Skip to content

Latest commit

 

History

History
20 lines (16 loc) · 575 Bytes

README.md

File metadata and controls

20 lines (16 loc) · 575 Bytes

movie-recommender

Practice of building personalized movie recommenadation service using Flask and Spark Referrence: A scalable on-line movie recommender using Spark and Flask

Usage

Submit Spark job

spark-submit --master 'local[*]'  --total-executor-cores 14 --executor-memory 6g  server.py

Get top ratings given user_id

http://0.0.0.0:<int:port>/<int:user_id>/ratings/top/<int:count>

Get ratings for moive_id given user_id

http://0.0.0.0:<int:port>/<int:user_id>/ratings/<int:movie_id>