results

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

• Loss: 4.5590
• Accuracy: 0.0625

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6612	1.0	40	3.9513	0.0
0.8129	2.0	80	3.9721	0.025
0.3799	3.0	120	4.3376	0.0125
0.0946	4.0	160	4.4142	0.0563
0.019	5.0	200	4.5590	0.0625
0.0062	6.0	240	4.9286	0.0437
0.0039	7.0	280	5.0577	0.0437
0.0028	8.0	320	5.1624	0.0437
0.0024	9.0	360	5.2316	0.0437
0.0023	10.0	400	5.2923	0.0437
0.0019	11.0	440	5.3317	0.0375
0.0017	12.0	480	5.3658	0.0375
0.0016	13.0	520	5.3915	0.0375
0.0016	14.0	560	5.4004	0.0375
0.0016	15.0	600	5.4022	0.0375

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1