vit-base-randaug
This model is a fine-tuned version of facebook/convnextv2-base-22k-384 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2287
- Accuracy: 0.9470
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6247 | 1.0 | 1099 | 0.3584 | 0.8982 |
| 0.4589 | 2.0 | 2198 | 0.2780 | 0.9229 |
| 0.3647 | 3.0 | 3297 | 0.2550 | 0.9264 |
| 0.3042 | 4.0 | 4396 | 0.2381 | 0.9400 |
| 0.2912 | 5.0 | 5495 | 0.2347 | 0.9419 |
| 0.2464 | 6.0 | 6594 | 0.2269 | 0.9459 |
| 0.2132 | 7.0 | 7693 | 0.2258 | 0.9483 |
| 0.1956 | 8.0 | 8792 | 0.2222 | 0.9495 |
| 0.1723 | 9.0 | 9891 | 0.2223 | 0.9499 |
| 0.1558 | 10.0 | 10990 | 0.2220 | 0.9507 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2