https://github.com/nakjun/hearing-loss-classification-vit

https://github.com/nakjun/hearing-loss-classification-vit

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/nakjun/hearing-loss-classification-vit
• Owner: nakjun
• Created: 2024-06-19T07:02:27.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-06-19T07:38:17.000Z (12 months ago)
• Last Synced: 2025-01-23T06:29:54.949Z (5 months ago)
• Language: Python
• Size: 48.8 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

        
# ViT Image Classification

This project implements an hearing loss  classification model using the Vision Transformer (ViT) architecture. The model is trained and evaluated on our custom dataset,

## Project Structure

- `dataset`: Directory containing training and testing data.

- `models`: Directory where trained model checkpoints are saved.

- `main.py` : Full Fine Tuning the ViT baseline model(`google/vit-base-patch16-224`)

## Track Training and Evaluation

- Training loss for each epoch

- Test accuracy for each epoch

- Accuracy graph

- Confusion matrix

## Dependencies

- torch

- torchvision

- transformers

- scikit-learn

- matplotlib

## Result of Evaluation

![Test Accuracy](./images/Figure_1.png)

![Test Confusion Matrix](./images/Figure_2.png)