9

I just upgraded to matplotlib 2.0, and I feel like I'm on crazy pills. I'm trying to make a log-linear plot, with the y-axis on a linear scale and the x-axis on a log10 scale. Previously, the following code would have allowed me to specify exactly where I want my ticks, and what I want their labels to be:

import matplotlib.pyplot as plt

plt.plot([0.0,5.0], [1.0, 1.0], '--', color='k', zorder=1, lw=2)

plt.xlim(0.4,2.0)
plt.ylim(0.0,2.0)

plt.xscale('log')

plt.tick_params(axis='x',which='minor',bottom='off',top='off')

xticks = [0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0]
ticklabels = ['0.4', '0.6', '0.8', '1.0', '1.2', '1.4', '1.6', '1.8', '2.0']
plt.xticks(xticks, ticklabels)

plt.show()

But in matplotlib 2.0, this now causes me to get a set of overlapping tick labels where matplotlib apparently wants to auto-create ticks:

enter image description here

But if I comment out the "plt.xlim(0.4,2.0)" line and let it automatically determine the axis limits, there are no overlapping tick labels and I just get the ones I want:

enter image description here

But that doesn't work because I now have useless x-axis limits.

Any ideas?

Edit: for people searching the internet in the future, I'm becoming more convinced that this is actually a bug in matplotlib itself. I reverted back to v. 1.5.3. to just avoid the issue.

Jonathan Leffler
  • 730,956
  • 141
  • 904
  • 1,278
Thomas
  • 93
  • 1
  • 1
  • 6

1 Answers1

8

The additional ticklabels that overlap originate from some minor ticklabels, which are present in the plot. To get rid of them, one can set the minor formatter to the NullFormatter:

plt.gca().xaxis.set_minor_formatter(matplotlib.ticker.NullFormatter())

The complete code from the question might then look like

import matplotlib.pyplot as plt
import matplotlib.ticker
import numpy as np

x = np.linspace(0,2.5)
y = np.sin(x*6)
plt.plot(x,y, '--', color='k', zorder=1, lw=2)

plt.xlim(0.4,2.0)

plt.xscale('log')

xticks = [0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0]
ticklabels = ['0.4', '0.6', '0.8', '1.0', '1.2', '1.4', '1.6', '1.8', '2.0']
plt.xticks(xticks, ticklabels)

plt.gca().xaxis.set_minor_formatter(matplotlib.ticker.NullFormatter())

plt.show()

enter image description here

A code that may be more intuitive as it is not setting the xticklabels as strings would be the following, where we use a FixedLocator and a ScalarFormatter.
This code produces the identical plot as the above.

import matplotlib.pyplot as plt
import matplotlib.ticker
import numpy as np

x = np.linspace(0,2.5)
y = np.sin(x*6)
plt.plot(x,y, '--', color='k', zorder=1, lw=2)

plt.xlim(0.4,2.0)
plt.xscale('log')

xticks = [0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0]

xmajorLocator = matplotlib.ticker.FixedLocator(locs=xticks) 
xmajorFormatter = matplotlib.ticker.ScalarFormatter()
plt.gca().xaxis.set_major_locator( xmajorLocator )
plt.gca().xaxis.set_major_formatter( xmajorFormatter )
plt.gca().xaxis.set_minor_formatter(matplotlib.ticker.NullFormatter())

plt.show()
ImportanceOfBeingErnest
  • 321,279
  • 53
  • 665
  • 712