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: 1.4301
• Accuracy: 0.4688

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
• 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0867	1.0	10	2.0602	0.1938
2.0294	2.0	20	1.9887	0.2562
1.9159	3.0	30	1.8738	0.3438
1.763	4.0	40	1.7523	0.375
1.6138	5.0	50	1.6505	0.4
1.5141	6.0	60	1.5861	0.4125
1.4328	7.0	70	1.5303	0.45
1.3357	8.0	80	1.4986	0.475
1.2833	9.0	90	1.4628	0.4688
1.2248	10.0	100	1.4501	0.5
1.1796	11.0	110	1.3972	0.4875
1.1526	12.0	120	1.4359	0.4813
1.1177	13.0	130	1.4077	0.4813
1.1006	14.0	140	1.3942	0.5
1.0679	15.0	150	1.3934	0.4875

Framework versions

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