vit_4090_7_3

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

• Loss: 0.0352
• Accuracy: 0.9942
• Precision: 0.9942
• Recall: 0.9942
• F1: 0.9942

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: 24
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.2467	1.0	1238	0.1080	0.9679	0.9677	0.9679	0.9670
0.098	2.0	2476	0.0927	0.9743	0.9754	0.9743	0.9740
0.0655	3.0	3714	0.0997	0.9749	0.9764	0.9749	0.9752
0.0494	4.0	4952	0.1050	0.9712	0.9715	0.9712	0.9706
0.0381	5.0	6190	0.1024	0.9776	0.9776	0.9776	0.9770
0.0348	6.0	7428	0.0528	0.9861	0.9866	0.9861	0.9861
0.0253	7.0	8666	0.1030	0.9800	0.9807	0.9800	0.9797
0.0215	8.0	9904	0.0515	0.9867	0.9870	0.9867	0.9866
0.0209	9.0	11142	0.0589	0.9870	0.9872	0.9870	0.9868
0.0123	10.0	12380	0.0783	0.9842	0.9846	0.9842	0.9842
0.0118	11.0	13618	0.0435	0.9915	0.9915	0.9915	0.9915
0.0084	12.0	14856	0.0472	0.9927	0.9927	0.9927	0.9927
0.0081	13.0	16094	0.0394	0.9927	0.9928	0.9927	0.9927
0.0056	14.0	17332	0.0537	0.9909	0.9909	0.9909	0.9908
0.0033	15.0	18570	0.0353	0.9933	0.9934	0.9933	0.9933
0.0022	16.0	19808	0.0386	0.9924	0.9924	0.9924	0.9924
0.0016	17.0	21046	0.0384	0.9930	0.9930	0.9930	0.9930
0.0009	18.0	22284	0.0342	0.9945	0.9945	0.9945	0.9945
0.0009	19.0	23522	0.0363	0.9936	0.9937	0.9936	0.9936
0.0007	20.0	24760	0.0352	0.9942	0.9942	0.9942	0.9942

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1
• Datasets 3.2.0
• Tokenizers 0.21.1