Add show_plot option to plot-functions in Python

[tsadakane]

Summary

A solution for #285

Add show_plot option to plot-functions in Python.

Modified

• if not set, plotImg, plotProj detects the matplotlib backend end and if it is a non-GUI backend, they don't call plt.show().
  • the values of backend are taken from https://matplotlib.org/stable/tutorials/introductory/usage.html#backends
  • The backend of jupyter and VSCode was "module://ipykernel.pylab.backend_inline"
    • In my environment, single page plot can be shown on VSCode and jupyter, but animations are not.
• User can force the functions not to call plt.show() by settting show_plot=False.

Test

Code

import tigre
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
from tigre.utilities import sl3d

geo = tigre.geometry_default(high_resolution=False)
angles = np.linspace(0, 2 * np.pi, 30)
head = sl3d.shepp_logan_3d(list(geo.nVoxel))
projections = tigre.Ax(head, geo, angles)

print(f"matplotlib.get_backend()= {matplotlib.get_backend()}")
backend_orig = matplotlib.get_backend() # "Qt5Agg" 

# Test plotImg
## Show
matplotlib.use(backend_orig)   # "Qt5Agg" is set.
tigre.plotImg(head, dim="Z", slice=head.shape[0]//4) # shown
## Set non-GUI matplotlib backend
matplotlib.use("agg")   #
tigre.plotImg(head, dim="Z", slice=head.shape[0]//2) # not shown 
tigre.plotImg(head, dim="Y", savegif="Yslice.gif")  # plot not shown on screen, but file is saved, 

# Test plotProj
## Show.
matplotlib.use(backend_orig)   # "Qt5Agg" is set.
tigre.plotProj(projections, angles)  # shown
## Set non-GUI matplotlib backend
matplotlib.use("agg")
tigre.plotSinogram(projections, projections.shape[1]//2)  # not shown

Environment

• Windows 10
• Python 3.8.5 (anaconda)
• matplotlib 3.3.1
• Visual studio 2019 (16.9.4)

[AnderBiguri]

Great! I'll test it and add it, after I test #290. Thanks a lot!