chrinide/glmnet_python

Glmnet for python

Contact

B. J. Balakumar [email protected]

Han Fang [email protected]

Install

Using pip (recommended)

pip install glmnet_python

Complied from source

git clone https://github.com/hanfang/glmnet_python.git
cd glmnet_python
python setup.py install

Requirement: Python 2/3, Linux

Currently, the checked-in version of GLMnet.so is compiled for the following config:

Linux: Linux version 2.6.32-573.26.1.el6.x86_64 (gcc version 4.4.7 20120313 (Red Hat 4.4.7-16) (GCC) ) OS: CentOS 6.7 (Final) Hardware: 8-core Intel(R) Core(TM) i7-2630QM gfortran: version 4.4.7 20120313 (Red Hat 4.4.7-17) (GCC)

Usage

import glmnet_python
from glmnet import glmnet

For more examples, see iPython notebook

Introduction

This is a python version of the popular glmnet library (beta release). Glmnet fits the entire lasso or elastic-net regularization path for linear regression, logistic and multinomial regression models, poisson regression and the cox model.

The underlying fortran codes are the same as the R version, and uses a cyclical path-wise coordinate descent algorithm as described in the papers linked below.

Currently, glmnet library methods for gaussian, multi-variate gaussian, binomial, multinomial, poisson and cox models are implemented for both normal and sparse matrices.

Additionally, cross-validation is also implemented for gaussian, multivariate gaussian, binomial, multinomial and poisson models. CV for cox models is yet to be implemented.

CV can be done in both serial and parallel manner. Parallellization is done using multiprocessing and joblib libraries.

During installation, the fortran code is compiled in the local machine using gfortran, and is called by the python code.

The best starting point to use this library is to start with the Jupyter notebooks in the test directory (iPython notebook). Detailed explanations of function calls and parameter values along with plenty of examples are provided there to get you started.

Authors:

Algorithm was designed by Jerome Friedman, Trevor Hastie and Rob Tibshirani. Fortran code was written by Jerome Friedman. R wrapper (from which the MATLAB wrapper was adapted) was written by Trevor Hastie.

The original MATLAB wrapper was written by Hui Jiang (14 Jul 2009), and was updated and is maintained by Junyang Qian (30 Aug 2013).

This python wrapper (which was adapted from the MATLAB and R wrappers) was originally written by B. J. Balakumar (5 Sep 2016), later modified by Han Fang.

B. J. Balakumar, [email protected] (Sep 5, 2016). Department of Statistics, Stanford University, Stanford, CA

Han Fang, [email protected] (Mar 1, 2017). Department of Applied Math & Statistics, Stony Brook University, Stony Brook, NY

REFERENCES:

• Friedman, J., Hastie, T. and Tibshirani, R. (2008) Regularization Paths for Generalized Linear Models via Coordinate Descent, http://www.jstatsoft.org/v33/i01/ Journal of Statistical Software, Vol. 33(1), 1-22 Feb 2010

• Simon, N., Friedman, J., Hastie, T., Tibshirani, R. (2011) Regularization Paths for Cox's Proportional Hazards Model via Coordinate Descent, http://www.jstatsoft.org/v39/i05/ Journal of Statistical Software, Vol. 39(5) 1-13

• Tibshirani, Robert., Bien, J., Friedman, J.,Hastie, T.,Simon, N.,Taylor, J. and Tibshirani, Ryan. (2010) Strong Rules for Discarding Predictors in Lasso-type Problems, http://www-stat.stanford.edu/~tibs/ftp/strong.pdf Stanford Statistics Technical Report