swin-tiny-patch4-window7-224_RCC_Classifierv3
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 7.6560
- F1: 0.0
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.0095 | 1.0 | 118 | 2.6285 | 0.0 |
| 0.0038 | 2.0 | 236 | 5.5074 | 0.0 |
| 0.0001 | 3.0 | 354 | 5.7253 | 0.0 |
| 0.0111 | 4.0 | 472 | 6.6322 | 0.0 |
| 0.003 | 5.0 | 590 | 7.3159 | 0.0 |
| 0.0002 | 6.0 | 708 | 7.8263 | 0.0 |
| 0.0112 | 7.0 | 826 | 7.8469 | 0.0 |
| 0.0021 | 8.0 | 944 | 7.4348 | 0.0 |
| 0.0 | 9.0 | 1062 | 7.6901 | 0.0 |
| 0.0002 | 10.0 | 1180 | 7.6560 | 0.0 |
Framework versions
- Transformers 4.34.1
- Pytorch 1.12.1
- Datasets 2.14.5
- Tokenizers 0.14.1