manunarayana99/FLOOD_PREDICTOR

🌊 Flood Peak Discharge Predictor

Predict annual peak flood discharge based on key climatic and geographical features using a KNN Regressor. Built using Streamlit, this project demonstrates how machine learning can be applied to environmental risk assessment.

🔍 About the Project

Floods are among the most devastating natural disasters, and early prediction plays a crucial role in minimizing damage. This tool takes in 25+ climate and terrain features and predicts the expected flood peak discharge (m³/s).

⚙️ Model Details

• —Model: KNN Regressor
• —Input Features:
• —Climate data: temperature, precipitation, seasonal variations
• —Geographical data: drainage area, flow length, basin metrics
• —Target Variable: Annual Peak Discharge

📌 Features

✅ Predict flood discharge based on input parameters ✅ Gauge-style risk visualization ✅ Dark-themed interactive interface ✅ Instant in-browser predictions with Streamlit ✅ Clean, responsive layout

🖥️ UI Snapshot

🧪 How to Use

1. 1.Launch the app on Hugging Face 🤗
2. 2.Enter the required climate and terrain parameters in the sidebar
3. 3.Click "Predict Flood Discharge"
4. 4.View the predicted value + flood risk insight

🚀 Tech Stack

• —Python
• —Streamlit
• —Scikit-learn
• —Pandas, NumPy
• —Matplotlib (for custom gauge bar)

🧠 Example Inputs

📈 Prediction Output

• —🔵 Discharge: 1750.25 m³/s
• —🧭 Risk Insight: High flood potential based on discharge

🧠 Future Improvements

• —Integrate real-time weather API
• —Add regional flood classification system
• —Expand dataset to include temporal flood patterns

🤝 Let's Collaborate!

Feel free to fork, extend, and improve the project. If you use this for your research, education, or deployment, please ⭐ the repo or mention the authors.

“Turn climate data into actionable insights. Predict floods, protect lives.”

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference