vit_idcard_classifier_x3_rot_ft
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the arvisioncode/vit_idcard_classifier_x3_rot dataset. It achieves the following results on the evaluation set:
- Loss: 0.9145
- Accuracy: 0.951
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
2.1646 | 0.06 | 100 | 2.1272 | 0.5833 |
1.7966 | 0.12 | 200 | 1.7785 | 0.834 |
1.4988 | 0.18 | 300 | 1.4818 | 0.8643 |
1.2142 | 0.24 | 400 | 1.2601 | 0.8973 |
1.1435 | 0.3 | 500 | 1.1162 | 0.921 |
1.1229 | 0.36 | 600 | 1.1424 | 0.9357 |
1.1081 | 0.41 | 700 | 1.1130 | 0.9393 |
1.0758 | 0.47 | 800 | 1.0606 | 0.9413 |
1.0139 | 0.53 | 900 | 1.0375 | 0.9467 |
1.0151 | 0.59 | 1000 | 1.0219 | 0.946 |
1.0024 | 0.65 | 1100 | 1.0174 | 0.938 |
0.9969 | 0.71 | 1200 | 1.0192 | 0.95 |
0.9716 | 0.77 | 1300 | 0.9799 | 0.9467 |
0.8884 | 0.83 | 1400 | 0.9490 | 0.9537 |
0.8483 | 0.89 | 1500 | 0.9274 | 0.952 |
0.9242 | 0.95 | 1600 | 0.9145 | 0.951 |
Framework versions
- Transformers 4.39.3
- Pytorch 1.10.1+cu113
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 23
Inference API (serverless) is not available, repository is disabled.
Model tree for arvisioncode/vit_idcard_classifier_x3_rot_ft
Base model
google/vit-base-patch16-224-in21k
Finetuned
this model