Skip to content

A Nonnegative Matrix Factorization algorithm based on alternating least squares and coordinate descent.

Notifications You must be signed in to change notification settings

panlanfeng/bignmf

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

bignmf

##Warning

This package is deprecated and no longer maintained!

This R package intends to solve Nonnegative Matrix Factorization efficiently. The algorithm bases on alternating least squares while solving nonnegative constrained regression via coordinate descent. The coordinate descent method is modified from glmnet (See Regularization Paths for Generalized Linear Models via Coordinate Descent, by J Friedman et al. 2010).

The inner part of the algorithm is implemented in C++. This package is currently tested only on Ubuntu.

##How to install

In Linux teminal, go to some directory:

cd  ~/mypackages/ 

Download it with:

git clone https://github.com/panlanfeng/bignmf.git

Build with:

R CMD build bignmf

Then in R:

install.packages("~/mypackages/bignmf_0.1.tar.gz", repos = NULL, type = "source")   
library(bignmf)  

About

A Nonnegative Matrix Factorization algorithm based on alternating least squares and coordinate descent.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published