End of training
Browse files- README.md +4 -2
- all_results.json +16 -0
- eval_results.json +11 -0
- logs/events.out.tfevents.1685337510.serv-9215.1440072.5 +3 -0
- train_results.json +8 -0
- trainer_state.json +348 -0
README.md
CHANGED
@@ -1,4 +1,6 @@
|
|
1 |
---
|
|
|
|
|
2 |
tags:
|
3 |
- generated_from_trainer
|
4 |
datasets:
|
@@ -13,7 +15,7 @@ model-index:
|
|
13 |
name: Text Classification
|
14 |
type: text-classification
|
15 |
dataset:
|
16 |
-
name:
|
17 |
type: glue
|
18 |
config: mrpc
|
19 |
split: validation
|
@@ -32,7 +34,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
# add_BERT_no_pretrain_mrpc
|
34 |
|
35 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on the
|
36 |
It achieves the following results on the evaluation set:
|
37 |
- Loss: 0.6240
|
38 |
- Accuracy: 0.6838
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
|
|
15 |
name: Text Classification
|
16 |
type: text-classification
|
17 |
dataset:
|
18 |
+
name: GLUE MRPC
|
19 |
type: glue
|
20 |
config: mrpc
|
21 |
split: validation
|
|
|
34 |
|
35 |
# add_BERT_no_pretrain_mrpc
|
36 |
|
37 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE MRPC dataset.
|
38 |
It achieves the following results on the evaluation set:
|
39 |
- Loss: 0.6240
|
40 |
- Accuracy: 0.6838
|
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"eval_accuracy": 0.6838235294117647,
|
4 |
+
"eval_combined_score": 0.7480253018237863,
|
5 |
+
"eval_f1": 0.8122270742358079,
|
6 |
+
"eval_loss": 0.6239899396896362,
|
7 |
+
"eval_runtime": 0.6442,
|
8 |
+
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 633.308,
|
10 |
+
"eval_steps_per_second": 6.209,
|
11 |
+
"train_loss": 0.6928667667340453,
|
12 |
+
"train_runtime": 412.0299,
|
13 |
+
"train_samples": 3668,
|
14 |
+
"train_samples_per_second": 445.113,
|
15 |
+
"train_steps_per_second": 3.519
|
16 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"eval_accuracy": 0.6838235294117647,
|
4 |
+
"eval_combined_score": 0.7480253018237863,
|
5 |
+
"eval_f1": 0.8122270742358079,
|
6 |
+
"eval_loss": 0.6239899396896362,
|
7 |
+
"eval_runtime": 0.6442,
|
8 |
+
"eval_samples": 408,
|
9 |
+
"eval_samples_per_second": 633.308,
|
10 |
+
"eval_steps_per_second": 6.209
|
11 |
+
}
|
logs/events.out.tfevents.1685337510.serv-9215.1440072.5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fd5919689c282f3f84b6aab5c2e254324bb6bc05ec7ad970a2c4acf820ca184a
|
3 |
+
size 467
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"train_loss": 0.6928667667340453,
|
4 |
+
"train_runtime": 412.0299,
|
5 |
+
"train_samples": 3668,
|
6 |
+
"train_samples_per_second": 445.113,
|
7 |
+
"train_steps_per_second": 3.519
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,348 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.6239899396896362,
|
3 |
+
"best_model_checkpoint": "add_BERT_no_pretrain_mrpc/checkpoint-406",
|
4 |
+
"epoch": 19.0,
|
5 |
+
"global_step": 551,
|
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": 0.0004910344827586207,
|
13 |
+
"loss": 1.154,
|
14 |
+
"step": 29
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.0,
|
18 |
+
"eval_accuracy": 0.6838235294117647,
|
19 |
+
"eval_combined_score": 0.7480253018237863,
|
20 |
+
"eval_f1": 0.8122270742358079,
|
21 |
+
"eval_loss": 0.6856198906898499,
|
22 |
+
"eval_runtime": 0.6445,
|
23 |
+
"eval_samples_per_second": 633.061,
|
24 |
+
"eval_steps_per_second": 6.206,
|
25 |
+
"step": 29
