使用离散刻度获取 matplotlib 中散点图的颜色条

Question

I would like to to get a scatter plot with color bar having discrete ticks that bound each of the colors used. Presently, the following code merges top two colors.

import numpy as np
np.random.seed(42)
import matplotlib.pyplot as plt
import matplotlib.colors as mcolors

a1 = np.random.randint(-100, 100, 10)
a2 = np.random.randint(-100, 100, 10)

fig = plt.figure(figsize=(5, 5))

limits1 = [min(a2), max(a2)]
bounds1 = np.linspace(limits1[0], limits1[1], 5)
colors1 = plt.get_cmap('jet')(np.linspace(0, 1, len(bounds1)))
cmap1 = mcolors.ListedColormap(colors1[1:-1])
norm1 = mcolors.BoundaryNorm(boundaries=bounds1, ncolors=len(bounds1)-1)

ax1 = fig.add_subplot(1, 1, 1)
im1 = ax1.scatter(a1, a2, c=a2, norm=norm1, cmap=cmap1)
fig.colorbar(im1, orientation='vertical', shrink=1, aspect=30, pad=0.03, extend='neither', ticks=bounds1, label='ampWT')

plt.show()

Answer 1

when running this line:

cmap1 = mcolors.ListedColormap(colors1[1:-1])

You're effectively removing the first and last colors from the colors1 list. Since you started with 5 colors (bounds1 is length 5) you're left with 3 colors, which explains why you don't have the correct number of colors.

If you want to fix it you have 2 solutions:

1. just add another color in the beginning to have 6 initial colors and remove 2

bounds1 = np.linspace(limits1[0], limits1[1], 5)
colors1 = plt.get_cmap('jet')(np.linspace(0, 1, len(bounds1)+1)) # add 1 color 
cmap1 = mcolors.ListedColormap(colors1[1:-1])
norm1 = mcolors.BoundaryNorm(boundaries=bounds1, ncolors=len(bounds1))

2. choose the final number of colors from the beginning (4)

bounds1 = np.linspace(limits1[0], limits1[1], 5)
colors1 = plt.get_cmap('jet')(np.linspace(0, 1, len(bounds1)-1)) # changed length of colors
cmap1 = mcolors.ListedColormap(colors1) # changed colors input
norm1 = mcolors.BoundaryNorm(boundaries=bounds1, ncolors=len(colors1)) # changed ncolors

The final colors are not the same, so pick the one you prefer