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: 1.1494
• Accuracy: 0.6062

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.8347	1.0	10	1.9052	0.3688
1.838	2.0	20	1.7999	0.375
1.7193	3.0	30	1.6869	0.35
1.5873	4.0	40	1.5855	0.4437
1.4919	5.0	50	1.4977	0.475
1.4049	6.0	60	1.4425	0.4875
1.3025	7.0	70	1.4254	0.45
1.238	8.0	80	1.3994	0.475
1.1704	9.0	90	1.3109	0.5312
1.1009	10.0	100	1.3309	0.525
1.0309	11.0	110	1.2941	0.5687
0.9705	12.0	120	1.2750	0.5188
0.9315	13.0	130	1.2402	0.55
0.8894	14.0	140	1.2425	0.5375
0.8374	15.0	150	1.2273	0.525
0.8	16.0	160	1.2454	0.5125
0.7597	17.0	170	1.2445	0.5125
0.7143	18.0	180	1.1750	0.5687
0.6832	19.0	190	1.2456	0.525
0.6573	20.0	200	1.2004	0.5938
0.639	21.0	210	1.1924	0.5563
0.635	22.0	220	1.1257	0.6
0.5982	23.0	230	1.1845	0.575
0.5675	24.0	240	1.2291	0.5625
0.5634	25.0	250	1.1837	0.5687
0.535	26.0	260	1.2384	0.5813
0.5233	27.0	270	1.1911	0.5875
0.529	28.0	280	1.2083	0.5875
0.5141	29.0	290	1.1813	0.5875
0.5166	30.0	300	1.1578	0.5938

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3