Installing pyqz

pyqz is available on pypi, which makes its installation easier than ever. In a terminal, type:

pip install pyqz

And that should take care of things.

The most recent release of pyqz is also available for download from its Github repository. Interested users can fork the pyqz repository if they want to get access to the latest updates not yet released. Push requests for bug fixes and new features are welcome and will be examined in detail.

Requirements

The following packages are required for pyqz to work properly:

  • numpy (1.12.1 or or above)
  • scipy (0.19.0 or above)
  • matplotlib (2.0.1 or above)

Optional (but strongly recommended):

  • statsmodels (0.6.1 or above)

The statsmodel package is required to perform the Kernel Density Estimations using statsmodel.nonparametric.KDEMultivariate(), which is often more suitable than the alternative scipy.stats.gaussian_kde().

Testing the installation

First, launch a Python shell and check that you can import pyqz, and that it is the intended version:

>>> import pyqz
>>> print pyqz.__version__

Next, as a quick test, try to fetch one of the diagnostic grid:

>>> a_grid = pyqz.get_grid('[NII]/[SII]+;[OIII]/[SII]+', sampling=1)
>>> print a_grid

More tests with unittest

A more complete set of tests, relying on the Python unittest module, are also available. The user willing to run them can do so as follows:

>>> import pyqz.tests
>>> pyqz.tests.run_all_tests(cleanup=True)

Note that the last test takes several seconds (~130s or so) to complete. It is designed to test the functions of pyqz_plots. If it succeeds, the demonstration plots will be deleted, unless the tests are run with cleanup=False. In that case, the plots will be stored in:

>>> print pyqz.tests.arena

Troubleshooting

1. If you get the following message when importing pyqz:

WARNING: Statsmodels module not found. KDE_method must be set to 'gauss' or else I will crash.

then pyqz could not import the statsmodels module. This module is required only if you want to use the KDEMultivariate function to construct the joint probability density function (which we suggest you do). To remove the warning, install statsmodels and try reloading pyqz.

2. If you encounter other errors when importing the module or running the example above, ensure that your numpy, scipy and matplotlib packages are up-to-date and try again.

3. If you still encounter errors after doing all that, check the FAQ.

4. If the FAQ doesn’t shine some light on your problem, try More tests with unittest.

5. Check if this is a known issue: https://github.com/fpavogt/pyqz/issues

6. If you still can’t figure out what’s wrong, please submit a new issue on the Github repository of the project. Provide as much detail as possible (error message, minimal example able to reproduce the error, operating system, Python version, etc …).

Note

Submitting a Github issue is the best way for you to get help rapidly, for us to keep track of the problems that need solving, and for future users to see what changes have been made over time (and look for existing solutions to their problem which may be the same as yours). Submitting a new issue on Github is rapid and easy, but if you are really against doing it (why would you ?), you can always email frederic.vogt@alumni.anu.edu.au for help.