Skip to content

Latest commit

 

History

History
26 lines (20 loc) · 1018 Bytes

README.md

File metadata and controls

26 lines (20 loc) · 1018 Bytes

SpotifAI

🎵 PlayList Curator Machine Learning Model

Often times, music can run dry after repeated listens. Thankfully, with SpotifAI, we can make you a playlist we KNOW you will like. What are you waiting for? Tell us some of your favorite songs!

More information can be found in this README.md

Current Status

Still a Work In Progress! Start Date: 1/25/22

Description

🔨 How I built it

  • Python
  • NumPy
  • Pandas
  • Scikit-learn
  • Dash
  • Spotify API

⚙️ What I did

  • Constructed an app with a Dash interface that allows users to input at least 5 favorite songs & in turn create custom target playlists catered for the user
  • Extracted user input searches from the Spotify API into a Pandas DataFrame that contains song traits to be spliced, filtered, and fed to ML models
  • Trained models with sklearn to execute algorithms that learn what songs the user likes based on the danceability, acousticness, energy, etc of the songs