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 2.0,
|
29 |
+
"learning_rate": 0.0004810344827586207,
|
30 |
+
"loss": 0.6781,
|
31 |
+
"step": 58
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 2.0,
|
35 |
+
"eval_accuracy": 0.6838235294117647,
|
36 |
+
"eval_combined_score": 0.7480253018237863,
|
37 |
+
"eval_f1": 0.8122270742358079,
|
38 |
+
"eval_loss": 0.6609018445014954,
|
39 |
+
"eval_runtime": 0.634,
|
40 |
+
"eval_samples_per_second": 643.583,
|
41 |
+
"eval_steps_per_second": 6.31,
|
42 |
+
"step": 58
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 3.0,
|
46 |
+
"learning_rate": 0.00047103448275862073,
|
47 |
+
"loss": 0.6458,
|
48 |
+
"step": 87
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 3.0,
|
52 |
+
"eval_accuracy": 0.6838235294117647,
|
53 |
+
"eval_combined_score": 0.7480253018237863,
|
54 |
+
"eval_f1": 0.8122270742358079,
|
55 |
+
"eval_loss": 0.634788990020752,
|
56 |
+
"eval_runtime": 0.6427,
|
57 |
+
"eval_samples_per_second": 634.811,
|
58 |
+
"eval_steps_per_second": 6.224,
|
59 |
+
"step": 87
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 4.0,
|
63 |
+
"learning_rate": 0.0004610344827586207,
|
64 |
+
"loss": 0.6395,
|
65 |
+
"step": 116
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 4.0,
|
69 |
+
"eval_accuracy": 0.31862745098039214,
|
70 |
+
"eval_combined_score": 0.16288515406162463,
|
71 |
+
"eval_f1": 0.0071428571428571435,
|
72 |
+
"eval_loss": 19.635393142700195,
|
73 |
+
"eval_runtime": 0.6439,
|
74 |
+
"eval_samples_per_second": 633.614,
|
75 |
+
"eval_steps_per_second": 6.212,
|
76 |
+
"step": 116
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 5.0,
|
80 |
+
"learning_rate": 0.0004510344827586207,
|
81 |
+
"loss": 1.1486,
|
82 |
+
"step": 145
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 5.0,
|
86 |
+
"eval_accuracy": 0.6838235294117647,
|
87 |
+
"eval_combined_score": 0.7480253018237863,
|
88 |
+
"eval_f1": 0.8122270742358079,
|
89 |
+
"eval_loss": 0.6657487750053406,
|
90 |
+
"eval_runtime": 0.6436,
|
91 |
+
"eval_samples_per_second": 633.968,
|
92 |
+
"eval_steps_per_second": 6.215,
|
93 |
+
"step": 145
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 6.0,
|
97 |
+
"learning_rate": 0.00044103448275862066,
|
98 |
+
"loss": 0.6446,
|
99 |
+
"step": 174
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 6.0,
|
103 |
+
"eval_accuracy": 0.6838235294117647,
|
104 |
+
"eval_combined_score": 0.7480253018237863,
|
105 |
+
"eval_f1": 0.8122270742358079,
|
106 |
+
"eval_loss": 0.627680778503418,
|
107 |
+
"eval_runtime": 0.6427,
|
108 |
+
"eval_samples_per_second": 634.78,
|
109 |
+
"eval_steps_per_second": 6.223,
|
110 |
+
"step": 174
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 7.0,
|
114 |
+
"learning_rate": 0.0004310344827586207,
|
115 |
+
"loss": 0.644,
|
116 |
+
"step": 203
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 7.0,
|
120 |
+
"eval_accuracy": 0.6838235294117647,
|
121 |
+
"eval_combined_score": 0.7480253018237863,
|
122 |
+
"eval_f1": 0.8122270742358079,
|
123 |
+
"eval_loss": 0.6241939663887024,
|
124 |
+
"eval_runtime": 0.6436,
|
125 |
+
"eval_samples_per_second": 633.911,
|
126 |
+
"eval_steps_per_second": 6.215,
|
127 |
+
"step": 203
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 8.0,
|
131 |
+
"learning_rate": 0.0004210344827586207,
|
132 |
+
"loss": 0.6337,
|
133 |
+
"step": 232
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 8.0,
