ChainsMakie.jl

Appearance

density ​

A density plot of Chains shows the kernel density estimate of the samples in each chain.

Density plots are a useful summary of both the shapes of distributions for each chain, and their location, highlighting potential convergence issues. Together with traceplots, they a part of the most common summary visualization of MCMC chains (see plot).

julia
using ChainsMakie, CairoMakie
import MCMCChains: Chains
chains = Chains(randn(300, 2, 4), [:A, :B])
fig = density(chains)
fig

It is possible to plot a subset of the parameters by passing their names as the second argument:

julia
using ChainsMakie, CairoMakie
import MCMCChains: Chains
chains = Chains(randn(300, 3, 4), [:A, :B, :C])
density(chains, [:A, :B])

Attributes ​

color ​

Controls which colors will be used to color the density plot for each chain.

Defaults to Makie's wong_colors palette and automatically switches to colormap = :viridis for more than seven chains.

julia
using ChainsMakie, CairoMakie
import MCMCChains: Chains
chains = Chains(randn(300, 2, 4), [:A, :B])
fig = density(chains; color = first(Makie.to_colormap(:tab20), 4))
fig

colormap ​

Controls which colormap will be used to color the density plot for each chain.

Defaults to the :viridis palette.

julia
using ChainsMakie, CairoMakie
import MCMCChains: Chains
chains = Chains(randn(300, 2, 8), [:A, :B])
fig = density(chains; colormap = :plasma)
fig

strokewidth ​

Controls the strokewidth of the density plot's contour.

Defaults to 1.0.

alpha ​

Controls the opacity of the density plot.

Defaults to 0.4.

julia
using ChainsMakie, CairoMakie
import MCMCChains: Chains
chains = Chains(randn(300, 2, 4), [:A, :B])
fig = density(chains; strokewidth = 0.0, alpha = 0.8)
fig