swin-base
This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0891
- Accuracy: 0.9804
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: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.5258 | 0.66 | 40 | 0.4284 | 0.9020 |
0.3085 | 1.31 | 80 | 0.1095 | 0.9608 |
0.3125 | 1.97 | 120 | 0.1925 | 0.9608 |
0.2637 | 2.62 | 160 | 0.1553 | 0.9608 |
0.0669 | 3.28 | 200 | 0.2901 | 0.9804 |
0.082 | 3.93 | 240 | 0.5031 | 0.9216 |
0.0426 | 4.59 | 280 | 0.2911 | 0.9608 |
0.0374 | 5.25 | 320 | 0.1936 | 0.9608 |
0.0002 | 5.9 | 360 | 0.4029 | 0.9412 |
0.0019 | 6.56 | 400 | 0.1289 | 0.9804 |
0.0025 | 7.21 | 440 | 0.2132 | 0.9412 |
0.0088 | 7.87 | 480 | 0.2070 | 0.9608 |
0.0133 | 8.52 | 520 | 0.2054 | 0.9804 |
0.0012 | 9.18 | 560 | 0.0864 | 0.9804 |
0.0 | 9.84 | 600 | 0.0891 | 0.9804 |
Framework versions
- Transformers 4.39.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 10