I recently started DataCamp's Statistical Thinking in Python (Part 1) course, and the instructor mentioned that when building plots in Python, the convention was often to assign plot objects to the NULL operator "_" while building the plot, prior to showing the plot.
Is this truly convention? I couldn't find much evidence for it, nor could I think of a justification as to why it would be convention. The code snippet below is taken from the course example and demonstrates what I mean.
# Plot all ECDFs on the same plot
_ = plt.plot(x_set, y_set, marker='.', linestyle='none')
_ = plt.plot(x_vers, y_vers, marker='.', linestyle='none')
_ = plt.plot(x_virg, y_virg, marker='.', linestyle='none')
# Annotate the plot
_ = plt.legend(('setosa', 'versicolor', 'virginica'), loc='lower right')
_ = plt.xlabel('petal length (cm)')
_ = plt.ylabel('ECDF')
# Display the plot
plt.show()