«  Welcome   ::   Contents   ::   Installation  »

Requirements

The standard requirements for MLZ are the following python libraries:

  • numpy
  • scipy
  • matplotlib
  • mpi4py (for parallel running)
  • healpy (for spherical coordinates) optional but recommended
  • f2py (a fortran wrapper) optional but recommended
  • pyfits (for storage) optional but recommended

Parallel MLZ

MLZ can run on a single node but it is strongly recommended to install MPI libraries and mpi4py.

In order to run in parallel MPI libraries must be present and the mpi4py module which can be obtained from http://mpi4py.scipy.org/ . A first you might get it easy by trying:

[sudo] easy_install install mpi4py

ot via pip:

[sudo] pip install mpi4py

with sudo permissions, or locally using:

easy_install --user install mpi4py

or:

pip install mpi4py --user

«  Welcome   ::   Contents   ::   Installation  »