[FIXED] Change colormap in SymPy's ``plot3d``

Issue

In SymPy you can automatically plot a surface from an expression, namely

from sympy import symbols
from sympy.plotting import plot3d

x, y = symbols('x y')
monkey_saddle = x**3 - 3*x*y**2
plot3d(monkey_saddle, cmap="RdYlBu")

to get

I could not find any kwarg that change the colormap. I know that I can create a lambdify and then directly make the plot using matplotlib. But I am curious if there is a way to do it directly in SymPy.

Solution

I read the source code of sympy.plotting.plot.py, it seems that the cmap is set to jet:

collection = self.ax.plot_surface(x, y, z, 
                                  cmap=self.cm.jet, 
                                  rstride=1, cstride=1, 
                                  linewidth=0.1)

You need to set the cmap of the collections object, and before calling plot3d() call unset_show() to disable calling pyplot.show():

from sympy import symbols
from sympy.plotting import plot3d
from sympy.plotting.plot import unset_show

unset_show()

x, y = symbols('x y')
monkey_saddle = x**3 - 3*x*y**2
p = plot3d(monkey_saddle)

p._backend.ax.collections[0].set_cmap("RdYlBu_r")

Answered By - HYRY