Skip to content

Latest commit

 

History

History
36 lines (19 loc) · 894 Bytes

README.md

File metadata and controls

36 lines (19 loc) · 894 Bytes

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This package is deprecated and no longer maintained!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)