带有 matplotlib 的颜色图

Question

I am trying to represent 3D scatter data with matplotlib.

I have 3 arrays to plot:

• an array with X coordinates of points
• an array with Y coordinates of points
• an array with a parameter values linked with each point (points of which the coordinates are described in X[] and Y[]

What I would like to have is:

• a 2D point cloud (scatter) plot for each (X,Y) couple
• each point has a color according to the corresponding value in the 3rd array
• a color bar representing the range and values of the linked parameter

I am having problems with the color mapping. Here is a minimal code with comments:

import numpy as np
import matplotlib.pyplot as plt
from matplotlib import cm
from matplotlib.colors import ListedColormap

X = [] # x coordinates array
Y = [] # y coordinates array
parameter = [] # linked parameter values array, each value represents a point. This array is ordered accordingly

# next : the arrays are populated ..
# now I need to plot the point cloud (X,Y) with each point having a color representing the value of the corresponding linked parameter.. 

I've read some tutorials and understood that I need to create a customized colormap for the parameter. I've tried to create a colormap according to the linked parameter by:

viridisBig = cm.get_cmap('viridis', 512)
newcmp = ListedColormap(viridisBig(np.linspace(np.min(PARAMETER), np.max(PARAMETER), 512)))

Then called the scatter plot:

plt.scatter(X, Y, cmap=newcmp)
plt.colorbar(cm.ScalarMappable(cmap=newcmp))
plt.show()

The above code is not working properly. It is plotting the scatter plot with a unique color (blue) and the colorbar on the side is yellow, as in the attached screenshot:

I guess I'm missing something related to the creation of the colormap, but I haven't found any clear examples for this simple case. Any ideas?

Thanks in advance!

Answer 1

Problem solved by normalizing the parameter before using it with a colormap as parameters for the scatter function:

colormap = plt.cm.get_cmap('cool', 512)
normalize = mpl.colors.Normalize(vmin=PARAMETER.min(), vmax=PARAMETER.max())

Then:

plt.scatter(X, Y, c=PARAMETER, s=5, cmap=colormap, norm=normalize)