End of training
Browse files- README.md +7 -5
- all_results.json +14 -0
- eval_results.json +9 -0
- logs/events.out.tfevents.1675039188.serv-3334.1119753.8 +3 -0
- train_results.json +8 -0
- trainer_state.json +250 -0
README.md
CHANGED
@@ -1,4 +1,6 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
tags:
|
4 |
- generated_from_trainer
|
@@ -13,7 +15,7 @@ model-index:
|
|
13 |
name: Text Classification
|
14 |
type: text-classification
|
15 |
dataset:
|
16 |
-
name:
|
17 |
type: glue
|
18 |
config: qnli
|
19 |
split: validation
|
@@ -21,7 +23,7 @@ model-index:
|
|
21 |
metrics:
|
22 |
- name: Accuracy
|
23 |
type: accuracy
|
24 |
-
value: 0.
|
25 |
---
|
26 |
|
27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
29 |
|
30 |
# mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_qnli
|
31 |
|
32 |
-
This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the
|
33 |
It achieves the following results on the evaluation set:
|
34 |
-
- Loss: 0.
|
35 |
-
- Accuracy: 0.
|
36 |
|
37 |
## Model description
|
38 |
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
license: apache-2.0
|
5 |
tags:
|
6 |
- generated_from_trainer
|
|
|
15 |
name: Text Classification
|
16 |
type: text-classification
|
17 |
dataset:
|
18 |
+
name: GLUE QNLI
|
19 |
type: glue
|
20 |
config: qnli
|
21 |
split: validation
|
|
|
23 |
metrics:
|
24 |
- name: Accuracy
|
25 |
type: accuracy
|
26 |
+
value: 0.8984074684239429
|
27 |
---
|
28 |
|
29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
31 |
|
32 |
# mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_qnli
|
33 |
|
34 |
+
This model is a fine-tuned version of [gokuls/mobilebert_sa_pre-training-complete](https://huggingface.co/gokuls/mobilebert_sa_pre-training-complete) on the GLUE QNLI dataset.
|
35 |
It achieves the following results on the evaluation set:
|
36 |
+
- Loss: 0.2158
|
37 |
+
- Accuracy: 0.8984
|
38 |
|
39 |
## Model description
|
40 |
|
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_accuracy": 0.8984074684239429,
|
4 |
+
"eval_loss": 0.21577584743499756,
|
5 |
+
"eval_runtime": 10.2259,
|
6 |
+
"eval_samples": 5463,
|
7 |
+
"eval_samples_per_second": 534.232,
|
8 |
+
"eval_steps_per_second": 4.205,
|
9 |
+
"train_loss": 0.6733751687471781,
|
10 |
+
"train_runtime": 7169.216,
|
11 |
+
"train_samples": 104743,
|
12 |
+
"train_samples_per_second": 730.505,
|
13 |
+
"train_steps_per_second": 5.712
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_accuracy": 0.8984074684239429,
|
4 |
+
"eval_loss": 0.21577584743499756,
|
5 |
+
"eval_runtime": 10.2259,
|
6 |
+
"eval_samples": 5463,
|
7 |
+
"eval_samples_per_second": 534.232,
|
8 |
+
"eval_steps_per_second": 4.205
|
9 |
+
}
|
logs/events.out.tfevents.1675039188.serv-3334.1119753.8
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:669381a26148522dac5131cd329c21e41c09aa09cd5ebde3365041933221bd86
|
3 |
+
size 363
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"train_loss": 0.6733751687471781,
|
4 |
+
"train_runtime": 7169.216,
|
5 |
+
"train_samples": 104743,
|
6 |
+
"train_samples_per_second": 730.505,
|
7 |
+
"train_steps_per_second": 5.712
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,250 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.21577584743499756,
|
3 |
+
"best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_pretrain_qnli/checkpoint-8190",
|
4 |
+
"epoch": 15.0,
|
5 |
+
"global_step": 12285,
|
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": 4.9e-05,
|
13 |
+
"loss": 1.1418,
