pyunicorn (Unified Complex Network and RecurreNce analysis toolbox) is a fully object-oriented Python package for the advanced analysis and modeling of complex networks. Above the standard measures of complex network theory such as degree, betweenness and clustering coefficient it provides some uncommon but interesting statistics like Newman's random walk betweenness. pyunicorn features novel node-weighted (node splitting invariant) network statistics as well as measures designed for analyzing networks of interacting/interdependent networks.

Moreover, pyunicorn allows to easily construct networks from uni- and multivariate time series data (functional (climate) networks and recurrence networks). This involves linear and nonlinear measures of time series analysis for constructing functional networks from multivariate data as well as modern techniques of nonlinear analysis of single time series like recurrence quantification analysis (RQA) and recurrence network analysis.

Stable releases, Development version

Changelog, Contributions

For extensive HTML documentation, jump right to the pyunicorn homepage. Recent PDF versions are also available.

On a local development version, HTML and PDF documentation can be generated using Sphinx:

$> pip install --user -e .
$> cd docs; make clean html latexpdf

pyunicorn relies on the following open source or freely available packages which have to be installed on your machine.

Required:

• Numpy 1.8+
• Scipy 0.14+
• Weave 0.15+
• igraph, python-igraph 0.7+

Optional (used only in certain classes and methods):

• PyNGL (for class NetCDFDictionary)
• netcdf4-python (for classes Data and NetCDFDictionary)
• Matplotlib 1.3+
• Matplotlib Basemap Toolkit (for drawing maps)
• mpi4py (for parallelizing costly computations)
• Sphinx (for generating documentation)
• Cython 0.21+ (for compiling code during development)

Stable release

Via the Python Package Index:

$> pip install pyunicorn

Development version

For a simple system-wide installation:

$> pip install .

Depending on your system, you may need root privileges. On UNIX-based operating systems (Linux, Mac OS X etc.) this is achieved with sudo.

For development, especially if you want to test pyunicorn from within the source directory:

$> pip install --user -e .

Before committing changes to the code base, please make sure that all tests pass. The test suite is managed by tox and configured to use system-wide packages when available. Thus to avoid frequent waiting, we recommend you to install the current versions of the following packages:

$> pip install tox nose networkx Sphinx
$> pip install pylint pytest pytest-xdist pytest-flakes pytest-pep8

The test suite can be run from anywhere in the project tree by issuing:

$> tox

To expose the defined test environments and target them independently:

$> tox -l
$> tox -e py27-units,py27-pylint

To test single files:

$> tests/test_doctests.py core.network      # doctests
$> nosetests -vs tests/core/TestNetwork.py  # unit tests
$> pylint pyunicorn/core/network.py         # code analysis
$> py.test pyunicorn/core/network.py        # style

Not implemented yet.

Please acknowledge and cite the use of this software and its authors when results are used in publications or published elsewhere. You can use the following reference:

J.F. Donges, J. Heitzig, B. Beronov, M. Wiedermann, J. Runge, Q.-Y. Feng, L. Tupikina, V. Stolbova, R.V. Donner, N. Marwan, H.A. Dijkstra, and J. Kurths, Unified functional network and nonlinear time series analysis for complex systems science: The pyunicorn package, Chaos 25, 113101 (2015), doi:10.1063/1.4934554, Preprint: arxiv.org:1507.01571 [physics.data-an].

pyunicorn is BSD-licensed (3 clause).