vit-base-1e-4-15ep

This model is a fine-tuned version of google/vit-base-patch16-224 on the vuongnhathien/30VNFoods dataset. It achieves the following results on the evaluation set:

• Loss: 0.3897
• Accuracy: 0.8867

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: 64
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5399	1.0	275	0.4756	0.8676
0.2126	2.0	550	0.4134	0.8875
0.0726	3.0	825	0.4687	0.8775
0.0345	4.0	1100	0.4552	0.8883
0.0123	5.0	1375	0.5129	0.8851
0.0068	6.0	1650	0.4877	0.8954
0.0063	7.0	1925	0.4667	0.9018
0.0055	8.0	2200	0.4697	0.9030
0.0021	9.0	2475	0.4620	0.9054
0.0039	10.0	2750	0.4652	0.9058
0.0027	11.0	3025	0.4658	0.9058
0.0024	12.0	3300	0.4668	0.9078
0.0021	13.0	3575	0.4671	0.9078
0.0019	14.0	3850	0.4681	0.9062
0.002	15.0	4125	0.4682	0.9062

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2