Closed matmel closed 6 years ago
There will not be a geom_contour
, it assumes too much about the format of the input data.
The master branch has geom_density_2d
& stat_density_2d
, and they will be in the next release.
Ok, well, my use case is not really related to computing statistical kernel densities though. I was really after some visualization of 3D surfaces projected on a 2D plane. The geom_contour input data format looked fine to me.
As a workaround, is there a possibility that I call matplotlib contour
function myself on the resulting figure generated by plotnine without to much hassle?
Yes you can do that. The draw
method returns the figure, from which you can get the axes.
p = ggplot() + ...
fig = p.draw()
axs = fig.get_axes()
for ax in axes:
# Plot more stuff
ax.contour(...)
Hi,
First of all, thank you 1000 times for
plotnine
, it works extremely well and is very nice to work with.Do you plan to implement geom_contour?
Thanks.