Hugging Face's logo

hansin91
/
emotion_classification

Image Classification
Transformers
PyTorch
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
hansin91 commited on
Commit
631d57a
•
1 Parent(s): 2a07913

End of training

Browse files
Files changed (3) hide show
  1. README.md +34 -34
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.4116
36
  - Accuracy: 0.5813
37
 
38
  ## Model description
@@ -52,12 +52,12 @@ More information needed
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
- - learning_rate: 5e-05
56
  - train_batch_size: 16
57
  - eval_batch_size: 16
58
  - seed: 42
59
- - gradient_accumulation_steps: 4
60
- - total_train_batch_size: 64
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_steps: 500
@@ -67,36 +67,36 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
- | No log | 1.0 | 10 | 1.3914 | 0.5312 |
71
- | No log | 2.0 | 20 | 1.3253 | 0.4875 |
72
- | No log | 3.0 | 30 | 1.4217 | 0.4813 |
73
- | No log | 4.0 | 40 | 1.3711 | 0.5062 |
74
- | No log | 5.0 | 50 | 1.3584 | 0.5 |
75
- | No log | 6.0 | 60 | 1.3163 | 0.5 |
76
- | No log | 7.0 | 70 | 1.3824 | 0.5188 |
77
- | No log | 8.0 | 80 | 1.3882 | 0.525 |
78
- | No log | 9.0 | 90 | 1.4126 | 0.5188 |
79
- | No log | 10.0 | 100 | 1.3213 | 0.5625 |
80
- | No log | 11.0 | 110 | 1.4385 | 0.5 |
81
- | No log | 12.0 | 120 | 1.3729 | 0.525 |
82
- | No log | 13.0 | 130 | 1.4603 | 0.4938 |
83
- | No log | 14.0 | 140 | 1.5326 | 0.4688 |
84
- | No log | 15.0 | 150 | 1.3687 | 0.5563 |
85
- | No log | 16.0 | 160 | 1.4537 | 0.55 |
86
- | No log | 17.0 | 170 | 1.5377 | 0.5188 |
87
- | No log | 18.0 | 180 | 1.6417 | 0.4688 |
88
- | No log | 19.0 | 190 | 1.5260 | 0.55 |
89
- | No log | 20.0 | 200 | 1.6854 | 0.4938 |
90
- | No log | 21.0 | 210 | 1.6457 | 0.5062 |
91
- | No log | 22.0 | 220 | 1.5855 | 0.5125 |
92
- | No log | 23.0 | 230 | 1.5083 | 0.5312 |
93
- | No log | 24.0 | 240 | 1.5656 | 0.525 |
94
- | No log | 25.0 | 250 | 1.5931 | 0.5125 |
95
- | No log | 26.0 | 260 | 1.4351 | 0.5687 |
96
- | No log | 27.0 | 270 | 1.5031 | 0.525 |
97
- | No log | 28.0 | 280 | 1.4129 | 0.55 |
98
- | No log | 29.0 | 290 | 1.5323 | 0.5125 |
99
- | No log | 30.0 | 300 | 1.5217 | 0.5625 |
100
 
101
 
102
  ### Framework versions
 
32
 
33
  This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 2.0973
36
  - Accuracy: 0.5813
37
 
38
  ## Model description
 
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
+ - learning_rate: 4e-05
56
  - train_batch_size: 16
57
  - eval_batch_size: 16
58
  - seed: 42
59
+ - gradient_accumulation_steps: 3
60
+ - total_train_batch_size: 48
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_steps: 500
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 0.0283 | 0.97 | 13 | 2.0768 | 0.575 |
71
+ | 0.0243 | 1.95 | 26 | 2.1315 | 0.5437 |
72
+ | 0.0174 | 3.0 | 40 | 2.1721 | 0.5437 |
73
+ | 0.0176 | 3.98 | 53 | 2.1687 | 0.525 |
74
+ | 0.0172 | 4.95 | 66 | 2.0882 | 0.5375 |
75
+ | 0.016 | 6.0 | 80 | 1.9676 | 0.5687 |
76
+ | 0.017 | 6.97 | 93 | 2.1717 | 0.5375 |
77
+ | 0.0169 | 7.95 | 106 | 2.0982 | 0.5375 |
78
+ | 0.0157 | 9.0 | 120 | 2.0893 | 0.5563 |
79
+ | 0.0169 | 9.97 | 133 | 1.7681 | 0.6188 |
80
+ | 0.0179 | 10.95 | 146 | 2.1032 | 0.5437 |
81
+ | 0.0186 | 12.0 | 160 | 2.1116 | 0.55 |
82
+ | 0.0196 | 12.97 | 173 | 2.2203 | 0.5625 |
83
+ | 0.018 | 13.95 | 186 | 2.0634 | 0.575 |
84
+ | 0.0193 | 15.0 | 200 | 2.2562 | 0.5312 |
85
+ | 0.026 | 15.97 | 213 | 2.3150 | 0.5062 |
86
+ | 0.0171 | 16.95 | 226 | 2.0457 | 0.5437 |
87
+ | 0.0194 | 18.0 | 240 | 1.9336 | 0.5938 |
88
+ | 0.0657 | 18.98 | 253 | 2.1706 | 0.5188 |
89
+ | 0.0537 | 19.95 | 266 | 2.0839 | 0.5437 |
90
+ | 0.0566 | 21.0 | 280 | 2.3004 | 0.4813 |
91
+ | 0.074 | 21.98 | 293 | 2.1488 | 0.5375 |
92
+ | 0.0394 | 22.95 | 306 | 2.4144 | 0.475 |
93
+ | 0.0228 | 24.0 | 320 | 2.2085 | 0.55 |
94
+ | 0.0514 | 24.98 | 333 | 2.1443 | 0.5312 |
95
+ | 0.0553 | 25.95 | 346 | 2.3013 | 0.5062 |
96
+ | 0.0436 | 27.0 | 360 | 1.9988 | 0.5813 |
97
+ | 0.1468 | 27.98 | 373 | 2.0166 | 0.5563 |
98
+ | 0.2184 | 28.95 | 386 | 2.4145 | 0.5 |
99
+ | 0.1519 | 29.25 | 390 | 2.2032 | 0.5375 |
100
 
101
 
102
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3c1fa90153b0d15b5b919f55ec7ed552c9a1a7601fecec2b491c54bd1d8d28b
3
  size 343287149
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:963f7a61c9820258773a4d0ae11937adaee6dacd31bfdc088dccfa3bd3bc0693
3
  size 343287149
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9a9ec2b909954ae15958e013288a5241dce7fcbc0981d02527ad5dff78c874a6
3
  size 4027
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7dd65146dabf2bdc422ad14fe08aed12e1f842da73ec31f2959ca7e75cf1cb6
3
  size 4027