<!-- 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-patch-32
This model is a fine-tuned version of google/vit-base-patch32-224-in21k on the ahishamm/HAM_db_sharpened dataset. It achieves the following results on the evaluation set:
- Loss: 0.4806
- Accuracy: 0.8369
- Recall: 0.8369
- F1: 0.8369
- Precision: 0.8369
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 |
---|---|---|---|---|---|---|---|
0.8099 | 0.2 | 100 | 0.8060 | 0.7247 | 0.7247 | 0.7247 | 0.7247 |
0.7437 | 0.4 | 200 | 0.7020 | 0.7541 | 0.7541 | 0.7541 | 0.7541 |
0.7982 | 0.6 | 300 | 0.7352 | 0.7411 | 0.7411 | 0.7411 | 0.7411 |
0.7646 | 0.8 | 400 | 0.6603 | 0.7626 | 0.7626 | 0.7626 | 0.7626 |
0.6141 | 1.0 | 500 | 0.6373 | 0.7771 | 0.7771 | 0.7771 | 0.7771 |
0.5934 | 1.2 | 600 | 0.6141 | 0.7820 | 0.7820 | 0.7820 | 0.7820 |
0.5524 | 1.4 | 700 | 0.5621 | 0.8030 | 0.8030 | 0.8030 | 0.8030 |
0.5057 | 1.6 | 800 | 0.6074 | 0.7855 | 0.7855 | 0.7855 | 0.7855 |
0.5519 | 1.8 | 900 | 0.5486 | 0.7990 | 0.7990 | 0.7990 | 0.7990 |
0.4784 | 2.0 | 1000 | 0.5382 | 0.8060 | 0.8060 | 0.8060 | 0.8060 |
0.2592 | 2.2 | 1100 | 0.5237 | 0.8165 | 0.8165 | 0.8165 | 0.8165 |
0.3872 | 2.4 | 1200 | 0.5345 | 0.8120 | 0.8120 | 0.8120 | 0.8120 |
0.2506 | 2.59 | 1300 | 0.5061 | 0.8214 | 0.8214 | 0.8214 | 0.8214 |
0.2907 | 2.79 | 1400 | 0.4940 | 0.8354 | 0.8354 | 0.8354 | 0.8354 |
0.2436 | 2.99 | 1500 | 0.4806 | 0.8369 | 0.8369 | 0.8369 | 0.8369 |
0.1472 | 3.19 | 1600 | 0.5231 | 0.8219 | 0.8219 | 0.8219 | 0.8219 |
0.1441 | 3.39 | 1700 | 0.5452 | 0.8329 | 0.8329 | 0.8329 | 0.8329 |
0.1327 | 3.59 | 1800 | 0.5410 | 0.8354 | 0.8354 | 0.8354 | 0.8354 |
0.0615 | 3.79 | 1900 | 0.5473 | 0.8424 | 0.8424 | 0.8424 | 0.8424 |
0.0943 | 3.99 | 2000 | 0.5490 | 0.8409 | 0.8409 | 0.8409 | 0.8409 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3