Rcolorbrewer and ggplot2 R: colormap for geom_hex

Question

I would like to make this type of figure below in R, which is a combination of marginal histograms and a geom_hex object I think. This is originally a matplotlib seaborn graph.

I can't get it to talk to RColorbrewer. Any thoughts why?

So far I've got:

require(ggplot2)
require(RColorBrewer)
require(ggExtra)
bl<-data.frame(beta=rnorm(100),lambda=rnorm(100))

p<-ggplot(bl,aes(x=beta,y=lambda))+
  stat_bin_hex()+
  #scale_fill_gradient(palette = "Greens") Neither of these work
  #scale_fill_continuous(palette = "Greens")+
  scale_fill_brewer()+
  theme_classic()


  ggExtra::ggMarginal(p, type = "histogram")

Original code:

x, y = np.random.multivariate_normal(mean, cov, 1000).T
with sns.axes_style("white"):

https://seaborn.pydata.org/tutorial/distributions.html

sns.jointplot(x=x, y=y, kind="hex", color="greens");

Answer 1

You can use scale_fill_gradientn and pass in the palette using brewer.pal. Then you just need to pass in the right fill and color to ggMarginal

library(ggplot2)
library(RColorBrewer)
library(ggExtra)

bl <- data.frame(beta=rnorm(10000),lambda=rnorm(10000))

p <- ggplot(bl, aes(x=beta, y = lambda))+
  stat_bin_hex() +
  scale_fill_gradientn(colors = brewer.pal(3,"Greens")) +
  theme_classic() +
  theme(legend.position = "bottom")

ggMarginal(p, type = "histogram", fill = brewer.pal(3,"Greens")[1], color = "white")

^{Created on 2018-11-20 by the reprex package (v0.2.1)}