Colormap centering & norms

[Phlya]

There is a beautiful way to center a colormap using seaborn in heatmap:

ax = sns.heatmap(np.linspace(-2, 5, 100).reshape((10, 10)), cmap='coolwarm', center=0, square=True)
ax.axis('off')

Unfortunately, there is not tool that does it in matplotlib itself. Is there a way to just get a colormap like this with specific vmin, center and vmax to use in matplotlib functions?

Also, is there a way to apply e.g. a LogNorm to the plot made with heatmap?

[mwaskom]

You can't really get "just a colormap .. with a specific vmin, center, and vmax" because the limits aren't properties of the colormap, they're properties of the visualization that uses it. But you could use/adapt this: https://gist.github.com/phobson/7916777

To answer your second question, you can use the norm= argument from pcolormesh, but it will probably interact in unpredictable ways with the centering.