vit

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: 0.0048
• Accuracy: 0.9986

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: 2e-05
• 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
0.2323	1.0	1595	0.0450	0.9859
0.095	2.0	3190	0.0332	0.9889
0.0648	3.0	4785	0.0256	0.9922
0.0568	4.0	6380	0.0145	0.9958
0.0493	5.0	7975	0.0248	0.9915
0.042	6.0	9570	0.0195	0.9939
0.0383	7.0	11165	0.0087	0.9969
0.0345	8.0	12760	0.0143	0.9960
0.0285	9.0	14355	0.0115	0.9972
0.0257	10.0	15950	0.0131	0.9965
0.0248	11.0	17545	0.0068	0.9979
0.0235	12.0	19140	0.0065	0.9979
0.0201	13.0	20735	0.0056	0.9976
0.0161	14.0	22330	0.0033	0.9988
0.017	15.0	23925	0.0041	0.9988
0.0144	16.0	25520	0.0032	0.9993
0.0121	17.0	27115	0.0055	0.9979
0.0105	18.0	28710	0.0052	0.9984
0.0103	19.0	30305	0.0054	0.9984
0.0092	20.0	31900	0.0048	0.9986

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1