vit-lr-0.01

This model is a fine-tuned version of google/vit-base-patch16-224 on the skin-cancer dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Accuracy: 0.4997
• Precision: 0.4902
• Recall: 0.4997
• F1: 0.4904

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: 0.01
• 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: cosine
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.4629	1.0	321	nan	0.4997	0.4902	0.4997	0.4904
0.0	2.0	642	nan	0.4997	0.4902	0.4997	0.4904
0.0096	3.0	963	nan	0.0316	0.0010	0.0316	0.0019
0.0	4.0	1284	nan	0.0316	0.0010	0.0316	0.0019
0.0	5.0	1605	nan	0.0316	0.0010	0.0316	0.0019
0.0	6.0	1926	nan	0.0316	0.0010	0.0316	0.0019
0.0	7.0	2247	nan	0.0316	0.0010	0.0316	0.0019
0.0	8.0	2568	nan	0.0316	0.0010	0.0316	0.0019
0.0	9.0	2889	nan	0.0316	0.0010	0.0316	0.0019
0.0	10.0	3210	nan	0.0316	0.0010	0.0316	0.0019
0.0	11.0	3531	nan	0.0316	0.0010	0.0316	0.0019

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2