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: 1.3340
• Accuracy: 0.575

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	1.5156	0.45
No log	2.0	80	1.4200	0.4562
No log	3.0	120	1.3790	0.5
No log	4.0	160	1.2859	0.525
No log	5.0	200	1.2592	0.5125
No log	6.0	240	1.3145	0.55
No log	7.0	280	1.3267	0.4813
No log	8.0	320	1.3288	0.5
No log	9.0	360	1.3073	0.5
No log	10.0	400	1.3066	0.5188
No log	11.0	440	1.2691	0.5563
No log	12.0	480	1.2809	0.5437
0.876	13.0	520	1.2963	0.5625
0.876	14.0	560	1.2965	0.5312
0.876	15.0	600	1.3542	0.5188
0.876	16.0	640	1.3489	0.5125
0.876	17.0	680	1.3146	0.5687
0.876	18.0	720	1.2442	0.575
0.876	19.0	760	1.3497	0.575
0.876	20.0	800	1.3316	0.5437

Framework versions

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