Hugging Face's logo

henryscheible
/
bert-large-uncased_winobias_finetuned

Text Classification
Transformers
PyTorch
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
Henry Scheible commited on
Commit
23ad26a
•
1 Parent(s): af86fa6

rollback to probed model version

Browse files
Files changed (3) hide show
  1. README.md +46 -73
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -16,12 +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: 0.6933
20
- - Accuracy: 0.5
21
- - Tp: 0.0
22
- - Tn: 0.5
23
- - Fp: 0.0
24
- - Fn: 0.5
25
 
26
  ## Model description
27
 
@@ -40,80 +36,57 @@ More information needed
40
  ### Training hyperparameters
41
 
42
  The following hyperparameters were used during training:
43
- - learning_rate: 0.0001
44
- - train_batch_size: 64
45
  - eval_batch_size: 64
46
  - seed: 42
47
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
  - lr_scheduler_type: linear
49
- - num_epochs: 50
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
54
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:---:|:---:|:---:|:---:|
55
- | 0.7205 | 0.8 | 20 | 0.7006 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
56
- | 0.7147 | 1.6 | 40 | 0.6985 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
57
- | 0.6975 | 2.4 | 60 | 0.6935 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
58
- | 0.7205 | 3.2 | 80 | 0.7013 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
59
- | 0.7022 | 4.0 | 100 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
60
- | 0.7095 | 4.8 | 120 | 0.6952 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
61
- | 0.7059 | 5.6 | 140 | 0.7019 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
62
- | 0.718 | 6.4 | 160 | 0.6988 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
63
- | 0.7053 | 7.2 | 180 | 0.7102 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
64
- | 0.7096 | 8.0 | 200 | 0.6936 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
65
- | 0.7041 | 8.8 | 220 | 0.6939 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
66
- | 0.6954 | 9.6 | 240 | 0.6952 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
67
- | 0.7084 | 10.4 | 260 | 0.6982 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
68
- | 0.7056 | 11.2 | 280 | 0.6937 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
69
- | 0.7012 | 12.0 | 300 | 0.7026 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
70
- | 0.7159 | 12.8 | 320 | 0.6942 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
71
- | 0.7059 | 13.6 | 340 | 0.6933 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
72
- | 0.6999 | 14.4 | 360 | 0.6938 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
73
- | 0.7091 | 15.2 | 380 | 0.6951 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
74
- | 0.7 | 16.0 | 400 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
75
- | 0.7048 | 16.8 | 420 | 0.6946 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
76
- | 0.703 | 17.6 | 440 | 0.6976 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
77
- | 0.711 | 18.4 | 460 | 0.6983 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
78
- | 0.7066 | 19.2 | 480 | 0.6997 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
79
- | 0.7109 | 20.0 | 500 | 0.6934 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
80
- | 0.6969 | 20.8 | 520 | 0.7001 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
81
- | 0.7026 | 21.6 | 540 | 0.6941 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
82
- | 0.7 | 22.4 | 560 | 0.6946 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
83
- | 0.7003 | 23.2 | 580 | 0.6977 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
84
- | 0.7154 | 24.0 | 600 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
85
- | 0.6976 | 24.8 | 620 | 0.6938 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
86
- | 0.7117 | 25.6 | 640 | 0.6944 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
87
- | 0.7023 | 26.4 | 660 | 0.6934 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
88
- | 0.7039 | 27.2 | 680 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
89
- | 0.7053 | 28.0 | 700 | 0.6950 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
90
- | 0.7001 | 28.8 | 720 | 0.6941 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
91
- | 0.7026 | 29.6 | 740 | 0.6935 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
92
- | 0.7007 | 30.4 | 760 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
93
- | 0.7058 | 31.2 | 780 | 0.6933 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
94
- | 0.6995 | 32.0 | 800 | 0.6940 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
95
- | 0.7053 | 32.8 | 820 | 0.6939 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
96
- | 0.7066 | 33.6 | 840 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
97
- | 0.7019 | 34.4 | 860 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
98
- | 0.7062 | 35.2 | 880 | 0.6952 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
99
- | 0.7061 | 36.0 | 900 | 0.6941 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
100
- | 0.7115 | 36.8 | 920 | 0.6933 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
101
- | 0.7025 | 37.6 | 940 | 0.6953 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
102
- | 0.6974 | 38.4 | 960 | 0.6936 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
103
- | 0.7077 | 39.2 | 980 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
104
- | 0.7038 | 40.0 | 1000 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
105
- | 0.6995 | 40.8 | 1020 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
106
- | 0.701 | 41.6 | 1040 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
107
- | 0.6972 | 42.4 | 1060 | 0.6934 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
108
- | 0.7038 | 43.2 | 1080 | 0.6936 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
109
- | 0.7047 | 44.0 | 1100 | 0.6934 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
110
- | 0.7087 | 44.8 | 1120 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
111
- | 0.7082 | 45.6 | 1140 | 0.6932 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
112
- | 0.7023 | 46.4 | 1160 | 0.6934 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
113
- | 0.6987 | 47.2 | 1180 | 0.6933 | 0.5 | 0.5 | 0.0 | 0.5 | 0.0 |
114
- | 0.6999 | 48.0 | 1200 | 0.6931 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
115
- | 0.701 | 48.8 | 1220 | 0.6932 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
116
- | 0.6995 | 49.6 | 1240 | 0.6933 | 0.5 | 0.0 | 0.5 | 0.0 | 0.5 |
117
 
