Hugging Face's logo

liuyanchen1015
/
roberta-base-mnli_MULTI

Text Classification
Transformers
PyTorch
roberta
generated_from_trainer
Inference Endpoints
Model card Files Community
1
liuyanchen1015 commited on
Commit
0f7f404
•
1 Parent(s): 34fdc2f

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -34
README.md CHANGED
@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [WillHeld/roberta-base-mnli](https://huggingface.co/WillHeld/roberta-base-mnli) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.7607
18
- - Acc: 0.8310
19
 
20
  ## Model description
21
 
@@ -35,8 +35,8 @@ More information needed
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 2e-05
38
- - train_batch_size: 32
39
- - eval_batch_size: 32
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
@@ -46,36 +46,21 @@ The following hyperparameters were used during training:
46
 
47
  | Training Loss | Epoch | Step | Validation Loss | Acc |
48
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
49
- | 0.4798 | 0.17 | 2000 | 0.4810 | 0.8211 |
50
- | 0.4494 | 0.33 | 4000 | 0.4606 | 0.8268 |
51
- | 0.431 | 0.5 | 6000 | 0.4623 | 0.8301 |
52
- | 0.4371 | 0.67 | 8000 | 0.4437 | 0.8297 |
53
- | 0.4297 | 0.83 | 10000 | 0.4548 | 0.8313 |
54
- | 0.4214 | 1.0 | 12000 | 0.4566 | 0.8337 |
55
- | 0.3123 | 1.17 | 14000 | 0.4893 | 0.8323 |
56
- | 0.3158 | 1.33 | 16000 | 0.4861 | 0.8342 |
57
- | 0.324 | 1.5 | 18000 | 0.4812 | 0.8308 |
58
- | 0.3161 | 1.66 | 20000 | 0.4630 | 0.8365 |
59
- | 0.32 | 1.83 | 22000 | 0.4630 | 0.8366 |
60
- | 0.3195 | 2.0 | 24000 | 0.4681 | 0.8355 |
61
- | 0.2274 | 2.16 | 26000 | 0.5347 | 0.8407 |
62
- | 0.2311 | 2.33 | 28000 | 0.5650 | 0.8310 |
63
- | 0.2293 | 2.5 | 30000 | 0.5408 | 0.8355 |
64
- | 0.2296 | 2.66 | 32000 | 0.5207 | 0.8374 |
65
- | 0.2274 | 2.83 | 34000 | 0.5696 | 0.8353 |
66
- | 0.23 | 3.0 | 36000 | 0.5332 | 0.8366 |
67
- | 0.1686 | 3.16 | 38000 | 0.6275 | 0.8343 |
68
- | 0.1632 | 3.33 | 40000 | 0.6457 | 0.8349 |
69
- | 0.1686 | 3.5 | 42000 | 0.5965 | 0.8338 |
70
- | 0.1634 | 3.66 | 44000 | 0.6272 | 0.8342 |
71
- | 0.1656 | 3.83 | 46000 | 0.6541 | 0.8312 |
72
- | 0.162 | 4.0 | 48000 | 0.6408 | 0.8317 |
73
- | 0.1288 | 4.16 | 50000 | 0.7237 | 0.8349 |
74
- | 0.1275 | 4.33 | 52000 | 0.7558 | 0.8295 |
75
- | 0.1291 | 4.5 | 54000 | 0.7730 | 0.8306 |
76
- | 0.1261 | 4.66 | 56000 | 0.7524 | 0.8300 |
77
- | 0.1272 | 4.83 | 58000 | 0.7572 | 0.8317 |
78
- | 0.1242 | 4.99 | 60000 | 0.7607 | 0.8310 |
79
 
80
 
81
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [WillHeld/roberta-base-mnli](https://huggingface.co/WillHeld/roberta-base-mnli) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.6448
18
+ - Acc: 0.8397
19
 
20
  ## Model description
21
 
 
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 2e-05
38
+ - train_batch_size: 64
39
+ - eval_batch_size: 64
40
  - seed: 42
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
 
46
 
47
  | Training Loss | Epoch | Step | Validation Loss | Acc |
48
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
49
+ | 0.4594 | 0.33 | 2000 | 0.4535 | 0.8292 |
50
+ | 0.4231 | 0.67 | 4000 | 0.4440 | 0.8323 |
51
+ | 0.4128 | 1.0 | 6000 | 0.4484 | 0.8352 |
52
+ | 0.3137 | 1.33 | 8000 | 0.4905 | 0.8338 |
53
+ | 0.3168 | 1.66 | 10000 | 0.4740 | 0.8372 |
54
+ | 0.3176 | 2.0 | 12000 | 0.4624 | 0.8415 |
55
+ | 0.2347 | 2.33 | 14000 | 0.5300 | 0.8361 |
56
+ | 0.2361 | 2.66 | 16000 | 0.5097 | 0.8378 |
57
+ | 0.238 | 3.0 | 18000 | 0.5016 | 0.8407 |
58
+ | 0.18 | 3.33 | 20000 | 0.5873 | 0.8382 |
59
+ | 0.179 | 3.66 | 22000 | 0.5917 | 0.8374 |
60
+ | 0.1786 | 4.0 | 24000 | 0.5868 | 0.8398 |
61
+ | 0.1409 | 4.33 | 26000 | 0.6468 | 0.8371 |
62
+ | 0.1395 | 4.66 | 28000 | 0.6423 | 0.8371 |
63
+ | 0.1401 | 4.99 | 30000 | 0.6448 | 0.8397 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
64
 
65
 
66
  ### Framework versions