Installation

This software is composed of two parts: a python package pastis, and two C++ programs, MDS_all, PM_all, based on the work of Duan and others. The installation is in two steps: first the python module pastis, then the two programs MDS_all and PM_all.

pastis

This package uses distutils, which is the default way of installing python modules.

The dependencies are:

  • python (>= 2.6)
  • setuptools
  • numpy (>= 1.3)
  • scipy (>= 0.7)
  • scikit-learn (>= 0.13)
  • argparse if python < 2.7

All of these dependencies can be installed at once using Anaconda

In addition, pyipopt is needed for pm2.

To install in your home directory, use:

python setup.py install --user

or using pip:

pip install --user pastis

To install for all users on Unix/Linux:

python setup.py build
sudo python setup.py install

or using pip:

pip install pastis

This will install a python package pastis, and four programs pastis-mds, pastis-nmds, pastis-pm1 and pastis-pm2. Calling any of those four programs will display the help.

MDS_all/PM_all

You also need to compile the two softwares in src/MDS and src/PM, and place them in the directory of your choice.

  • First install IPOPT: https://projects.coin-or.org/Ipopt IPOPT can be installed anywhere. IPOPT depends on several external packages that are not included directly with IPOPT. IPOPT requires at least one linear solver for sparse symmetric indefinite matrices. You have the choice between HSL, MUMPS or Pardiso. The makefile for MDS and PM are written to use with the HSL subroutines: you will need to edit the makefiles if you choose to use another linear solver.
  • You have to edit the file Makefile in both the src/MDS and src/PM, and set the following 3 variables at the top: IPOPTPATH, IPOPTINCDIR, IPOPTLIBDIR.
  • Type make.