|
137 |
+
"eval_accuracy": 0.6838235294117647,
|
138 |
+
"eval_combined_score": 0.7480253018237863,
|
139 |
+
"eval_f1": 0.8122270742358079,
|
140 |
+
"eval_loss": 0.6242448687553406,
|
141 |
+
"eval_runtime": 0.6458,
|
142 |
+
"eval_samples_per_second": 631.822,
|
143 |
+
"eval_steps_per_second": 6.194,
|
144 |
+
"step": 232
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 9.0,
|
148 |
+
"learning_rate": 0.0004110344827586207,
|
149 |
+
"loss": 0.6388,
|
150 |
+
"step": 261
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 9.0,
|
154 |
+
"eval_accuracy": 0.6838235294117647,
|
155 |
+
"eval_combined_score": 0.7480253018237863,
|
156 |
+
"eval_f1": 0.8122270742358079,
|
157 |
+
"eval_loss": 0.6252579092979431,
|
158 |
+
"eval_runtime": 0.6457,
|
159 |
+
"eval_samples_per_second": 631.91,
|
160 |
+
"eval_steps_per_second": 6.195,
|
161 |
+
"step": 261
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 10.0,
|
165 |
+
"learning_rate": 0.0004010344827586207,
|
166 |
+
"loss": 0.634,
|
167 |
+
"step": 290
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 10.0,
|
171 |
+
"eval_accuracy": 0.6838235294117647,
|
172 |
+
"eval_combined_score": 0.7480253018237863,
|
173 |
+
"eval_f1": 0.8122270742358079,
|
174 |
+
"eval_loss": 0.6241927742958069,
|
175 |
+
"eval_runtime": 0.6462,
|
176 |
+
"eval_samples_per_second": 631.361,
|
177 |
+
"eval_steps_per_second": 6.19,
|
178 |
+
"step": 290
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 11.0,
|
182 |
+
"learning_rate": 0.0003910344827586207,
|
183 |
+
"loss": 0.6346,
|
184 |
+
"step": 319
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 11.0,
|
188 |
+
"eval_accuracy": 0.6838235294117647,
|
189 |
+
"eval_combined_score": 0.7480253018237863,
|
190 |
+
"eval_f1": 0.8122270742358079,
|
191 |
+
"eval_loss": 0.6264073848724365,
|
192 |
+
"eval_runtime": 0.6465,
|
193 |
+
"eval_samples_per_second": 631.046,
|
194 |
+
"eval_steps_per_second": 6.187,
|
195 |
+
"step": 319
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 12.0,
|
199 |
+
"learning_rate": 0.00038103448275862066,
|
200 |
+
"loss": 0.6338,
|
201 |
+
"step": 348
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 12.0,
|
205 |
+
"eval_accuracy": 0.6838235294117647,
|
206 |
+
"eval_combined_score": 0.7480253018237863,
|
207 |
+
"eval_f1": 0.8122270742358079,
|
208 |
+
"eval_loss": 0.6273019909858704,
|
209 |
+
"eval_runtime": 0.6464,
|
210 |
+
"eval_samples_per_second": 631.183,
|
211 |
+
"eval_steps_per_second": 6.188,
|
212 |
+
"step": 348
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 13.0,
|
216 |
+
"learning_rate": 0.0003710344827586207,
|
217 |
+
"loss": 0.6343,
|
218 |
+
"step": 377
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 13.0,
|
222 |
+
"eval_accuracy": 0.6838235294117647,
|
223 |
+
"eval_combined_score": 0.7480253018237863,
|
224 |
+
"eval_f1": 0.8122270742358079,
|
225 |
+
"eval_loss": 0.6262201070785522,
|
226 |
+
"eval_runtime": 0.6462,
|
227 |
+
"eval_samples_per_second": 631.42,
|
228 |
+
"eval_steps_per_second": 6.19,
|
229 |
+
"step": 377
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 14.0,
|
233 |
+
"learning_rate": 0.0003610344827586207,
|
234 |
+
"loss": 0.6339,
|
235 |
+
"step": 406
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 14.0,
|
239 |
+
"eval_accuracy": 0.6838235294117647,
|
240 |
+
"eval_combined_score": 0.7480253018237863,
|
241 |
+
"eval_f1": 0.8122270742358079,