|
14 |
+
"step": 819
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.0,
|
18 |
+
"eval_accuracy": 0.5053999633900788,
|
19 |
+
"eval_loss": 1.0623146295547485,
|
20 |
+
"eval_runtime": 10.2085,
|
21 |
+
"eval_samples_per_second": 535.141,
|
22 |
+
"eval_steps_per_second": 4.212,
|
23 |
+
"step": 819
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 2.0,
|
27 |
+
"learning_rate": 4.8e-05,
|
28 |
+
"loss": 1.1397,
|
29 |
+
"step": 1638
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 2.0,
|
33 |
+
"eval_accuracy": 0.5053999633900788,
|
34 |
+
"eval_loss": 1.0616631507873535,
|
35 |
+
"eval_runtime": 10.1836,
|
36 |
+
"eval_samples_per_second": 536.452,
|
37 |
+
"eval_steps_per_second": 4.222,
|
38 |
+
"step": 1638
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 3.0,
|
42 |
+
"learning_rate": 4.7e-05,
|
43 |
+
"loss": 1.1439,
|
44 |
+
"step": 2457
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 3.0,
|
48 |
+
"eval_accuracy": 0.5053999633900788,
|
49 |
+
"eval_loss": 1.063395380973816,
|
50 |
+
"eval_runtime": 10.2401,
|
51 |
+
"eval_samples_per_second": 533.492,
|
52 |
+
"eval_steps_per_second": 4.199,
|
53 |
+
"step": 2457
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 4.0,
|
57 |
+
"learning_rate": 4.600000000000001e-05,
|
58 |
+
"loss": 1.1397,
|
59 |
+
"step": 3276
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 4.0,
|
63 |
+
"eval_accuracy": 0.5053999633900788,
|
64 |
+
"eval_loss": 1.0635251998901367,
|
65 |
+
"eval_runtime": 10.2624,
|
66 |
+
"eval_samples_per_second": 532.329,
|
67 |
+
"eval_steps_per_second": 4.19,
|
68 |
+
"step": 3276
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 5.0,
|
72 |
+
"learning_rate": 4.5e-05,
|
73 |
+
"loss": 1.14,
|
74 |
+
"step": 4095
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 5.0,
|
78 |
+
"eval_accuracy": 0.5053999633900788,
|
79 |
+
"eval_loss": 1.0643025636672974,
|
80 |
+
"eval_runtime": 10.1907,
|
81 |
+
"eval_samples_per_second": 536.078,
|
82 |
+
"eval_steps_per_second": 4.22,
|
83 |
+
"step": 4095
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 6.0,
|
87 |
+
"learning_rate": 4.4000000000000006e-05,
|
88 |
+
"loss": 1.1399,
|
89 |
+
"step": 4914
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 6.0,
|
93 |
+
"eval_accuracy": 0.5053999633900788,
|
94 |
+
"eval_loss": 1.0611162185668945,
|
95 |
+
"eval_runtime": 10.1973,
|
96 |
+
"eval_samples_per_second": 535.731,
|
97 |
+
"eval_steps_per_second": 4.217,
|
98 |
+
"step": 4914
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 7.0,
|
102 |
+
"learning_rate": 4.3e-05,
|
103 |
+
"loss": 1.14,
|
104 |
+
"step": 5733
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 7.0,
|
108 |
+
"eval_accuracy": 0.5053999633900788,
|
109 |
+
"eval_loss": 1.06246018409729,
|
110 |
+
"eval_runtime": 10.1876,
|
111 |
+
"eval_samples_per_second": 536.238,
|
112 |
+
"eval_steps_per_second": 4.221,
|
113 |
+
"step": 5733
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 8.0,
|
117 |
+
"learning_rate": 4.2e-05,
|
118 |
+
"loss": 1.0013,
|
119 |
+
"step": 6552
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 8.0,
|
123 |
+
"eval_accuracy": 0.8420281896393923,
|
124 |
+
"eval_loss": 0.380109041929245,
|
125 |
+
"eval_runtime": 10.1848,
|
126 |
+
"eval_samples_per_second": 536.385,
|
127 |
+
"eval_steps_per_second": 4.222,
|
128 |
+
"step": 6552
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 9.0,
|
132 |
+
"learning_rate": 4.1e-05,
|
133 |
+
"loss": 0.3353,
|
134 |
+
