Skip to content

Photon ML Tutorial

Alex Shelkovnykov edited this page Nov 1, 2017 · 12 revisions

The Photon ML Tutorial is a Jupyter notebook and web app developed to demonstrate how you can use generalized linear mixed models (GLMix) to personalize a recommendation system. We originally presented this tutorial at KDD 2016, and prepared the following introductory slides providing an overview of the recommender system at LinkedIn and how is GLMix implemented within Photon's GAME module:

Setup

Our tutorial runs in Docker. You will need to install Docker or Docker Toolbox on your system to use it. To install Docker, visit https://docs.docker.com/engine/installation/ and follow the instructions to download and install Docker for your operating system. After you have installed Docker, launch the Docker daemon (this happens automatically on some systems).

The next step is to install and launch the Photon ML Tutorial Docker image. In your terminal, run docker pull linkedin/photon-ml-tutorial. Launch a container for the tutorial image by running docker run -dit -p 5334:80 --name photon-ml-tutorial linkedin/photon-ml-tutorial. You can verify that the container is running using docker ps.

To access the Photon ML Tutorial, navigate to http://localhost:5334/tutorial. If you are using Docker Toolbox and the app does not load using the above URL, run docker-machine ls to get the ip for the tutorial app and navigate to http://<ip>:5334/tutorial. The tutorial contains detailed instructions on how to run it within the Jupyter notebook.

Teardown

Shut down the tutorial container using the command docker stop photon-ml-tutorial. Next, remove the tutorial container by running docker rm photon-ml-tutorial. You can also remove the tutorial image by running docker rmi linkedin/photon-ml-tutorial.

Video

Follow this link to view a recording of an older version of this tutorial presented at KDD 2016.

Screenshots

Jupyter Notebook Rating App

Clone this wiki locally