118
 
119
  ### 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: 0.4783
20
+ - Accuracy: 0.7986
 
 
 
 
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 1e-05
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
44
  - lr_scheduler_type: linear
45
+ - num_epochs: 15
46
 
47
  ### Training results
48
 
49
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
50
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
51
+ | No log | 0.38 | 5 | 0.7011 | 0.4994 |
52
+ | No log | 0.77 | 10 | 0.6942 | 0.4987 |
53
+ | No log | 1.15 | 15 | 0.6941 | 0.5063 |
54
+ | No log | 1.54 | 20 | 0.6936 | 0.4924 |
55
+ | No log | 1.92 | 25 | 0.6928 | 0.5114 |
56
+ | No log | 2.31 | 30 | 0.6925 | 0.5196 |
57
+ | No log | 2.69 | 35 | 0.6925 | 0.5215 |
58
+ | No log | 3.08 | 40 | 0.6923 | 0.5227 |
59
+ | No log | 3.46 | 45 | 0.6922 | 0.5259 |
60
+ | No log | 3.85 | 50 | 0.6922 | 0.5202 |
61
+ | No log | 4.23 | 55 | 0.6918 | 0.5316 |
62
+ | No log | 4.62 | 60 | 0.6912 | 0.5499 |
63
+ | No log | 5.0 | 65 | 0.6904 | 0.5574 |
64
+ | No log | 5.38 | 70 | 0.6899 | 0.5492 |
65
+ | No log | 5.77 | 75 | 0.6894 | 0.5417 |
66
+ | No log | 6.15 | 80 | 0.6890 | 0.5290 |
67
+ | No log | 6.54 | 85 | 0.6883 | 0.5366 |
68
+ | No log | 6.92 | 90 | 0.6863 | 0.5726 |
69
+ | No log | 7.31 | 95 | 0.6837 | 0.5909 |
70
+ | No log | 7.69 | 100 | 0.6812 | 0.5890 |
71
+ | No log | 8.08 | 105 | 0.6788 | 0.5915 |
72
+ | No log | 8.46 | 110 | 0.6738 | 0.6225 |
73
+ | No log | 8.85 | 115 | 0.6685 | 0.6503 |
74
+ | No log | 9.23 | 120 | 0.6616 | 0.6698 |
75
+ | No log | 9.62 | 125 | 0.6533 | 0.6799 |
76
+ | No log | 10.0 | 130 | 0.6403 | 0.7027 |
77
+ | No log | 10.38 | 135 | 0.6282 | 0.7077 |
78
+ | No log | 10.77 | 140 | 0.6142 | 0.7235 |
79
+ | No log | 11.15 | 145 | 0.5967 | 0.7355 |
80
+ | No log | 11.54 | 150 | 0.5814 | 0.7437 |
81
+ | No log | 11.92 | 155 | 0.5662 | 0.7513 |
82
+ | No log | 12.31 | 160 | 0.5454 | 0.7607 |
83
+ | No log | 12.69 | 165 | 0.5251 | 0.7771 |
84
+ | No log | 13.08 | 170 | 0.5091 | 0.7872 |
85
+ | No log | 13.46 | 175 | 0.4975 | 0.7942 |
86
+ | No log | 13.85 | 180 | 0.4892 | 0.7967 |
87
+ | No log | 14.23 | 185 | 0.4832 | 0.7992 |
88
+ | No log | 14.62 | 190 | 0.4797 | 0.8005 |
89
+ | No log | 15.0 | 195 | 0.4783 | 0.7986 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
 
91
 
92
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c17dc0f170656d33d54ef1274f6edbfc2cca2a723bb23d139fb7f936b9bcb37
3
  size 1340714613
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:964f400ea7dec1a3b0cfb382d401bb0b9fcb57945bd02e811849d5eae34ce86a
3
  size 1340714613
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3482324b6d141994957c64d9984d50c4aacfd13b778fc88d7605ee89b2a578c3
3
  size 3515
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e22720a57ff1706e69bbda9b52a864205ed403d4e3fd366dce8bbecb9fdf1d6
3
  size 3515