Skip to content

This is an implementation of a feed-forward neural network trained on data supplied from a users Spotify library to allow intelligent song recommendation. We used a multilayer perceptron with back-propagation.

Notifications You must be signed in to change notification settings

nick-bigger/SpotifySongRecommender

Repository files navigation

SpotifySongRecommender

This is an implementation of a neural network to create better playlists based on training data supplied from a users Spotify library.

Getting Started

First, use the package manager pip to install Spotipy. Then, create a Spotify Developer Account and retrieve your unique client id and secret. Next, pick a redirect uri for the API to redirect to after you've been logged in. http://localhost/ will work fine. Set these values as enviornmental variables before you run.

brew install python3
pip3 install spotipy
pip3 install sklearn
pip3 install pandas

export SPOTIPY_CLIENT_ID='your-spotify-client-id'
export SPOTIPY_CLIENT_SECRET='your-spotify-client-secret'
export SPOTIPY_REDIRECT_URI='your-app-redirect-url'

Usage

python3 connectSpotify.py <username>

Authors

Acknowledgments

  • Thank you to the Spotipy team for use of their product, as well as the use of sklearn and pandas.
  • Additionally, thank you to Wes Doyle and his YouTube series on how to do machine learning in Python.

About

This is an implementation of a feed-forward neural network trained on data supplied from a users Spotify library to allow intelligent song recommendation. We used a multilayer perceptron with back-propagation.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages