<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
vit-large-modified-augmented-ph2-patch-16
This model is a fine-tuned version of google/vit-large-patch16-224-in21k on the ahishamm/Modified_Augmented_PH2_db_sharpened dataset. It achieves the following results on the evaluation set:
- Loss: 0.0827
- Accuracy: 0.9709
- Recall: 0.9709
- F1: 0.9709
- Precision: 0.9709
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | F1 | Precision |
|---|---|---|---|---|---|---|---|
| 0.3402 | 0.29 | 50 | 0.6269 | 0.7945 | 0.7945 | 0.7945 | 0.7945 |
| 0.1387 | 0.59 | 100 | 0.2957 | 0.8921 | 0.8921 | 0.8921 | 0.8921 |
| 0.2921 | 0.88 | 150 | 0.3157 | 0.8836 | 0.8836 | 0.8836 | 0.8836 |
| 0.1268 | 1.18 | 200 | 0.4557 | 0.8527 | 0.8527 | 0.8527 | 0.8527 |
| 0.2071 | 1.47 | 250 | 0.2690 | 0.8818 | 0.8818 | 0.8818 | 0.8818 |
| 0.1238 | 1.76 | 300 | 0.2999 | 0.9178 | 0.9178 | 0.9178 | 0.9178 |
| 0.1327 | 2.06 | 350 | 0.6026 | 0.7877 | 0.7877 | 0.7877 | 0.7877 |
| 0.1453 | 2.35 | 400 | 0.2887 | 0.8990 | 0.8990 | 0.8990 | 0.8990 |
| 0.0686 | 2.65 | 450 | 0.2049 | 0.9503 | 0.9503 | 0.9503 | 0.9503 |
| 0.0414 | 2.94 | 500 | 0.3040 | 0.9195 | 0.9195 | 0.9195 | 0.9195 |
| 0.0851 | 3.24 | 550 | 0.2244 | 0.9298 | 0.9298 | 0.9298 | 0.9298 |
| 0.0054 | 3.53 | 600 | 0.1356 | 0.9555 | 0.9555 | 0.9555 | 0.9555 |
| 0.0029 | 3.82 | 650 | 0.0827 | 0.9709 | 0.9709 | 0.9709 | 0.9709 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
