 |
+
"eval_steps_per_second": 45.496,
|
42 |
+
"step": 230
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 3.0,
|
46 |
+
"learning_rate": 1.7500000000000002e-05,
|
47 |
+
"loss": 0.3019,
|
48 |
+
"step": 345
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 3.0,
|
52 |
+
"eval_accuracy": 0.8406862745098039,
|
53 |
+
"eval_combined_score": 0.8653515974917887,
|
54 |
+
"eval_f1": 0.8900169204737733,
|
55 |
+
"eval_loss": 0.3739049434661865,
|
56 |
+
"eval_runtime": 1.1236,
|
57 |
+
"eval_samples_per_second": 363.103,
|
58 |
+
"eval_steps_per_second": 45.388,
|
59 |
+
"step": 345
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 4.0,
|
63 |
+
"learning_rate": 1.5000000000000002e-05,
|
64 |
+
"loss": 0.1395,
|
65 |
+
"step": 460
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 4.0,
|
69 |
+
"eval_accuracy": 0.8578431372549019,
|
70 |
+
"eval_combined_score": 0.8790933899332929,
|
71 |
+
"eval_f1": 0.9003436426116839,
|
72 |
+
"eval_loss": 0.48073676228523254,
|
73 |
+
"eval_runtime": 1.1302,
|
74 |
+
"eval_samples_per_second": 360.994,
|
75 |
+
"eval_steps_per_second": 45.124,
|
76 |
+
"step": 460
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 5.0,
|
80 |
+
"learning_rate": 1.25e-05,
|
81 |
+
"loss": 0.0603,
|
82 |
+
"step": 575
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 5.0,
|
86 |
+
"eval_accuracy": 0.8357843137254902,
|
87 |
+
"eval_combined_score": 0.8594279334770557,
|
88 |
+
"eval_f1": 0.8830715532286212,
|
89 |
+
"eval_loss": 0.6254621148109436,
|
90 |
+
"eval_runtime": 1.1347,
|
91 |
+
"eval_samples_per_second": 359.579,
|
92 |
+
"eval_steps_per_second": 44.947,
|
93 |
+
"step": 575
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 6.0,
|
97 |
+
"learning_rate": 1e-05,
|
98 |
+
"loss": 0.0356,
|
99 |
+
"step": 690
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 6.0,
|
103 |
+
"eval_accuracy": 0.8431372549019608,
|
104 |
+
"eval_combined_score": 0.8676965735789266,
|
105 |
+
"eval_f1": 0.8922558922558923,
|
106 |
+
"eval_loss": 0.8049449324607849,
|
107 |
+
"eval_runtime": 1.1199,
|
108 |
+
"eval_samples_per_second": 364.325,
|
109 |
+
"eval_steps_per_second": 45.541,
|
110 |
+
"step": 690
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 7.0,
|
114 |
+
"learning_rate": 7.500000000000001e-06,
|
115 |
+
"loss": 0.0267,
|
116 |
+
"step": 805
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 7.0,
|
120 |
+
"eval_accuracy": 0.8455882352941176,
|
121 |
+
"eval_combined_score": 0.8668438512172187,
|
122 |
+
"eval_f1": 0.8880994671403197,
|
123 |
+
"eval_loss": 0.7565263509750366,
|
124 |
+
"eval_runtime": 1.1371,
|
125 |
+
"eval_samples_per_second": 358.818,
|
126 |
+
"eval_steps_per_second": 44.852,
|
127 |
+
"step": 805
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 8.0,
|
131 |
+
"learning_rate": 5e-06,
|
132 |
+
"loss": 0.0117,
|
133 |
+
"step": 920
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 8.0,
|
137 |
+
"eval_accuracy": 0.8382352941176471,
|
138 |
+
"eval_combined_score": 0.8622210953346856,
|
139 |
+
"eval_f1": 0.8862068965517241,
|
140 |
+
"eval_loss": 0.8377837538719177,
|
141 |
+
"eval_runtime": 1.1383,
|
142 |
+
"eval_samples_per_second": 358.436,
|
143 |
+
"eval_steps_per_second": 44.805,
|
144 |
+
"step": 920
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 9.0,
|
148 |
+
"learning_rate": 2.5e-06,
|
149 |
+
"loss": 0.0086,
|
150 |
+
"step": 1035
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 9.0,
|
154 |
+
"eval_accuracy": 0.8382352941176471,
|
155 |
+
"eval_combined_score": 0.8614253393665159,
|
156 |
+
"eval_f1": 0.8846153846153846,
|
157 |
+
"eval_loss": 0.870381772518158,
|
158 |
+
"eval_runtime": 1.1281,
|
159 |
+
"eval_samples_per_second": 361.659,
|
160 |
+
"eval_steps_per_second": 45.207,
|
161 |
+
"step": 1035
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 10.0,
|
165 |
+
"learning_rate": 0.0,
|
166 |
+
"loss": 0.0075,
|
167 |
+
"step": 1150
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 10.0,
|
171 |
+
"eval_accuracy": 0.8406862745098039,
|
172 |
+
"eval_combined_score": 0.8640173140313319,
|
173 |
+
"eval_f1": 0.8873483535528597,
|
174 |
+
"eval_loss": 0.8665909171104431,
|
175 |
+
"eval_runtime": 1.1293,
|
176 |
+
"eval_samples_per_second": 361.28,
|
177 |
+
"eval_steps_per_second": 45.16,
|
178 |
+
"step": 1150
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 10.0,
|
182 |
+
"step": 1150,
|
183 |
+
"total_flos": 2412728377651200.0,
|
184 |
+
"train_loss": 0.17106628703034443,
|
185 |
+
"train_runtime": 327.4394,
|
186 |
+
"train_samples_per_second": 112.021,
|
187 |
+
"train_steps_per_second": 3.512
|
188 |
+
}
|
189 |
+
],
|
190 |
+
"max_steps": 1150,
|
191 |
+
"num_train_epochs": 10,
|
192 |
+
"total_flos": 2412728377651200.0,
|
193 |
+
"trial_name": null,
|
194 |
+
"trial_params": null
|
195 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0456efef3913b90541d5850bdca4a25220c2f1241bfb896cf4767ef6570abcf4
|
3 |
+
size 2671
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|