Real probability scales for matplotlib
Official releases are available through the conda-forge channel or pip
conda install mpl-probscale --channel=conda-forge
pip install probscale
This is a pure-python package, so building from source is easy on all platforms:
git clone git@github.com:matplotlib/mpl-probscale.git
cd mpl-probscale
pip install -e .
This library depends on pytest framework.
The current release version does not have it listed as a hard dependency, however.
So for now you will need to install pytest yourself to use mpl-probscale:
pip install pytest
or
conda install pytest --channel=conda-forge
In the next release, this depedency will be made optional.
Simply importing probscale lets you use probability scales in your matplotlib figures:
import matplotlib.pyplot as plt
import probscale
import seaborn
clear_bkgd = {'axes.facecolor':'none', 'figure.facecolor':'none'}
seaborn.set(style='ticks', context='notebook', rc=clear_bkgd)
fig, ax = plt.subplots(figsize=(8, 4))
ax.set_ylim(1e-2, 1e2)
ax.set_yscale('log')
ax.set_xlim(0.5, 99.5)
ax.set_xscale('prob')
seaborn.despine(fig=fig)Testing is generally done via the pytest and numpy.testing modules.
The best way to run the tests is in an interactive python session:
import matplotlib
matplotlib.use('agg')
from probscale import tests
tests.test()