vit-base-patch16-224-ethosrealdata
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2117
- Accuracy: 0.9340
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: 0.0002
- 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 | Accuracy |
|---|---|---|---|---|
| 0.9707 | 0.9913 | 57 | 0.6825 | 0.8160 |
| 0.3507 | 2.0 | 115 | 0.3680 | 0.8909 |
| 0.2002 | 2.9913 | 172 | 0.3121 | 0.9023 |
| 0.1249 | 4.0 | 230 | 0.2951 | 0.9150 |
| 0.1002 | 4.9913 | 287 | 0.2596 | 0.9251 |
| 0.1014 | 6.0 | 345 | 0.2615 | 0.9251 |
| 0.1261 | 6.9913 | 402 | 0.2437 | 0.9365 |
| 0.0556 | 8.0 | 460 | 0.2198 | 0.9416 |
| 0.0415 | 8.9913 | 517 | 0.2119 | 0.9416 |
| 0.0294 | 9.9130 | 570 | 0.2117 | 0.9340 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1