How aman-mitra/house-pricing-prediction-model works
Project type and purpose
• This is a machine-learning-powered web application: a Flask-based service that exposes a house-pricing prediction model (regression) via a simple HTML front end.
• The Jupyter notebook (Linear_Reg.ipynb) contains the offline model training; the app uses pickled artifacts (regmodel.pkl, scaling.pkl) for inference.
File structure
• app.py: Flask application entry point, handles HTTP requests, loads scaler and model, renders templates.
• templates/home.html: front-end form for collecting input features; displays predicted price.
• regmodel.pkl & scaling.pkl: serialized scikit-learn model and standard scaler.
• Linear_Reg.ipynb: data exploration and model training pipeline.
• requirements.txt: Python dependencies (Flask, scikit-learn, etc.).
• Dockerfile: containerizes the Flask app for local testing or deployment.
• Procfile: declares Heroku process type (web: gunicorn app:app).
• .github/workflows/main.yaml: CI pipeline (build, test, container build, deploy).
Final advice
• Be detailed: include each file/artifact group (templates, pickles, Dockerfile, Procfile, workflows).
• Both the runtime request flow and the code-to-deploy pipeline.
• Keep it clear: group related elements, label data formats (HTTP, JSON, pickle), and use consistent iconography for services.