Hugging Face's logo

henryscheible
/
bert-large-uncased_winobias_classifieronly

Text Classification
Transformers
PyTorch
bert
Generated from Trainer
text-embeddings-inference
Model card Files
xet
 Community
2
henryscheible commited on
Commit
1f55bf8
·
1 Parent(s): ef93e01

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +54 -30
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.6939
20
- - Accuracy: 0.5025
21
 
22
  ## Model description
23
 
@@ -36,8 +36,8 @@ More information needed
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
- - learning_rate: 0.01
40
- - train_batch_size: 128
41
  - eval_batch_size: 64
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -48,32 +48,56 @@ The following hyperparameters were used during training:
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
50
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
51
- | No log | 0.77 | 10 | 3.5262 | 0.5006 |
52
- | No log | 1.54 | 20 | 0.7323 | 0.4931 |
53
- | No log | 2.31 | 30 | 1.3535 | 0.4987 |
54
- | No log | 3.08 | 40 | 0.9251 | 0.5006 |
55
- | No log | 3.85 | 50 | 1.0468 | 0.5019 |
56
- | No log | 4.62 | 60 | 0.9954 | 0.4994 |
57
- | No log | 5.38 | 70 | 0.7594 | 0.5013 |
58
- | No log | 6.15 | 80 | 0.8031 | 0.5006 |
59
- | No log | 6.92 | 90 | 0.9357 | 0.4987 |
60
- | No log | 7.69 | 100 | 0.7557 | 0.5025 |
61
- | No log | 8.46 | 110 | 0.7006 | 0.5114 |
62
- | No log | 9.23 | 120 | 1.2070 | 0.5006 |
63
- | No log | 10.0 | 130 | 0.7082 | 0.5114 |
64
- | No log | 10.77 | 140 | 0.8428 | 0.4994 |
65
- | No log | 11.54 | 150 | 0.7132 | 0.4975 |
66
- | No log | 12.31 | 160 | 0.7783 | 0.5019 |
67
- | No log | 13.08 | 170 | 0.7105 | 0.5069 |
68
- | No log | 13.85 | 180 | 0.7162 | 0.4994 |
69
- | No log | 14.62 | 190 | 0.7015 | 0.5006 |
70
- | No log | 15.38 | 200 | 0.7163 | 0.5032 |
71
- | No log | 16.15 | 210 | 0.7080 | 0.5 |
72
- | No log | 16.92 | 220 | 0.7436 | 0.4994 |
73
- | No log | 17.69 | 230 | 0.6916 | 0.5278 |
74
- | No log | 18.46 | 240 | 0.7042 | 0.4994 |
75
- | No log | 19.23 | 250 | 0.6956 | 0.5101 |
76
- | No log | 20.0 | 260 | 0.6939 | 0.5025 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.6930
20
+ - Accuracy: 0.5095
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 5e-05
40
+ - train_batch_size: 64
41
  - eval_batch_size: 64
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
50
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
51
+ | 0.696 | 0.4 | 10 | 0.6953 | 0.4962 |
52
+ | 0.6961 | 0.8 | 20 | 0.6933 | 0.5133 |
53
+ | 0.7037 | 1.2 | 30 | 0.6936 | 0.5082 |
54
+ | 0.6972 | 1.6 | 40 | 0.6934 | 0.5044 |
55
+ | 0.6923 | 2.0 | 50 | 0.6939 | 0.5095 |
56
+ | 0.7141 | 2.4 | 60 | 0.6955 | 0.4975 |
57
+ | 0.7108 | 2.8 | 70 | 0.6933 | 0.5170 |
58
+ | 0.6899 | 3.2 | 80 | 0.6936 | 0.5038 |
59
+ | 0.702 | 3.6 | 90 | 0.6935 | 0.5025 |
60
+ | 0.7002 | 4.0 | 100 | 0.6932 | 0.5044 |
61
+ | 0.6997 | 4.4 | 110 | 0.6934 | 0.5057 |
62
+ | 0.7002 | 4.8 | 120 | 0.6932 | 0.5170 |
63
+ | 0.7112 | 5.2 | 130 | 0.6942 | 0.5006 |
64
+ | 0.7054 | 5.6 | 140 | 0.6938 | 0.4956 |
65
+ | 0.7084 | 6.0 | 150 | 0.6932 | 0.5152 |
66
+ | 0.6983 | 6.4 | 160 | 0.6935 | 0.5019 |
67
+ | 0.7123 | 6.8 | 170 | 0.6932 | 0.5126 |
68
+ | 0.6932 | 7.2 | 180 | 0.6932 | 0.5082 |
69
+ | 0.6979 | 7.6 | 190 | 0.6942 | 0.5019 |
70
+ | 0.7062 | 8.0 | 200 | 0.6937 | 0.4987 |
71
+ | 0.7083 | 8.4 | 210 | 0.6931 | 0.5164 |
72
+ | 0.7063 | 8.8 | 220 | 0.6935 | 0.5032 |
73
+ | 0.6969 | 9.2 | 230 | 0.6931 | 0.5170 |
74
+ | 0.7051 | 9.6 | 240 | 0.6931 | 0.5152 |
75
+ | 0.6969 | 10.0 | 250 | 0.6933 | 0.4987 |
76
+ | 0.7003 | 10.4 | 260 | 0.6935 | 0.4962 |
77
+ | 0.7045 | 10.8 | 270 | 0.6932 | 0.5076 |
78
+ | 0.7003 | 11.2 | 280 | 0.6931 | 0.5126 |
79
+ | 0.7049 | 11.6 | 290 | 0.6931 | 0.5133 |
80
+ | 0.7117 | 12.0 | 300 | 0.6932 | 0.5152 |
81
+ | 0.7016 | 12.4 | 310 | 0.6933 | 0.5063 |
82
+ | 0.6961 | 12.8 | 320 | 0.6930 | 0.5082 |
83
+ | 0.6992 | 13.2 | 330 | 0.6932 | 0.5063 |
84
+ | 0.703 | 13.6 | 340 | 0.6934 | 0.5063 |
85
+ | 0.7112 | 14.0 | 350 | 0.6930 | 0.5114 |
86
+ | 0.7036 | 14.4 | 360 | 0.6938 | 0.5063 |
87
+ | 0.695 | 14.8 | 370 | 0.6939 | 0.5057 |
88
+ | 0.7051 | 15.2 | 380 | 0.6932 | 0.5076 |
89
+ | 0.7019 | 15.6 | 390 | 0.6930 | 0.5120 |
90
+ | 0.701 | 16.0 | 400 | 0.6930 | 0.5126 |
91
+ | 0.6988 | 16.4 | 410 | 0.6931 | 0.5101 |
92
+ | 0.7049 | 16.8 | 420 | 0.6931 | 0.5107 |
93
+ | 0.7073 | 17.2 | 430 | 0.6931 | 0.5101 |
94
+ | 0.6995 | 17.6 | 440 | 0.6930 | 0.5088 |
95
+ | 0.7142 | 18.0 | 450 | 0.6930 | 0.5076 |
96
+ | 0.7041 | 18.4 | 460 | 0.6930 | 0.5101 |
97
+ | 0.7041 | 18.8 | 470 | 0.6931 | 0.5107 |
98
+ | 0.7064 | 19.2 | 480 | 0.6930 | 0.5095 |
99
+ | 0.7019 | 19.6 | 490 | 0.6930 | 0.5107 |
100
+ | 0.7045 | 20.0 | 500 | 0.6930 | 0.5095 |
101
 
102
 
103
  ### Framework versions