nickhand / classylss Goto Github PK

A lightweight Python binding of the CMB Boltzmann code CLASS, with an emphasis on the routines that are important for large-scale structure calculations. The main modules of the CLASS code are exposed to the user via a Cython wrapper.

Installation instructions, examples, and API reference are availabe on ReadTheDocs: http://classylss.readthedocs.io/.

The package is lightweight and the only dependencies are:

• numpy
• cython

The CLASS code will automatically be downloaded and compiled, and is thus, not an external dependency for the user. However, the user will need a valid C compiler to compile the CLASS code. The version of CLASS compiled by the code is stored in the variable classylss.class_version.

The package can be installed as part of the Anaconda package manager using

conda install -c bccp classylss

The package can also be installed via the pip command

pip install classylss

The package can be also be downloaded from github using

git clone https://github.com/nickhand/classylss.git
cd classylss

If CLASS is not built succesfully, the user can edit the default configuration variables in depends/class.cfg, which are used when building the CLASS library.

To verify that the installation has succeeded, run:

import classylss

See the tests of the code in classylss/tests/ for examples of using each of the main CLASS modules.

Additional features of the CLASS code that are not yet implemented can be exposed via the Cython wrapper relatively easily. We encourage users to open up a GitHub issue with any feature requests.