Matplotlib scatter plot colors

Question

I am trying to set the colors on a scatter plot using meshgrid input data. The code I am using is the following one :

xmesh = np.linspace(-5, 5, 30)
ymesh = np.linspace(-5, 5, 30)
xv, yv = np.meshgrid(xmesh, ymesh)
zv = a*xv+b*yv+c #a,b,c are some scalar constants
col = np.where(zv<0.5,'b','r')
plt.scatter(xv,yv,c=col)
plt.show()

Executing this code returns the following error :

could not convert string to float: 'b'

While if I chane the color map to float values :

col = np.where(zv<0.5,0.1,0.2) 

This works without any issue. Any idea why ?

When looking at the part of the code throwing the error, I see the following remarks :

# tuple color.
# Python 2.7 / numpy 1.6 apparently require this to return builtin floats,
# not numpy floats.
try:
c = tuple(map(float, c))

Maybe there is something to understand there to find the solution but I could not get it

Answer 1

From the scatter documentation

c can be a 2-D array in which the rows are RGB or RGBA, however, including the case of a single row to specify the same color for all points.

So the array given to c must either be a 1D array or a 2D array of RGB or RGBA values. It cannot be an arbitrary 2D array.

Hence you need to flatten your array,

col = np.where(zv<0.5,'b','r').flatten()
plt.scatter(xv,yv,c=col)