Movie recommender system in a stream lit app

This github is part of the data science bootcamp at WBS. The project goal was recommend movies based on popularity, user or movies similar to a selected movie using a modified version of the movielens 100k data.

1. setting up the repository locally

1.1 install dependencies

It's recommended to create a new virtualenv or conda environment and install the required packages inside that environment.

To install the required packages please run:

pip install -r requirements.txt

1.2 setting up your environment variables

please start with making a copy of the .env-dist file and rename it to .env. Inside the created .env file you will need to add your API from the movie database.

getting an tmdb API key

You will need to create an account on the movie database website by clicking on the join TMDB button in the top right.

after signing up, log into your account and go to settings

in settings go to API

This screenshot isn't accurate! I already requested my API key, you will likely see something like along the lines of request API key.

During the API key request you will asked about the purpose, here I selected Education, and to give a Application url. Because I had no intention of actually deploying it, I added here http://localhost:7000.

You should get a confirmation email regarding your API key request and now you can find your API Key and API Read Access Token under the API tab of your accounts settings (as seen in the screenshot above).

Copy and paste the API Read Access Token into your created .env file

1.3 getting the data and models

downloading the data

create a data folder and place inside download the data

download the models

create a model folder and place inside download the models

2. Running the streamlit app

I run the app from inside vscode using the run Start debugging (F5 key) or run without debugging (Ctrl + F5) command.

Running the streamlit app from the terminal caused Invalid API key errors as the API variables in the .env file aren't accessed.