swin-tiny-patch4-window7-224-finetuned-car-models-classification-car-parts-v001
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: 1.0027
- Accuracy: 0.8
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 2 | 3.0537 | 0.2 |
| No log | 2.0 | 4 | 2.8544 | 0.2 |
| No log | 3.0 | 6 | 2.5904 | 0.2 |
| No log | 4.0 | 8 | 2.4008 | 0.2 |
| 1.5975 | 5.0 | 10 | 2.2402 | 0.0 |
| 1.5975 | 6.0 | 12 | 2.0655 | 0.0 |
| 1.5975 | 7.0 | 14 | 1.8575 | 0.2 |
| 1.5975 | 8.0 | 16 | 1.6549 | 0.2 |
| 1.5975 | 9.0 | 18 | 1.4673 | 0.2 |
| 0.6346 | 10.0 | 20 | 1.3287 | 0.4 |
| 0.6346 | 11.0 | 22 | 1.2002 | 0.6 |
| 0.6346 | 12.0 | 24 | 1.0911 | 0.6 |
| 0.6346 | 13.0 | 26 | 1.0027 | 0.8 |
| 0.6346 | 14.0 | 28 | 0.9490 | 0.8 |
| 0.2881 | 15.0 | 30 | 0.9202 | 0.8 |
| 0.2881 | 16.0 | 32 | 0.9148 | 0.8 |
| 0.2881 | 17.0 | 34 | 0.8827 | 0.8 |
| 0.2881 | 18.0 | 36 | 0.8587 | 0.8 |
| 0.2881 | 19.0 | 38 | 0.8445 | 0.8 |
| 0.2595 | 20.0 | 40 | 0.8331 | 0.8 |
Framework versions
- Transformers 4.54.1
- Pytorch 2.7.1+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4