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: 1.9492
• Accuracy: 0.5312

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: 0.0001
• 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	1.3627	0.4313
No log	2.0	80	1.3275	0.4875
No log	3.0	120	1.2246	0.5188
No log	4.0	160	1.3181	0.5437
No log	5.0	200	1.2843	0.55
No log	6.0	240	1.3726	0.4938
No log	7.0	280	1.4959	0.475
No log	8.0	320	1.4542	0.4875
No log	9.0	360	1.7002	0.4625
No log	10.0	400	1.5043	0.5
No log	11.0	440	1.5684	0.5062
No log	12.0	480	1.6611	0.5
0.5862	13.0	520	1.7354	0.4688
0.5862	14.0	560	1.7357	0.4813
0.5862	15.0	600	1.7006	0.4875
0.5862	16.0	640	1.8564	0.4938
0.5862	17.0	680	1.8633	0.475
0.5862	18.0	720	1.7142	0.5062
0.5862	19.0	760	1.9792	0.4562
0.5862	20.0	800	1.8761	0.5
0.5862	21.0	840	2.0587	0.45
0.5862	22.0	880	2.0288	0.4813
0.5862	23.0	920	1.6472	0.5563
0.5862	24.0	960	2.0372	0.5
0.1675	25.0	1000	1.8781	0.5312
0.1675	26.0	1040	2.0097	0.5062
0.1675	27.0	1080	1.8897	0.5188
0.1675	28.0	1120	1.8845	0.5188
0.1675	29.0	1160	1.9099	0.5312
0.1675	30.0	1200	1.9492	0.5312

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.1