Pfthb

In Mathematica you can store plots in variables and then overlay them at some later time. For example,

plt1 = Plot[Cos[x],x,0,Pi];
plt2 = Plot[Sin[x],x,0,Pi];
plt3 = Plot[x,x,0,Pi];

Show[plt1,plt2]
Show[plt1,plt3]

gives two plots, one overlays cos(x) and sin(x) plots, and the other overlays cos(x) and x plots. Therefore, I do not need to replot cos(x) for the second overlay since it is already saved in plt1.

I am wondering the same thing can happen in python too. I have a 2D function that is time-consuming to plot, and I need to replot it and overlay it with some other data every time. Can I plot it only once and then overlay it in with plots of other data?

I will interprete the question as to ask about matplotlib (since it is tagged by that), but there are of course other python plotting tools, which might behave differently.

In matplotlib an artist (e.g. a line) is necessarily part of exactly one figure. You cannot add the same artist to more than one figure.

So the usual solution is to not wanting to duplicate the artist itself, but rather the procedure to create the artist.

def mycos(x, ax=None, **kwargs):
 ax = ax or plt.gca()
 ax.plot(x, np.cos(x), **kwargs)

def mysin(x, ax=None, **kwargs):
 ax = ax or plt.gca()
 ax.plot(x, np.sin(x), **kwargs)

x = np.linspace(0,2*np.pi)

# Create one figure with two subplots, plot one function in each subplot
fig, axes = plt.subplots(2)
mycos(x, ax=axes[0])
mysin(x, ax=axes[1])

# Create another figure with one subplot, plot both functions
fig, ax = plt.subplots(1)
mycos(x, ax=ax)
mysin(x, ax=ax)