README.md

---
title: GoGenix MRI Brain Tumor Classifier
emoji: ⚡
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 6.8.0
python_version: '3.10'
app_file: app.py
pinned: true
license: mit
short_description: quick-predictions
---

# GoGenix MRI Brain Tumor Classifier

Fine-tuned version of `Falconsai/nsfw_image_detection` for Brain Tumor MRI classification.

## 🚀 Model Information
- **Base Model**: [Falconsai/nsfw_image_detection](https://huggingface.co/Falconsai/nsfw_image_detection)
- **Custom Model**: GoGenix_MRI_Brain
- **Number of Classes**: 4 (Glioma, Meningioma, No Tumor, Pituitary Tumor)
- **Dataset**: Brain Tumor MRI 4-class Dataset

## 📊 Dataset Information
The model is trained on a comprehensive Brain Tumor MRI dataset containing:
- **Glioma Tumors**
- **Meningioma Tumors** 
- **Pituitary Tumors**
- **No Tumor** (Healthy brains)

## 🎯 Usage
### Training
1. Navigate to the **"🚀 Train Model"** tab
2. Click **"Start Training"** to begin fine-tuning
3. Monitor the training progress in the status box

### Classification
1. Go to the **"🔍 Classify MRI"** tab
2. Upload a Brain MRI image
3. Click **"Classify"** to get tumor classification results

## 🔧 Technical Details
- **Framework**: PyTorch + Hugging Face Transformers
- **Image Size**: 224x224 pixels
- **Architecture**: Based on Vision Transformer (ViT)
- **Training Time**: ~10-30 minutes on T4 GPU

## 📁 Files Structure