SwinV2-30VNFood / README.md
vuongnhathien's picture
Hehe
b4121b6 verified
metadata
license: apache-2.0
base_model: microsoft/swinv2-tiny-patch4-window16-256
tags:
  - image-classification
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: SwinV2-30VNFood
    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: 0.8771825396825397

SwinV2-30VNFood

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window16-256 on the vuongnhathien/30VNFoods dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4561
  • Accuracy: 0.8772

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: 0.0003
  • 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: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.7587 1.0 275 0.5447 0.8477
0.4341 2.0 550 0.4809 0.8640
0.2737 3.0 825 0.4703 0.8763
0.1704 4.0 1100 0.5040 0.8791
0.1225 5.0 1375 0.4893 0.8879
0.0886 6.0 1650 0.5733 0.8863
0.0568 7.0 1925 0.5986 0.8803
0.0407 8.0 2200 0.5664 0.8998
0.0175 9.0 2475 0.5790 0.8998
0.0175 10.0 2750 0.5754 0.9038

Framework versions

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