Smooth Gaussians in corner plot about corner.py HOT 6 CLOSED

Oh - I see what you're looking for now! No, there's no KDE option. These parameters smooth out sampling noise but you're always going to see a histogram with bin sizes set by the bins parameter. If you want something different, you'll need to plot it manually using matplotlib directly. To do that, you can use the method described here to access the axis objects directly and then plot whatever you want: https://corner.readthedocs.io/en/latest/pages/custom.html

Unfortunately I can't give you more concrete suggestions.

from corner.py.

You can smooth the 1D histograms using the smooth_1d parameter (I agree that the docstring is a little ambiguous): https://corner.readthedocs.io/en/latest/api.html#corner.corner

The shading is performing as expected. The size of the "pixels" are set by the bins parameter.

from corner.py.

Thanks for the reply. Increasing the number of bins solved the shading problem, but I still could not obtain smooth Gaussians. I set both the options, smooth=True and smooth_1d=True, but it's not working. In fact, there was no change on setting smooth_1d=True. Is it not possible to obtain a smooth plot like this?

from corner.py.

As described in the documentation linked above, these parameters are not booleans; the should be floats. Try adjusting those parameters to get what you want.

from corner.py.

Sorry, but this did not work either. I tried something like
smooth=10.0, smooth_1d=10.0
but the histogram is still step-like.

from corner.py.

Thanks.

from corner.py.