|
242 |
+
"eval_loss": 0.6239899396896362,
|
243 |
+
"eval_runtime": 0.6481,
|
244 |
+
"eval_samples_per_second": 629.552,
|
245 |
+
"eval_steps_per_second": 6.172,
|
246 |
+
"step": 406
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 15.0,
|
250 |
+
"learning_rate": 0.0003510344827586207,
|
251 |
+
"loss": 0.635,
|
252 |
+
"step": 435
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 15.0,
|
256 |
+
"eval_accuracy": 0.6838235294117647,
|
257 |
+
"eval_combined_score": 0.7480253018237863,
|
258 |
+
"eval_f1": 0.8122270742358079,
|
259 |
+
"eval_loss": 0.6244243383407593,
|
260 |
+
"eval_runtime": 0.6472,
|
261 |
+
"eval_samples_per_second": 630.443,
|
262 |
+
"eval_steps_per_second": 6.181,
|
263 |
+
"step": 435
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 16.0,
|
267 |
+
"learning_rate": 0.0003410344827586207,
|
268 |
+
"loss": 0.6331,
|
269 |
+
"step": 464
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 16.0,
|
273 |
+
"eval_accuracy": 0.6838235294117647,
|
274 |
+
"eval_combined_score": 0.7480253018237863,
|
275 |
+
"eval_f1": 0.8122270742358079,
|
276 |
+
"eval_loss": 0.6240312457084656,
|
277 |
+
"eval_runtime": 0.6454,
|
278 |
+
"eval_samples_per_second": 632.149,
|
279 |
+
"eval_steps_per_second": 6.198,
|
280 |
+
"step": 464
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 17.0,
|
284 |
+
"learning_rate": 0.0003310344827586207,
|
285 |
+
"loss": 0.6328,
|
286 |
+
"step": 493
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 17.0,
|
290 |
+
"eval_accuracy": 0.6838235294117647,
|
291 |
+
"eval_combined_score": 0.7480253018237863,
|
292 |
+
"eval_f1": 0.8122270742358079,
|
293 |
+
"eval_loss": 0.6266641020774841,
|
294 |
+
"eval_runtime": 0.646,
|
295 |
+
"eval_samples_per_second": 631.583,
|
296 |
+
"eval_steps_per_second": 6.192,
|
297 |
+
"step": 493
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 18.0,
|
301 |
+
"learning_rate": 0.00032103448275862067,
|
302 |
+
"loss": 0.6338,
|
303 |
+
"step": 522
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 18.0,
|
307 |
+
"eval_accuracy": 0.6838235294117647,
|
308 |
+
"eval_combined_score": 0.7480253018237863,
|
309 |
+
"eval_f1": 0.8122270742358079,
|
310 |
+
"eval_loss": 0.6256989240646362,
|
311 |
+
"eval_runtime": 0.6466,
|
312 |
+
"eval_samples_per_second": 630.959,
|
313 |
+
"eval_steps_per_second": 6.186,
|
314 |
+
"step": 522
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 19.0,
|
318 |
+
"learning_rate": 0.0003110344827586207,
|
319 |
+
"loss": 0.6321,
|
320 |
+
"step": 551
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 19.0,
|
324 |
+
"eval_accuracy": 0.6838235294117647,
|
325 |
+
"eval_combined_score": 0.7480253018237863,
|
326 |
+
"eval_f1": 0.8122270742358079,
|
327 |
+
"eval_loss": 0.623990535736084,
|
328 |
+
"eval_runtime": 0.6488,
|
329 |
+
"eval_samples_per_second": 628.807,
|
330 |
+
"eval_steps_per_second": 6.165,
|
331 |
+
"step": 551
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 19.0,
|
335 |
+
"step": 551,
|
336 |
+
"total_flos": 9953687707320320.0,
|
337 |
+
"train_loss": 0.6928667667340453,
|
338 |
+
"train_runtime": 412.0299,
|
339 |
+
"train_samples_per_second": 445.113,
|
340 |
+
"train_steps_per_second": 3.519
|
341 |
+
}
|
342 |
+
],
|
343 |
+
"max_steps": 1450,
|
344 |
+
"num_train_epochs": 50,
|
345 |
+
"total_flos": 9953687707320320.0,
|
346 |
+
"trial_name": null,
|
347 |
+
"trial_params": null
|
348 |
+
}
|