CODEWITHSUNDEEP
pythonmatplotlib
fishbacp
fishbacp

Reputation: 1263

How do I set equally-spaced xtickmarks using a specified list of floats that are not uniformly spaced?

I'm creating a heatplot and want to label the equally-spaced x-axis tickmarks using values from a list. The list values are floats, given in increasing order, but not uniformly spaced.

Here's what I've tried for a heatmap corresponding to a 5-by-20 matrix using a list of four tickmark labels: [1.2,2.3,4.7,8.3]:

import numpy as np
import matplotlib.pyplot as plt

M=np.random.rand(5,20)

fig,ax=plt.subplots()
pos=ax.imshow(M,aspect='auto',cmap='jet')
fig.colorbar(pos, ax=ax)

labels=[1.2, 2.3, 4.7,8.3]
ax.xaxis.set_major_locator(plt.MaxNLocator(len(labels)))

ax.set_xticklabels([t for t in labels])  
plt.show()

The resulting graph has four equally-spaced tickmarks on the horizontal axis. The first three are labelled, from left to right, as 2.3, 4.7, and 8.3. The last tickmark has no value, and the first value of 1.2 does not appear in the lower-left corner as desired. I can't seem to determine how to correct this.

enter image description here

Upvotes: 1

Views: 69

Answers (1)

Abstract
Abstract

Reputation: 995

Looks like set_xticks is what you're missing. The positions can be created equidistant by just using range and setting the step size:

import numpy as np
import matplotlib.pyplot as plt

num_cols = 20
M = np.random.rand(5, num_cols)

fig,ax=plt.subplots()
pos=ax.imshow(M, aspect='auto',cmap='jet')
fig.colorbar(pos, ax=ax)

labels = [1.2, 2.3, 4.7, 8.3]
pos = range(0, num_cols, num_cols // len(labels))

ax.xaxis.set_major_locator(plt.MaxNLocator(len(labels)))
ax.set_xticks(pos)
ax.set_xticklabels(labels)
plt.show()

Output:

output1

Per your comment above, for a 5-by-400 matrix, you'd just change the following:

num_cols = 400
labels = [60, 280, 410, 756]

which produces:

output2

Upvotes: 1

Related Questions