<!-- 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-base-HAM-10000-sharpened-large-patch-32
This model is a fine-tuned version of google/vit-large-patch32-224-in21k on the ahishamm/HAM_db_sharpened dataset. It achieves the following results on the evaluation set:
- Loss: 0.4582
- Accuracy: 0.8404
- Recall: 0.8404
- F1: 0.8404
- Precision: 0.8404
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 |
---|---|---|---|---|---|---|---|
0.6739 | 0.2 | 100 | 0.7775 | 0.7257 | 0.7257 | 0.7257 | 0.7257 |
0.6922 | 0.4 | 200 | 0.6455 | 0.7711 | 0.7711 | 0.7711 | 0.7711 |
0.8219 | 0.6 | 300 | 0.7582 | 0.7426 | 0.7426 | 0.7426 | 0.7426 |
0.6801 | 0.8 | 400 | 0.6363 | 0.7651 | 0.7651 | 0.7651 | 0.7651 |
0.5499 | 1.0 | 500 | 0.6231 | 0.7751 | 0.7751 | 0.7751 | 0.7751 |
0.5156 | 1.2 | 600 | 0.6399 | 0.7761 | 0.7761 | 0.7761 | 0.7761 |
0.4478 | 1.4 | 700 | 0.5324 | 0.8020 | 0.8020 | 0.8020 | 0.8020 |
0.4364 | 1.6 | 800 | 0.5597 | 0.7970 | 0.7970 | 0.7970 | 0.7970 |
0.4545 | 1.8 | 900 | 0.5212 | 0.8115 | 0.8115 | 0.8115 | 0.8115 |
0.4294 | 2.0 | 1000 | 0.4926 | 0.8264 | 0.8264 | 0.8264 | 0.8264 |
0.135 | 2.2 | 1100 | 0.5448 | 0.8204 | 0.8204 | 0.8204 | 0.8204 |
0.2628 | 2.4 | 1200 | 0.4916 | 0.8304 | 0.8304 | 0.8304 | 0.8304 |
0.2577 | 2.59 | 1300 | 0.4582 | 0.8404 | 0.8404 | 0.8404 | 0.8404 |
0.2093 | 2.79 | 1400 | 0.5079 | 0.8344 | 0.8344 | 0.8344 | 0.8344 |
0.1415 | 2.99 | 1500 | 0.4760 | 0.8439 | 0.8439 | 0.8439 | 0.8439 |
0.0686 | 3.19 | 1600 | 0.5379 | 0.8444 | 0.8444 | 0.8444 | 0.8444 |
0.1031 | 3.39 | 1700 | 0.5572 | 0.8384 | 0.8384 | 0.8384 | 0.8384 |
0.102 | 3.59 | 1800 | 0.5343 | 0.8464 | 0.8464 | 0.8464 | 0.8464 |
0.0531 | 3.79 | 1900 | 0.5482 | 0.8479 | 0.8479 | 0.8479 | 0.8479 |
0.0756 | 3.99 | 2000 | 0.5454 | 0.8454 | 0.8454 | 0.8454 | 0.8454 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3