Skip to content

Matrix factorization for data with very few observations

License

Notifications You must be signed in to change notification settings

dpmerrell/SparseMatFac.jl

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SparseMatFac.jl

Matrix factorization for rectangular array data with very few observations.

This is potentially useful for a variety of situations where you want to model pairwise interactions between entities:

  • Game scores between sports teams
  • Uses of products by customers (as in a recommender system)
  • Interactions between drugs and biological samples (as in a drug discovery setting)
  • etc.

WARNING: this package is under development -- its API is almost sure to change!

See Bracketology.jl for example usage of this package in a sports analytics application.

SparseMatFac.jl is intended to be the "sparse analog" of MatFac.jl.

About

Matrix factorization for data with very few observations

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages