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: 1.1383
• Accuracy: 0.6312

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: 0.0002
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.925	1.0	10	1.3570	0.4688
0.8379	2.0	20	1.1685	0.5875
0.6737	3.0	30	1.1795	0.6
0.4606	4.0	40	1.1383	0.6312
0.3416	5.0	50	1.2393	0.5687
0.2493	6.0	60	1.3971	0.5938
0.2341	7.0	70	1.3546	0.6062
0.1797	8.0	80	1.3681	0.5938
0.1221	9.0	90	1.6936	0.525
0.1077	10.0	100	1.7008	0.5375
0.0966	11.0	110	1.7380	0.525
0.1073	12.0	120	1.5617	0.575
0.0849	13.0	130	1.6178	0.6125
0.0704	14.0	140	1.6144	0.6125
0.0568	15.0	150	1.6111	0.6188
0.0555	16.0	160	1.5946	0.6
0.0498	17.0	170	1.6291	0.625
0.0464	18.0	180	1.6574	0.6188
0.0443	19.0	190	1.6740	0.6125
0.0429	20.0	200	1.6781	0.6125

Framework versions

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