vit-base-patch16-224-in21k-shiba-inu-detector

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on dataset with 4 dog types including Shiba Inu.

It achieves the following results on the evaluation set:

• Loss: 0.6511
• Accuracy: 1.0

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• 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	0.94	4	1.3875	0.1667
No log	1.94	8	1.2712	0.7833
1.4176	2.94	12	1.0972	0.9
1.4176	3.94	16	0.9365	0.95
1.0144	4.94	20	0.7836	0.9833
1.0144	5.94	24	0.6511	1.0
1.0144	6.94	28	0.5329	1.0
0.6329	7.94	32	0.4403	1.0
0.6329	8.94	36	0.3777	1.0
0.3821	9.94	40	0.3273	1.0
0.3821	10.94	44	0.2886	1.0
0.3821	11.94	48	0.2622	1.0
0.2655	12.94	52	0.2397	1.0
0.2655	13.94	56	0.2250	1.0
0.202	14.94	60	0.2152	1.0
0.202	15.94	64	0.2074	1.0
0.202	16.94	68	0.2003	1.0
0.1785	17.94	72	0.1960	1.0
0.1785	18.94	76	0.1936	1.0
0.1618	19.94	80	0.1930	1.0

Framework versions

• Transformers 4.17.0
• Pytorch 1.11.0+cu113
• Datasets 2.0.0
• Tokenizers 0.11.6