finetuned-xray

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

• Loss: 0.0206
• Accuracy: 0.9940

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: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7605	0.2123	100	0.6921	0.5030
0.5794	0.4246	200	0.5186	0.7519
0.4215	0.6369	300	0.4524	0.7962
0.3343	0.8493	400	0.3058	0.8707
0.2716	1.0616	500	0.3114	0.8880
0.2027	1.2739	600	0.1684	0.9346
0.1439	1.4862	700	0.1283	0.9579
0.1269	1.6985	800	0.0816	0.9737
0.1247	1.9108	900	0.0920	0.9692
0.1696	2.1231	1000	0.0655	0.9767
0.1004	2.3355	1100	0.0612	0.9857
0.0748	2.5478	1200	0.0764	0.9797
0.1652	2.7601	1300	0.0355	0.9902
0.1129	2.9724	1400	0.0341	0.9917
0.0816	3.1847	1500	0.0511	0.9872
0.091	3.3970	1600	0.0475	0.9850
0.0778	3.6093	1700	0.0329	0.9902
0.111	3.8217	1800	0.0206	0.9940

Framework versions

• Transformers 4.56.2
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1