How nodirbek-ai/movie-recommender-ai works
This is a command-line movie recommender. Its data-preparation workflow fetches popular movie metadata from TMDB, saves a local catalog, and builds text embeddings; the CLI loads those embeddings and supports natural-language search and finding films similar to a title. The graph separates catalog ingestion, embedding preparation, and interactive recommendation. Source excerpts show these stages and their file-based handoffs; the README is empty, so no further documented capabilities are available.
Main entry points: build_embeddings.py Embedding builder · recommender.py Recommendation engine · fetch_movies.py Movie fetcher
Components 9
Catalog ingestion
- Movie catalog
src/fetch_movies.py - Movie fetcher
src/fetch_movies.py
Embedding preparation
- Embedding builder
src/build_embeddings.py - Embedding data
src/build_embeddings.py
Interactive recommendations
- Menu interface
src/recommender.py - Recommendation engine
src/recommender.py
Other components
- CLI user
- TMDB API
- Sentence Transformer
Connections 9
- CLI user → Menu interface: uses
src/recommender.py - Menu interface → Recommendation engine: queries
src/recommender.py - Movie fetcher → TMDB API: fetches popular films
src/fetch_movies.py - Movie fetcher → Movie catalog: writes JSON
src/fetch_movies.py - Embedding builder → Movie catalog: reads JSON
src/build_embeddings.py - Embedding builder → Sentence Transformer: encodes film text
src/build_embeddings.py - Embedding builder → Embedding data: writes JSON
src/build_embeddings.py - Recommendation engine → Embedding data: loads vectors
src/recommender.py - Recommendation engine → Sentence Transformer: embeds query
src/recommender.py