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: 1.2851
• Accuracy: 0.5563

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.9706	1.0	20	1.9258	0.35
1.672	2.0	40	1.7025	0.4625
1.4489	3.0	60	1.5581	0.4313
1.2031	4.0	80	1.4534	0.5
0.9503	5.0	100	1.3794	0.5
0.758	6.0	120	1.3283	0.5312
0.6021	7.0	140	1.3007	0.5125
0.4784	8.0	160	1.2851	0.5563
0.3682	9.0	180	1.2815	0.525
0.3117	10.0	200	1.3074	0.5125
0.2753	11.0	220	1.2945	0.525
0.2585	12.0	240	1.2903	0.5375
0.2483	13.0	260	1.2903	0.5437
0.245	14.0	280	1.2927	0.5375
0.2459	15.0	300	1.2925	0.5375

Framework versions

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