Hugging Face's logo

qinweijia
/
distilbert-base-uncased-finetuned-cola

Text Classification
Transformers
Safetensors
distilbert
Generated from Trainer
Inference Endpoints
Model card Files Community
qinweijia commited on
Commit
47cf24b
·
verified ·
1 Parent(s): 4d8bcf4

End of training

Browse files
Files changed (3) hide show
  1. README.md +18 -55
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -5,6 +5,10 @@ base_model: distilbert-base-uncased
5
  tags:
6
  - generated_from_trainer
7
  metrics:
 
 
 
 
8
  - matthews_correlation
9
  model-index:
10
  - name: distilbert-base-uncased-finetuned-cola
@@ -18,8 +22,12 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 1.6117
22
- - Matthews Correlation: 0.5489
 
 
 
 
23
 
24
  ## Model description
25
 
@@ -44,62 +52,17 @@ The following hyperparameters were used during training:
44
  - seed: 42
45
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
- - num_epochs: 50
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
52
- |:-------------:|:-----:|:-----:|:---------------:|:--------------------:|
53
- | 0.5247 | 1.0 | 535 | 0.4696 | 0.4358 |
54
- | 0.3581 | 2.0 | 1070 | 0.4678 | 0.5081 |
55
- | 0.2448 | 3.0 | 1605 | 0.5805 | 0.5089 |
56
- | 0.1893 | 4.0 | 2140 | 0.7913 | 0.5332 |
57
- | 0.1452 | 5.0 | 2675 | 0.8406 | 0.5127 |
58
- | 0.1244 | 6.0 | 3210 | 0.9495 | 0.5406 |
59
- | 0.1044 | 7.0 | 3745 | 0.9900 | 0.5089 |
60
- | 0.0767 | 8.0 | 4280 | 1.1411 | 0.5235 |
61
- | 0.0618 | 9.0 | 4815 | 1.1395 | 0.5181 |
62
- | 0.0607 | 10.0 | 5350 | 1.2811 | 0.5081 |
63
- | 0.0617 | 11.0 | 5885 | 1.2854 | 0.5207 |
64
- | 0.0455 | 12.0 | 6420 | 1.3853 | 0.5174 |
65
- | 0.0377 | 13.0 | 6955 | 1.5257 | 0.5242 |
66
- | 0.0308 | 14.0 | 7490 | 1.6128 | 0.5159 |
67
- | 0.0214 | 15.0 | 8025 | 1.6524 | 0.4882 |
68
- | 0.0292 | 16.0 | 8560 | 1.4723 | 0.5331 |
69
- | 0.0269 | 17.0 | 9095 | 1.5875 | 0.5154 |
70
- | 0.0232 | 18.0 | 9630 | 1.6217 | 0.5171 |
71
- | 0.0334 | 19.0 | 10165 | 1.5183 | 0.5282 |
72
- | 0.0247 | 20.0 | 10700 | 1.6117 | 0.5489 |
73
- | 0.0168 | 21.0 | 11235 | 1.6357 | 0.5452 |
74
- | 0.0172 | 22.0 | 11770 | 1.7956 | 0.5378 |
75
- | 0.0187 | 23.0 | 12305 | 1.5658 | 0.5260 |
76
- | 0.0206 | 24.0 | 12840 | 1.6623 | 0.5246 |
77
- | 0.0188 | 25.0 | 13375 | 1.8499 | 0.5195 |
78
- | 0.0096 | 26.0 | 13910 | 1.8293 | 0.5229 |
79
- | 0.0083 | 27.0 | 14445 | 1.5991 | 0.5403 |
80
- | 0.0218 | 28.0 | 14980 | 1.8168 | 0.4840 |
81
- | 0.0096 | 29.0 | 15515 | 1.7009 | 0.5215 |
82
- | 0.0114 | 30.0 | 16050 | 1.7343 | 0.5426 |
83
- | 0.0095 | 31.0 | 16585 | 1.8909 | 0.5073 |
84
- | 0.0094 | 32.0 | 17120 | 1.7644 | 0.5328 |
85
- | 0.0106 | 33.0 | 17655 | 1.6923 | 0.5348 |
86
- | 0.0089 | 34.0 | 18190 | 1.9452 | 0.5097 |
87
- | 0.0075 | 35.0 | 18725 | 1.7923 | 0.5343 |
88
- | 0.0078 | 36.0 | 19260 | 1.7430 | 0.5365 |
89
- | 0.0097 | 37.0 | 19795 | 1.8556 | 0.5147 |
90
- | 0.0058 | 38.0 | 20330 | 1.9414 | 0.5084 |
91
- | 0.0055 | 39.0 | 20865 | 1.8769 | 0.5400 |
92
- | 0.0037 | 40.0 | 21400 | 1.9577 | 0.5288 |
93
- | 0.0035 | 41.0 | 21935 | 2.0144 | 0.5227 |
94
- | 0.0034 | 42.0 | 22470 | 2.0151 | 0.5150 |
95
- | 0.0033 | 43.0 | 23005 | 2.0254 | 0.5238 |
96
- | 0.002 | 44.0 | 23540 | 2.0601 | 0.5262 |
97
- | 0.0048 | 45.0 | 24075 | 1.9922 | 0.5392 |
98
- | 0.0054 | 46.0 | 24610 | 2.0499 | 0.5285 |
99
- | 0.0055 | 47.0 | 25145 | 2.0038 | 0.5466 |
100
- | 0.0044 | 48.0 | 25680 | 2.0758 | 0.5206 |
101
- | 0.0033 | 49.0 | 26215 | 2.0514 | 0.5219 |
102
- | 0.0042 | 50.0 | 26750 | 2.0506 | 0.5245 |
103
 
