ehtplot is a python module that assists scientists in the Event
Horizon Telescope (EHT) Collaboration
to create publication quality, elegant, and consistent plots. It
provides a set of easy-to-use plotting functions for EHT and
Very-Long-Baseline Interferometry (VLBI) specific figures (see the
documentation of the Figure class). This includes plotting visibility
and images for both synthetic and real data, adding uv-tracks to the
plots, adding the expected event horizon size to the plots, etc.
By introducing a logical layer to hierarchically organize subplots and
manage subplot properties, ehtplot is capable to combining many
simple plots in a complex figure (see the documentation of the Panel
class). For example, the following code place multiple figures
side-by-side:
import ehtplot as ep
fig = ep.Figure(ep.Panel(image="1.fits"),
ep.Panel(image="2.fits"),
ep.Panel(image="3.fits"),
ep.Panel(image="4.fits"))
fig.show()
This feature can help generating comparative plots for the EHT Imaging Challenge.
In a typical scientific workflow, a publication quality plot often
goes to multiple places, which may have very different typesetting and
coloring requirements. For example, a plot first needs to be rendered
clearly and accurately onscreen; then, when exported into vector
graphic formats for publications, the lines need to be visible, the
font sizes of labels should match the caption font size, etc; finally,
important plots also go into talks, where larger fonts and wider lines
are usually preferred; in addition, slides come with different
theme—dark and light backgrounds—which may require
changing the color theme in a plot. In order to make creating these
multi-style-multi-destination figures easy, ehtplot is theme based
(see the documentation in "ehtplot/theme.py"). After creating a plot,
its presentation and rendering depend on the user selected targeted
outputs.
fig.save("figure.eps")
fig.save("figure-seaborn.eps", style="seaborn")
fig.save("figure-talk-dark.png", format="talk", theme="dark")
It is straightforward to enable ehtplot's theme even without using
ehtplot for plotting. One simply loads the theme submodule:
import ehtplot.theme
and all ehtplot theme will be registered to matplotlib.
We also spent a lot of time to create perceptually uniform colormaps
based on the CAM02-UCS color appearance model. For more information,
see ehtplot's color submodule documentation.
It is also straightforward to enable ehtplot's colormaps even
without using ehtplot for plotting. One simply loads the color
submodule:
import ehtplot.color
and all ehtplot colormaps will be registered to matplotlib.