How tejas1730/movie-recommandation-ai works
Project type and purpose
• This is a small-scale, full-stack machine-learning-driven web application (“Movie Recommendation System”).
• It combines a data pipeline and ML model (in a Jupyter notebook) with a lightweight frontend server (Streamlit) to deliver personalized movie recommendations.
File structure
• Top-level files:
• app.py – Streamlit app that serves predictions
• movierecommend.ipynb – notebook containing data loading, preprocessing, model training/evaluation
• movies.csv – raw dataset of movie metadata and ratings
• images/ – static assets (movie poster images)
• .gitattributes & README.md – repo metadata and documentation
• No separate “backend” directory, but the notebook is the training pipeline and app.py is the serving component.
README for additional insights
• Confirms Python 3.x, pandas, numpy, scikit-learn, Streamlit, Jupyter Notebook
• Describes high-level flow: data → ML algorithm → model → user-facing recommendation
• Mentions use of Git LFS for large artifacts (e.g., trained model pickles)
Detail without overthinking
• Every file-backed component (CSV, .ipynb, images folder)
• Clearly distinguish offline training vs. online serving
• How the model artifact bridges those two worlds