---
title: Radio Galaxy Classifier
emoji: 🌌
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: "6.20.0"
python_version: "3.12"
app_file: app.py
pinned: false
---

# Radio Galaxy Classifier

A web application for classifying radio galaxy morphology into:

- FRI
- FRII
- Compact
- Bent

Built with:
- PyTorch
- Gradio
- Docker

## Live Demo

[Hugging Face Spaces](https://huggingface.co/spaces/huma-03/radio-galaxy-classifier)

## Features

- Upload a radio galaxy image
- Predict morphology
- Confidence score
- Grad-CAM visualization
- Morphology explanations

## Model

The model was trained on the RadioGalaxyDataset containing 2,158 FIRST survey images.

For the complete training pipeline, experiments, and implementation details, see the companion repository:

https://github.com/huma-m/radio-galaxy-classifier

## Local Run
```
pip install -r requirements.txt
python app.py
```