3

I encountered a problem on creating a difference map between two matplotlib.pyplot hexbin plots, which means to get the value differences of each corresponding hexbin first and then create a difference hexbin map.

To give a simple example of my problem here, say the value of one hexbin in Map 1 is 3 and the value of the corresponding hexbin in Map 2 is 2, what I would like to do is to get the difference 3 – 2 = 1 first and then plot it in a new hexbin map, i.e. a difference map, at the same positions as Map 1 and Map 2.

My input codes and output plots are as follows. Could anyone please give me a solution to this problem?

Thanks for your time!

In [1]: plt.hexbin(lon_origin_df, lat_origin_df)
Out[1]: <matplotlib.collections.PolyCollection at 0x13ff40610>

enter image description here

In [2]: plt.hexbin(lon_termination_df, lat_termination_df)
Out[2]: <matplotlib.collections.PolyCollection at 0x13fff49d0>

enter image description here

Bart
  • 9,825
  • 5
  • 47
  • 73
geosciz
  • 161
  • 1
  • 4
  • 11
  • Is [rectangular histogram](http://docs.scipy.org/doc/numpy-1.10.1/reference/generated/numpy.histogram2d.html) out of the question? `hexbin`'s binning logic is not detachable from the creation of a polygon collection in the current code. – fjarri Dec 14 '15 at 04:44
  • The answer below works. Thanks all the same! – geosciz Dec 15 '15 at 00:44

1 Answers1

4

It's possible to get the values from a h=hexbin() using h.get_values(), and set the values using h.set_values(), so you could create a new hexbin and set its values to the difference between the other two. For example:

import numpy as np
import matplotlib.pylab as pl

x  = np.random.random(200)
y1 = np.random.random(200)
y2 = np.random.random(200)

pl.figure()
pl.subplot(131)
h1=pl.hexbin(x, y1, gridsize=3, vmin=0, vmax=40, cmap=pl.cm.RdBu_r)
pl.colorbar()

pl.subplot(132)
h2=pl.hexbin(x, y2, gridsize=3, vmin=0, vmax=40, cmap=pl.cm.RdBu_r)
pl.colorbar()

pl.subplot(133)
# Create dummy hexbin using whatever data..:
h3=pl.hexbin(x, y2, gridsize=3, vmin=-10, vmax=10, cmap=pl.cm.RdBu_r)
h3.set_array(h1.get_array()-h2.get_array())
pl.colorbar()

enter image description here

Bart
  • 9,825
  • 5
  • 47
  • 73