https://github.com/ruuda/deepnote

Generate track embeddings based on listening history
https://github.com/ruuda/deepnote

Last synced: 12 months ago
JSON representation

Generate track embeddings based on listening history

• Host: GitHub
• URL: https://github.com/ruuda/deepnote
• Owner: ruuda
• License: gpl-3.0
• Created: 2018-12-16T22:16:01.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2018-12-26T21:01:43.000Z (over 7 years ago)
• Last Synced: 2025-03-30T10:30:22.258Z (about 1 year ago)
• Language: Python
• Size: 22.5 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md
  • License: license

Awesome Lists containing this project

README

          
# Deepnote

Generate embeddings of music tracks, albums, and artists, based on listening

histories, for similarity comparisons and ultimately music recommendation.

Basically, word2vec applied to listening histories.

There is nothing deep about the model; in fact it is extremely shallow. But

“deepnote” sounds more catchy than “shallownote”.