Skip to content

ChiroK0407/Corrosion-predictor

Repository files navigation

🔩 Corrosion Predictor Dashboard

Welcome to the Corrosion Intelligence Platform — a research-driven tool that blends ISO 9223 standards with machine learning models to estimate corrosion rates.

🚀 Features

  • 🌍 Fetches weather & pollution data automatically via APIs
  • ⚙️ Combines ISO logic + ML (SVR, Random Forest, XGBoost)
  • 📊 Visualizes predictions with charts & maps
  • 🏛️ Benchmarks against CECRI archival datasets
  • 🔍 Offers explainability via feature importance plots

🧭 Navigation

  • Model 1 (API-driven) → Predict corrosion rates using live weather & pollution data
  • Model 2 (CECRI archival) → Train & benchmark ML models against CECRI datasets
  • Map Views → Visualize corrosion intensity across regions

📖 References & Standards

  • ISO 9223: Corrosivity of atmospheres
  • ISO 9224: Guiding values for corrosion rates
  • Indian Standard IS 1786:2008
  • CECRI archival datasets

🛠️ Tech Stack

  • Streamlit (frontend)
  • FastAPI (backend services)
  • XGBoost, Random Forest, SVR (ML models)
  • Plotly, Matplotlib (visualizations)

▶️ Running Locally

git clone https://github.com/<your-username>/Corrosion-predictor.git
cd Corrosion-predictor
pip install -r requirements.txt
streamlit run frontend/Home.py

About

A Streamlit dashboard for corrosion prediction using ISO standards, ML models, and CECRI archival datasets.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors