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: 1.2966
• Accuracy: 0.525

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	1.4307	0.475
No log	2.0	80	1.3231	0.5125
No log	3.0	120	1.3044	0.5437
No log	4.0	160	1.3204	0.525
No log	5.0	200	1.2457	0.5875
No log	6.0	240	1.3604	0.5125
No log	7.0	280	1.2296	0.5813
No log	8.0	320	1.3598	0.525
No log	9.0	360	1.3343	0.5188
No log	10.0	400	1.4003	0.5625
No log	11.0	440	1.3580	0.5563
No log	12.0	480	1.3214	0.5687
0.4908	13.0	520	1.3713	0.5312
0.4908	14.0	560	1.3820	0.55
0.4908	15.0	600	1.3384	0.5813
0.4908	16.0	640	1.4905	0.5375
0.4908	17.0	680	1.3985	0.5687
0.4908	18.0	720	1.4733	0.5312
0.4908	19.0	760	1.3403	0.5813
0.4908	20.0	800	1.3991	0.5563

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3