vit-augment-v3

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the SkinCancerClassification dataset. It achieves the following results on the evaluation set:

• Loss: 0.4122
• Accuracy: {'accuracy': 0.8682896379525593}
• Precision: {'precision': 0.7891363629481224}
• Recall: {'recall': 0.702147611641084}
• F1: {'f1': 0.7241023904492959}

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: cosine
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.3588	1.0	321	0.5097	{'accuracy': 0.8177278401997503}	{'precision': 0.6458951735407427}	{'recall': 0.6505053236299779}	{'f1': 0.638516645464825}
0.3331	2.0	642	0.4549	{'accuracy': 0.8445692883895131}	{'precision': 0.7630341246189177}	{'recall': 0.6706409138231197}	{'f1': 0.6927226829652856}
0.2535	3.0	963	0.4268	{'accuracy': 0.8651685393258427}	{'precision': 0.792673703587323}	{'recall': 0.6988841550092643}	{'f1': 0.7201165886292592}
0.1469	4.0	1284	0.4122	{'accuracy': 0.8682896379525593}	{'precision': 0.7891363629481224}	{'recall': 0.702147611641084}	{'f1': 0.7241023904492959}

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2