New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Compress lsa #1

Merged
merged 11 commits into from Jul 1, 2018

Conversation

Projects
None yet
1 participant
@dougturnbull
Contributor

dougturnbull commented Jul 1, 2018

Final Submission

dougturnbull added some commits May 12, 2018

Updated DataManager to create matrices without using rare tracks.
Added top album tracks, and top artist tracks.
Created LSA with limited corpus of tracks before sending to LSA
Need to reimplement so that Data Manager loads data matrices X
and lsa uses a column mask to reduce the number of tracks in corpus
Refactored DataManager and predict_with_LSA so all LSA parameters are…
… now in LSA code. Useful for running multiple experiements without reloading data.

Cleaned up LSA, related top tracks for artists and albums, and overall popularity as a linear combination.
Next up: prepare for submission with new code, then languague model.

@dougturnbull dougturnbull merged commit 9493b85 into master Jul 1, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment