Search code examples
matplotliblogarithmcolorbar

How can I see all the colorbar labels with custom ticks and logarithmic transformation?

When using custom ticks and LogNorm(), only power of 10 labels are displayed, not all labels.

import matplotlib
import matplotlib.pyplot as plt
import numpy as np
from matplotlib import cm
from numpy.random import random

# Make plot with horizontal colorbar
fig, ax = plt.subplots()

data = random((250,250)) + 3.5

norm = matplotlib.colors.LogNorm(vmin=0.1,vmax=10)

cax = ax.imshow(data, interpolation='nearest', cmap=cm.afmhot, norm=norm)
ax.set_title('Gaussian noise with horizontal colorbar')

cbar = fig.colorbar(cax, ticks=[0.1,0.2,0.5,1,2,5,10], orientation='horizontal')

plt.savefig("example.png")

enter image description here

Solution

  • You have to use the format option: 

    cbar = fig.colorbar(cax, ticks=[0.1,0.2,0.5,1,2,5,10], 
                    orientation='horizontal', format='%.1f')

    This moves away from the representation as 10^n (which struggles with 0.5 for example) to floating point numbers:

    all ticks labeled