swin-tiny-patch4-window7-224-finetunedo

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.3710
• Roc Auc: 0.8606

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Roc Auc
No log	1.0	3	0.5066	0.7647
No log	2.0	6	0.4204	0.7941
No log	3.0	9	0.4298	0.7353
0.4868	4.0	12	0.4040	0.8018
0.4868	5.0	15	0.3925	0.7724
0.4868	6.0	18	0.3674	0.8235
0.4096	7.0	21	0.3673	0.8606
0.4096	8.0	24	0.3710	0.8606

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1