How do i set the width of an heatmap in Matplotlib?

Question

I created a simple heatmap with matplotlib with the following code:

fig,ax = plt.subplots(1)

# Display the image
ax.imshow(im)

a = np.array([[0.8, 2.4, 2.5, 3.9],
              [2.4, 0.0, 4.0, 1.0],
              [1.1, 2.4, 0.8, 4.3],
              [0.6, 0.0, 0.3, 0.0],
              [0.7, 1.7, 0.6, 2.6]])

ax1 = fig.add_subplot(111)
ax1.imshow(a, cmap='hot')
pos1 = ax1.get_position() # get the original position 
pos2 = [0.13, 0.15,  0.3, 0.3] 
ax1.set_position(pos2) # set a new position

This code works, the only problem is that i don't know how to make my heatmap wider. How can i set the width of an heatmap on MPL?

Answer 1

The image can be positioned using the extent=[x0, x1, y0, y1] parameter of imshow . Without explicitly setting extent , the x goes from -0.5 to width-0.5 . This puts the ticks at integer positions nicely in the center of the cells.

As imshow resets the xlim to the last image drawn, these need to be set explicitly.

Optionally the aspect ratio can be set to 'auto' to make the image stretch with the dimensions of the surrounding figure.

from matplotlib import pyplot  as plt
import numpy as np

fig, ax = plt.subplots(1)

# Display the image
im = np.random.randn(5, 10).cumsum(axis=0).cumsum(axis=1)
ax.imshow(im)

a = np.array([[0.8, 2.4, 2.5, 3.9],
              [2.4, 0.0, 4.0, 1.0],
              [1.1, 2.4, 0.8, 4.3],
              [0.6, 0.0, 0.3, 0.0],
              [0.7, 1.7, 0.6, 2.6]])
x0 = im.shape[1] - 0.5
x1 = x0 + a.shape[1]
ax.imshow(a, cmap='hot', extent=[x0, x1, -0.5, a.shape[0] - 0.5])
ax.set_xlim(-0.5, x1)
# ax.set_aspect('auto')

plt.show()