irlba: Fast and memory efficient partial SVD for R
The implicitly restarted Lanczos bidiagonalization algorithm (IRLBA) of
Jim Baglama and Lothar Reichel is a state of the art method for computing
a few singular vectors and corresponding singular values of huge matrices.
Code available from:
http://rforge.net/irbla.
The R package is on CRAN:
http://cran.r-project.org/web/packages/irlba/index.html.
The vignette documentation is here:
irlba.pdf.
Here is a quick video presentation of the package that uses the Netflix
prize data set:
And the slides used in the video can be found here:
http://goo.gl/fRech.
Enjoy!