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: 0.5326
Accuracy: 0.8235
Precision: 0.8344
Recall: 0.8235
F1: 0.8208

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: 0.0001
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
lr_scheduler_warmup_steps: 770
num_epochs: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.8606	1.0	321	0.5784	0.7930	0.7895	0.7930	0.7809
0.5095	2.0	642	0.5439	0.8048	0.8197	0.8048	0.7949
0.4085	3.0	963	0.5629	0.8228	0.8183	0.8228	0.8064
0.2672	4.0	1284	0.5326	0.8235	0.8344	0.8235	0.8208
0.1578	5.0	1605	0.6352	0.8422	0.8433	0.8422	0.8379
0.0921	6.0	1926	0.7425	0.8232	0.8397	0.8232	0.8261
0.0537	7.0	2247	0.8936	0.8336	0.8288	0.8336	0.8188
0.0481	8.0	2568	0.9522	0.8509	0.8451	0.8509	0.8409
0.0291	9.0	2889	0.9770	0.8450	0.8454	0.8450	0.8429
0.04	10.0	3210	0.9303	0.8471	0.8478	0.8471	0.8445
0.0235	11.0	3531	0.9866	0.8454	0.8439	0.8454	0.8395
0.0164	12.0	3852	1.0983	0.8408	0.8473	0.8408	0.8346
0.005	13.0	4173	1.1124	0.8429	0.8433	0.8429	0.8376
0.0064	14.0	4494	1.0629	0.8575	0.8519	0.8575	0.8534

Framework versions

Transformers 4.40.0.dev0
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2

sharren/vit-ds-processed

vit-ds-processed

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

sharren/
vit-ds-processed