pyEntropy requires a recent version of Python and NumPy. Any Python >2.5 (but < 3.0) and NumPy > 1.3 should be fine, but the most recent releases are recommend. If you have any problems please email (or open an issue). From version 0.5.0, C/C++ implementations of the NSB and BUB methods (from the Spike Train Analysis Toolkit) are included. These require a compiler, and also the Gnu Scientific Library (GSL). On Linux, you should be able to install these through your package manager. E.g. for Ubuntu:
sudo aptitude install build-essential python-dev libgsl0-dev
should provide all the requirements. On a Mac, XCode and (with MacPorts installed):
sudo port install gsl
should be enough. On Posix platforms (Mac/Linux) you should have the gsl-config program in your path. For Windows, I am currently investigating the best way, and will hopefully provide binaries.
Note
The pyentropy.statk.wrap module is optional. I intend to keep pyEntropy installable and usable as a pure Python package with minimal dependencies. If the build fails for any reason, warnings will be displayed, but the pyEntropy package should still be installable. The nsb and bub entropy methods will be unavailable but if you have the nsb-entropy program from http://nsb-entropy.sourceforge.net/ on your path you should still be able to use the nsb-ext method.
The pyentropy.maxent module also requires SciPy.
For windows, running the installer should be all that is needed. On other platforms, uncompress the archive and run the following command:
python setup.py install
This will install the package to the appropriate location of your Python installation. This package uses distutils; more details are available in the python documentation. Note that from Python 2.6 you can run:
python setup.py install --user
to install to .local directory in the home directory. This is automatically added to PYTHONPATH and is the recommended way to install for a single user on UNIX systems (without root access).
Note
You can test your installation by running pyentropy.test(). This runs a series of unit tests found in the tests directory. Since the tests of the NSB method can take a long time (around 2 minutes on a fast computer) they are not run by default, but can be included by running pyentropy.test('full').
pyentropy.maxent stores generated transformation matrices for a given set of parameters are to disk. These files can get large so you should be aware that they are there. The default location for the cache is a .pyentropy (_pyentropy on windows) directory in the users home directory. To override this and use a custom location (for example to share the folder between users) you can put a configuration file .pyentropy.cfg (pyentropy.cfg on windows) file in the home directory with the following format:
[maxent]
cache_dir = /path/to/cache
pyentropy.maxent.get_config_file() will show where it is looking for the config file.