I have a 2x2x3 numpy array like this

import matplotlib.pyplot as plt
import numpy as np

red = np.array([[0, 50], [100, 200]])
green = np.array([[0, 185], [100, 255]])
blue = np.array([[0, 129], [0, 255]])

combined = np.stack((red, green, blue), -1)
# [[[  0,   0,   0],
#   [ 50, 185, 129]],
#  [[100, 100,   0],
#   [200, 255, 255]]]

I can plot the array with color just fine.

plt.figure()
im = plt.imshow(
  X=combined,
  vmin=0, 
  vmax=255, 
  interpolation='none',
  aspect='equal',
)
plt.colorbar()
plt.show()

I can also plot just the red channel, like this

reds = combined.copy()
reds[:, :, 1:] = 0

plt.figure()
im = plt.imshow(
  X=reds,
  vmin=0, 
  vmax=255, 
  interpolation='none',
  aspect='equal',
)
plt.colorbar()
plt.show()

But I’m struggling with two questions..

1. Is the colorbar in the first plot event valid. I’m not sure how to interpret it.
2. How do change the colorbar in the second plot to go from 0 (black) to 255 (red), as it should? Obviously I want the colorbar values and scale to match the actual plot behavior.