1

I am trying to create a heatmap with pcolor using the example from Heatmap in matplotlib with pcolor? but I run into trouble.

First the example from the accepted anwser doesn't work for me, and I get the following error message:

Traceback (most recent call last):
  File "/home/knorrs/temp/HeatMapTest.py", line 22, in <module>
    heatmap = ax.pcolor(nba_sort, cmap=plt.cm.Blues, alpha=0.8)
  File "/home/martin/pybin/lib/python2.7/site-packages/matplotlib/axes.py", line 7309, in pcolor
    X, Y, C = self._pcolorargs('pcolor', *args)
  File "/home/martin/pybin/lib/python2.7/site-packages/matplotlib/axes.py", line 7132, in _pcolorargs
    numRows, numCols = C.shape
AttributeError: 'NoneType' object has no attribute 'shape'

Creating the plot "my" way leads to a good result with one exception: the labels/ticks on the y-axis stop after a while. It seems that the number of labels on the y-axis has to be the same as the number of labels on the x-axis.

Here is the part of my code responsible for the creation of the heatmap. Note that data_test is a two-dimensional numpy array (data_test.shape yields (20, 13)).

import matplotlib.pyplot as pl
import numpy as np

def plot_heatmap(data):
    x_min = ((data.shape[1]-1)/2)*-1    #if the shape is for example 13 (has to be odd) we set the x_min to -6
    x_max = (data.shape[1]-1)/2         #and the x_max to +6
    x_labels = range(x_min,x_max+1,1)   #this way we create the x_labels going from -6 over 0 to +6

    fig = pl.figure(figsize=(24,18))
    ax = fig.add_subplot(1,1,1)
    plot = ax.pcolor(data, cmap=pl.cm.Blues, edgecolors='k')

    # put the major ticks at the middle of each cell
    ax.set_xticks(np.arange(data.shape[0])+0.5, minor=False)
    ax.set_yticks(np.arange(data.shape[1])+0.5, minor=False)
    ax.set_ybound(lower = 0, upper = data.shape[0])
    ax.set_xbound(lower = 0, upper = data.shape[1])

    # want a more natural, table-like display
    ax.invert_yaxis()
    #ax.xaxis.tick_top()

    ax.set_xticklabels(x_labels, minor=False)
    ax.set_yticklabels(AA, minor=False)
    fig.colorbar(plot)

    pl.show()

AA = ['G', 'A', 'V', 'S', 'T', 'C', 'M', 'L', 'I', 'K', 'R', 'E', 'D', 'Q', 'N', 'F', 'Y', 'W', 'P', 'H']

data_test = np.random.rand(20,13)
plot_heatmap(data_test)

Resulting plot

If you could tell me the reason for this behavior and the how to change it, I would be very happy.

Community
  • 1
  • 1
user2720752
  • 13
  • 3

1 Answers1

3

This behaviour is due to the following line of your code:

ax.set_yticks(np.arange(data.shape[1])+0.5, minor=False)

You are defining your yticks with the wrong dimension of data!

If you correct to:

ax.set_xticks(np.arange(data.shape[1])+0.5, minor=False)
ax.set_yticks(np.arange(data.shape[0])+0.5, minor=False)

It will work.

carla
  • 2,181
  • 19
  • 26