README.md

---
title: GERD Lightweight ViT Classifier
emoji: 🏥
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.0.0
app_file: app.py
pinned: false
---

# GERD Lightweight Vision Transformer Classifier

This is a lightweight Vision Transformer model for GERD classification.

## Model Architecture

- **Image size**: 224×224×3
- **Patch size**: 8×8
- **Projection dimension**: 64
- **Transformer layers**: 4
- **Attention heads**: 4
- **MLP head units**: [128, 64]

## Usage

Upload an image and the model will classify it into one of the predefined classes.

## Preprocessing

Images are:

1. Resized to 224×224
2. Converted to RGB
3. Normalized to [0, 1] range