quick plot

The scripting interface of matplotlib is stateful, therefore dangerous. The object oriented interface to the plotting library is polluted. This tiny library combines the two interfaces into one context manager.

fundament

If this file is somewhere in your =$PYTHONPATH=, the following code should produce a plot.

´´´python import numpy as np import quick

with quick.Plot() as qp: qp.plot(np.sin(np.linspace(0, 2 * np.pi))) ´´´

The =Plot= class is the only available class in the =quick= module. It should be instantiated exactly once. Its instance can do everything the matplotlib axis object can do. For some methods, a default configuration is supplied. There are some methods, that are not contained in the axis object, for example =trajectory=, =field=, =points=, =remove_border=, =remove_xticks=. When a method is called, that is neither a method of the =axis= object, nor one of the additional methods of this module, then the module =matplotlib.pyplot= is searched for a function with that name, and, if it exists, it is called. When the context manager is left, the plot shows up.

arguments

=Plot= takes optional arguments. | =filename= | given that, the plot is not shown interactively, but saved as a file, default =None= | | =figsize= | a tuple, representing (width, height), default =(3, 3)= | | =grid= | if =True= shows a grid, default =False | all other optional arguments are directly passed to =matplotlib.pyplot.subplot=.

short form

If you just want to quickly see some of your data, and you do not want to do any customization of the plotting result, the context manager is a bit of a boilerplate. So there is the following shorthand for the previous script.

´´´python import numpy as np import quick

quick.plot(np.sin(np.linspace(0, 2 * np.pi))) ´´´

The context manager is called implicitly.

more convenience

The quick module additionally provides a few methods that are not directly related to the =matplotlib= plotting facility, e.g.

• =colored=: takes a list of array and return an iterator with tuple, with the data and a color, that can be past to the color argument of plotting methods. Optional argument is the name of a colormap
• =landscape=, =portrait=: they return tuples that can be past to the =figsize= argument of =Plot=, internally using the golden ratio.
• =tex=: surround a string by dollar signs.

drawback

There is no interactive workflow anymore with this approach. I like that way of working, because due to script restarts you eliminate all errors in your script, that may occur from state changes from the past, that confuses the (fragile) python module system.