README.md

---
title: Energy ML Prediction System
emoji: ⚡
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.42.0
app_file: app.py
pinned: false
license: mit
---

# Energy ML Prediction System

AI-powered energy consumption prediction and threshold exceedance detection.

## Models

- **Threshold Detection**: Random Forest classifier (AUC = 0.94)
- **Energy Prediction**: Random Forest & XGBoost regressors
- **Performance**: R² up to 0.72, MAE as low as 0.53 MWh

## Features

- Real-time predictions via simple JSON interface
- Multiple model comparison (RF vs XGBoost)
- Threshold exceedance probability (8.3 and 9.0 MWh)
- Secure access with authentication

## Usage

1. Select your model
2. Input JSON configuration
3. Get instant predictions

## Authentication

Login required for access.
- Username: admin
- Password: energy123

---
*Built with Gradio • Powered by scikit-learn & XGBoost*