<!-- 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-patch-32
This model is a fine-tuned version of google/vit-base-patch32-224-in21k on the ahishamm/HAM_db dataset. It achieves the following results on the evaluation set:
- Loss: 0.5210
- Accuracy: 0.8040
- Recall: 0.8040
- F1: 0.8040
- Precision: 0.8040
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 |
---|---|---|---|---|---|---|---|
0.7 | 0.2 | 100 | 0.7878 | 0.7307 | 0.7307 | 0.7307 | 0.7307 |
0.8248 | 0.4 | 200 | 0.7338 | 0.7476 | 0.7476 | 0.7476 | 0.7476 |
0.6647 | 0.6 | 300 | 0.6417 | 0.7541 | 0.7541 | 0.7541 | 0.7541 |
0.6755 | 0.8 | 400 | 0.6682 | 0.7576 | 0.7576 | 0.7576 | 0.7576 |
0.7443 | 1.0 | 500 | 0.6037 | 0.7890 | 0.7890 | 0.7890 | 0.7890 |
0.5316 | 1.2 | 600 | 0.5963 | 0.7915 | 0.7915 | 0.7915 | 0.7915 |
0.4404 | 1.4 | 700 | 0.5626 | 0.7955 | 0.7955 | 0.7955 | 0.7955 |
0.4431 | 1.6 | 800 | 0.5719 | 0.8005 | 0.8005 | 0.8005 | 0.8005 |
0.5011 | 1.8 | 900 | 0.5581 | 0.7880 | 0.7880 | 0.7880 | 0.7880 |
0.4692 | 2.0 | 1000 | 0.5210 | 0.8040 | 0.8040 | 0.8040 | 0.8040 |
0.2648 | 2.2 | 1100 | 0.5776 | 0.8070 | 0.8070 | 0.8070 | 0.8070 |
0.2723 | 2.4 | 1200 | 0.5317 | 0.8180 | 0.8180 | 0.8180 | 0.8180 |
0.2325 | 2.59 | 1300 | 0.5223 | 0.8170 | 0.8170 | 0.8170 | 0.8170 |
0.2547 | 2.79 | 1400 | 0.5314 | 0.8244 | 0.8244 | 0.8244 | 0.8244 |
0.146 | 2.99 | 1500 | 0.5583 | 0.8274 | 0.8274 | 0.8274 | 0.8274 |
0.1224 | 3.19 | 1600 | 0.5960 | 0.8289 | 0.8289 | 0.8289 | 0.8289 |
0.0313 | 3.39 | 1700 | 0.6081 | 0.8304 | 0.8304 | 0.8304 | 0.8304 |
0.104 | 3.59 | 1800 | 0.5770 | 0.8339 | 0.8339 | 0.8339 | 0.8339 |
0.0538 | 3.79 | 1900 | 0.5364 | 0.8464 | 0.8464 | 0.8464 | 0.8464 |
0.0827 | 3.99 | 2000 | 0.5414 | 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