User’s Guide

This page contains information on how to install DSO699-SpLCM and how to build the documentation locally.
Author

Yichen Zhou

Published

2025-05-21

Usage

Dependency

The package uses nbdev, numpy, torch, matplotlib, cvxpy, and rpy2 (which requires an R installation).

Besides standard python packages, the package rely on the R package clime1, which can be installed by running the following command in R console:

$ install.packages("clime")

Installation

In the root folder of the exported package, run:

$ pip install -e .

Compiling the documentation locally

In the root folder of the package, run:

$ make docs
$ chmod +x preview.sh
$ ./preview.sh

Note chmod only need to be run once.

Using the package

Simply write:

$ import DSO699.SpLCM.core as splcm

and then we can for example do

$ splcm.SpLCM_PGD(np.identity(60))

and the complete list of callable functions can be seen in the documentation.

Footnotes

  1. T. Cai, W. Liu, and X. Luo, ‘A Constrainedℓ1Minimization approach to sparse precision matrix estimation’, J. Am. Stat. Assoc., vol. 106, no. 494, pp. 594–607, Jun. 2011.↩︎