image_classification

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

• Loss: 1.1196
• Accuracy: 0.6188

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	1.8353	0.4
No log	2.0	80	1.5508	0.5062
No log	3.0	120	1.4576	0.45
No log	4.0	160	1.3557	0.5062
No log	5.0	200	1.2596	0.5687
No log	6.0	240	1.2627	0.5125
No log	7.0	280	1.1809	0.5687
No log	8.0	320	1.1993	0.575
No log	9.0	360	1.2327	0.5375
No log	10.0	400	1.1796	0.5938
No log	11.0	440	1.1102	0.6062
No log	12.0	480	1.2828	0.5375
1.0547	13.0	520	1.1936	0.5875
1.0547	14.0	560	1.2056	0.5563
1.0547	15.0	600	1.2566	0.5625
1.0547	16.0	640	1.1936	0.6438
1.0547	17.0	680	1.1728	0.5813
1.0547	18.0	720	1.2201	0.5875
1.0547	19.0	760	1.2704	0.5563
1.0547	20.0	800	1.2232	0.5375

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1