How to plot pcolor colorbar in a different subplot - matplotlib

gc5 picture gc5 · Sep 18, 2013 · Viewed 20.4k times · Source

I am trying to split my plots in different subplots.. What I want to achieve is to put a colorbar for a subplot in a different subplot. Right now I am using:

# first graph
axes = plt.subplot2grid((4, 2), (0, 0), rowspan=3)
pc = plt.pcolor(df1, cmap='jet')

# second graph
axes = plt.subplot2grid((4, 2), (3, 0))
plt.pcolor(df2, cmap='Greys')

# colorbar
plt.subplot2grid((4, 2), (0, 1), rowspan=3)
plt.colorbar(pc) 

But the result is the following (notice the unwanted empty graph left to the colorbar): Wrong graph

What can I do to print only the colorbar without the left plot?

Thanks

Answer

tbekolay picture tbekolay · Sep 18, 2013

colorbar() accepts a cax keyword argument that allows you to specify the axes object that the colorbar will be drawn on.

In your case, you would change your colorbar call to the following:

# colorbar
axes = plt.subplot2grid((4, 2), (0, 1), rowspan=3)
plt.colorbar(pc, cax=axes)

This will take up the whole space given by subplot2grid; you can adjust this to be more reasonable either by having the main axes take up many more columns than the colorbar axes, or by setting up the gridspec explicitly. For example, your figure may be easier to tweak with the following:

from matplotlib import gridspec
gs = gridspec.GridSpec(2, 2, height_ratios=(3, 1), width_ratios=(9, 1))

# first graph
axes = plt.subplot(gs[0,0])
pc = plt.pcolor(df1, cmap='jet')

# second graph
axes = plt.subplot(gs[1,0])
plt.pcolor(df2, cmap='Greys')

# colorbar
axes = plt.subplot(gs[0,1])
plt.colorbar(pc, cax=axes)

Then you can just change height_ratios and width_ratios to your liking.