Bug report

Bug summary

imshow behaviour change between matplotlib 3.1.0 and 3.1.1; y-axis is chopped-off in 3.1.1.

Code for reproduction

Run this in matplotlib 3.1.0 and 3.1.1.

import numpy as np
import matplotlib.pyplot as plt
xy = ["1", "2"]
fig, ax = plt.subplots()
im = ax.imshow(np.array([[0.8, 0.0], [0.1, 6.3]]))
ax.set_xticks(np.arange(len(xy)))
ax.set_yticks(np.arange(len(xy)))
ax.set_xticklabels(xy)
ax.set_yticklabels(xy)
plt.show()

This is a simplified version from the first figure on

• https://matplotlib.org/3.1.0/gallery/images_contours_and_fields/image_annotated_heatmap.html
• https://matplotlib.org/3.1.1/gallery/images_contours_and_fields/image_annotated_heatmap.html

where the difference can also be observed.

Actual outcome

Expected outcome

Matplotlib version

• Operating system: Linux
• Matplotlib version: v3.1.0 vs v3.1.1
• Matplotlib backend: Qt5Agg
• Python version: 3.7