metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - image-classification
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-base-1e-4-20ep
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: vuongnhathien/30VNFoods
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8873015873015873

vit-base-1e-4-20ep

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: 0.4034
Accuracy: 0.8873

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5376	1.0	275	0.4677	0.8640
0.2085	2.0	550	0.4375	0.8811
0.0755	3.0	825	0.4605	0.8899
0.0429	4.0	1100	0.4784	0.8879
0.0146	5.0	1375	0.5386	0.8799
0.0176	6.0	1650	0.5524	0.8803
0.0137	7.0	1925	0.5249	0.8887
0.0076	8.0	2200	0.5401	0.8942
0.0026	9.0	2475	0.5477	0.8934
0.0054	10.0	2750	0.5417	0.8946
0.0034	11.0	3025	0.5430	0.8974
0.0033	12.0	3300	0.5443	0.8954
0.0027	13.0	3575	0.5423	0.8986
0.0024	14.0	3850	0.5434	0.8990
0.0027	15.0	4125	0.5483	0.8962
0.0027	16.0	4400	0.5485	0.8998
0.0019	17.0	4675	0.5502	0.8998
0.0022	18.0	4950	0.5508	0.8998
0.0015	19.0	5225	0.5509	0.9002
0.002	20.0	5500	0.5510	0.9010

Framework versions

Transformers 4.39.3
Pytorch 2.1.2
Datasets 2.18.0
Tokenizers 0.15.2