DiscJockey

We setout to build a content based recommender system by training and selecting an ML classification model that could determine whether a song is fit enough for your party preferences. We encapsulated this recommender in a DiscJockey class that can take your party preferences and use them to determine whether other songs are fit enough for the party of your liking.

For info on how we created and selected our model look at models.ipynb For the Disc Jockey class look at discjockey.py For all the data we used to train and test our recommender look at the musicdata folder To test out our recommender look at main.py

Note: In order to enter a track id: Go to Spotify, click on the three dots next to the song, navigate to share, and then copy Spotify URI and take the end digits/numbers to input as a track id.

In order to enter a playlist: Go to Spotify, click on the three dots next to the playlist, navigate to share, and then copy Spotify URI and take the end digits/numbers to input as a playlist id.

Built With

• anaconda
• jupyter-notebook
• python
• scklean
• spotipy