sgd: Stochastic Gradient Descent for Scalable Estimation

A fast and flexible set of tools for large scale estimation. It features many stochastic gradient methods, built-in models, visualization tools, automated hyperparameter tuning, model checking, interval estimation, and convergence diagnostics.

Version: 1.1
Imports: ggplot2, MASS, methods, Rcpp (≥ 0.11.3)
LinkingTo: BH, bigmemory, Rcpp, RcppArmadillo
Suggests: bigmemory, gridExtra, R.rsp, testthat
Published: 2016-01-05
Author: Dustin Tran [aut, cre], Panos Toulis [aut], Tian Lian [ctb], Ye Kuang [ctb], Edoardo Airoldi [ctb]
Maintainer: Dustin Tran <dustin at cs.columbia.edu>
BugReports: https://github.com/airoldilab/sgd/issues
License: GPL-2
URL: https://github.com/airoldilab/sgd
NeedsCompilation: yes
Materials: README
CRAN checks: sgd results

Downloads:

Reference manual: sgd.pdf
Vignettes: Stochastic gradient decent methods for estimation with large data sets
Package source: sgd_1.1.tar.gz
Windows binaries: r-devel: sgd_1.1.zip, r-release: sgd_1.1.zip, r-oldrel: sgd_1.1.zip
OS X El Capitan binaries: r-release: sgd_1.1.tgz
OS X Mavericks binaries: r-oldrel: sgd_1.1.tgz
Old sources: sgd archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=sgd to link to this page.