sgd: Stochastic Gradient Descent for Scalable Estimation

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

Version: 0.1
Imports: Rcpp (≥ 0.11.3), MASS
LinkingTo: Rcpp, RcppArmadillo, BH
Published: 2015-05-08
Author: Dustin Tran [aut, cre], Tian Lian [aut], Panos Toulis [aut], Ye Kuang [ctb], Edoardo Airoldi [ctb]
Maintainer: Dustin Tran <dtran at>
License: GPL-2
NeedsCompilation: yes
Materials: README
CRAN checks: sgd results


Reference manual: sgd.pdf
Package source: sgd_0.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: sgd_0.1.tgz, r-oldrel: sgd_0.1.tgz
OS X Mavericks binaries: r-release: sgd_0.1.tgz