Skip to main content

Python wrapper for sparseSEM

Project description

Elastic Net for Structural Equation Models (SEM)

Anhui Huang | Ph.D. Electrical and Computer Engineering

https://scholar.google.com/citations?user=WhDMZEIAAAAJ&hl=en

Summary

Provides elastic net penalized maximum likelihood for structural equation models (SEM). The package implements lasso and elastic net (l1/l2) penalized SEM and estimates the model parameters with an efficient block coordinate ascent algorithm that maximizes the penalized likelihood of the SEM. Hyperparameters are inferred from cross-validation (CV). A Stability Selection (STS) function is also available to provide accurate causal effect selection.

The experimental study and vignettes are also available in the doc/ folder in the package.

PyPI installation

sparseSEM is available on PyPI: https://pypi.org/project/sparseSEM/. Run command pip install sparseSEM to install from PyPI.

test/ folder contains examples using data packed along with this package in data/ folder. To run test/ examples, clone this repo, and run from test/ directory.

Configuration

This package was originally developed to leverage high performance computer clusters to enable parallel computation through openMPI. Users who have access to large scale computational resources can explore the functionality and checkout the openMPI module in this package.

Current package utilizes blas/lapack for high speed computation. To build the C/C++ code, the intel OneMKL library is specified in the package setup.

Release Note

  • V2.0: add more output information include CV results, hyperparameter, and details of model fit. V2.0 is a major release with stability selection added.
  • V1: initial release with corresponding to R package v2.

Package for other platforms

R package

An R package for sparseSEM is also available at CRAN: https://cran.r-project.org/web/packages/sparseSEM/index.html

OpenMPI

C/C++ implementation of sparseSEM with openMPI for parallel computation is available in openMPI branch (https://github.com/anhuihng/pySparseSEM/tree/openMPI).

Reference

- Huang A. (2014) Sparse Model Learning for Inferring Genotype and Phenotype Associations. Ph.D Dissertation,
University of Miami, Coral Gables, FL, USA.
- Huang A. (2014) sparseSEM: Sparse-Aware Maximum Likelihood for Structural Equation Models. Rpackage
(https://cran.r-project.org/web/packages/sparseSEM/index.html)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sparseSEM-2.0.tar.gz (4.9 MB view hashes)

Uploaded Source

Built Distribution

sparseSEM-2.0-cp310-cp310-macosx_10_9_x86_64.whl (109.3 kB view hashes)

Uploaded CPython 3.10 macOS 10.9+ x86-64

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page