metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: vit-base-skin
results: []
vit-base-skin
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6813
- Accuracy: 0.8497
- F1: 0.8487
- Precision: 0.8530
- Recall: 0.8497
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.4952 | 1.0 | 626 | 0.5520 | 0.8083 | 0.8152 | 0.8906 | 0.8083 |
| 0.406 | 2.0 | 1252 | 0.4405 | 0.8808 | 0.8740 | 0.8807 | 0.8808 |
| 0.3248 | 3.0 | 1878 | 0.4204 | 0.8446 | 0.8349 | 0.8381 | 0.8446 |
| 0.1064 | 4.0 | 2504 | 0.6135 | 0.8446 | 0.8375 | 0.8336 | 0.8446 |
| 0.0053 | 5.0 | 3130 | 0.5843 | 0.8705 | 0.8716 | 0.8762 | 0.8705 |
| 0.0024 | 6.0 | 3756 | 0.6813 | 0.8497 | 0.8487 | 0.8530 | 0.8497 |
Framework versions
- Transformers 4.29.2
- Pytorch 1.13.1
- Datasets 2.14.5
- Tokenizers 0.13.3