vit-xray-pneumonia-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: 0.0740
• Accuracy: 0.9734

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4843	0.9882	63	0.1954	0.9408
0.1986	1.9922	127	0.1483	0.9494
0.1694	2.9961	191	0.1316	0.9459
0.1368	4.0	255	0.1207	0.9554
0.1399	4.9882	318	0.1738	0.9296
0.1203	5.9922	382	0.0966	0.9631
0.1085	6.9961	446	0.0956	0.9631
0.1046	8.0	510	0.0952	0.9665
0.0883	8.9882	573	0.0990	0.9665
0.0773	9.9922	637	0.0896	0.9717
0.0815	10.9961	701	0.1084	0.9605
0.0793	12.0	765	0.0767	0.9742
0.0778	12.9882	828	0.0885	0.9691
0.0609	13.9922	892	0.0778	0.9708
0.0685	14.8235	945	0.0740	0.9734

Framework versions

• Transformers 4.40.1
• Pytorch 2.3.0
• Datasets 2.19.0
• Tokenizers 0.19.1