"step": 7371
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 9.0,
|
138 |
+
"eval_accuracy": 0.9029837085850265,
|
139 |
+
"eval_loss": 0.21630515158176422,
|
140 |
+
"eval_runtime": 10.2676,
|
141 |
+
"eval_samples_per_second": 532.064,
|
142 |
+
"eval_steps_per_second": 4.188,
|
143 |
+
"step": 7371
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 10.0,
|
147 |
+
"learning_rate": 4e-05,
|
148 |
+
"loss": 0.2165,
|
149 |
+
"step": 8190
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 10.0,
|
153 |
+
"eval_accuracy": 0.8984074684239429,
|
154 |
+
"eval_loss": 0.21577584743499756,
|
155 |
+
"eval_runtime": 10.195,
|
156 |
+
"eval_samples_per_second": 535.853,
|
157 |
+
"eval_steps_per_second": 4.218,
|
158 |
+
"step": 8190
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 11.0,
|
162 |
+
"learning_rate": 3.9000000000000006e-05,
|
163 |
+
"loss": 0.1593,
|
164 |
+
"step": 9009
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 11.0,
|
168 |
+
"eval_accuracy": 0.9057294526816767,
|
169 |
+
"eval_loss": 0.22050638496875763,
|
170 |
+
"eval_runtime": 10.1795,
|
171 |
+
"eval_samples_per_second": 536.669,
|
172 |
+
"eval_steps_per_second": 4.224,
|
173 |
+
"step": 9009
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 12.0,
|
177 |
+
"learning_rate": 3.8e-05,
|
178 |
+
"loss": 0.126,
|
179 |
+
"step": 9828
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 12.0,
|
183 |
+
"eval_accuracy": 0.9077429983525536,
|
184 |
+
"eval_loss": 0.22907224297523499,
|
185 |
+
"eval_runtime": 10.1911,
|
186 |
+
"eval_samples_per_second": 536.053,
|
187 |
+
"eval_steps_per_second": 4.219,
|
188 |
+
"step": 9828
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 13.0,
|
192 |
+
"learning_rate": 3.7e-05,
|
193 |
+
"loss": 0.1049,
|
194 |
+
"step": 10647
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 13.0,
|
198 |
+
"eval_accuracy": 0.9071938495332235,
|
199 |
+
"eval_loss": 0.23231425881385803,
|
200 |
+
"eval_runtime": 10.2915,
|
201 |
+
"eval_samples_per_second": 530.824,
|
202 |
+
"eval_steps_per_second": 4.178,
|
203 |
+
"step": 10647
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 14.0,
|
207 |
+
"learning_rate": 3.6e-05,
|
208 |
+
"loss": 0.0903,
|
209 |
+
"step": 11466
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 14.0,
|
213 |
+
"eval_accuracy": 0.8984074684239429,
|
214 |
+
"eval_loss": 0.2675889730453491,
|
215 |
+
"eval_runtime": 10.2001,
|
216 |
+
"eval_samples_per_second": 535.585,
|
217 |
+
"eval_steps_per_second": 4.216,
|
218 |
+
"step": 11466
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 15.0,
|
222 |
+
"learning_rate": 3.5e-05,
|
223 |
+
"loss": 0.0819,
|
224 |
+
"step": 12285
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 15.0,
|
228 |
+
"eval_accuracy": 0.900604063701263,
|
229 |
+
"eval_loss": 0.23767022788524628,
|
230 |
+
"eval_runtime": 10.1798,
|
231 |
+
"eval_samples_per_second": 536.652,
|
232 |
+
"eval_steps_per_second": 4.224,
|
233 |
+
"step": 12285
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 15.0,
|
237 |
+
"step": 12285,
|
238 |
+
"total_flos": 4.989595309572096e+16,
|
239 |
+
"train_loss": 0.6733751687471781,
|
240 |
+
"train_runtime": 7169.216,
|
241 |
+
"train_samples_per_second": 730.505,
|
242 |
+
"train_steps_per_second": 5.712
|
243 |
+
}
|
244 |
+
],
|
245 |
+
"max_steps": 40950,
|
246 |
+
"num_train_epochs": 50,
|
247 |
+
"total_flos": 4.989595309572096e+16,
|
248 |
+
"trial_name": null,
|
249 |
+
"trial_params": null
|
250 |
+
}
|