pbugnion / scikit-monaco

Python routines for Monte Carlo integration

scikit-monaco

scikit-monaco is a library for Monte Carlo integration in python. The core is written in Cython, with process-level parallelism to squeeze the last bits of speed out of the python interpreter.

A code snippet is worth a thousand words. Let's look at integrating sqrt(x**2 + y**2 + z**2) in the unit square:

>>> from skmonaco import mcquad
>>> from math import sqrt
>>> result, error = mcquad(
...     lambda xs: sqrt(xs[0]**2+xs[1]**2+xs[2]**2),
...     npoints=1e6, xl=[0.,0.,0.], xu=[1.,1.,1.])
>>> print "{} +/- {}".format(result,error)
0.960695982212 +/- 0.000277843266684

Links

• Home page: https://pypi.python.org/pypi/scikit-monaco
• Documentation: http://scikit-monaco.readthedocs.org/en/latest/
• Source code: https://github.com/scikit-monaco/scikit-monaco
• Issues: https://github.com/scikit-monaco/scikit-monaco/issues

Installation

From Pypi

The easiest way to download and install scikit-monaco is from the Python package index (pypi). Just run:

$ python easy_install scikit-monaco

Or, if you have pip:

$ pip install scikit-monaco

From source

Clone the repository using:

$ git clone https://github.com/scikit-monaco/scikit-monaco.git

And run:

$ python setup.py install

in the project's root directory.

Testing

After the installation, run $ python runtests.py in the package's root directory.

Issue reporting and contributing

Report issues using the github issue tracker.

Read the CONTRIBUTING guide to learn how to contribute.