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: 1.2378
• Accuracy: 0.5875

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: 2e-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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	2.0656	0.125
No log	2.0	80	2.0558	0.1938
No log	3.0	120	2.0177	0.2375
No log	4.0	160	1.9156	0.3438
No log	5.0	200	1.7849	0.3063
No log	6.0	240	1.6961	0.3187
No log	7.0	280	1.6026	0.3937
No log	8.0	320	1.5455	0.3688
No log	9.0	360	1.4723	0.4562
No log	10.0	400	1.3931	0.5
No log	11.0	440	1.4418	0.4375
No log	12.0	480	1.3306	0.4437
1.5855	13.0	520	1.2437	0.575
1.5855	14.0	560	1.3712	0.4875
1.5855	15.0	600	1.2102	0.55
1.5855	16.0	640	1.3217	0.5188
1.5855	17.0	680	1.3656	0.4938
1.5855	18.0	720	1.3261	0.525
1.5855	19.0	760	1.5611	0.4625
1.5855	20.0	800	1.4503	0.5125

Framework versions

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