Wound-Image-classification

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: 0.1209
• Accuracy: 0.965

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0919	1.0	200	0.7780	0.76
0.6157	2.0	400	0.5695	0.7925
0.4894	3.0	600	0.3667	0.8775
0.3786	4.0	800	0.4436	0.8625
0.3142	5.0	1000	0.4412	0.8625
0.2636	6.0	1200	0.4430	0.86
0.198	7.0	1400	0.2760	0.9175
0.1456	8.0	1600	0.2211	0.93
0.1586	9.0	1800	0.3520	0.905
0.1307	10.0	2000	0.3188	0.9175
0.106	11.0	2200	0.3167	0.925
0.0975	12.0	2400	0.2633	0.92
0.0734	13.0	2600	0.1813	0.9525
0.0994	14.0	2800	0.2150	0.945
0.0622	15.0	3000	0.1757	0.955
0.0609	16.0	3200	0.1209	0.965

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2