Wangelev/southafrica-water-quality

data science project on quality of water in south africa # South Africa Water Quality Prediction This repository contains a machine learning pipeline for predicting water quality indicators in South Africa using environmental, Landsat, and TerraClimate data. The project demonstrates **data processing, feature engineering, model training, and evaluation**. The dataset spans a five-year period from 2011 to 2015. Using API-based data extraction methods, both Landsat and TerraClimate features were retrieved directly from the Microsoft Planetary Computer portal. These combined spectral, index-based, and climatic features were used as predictors in a regression model to estimate three key water quality parameters: Total Alkalinity (TA), Electrical Conductance (EC), and Dissolved Reactive Phosphorus (DRP). --- ## Project Structure south_africa-water_quality-ml/ │ ├── data/ │ ├── raw/ # Raw datasets (not included in repo due to size) │ ├── processed/ │ └── example/ │ ├── src/ │ ├── data_processing.py │ ├── feature_engineering.py │ ├── train_model.py │ └── evaluate_model.py │ ├── models/ │ └── trained_models/ │ ├── .gitignore └── README.md --- ## Setup 1. **Clone the repository** ```bash git clone cd south_africa-water_quality-ml python -m venv .venv source .venv/bin/activate # Linux/macOS .venv\Scripts\activate # Windows pip install -r requirements.txt # Running the pipeline python src/data_processing.py python src/feature_engineering.py python src/train_model.py python src/evaluate_model.py