swinv2-tiny-patch4-window8-256-finetuned-pokemon-classification
This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on the pokemon-classification dataset. It achieves the following results on the evaluation set:
- Loss: 0.2696
- Accuracy: 0.9343
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 |
|---|---|---|---|---|
| 4.874 | 0.99 | 34 | 4.5972 | 0.0431 |
| 3.6502 | 1.99 | 68 | 2.5463 | 0.4990 |
| 1.7664 | 2.98 | 102 | 1.0859 | 0.7762 |
| 1.0898 | 4.0 | 137 | 0.6180 | 0.8542 |
| 0.8125 | 4.99 | 171 | 0.4411 | 0.9035 |
| 0.7437 | 5.99 | 205 | 0.3597 | 0.9076 |
| 0.6117 | 6.98 | 239 | 0.3174 | 0.9302 |
| 0.5581 | 8.0 | 274 | 0.2878 | 0.9281 |
| 0.5178 | 8.99 | 308 | 0.2765 | 0.9302 |
| 0.4802 | 9.93 | 340 | 0.2696 | 0.9343 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3