Skip to content

Latest commit

 

History

History
17 lines (14 loc) · 964 Bytes

README.md

File metadata and controls

17 lines (14 loc) · 964 Bytes

Deploying a recommender system on the cloud - Sourcecode

This directory contains the code of the algorithm and API configuration files.

Libraries

  • Scikit-Surprise: Used to train the recommender model and to make predictions of the ratings.
  • Flask: Used to build the REST API method for generating a predicted rating given a user and a movie id.
  • Nginx: Used as a front-end reverse proxy
  • Gunicorn: Application server to communicate with the Flask application.

Code files

  • lambda_function.py: Code for the handler function of AWS Lambda.
  • nginx.conf: Configuration of the reverse proxy.
  • predictor.py: Flask application with the API methods for predicting movie ratings with the loaded model.
  • serve: Scoring service shell. Generally there is no need to modify it.
  • train: Algorithm for training the recommender model after loading the ratings dataset.
  • wsgi.py: Wrapper for Gunicorn to find the app.