https://github.com/z-fran/pytorch-cifar10-image-classification

A template and Jupyter Notebook tutorial for Image Classification with PyTorch.
https://github.com/z-fran/pytorch-cifar10-image-classification

cifar10 image-classification pytorch resnet vision-transformer

Last synced: 3 months ago
JSON representation

A template and Jupyter Notebook tutorial for Image Classification with PyTorch.

• Host: GitHub
• URL: https://github.com/z-fran/pytorch-cifar10-image-classification
• Owner: Z-Fran
• Created: 2024-11-13T14:20:00.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2024-11-16T13:36:31.000Z (7 months ago)
• Last Synced: 2025-01-19T14:43:02.928Z (5 months ago)
• Topics: cifar10, image-classification, pytorch, resnet, vision-transformer
• Language: Jupyter Notebook
• Homepage:
• Size: 171 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Pytorch-CIFAR10-Image-Classification

This can be a template and Jupyter Notebook tutorial for Image Classification with PyTorch.

Highlights:

+ Customize your data augmentation process based on the notebook.

+ Define your own dataset based on the notebook.

+ Define your own deep learning model referring to classic ResNet and ViT models.

+ Provide you unified train and test pipleline.

+ Provide you various popular metrics.

### 1. Data preprocessing & augmentation

- RandomResizedCrop: and RandomHorizontalFlip are useful.

- RandomRotation, RandomVerticalFlip and Normalizea are unuseful.

### 2.Define the model

- ResNet

  - Classic CNN network.

  - Very easy to achieve 90%+ score on this dataset.

- ViT

  - Use transformer framework on vision tasks.

  - Training more slowly than ResNet.

### 3. Pipelines

- Design unified pipelines to train model and test on test dataset.

### 4. Train model and hyperparameters tuning

- ResNet

  - larger framework is not useful.

  - Adam and StepLR is better.

  - larger learning rate is batter (1e-3 > 1e-4).

- ViT

  - Because of slow convergence speed, MultiStepLr can set a large learning rate in the beginning to accelarate training.

  

### 5. Evaluation

  - Compare ResNet and ViT with following metrics:

    - Precision

    - Recall

    - F1 score

    - Overall Accuracy

    - ROC curve

    - AUC

    - Confusion matrix

- Accuracy and AUC of Resnet are higher than ViT.

- Both of models are more likely to confuse airplane and dog.

### Ensemble Learning

- Using Ensemble Learning method, use Resnet and ViT vote for the pridiction result.

- Weights of them is Resnet:ViT = 3:1.