Recipes

Plots for Publication

Style

See also

In order to produce multiple plots with consistent style it is useful to set some matplotlib options globally. One way to do this is by updating rcParams inside a script, e.g.:

import matplotlib

matplotlib.rcParams['font.family'] = 'sans-serif'
matplotlib.rcParams['font.size'] = 9
# The size of saved figures depends on the figure size (w, h) and DPI
matplotlib.rcParams['savefig.dpi'] = 300

Plot sizes

Matplotlib’s tight_layout() functionality provides an easy way for plots to fill the available space, and most Eelbrain plots use it by default. However, when trying to generate multiple figures with identical scaling it can lead to unwanted differences between figures.

For a consistent layout, set the font size through rcParams, and then control the size of plots and figures using the corresponding parameters (w, h, axw, axh). Subplot placement can also be specified in absolute scale through the margins parameter (see General layout parameters).

Various

If a script produces several plots and there is no need to show them on the screen, they can be kept hidden through the show=False argument:

p = plot.UTSStat('uts', 'A', ds=ds, w=5, tight=False, show=False)

When working on more complex figures, it is often desirable to save the legend separately and combine it in a layout application:

p = plot.UTSStat('uts', 'A', ds=ds, w=5, tight=False, show=False, legend=False)
p.save('plot.pdf', transparent=True)
p.save_legend('legend.pdf', transparent=True)