104
 
105
  ### Framework versions
 
5
  tags:
6
  - generated_from_trainer
7
  metrics:
8
+ - accuracy
9
+ - f1
10
+ - precision
11
+ - recall
12
  - matthews_correlation
13
  model-index:
14
  - name: distilbert-base-uncased-finetuned-cola
 
22
 
23
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
24
  It achieves the following results on the evaluation set:
25
+ - Loss: 0.8321
26
+ - Accuracy: 0.8178
27
+ - F1: 0.8758
28
+ - Precision: 0.8282
29
+ - Recall: 0.9293
30
+ - Matthews Correlation: 0.5510
31
 
32
  ## Model description
33
 
 
52
  - seed: 42
53
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
  - lr_scheduler_type: linear
55
+ - num_epochs: 5
56
 
57
  ### Training results
58
 
59
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | Matthews Correlation |
60
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:--------------------:|
61
+ | 0.5185 | 1.0 | 535 | 0.4731 | 0.7756 | 0.8536 | 0.7777 | 0.9459 | 0.4298 |
62
+ | 0.3525 | 2.0 | 1070 | 0.4631 | 0.8044 | 0.8622 | 0.8406 | 0.8849 | 0.5284 |
63
+ | 0.2374 | 3.0 | 1605 | 0.6060 | 0.8102 | 0.8699 | 0.8265 | 0.9182 | 0.5324 |
64
+ | 0.1694 | 4.0 | 2140 | 0.7755 | 0.8140 | 0.8734 | 0.8249 | 0.9279 | 0.5408 |
65
+ | 0.1297 | 5.0 | 2675 | 0.8321 | 0.8178 | 0.8758 | 0.8282 | 0.9293 | 0.5510 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
 
67
 
68
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbb9e1012c5f8a9b6db4bbc26e9a37645a62a6d27a1f305560215b48cf0492c5
3
  size 267832560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0163881a00ad8577e429caa3a2244fa985662fa940f9eaf6a1ebcd77613bd1f
3
  size 267832560
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:013c83847a196d63623f8b83a65aad197246d89b09d192d772a7977245a1b8a5
3
  size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:583c56866d1946e90679e03f49ddeb6c8178167a4b3bd6e40a547744de2ff2d9
3
  size 5368