swin-tiny-patch4-window7-224-finetuned-trash_classification
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: 0.3372
- Accuracy: 0.8827
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- 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 | Accuracy |
|---|---|---|---|---|
| 0.4991 | 1.0 | 22 | 0.5482 | 0.7911 |
| 0.4008 | 2.0 | 44 | 0.5193 | 0.7954 |
| 0.3659 | 3.0 | 66 | 0.4464 | 0.8398 |
| 0.372 | 4.0 | 88 | 0.4384 | 0.8398 |
| 0.3388 | 5.0 | 110 | 0.4281 | 0.8455 |
| 0.2654 | 6.0 | 132 | 0.3618 | 0.8712 |
| 0.2326 | 7.0 | 154 | 0.3550 | 0.8755 |
| 0.2354 | 8.0 | 176 | 0.3401 | 0.8798 |
| 0.1774 | 9.0 | 198 | 0.3372 | 0.8827 |
| 0.1849 | 10.0 | 220 | 0.3380 | 0.8827 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2