<!-- 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. -->
Output-prova_melanoma
This model is a fine-tuned version of UnipaPolitoUnimore/vit-large-patch32-384-melanoma on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.7684
- Accuracy: 0.6852
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: 1e-05
- 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: 40
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3691 | 0.98 | 13 | 1.2635 | 0.4074 |
| 1.0847 | 1.96 | 26 | 1.0838 | 0.5 |
| 0.9902 | 2.94 | 39 | 1.0133 | 0.5185 |
| 1.081 | 4.0 | 53 | 0.9913 | 0.5370 |
| 0.9947 | 4.98 | 66 | 0.9546 | 0.5556 |
| 0.893 | 5.96 | 79 | 0.9383 | 0.5556 |
| 1.0001 | 6.94 | 92 | 0.9344 | 0.5926 |
| 0.8996 | 8.0 | 106 | 0.9466 | 0.5370 |
| 0.8788 | 8.98 | 119 | 0.9110 | 0.5556 |
| 0.9626 | 9.96 | 132 | 0.8969 | 0.5926 |
| 0.8905 | 10.94 | 145 | 0.8685 | 0.5926 |
| 0.9012 | 12.0 | 159 | 0.8696 | 0.5926 |
| 1.0567 | 12.98 | 172 | 0.8708 | 0.4815 |
| 0.8742 | 13.96 | 185 | 0.8362 | 0.6111 |
| 0.8001 | 14.94 | 198 | 0.8310 | 0.5926 |
| 0.9412 | 16.0 | 212 | 0.8012 | 0.7222 |
| 0.909 | 16.98 | 225 | 0.8124 | 0.6481 |
| 0.9637 | 17.96 | 238 | 0.8133 | 0.6481 |
| 0.8025 | 18.94 | 251 | 0.7845 | 0.7037 |
| 0.855 | 20.0 | 265 | 0.7631 | 0.6481 |
| 0.8325 | 20.98 | 278 | 0.7550 | 0.6667 |
| 0.7347 | 21.96 | 291 | 0.7869 | 0.6296 |
| 0.8302 | 22.94 | 304 | 0.8143 | 0.6481 |
| 0.9766 | 24.0 | 318 | 0.7929 | 0.6852 |
| 0.7516 | 24.98 | 331 | 0.8089 | 0.6296 |
| 0.7784 | 25.96 | 344 | 0.7878 | 0.6296 |
| 0.6438 | 26.94 | 357 | 0.7569 | 0.7037 |
| 0.5649 | 28.0 | 371 | 0.7707 | 0.7037 |
| 0.8274 | 28.98 | 384 | 0.7648 | 0.6852 |
| 0.7754 | 29.96 | 397 | 0.7532 | 0.7037 |
| 0.7434 | 30.94 | 410 | 0.7607 | 0.6852 |
| 0.8634 | 32.0 | 424 | 0.7515 | 0.6852 |
| 0.7391 | 32.98 | 437 | 0.7634 | 0.6852 |
| 0.7261 | 33.96 | 450 | 0.7665 | 0.6852 |
| 0.6929 | 34.94 | 463 | 0.7705 | 0.6852 |
| 0.655 | 36.0 | 477 | 0.7655 | 0.6667 |
| 0.707 | 36.98 | 490 | 0.7612 | 0.6852 |
| 0.7946 | 37.96 | 503 | 0.7685 | 0.6481 |
| 0.7646 | 38.94 | 516 | 0.7680 | 0.6852 |
| 0.7836 | 39.25 | 520 | 0.7684 | 0.6852 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.3
