Skip to content

README.md

Latest commit

 

History

History
35 lines (24 loc) · 1.27 KB

README.md

File metadata and controls

35 lines (24 loc) · 1.27 KB

🌊 RippleWorks – AquaPredict Pro

An interactive Water Quality Intelligence System built with machine learning and Streamlit.
This project predicts Water Quality Index (WQI) (numeric score) and Water Quality Classification (WQC) (Excellent / Good / Medium / Poor / Very Poor) based on water parameters.

image

Model notebook of this project is in private repo!

Features

  • End-to-end pipeline: cleaning → feature engineering → modeling → deployment
  • Handles missing values (KNN/median strategies) and outliers (IQR capping + domain thresholds)
  • Computes WQI dynamically and categorizes into classes (WQC)
  • Supports both regression (WQI prediction) and classification (WQC prediction)
  • Deployed as a sleek Streamlit web app

Tech Stack

  • Python (Pandas, NumPy, Scikit-learn, XGBoost)
  • Streamlit for web UI
  • Joblib for model persistence
  • GitHub + Streamlit Cloud for deployment

Dataset

Source: Kaggle – Indian River Water Quality dataset

  • 1991 rows, 8 key water quality parameters
  • Target variables engineered: WQI and WQC