Colors

Makie has support for you to color your plots however you want to. You can manipulate the color of a plot by using the color keyword, and change the colormap by using the colormap keyword.

Colors

For line plots, you can provide a single color or symbol that will color the entire line; or, you can provide an array of values that map to colors using a colormap.

Any color symbol supported by Color.jl is supported, check out their page on named colors to see what you can get away with! You can also pass RGB or RGBA values.

Colormaps are mappings of values to colors. You can supply the coloring values using the color keyword argument, and the colormap will automatically be adjusted to fit those values. THe default colormap is viridis, which looks like this:

using Makie

 c = to_colormap(:viridis) # get colors of colormap

 image(         # to plot colors, an image is best
    0..10,      # x range
    0..1,       # y range
    hcat(c, c), # reshape this to a matrix for the colors
    show_axis = false # don't show axes
 )

You can copy this code and substitute cmap with any Colormap to show the colormap.

Makie supports multiple colormap libraries. Currently, support for colormaps provided by PlotUtils is inbuilt, meaning that any colormap symbol that works with Plots will also work with Makie. Colormaps from the ColorSchemes package can be used by colormap = ColorSchemes.<name of colormap>.colors. Similarly, colormaps from the PerceptualColourMaps package (which is a superset of the colorcet library) can be used by colormap = PerceptualColourMaps.cgrad("<name of colormap>"). In principle, any Array of RGB values can be used as a colormap.

Builtins

Color gradients are arranged into color libraries. To get a list of color libraries, use the clibraries function. To get a list of color gradients in each library, call cgradients(library). showlibrary(library) creates a visual representation of color schemes. To change the active library, use clibrary(library). This is only necessary in the case of namespace clashes, e.g. if there are multiple :blues. The gradients can be reversed by Reverse(:<gradient_name>). The clims::NTuple{2,Number} attribute can be used to define the data values that correspond with the ends of the colormap.

PlotUtils bundles with it colormaps from many libraries. As of the 16th of March, 2019, those are:

:Plots    # default
:cmocean
:colorbrewer
:colorcet

Again, the clibrary function can be used to change the preferred colour library in case of namespae conflict. For example, to prefer the use of cmocean colourmaps if available, you might call clibrary(:cmocean) before plotting.

Libraries

PLOTS

The default library. Created by Nathaniel J. Smith, Stefan van der Walt, and (in the case of viridis) Eric Firing. Released under CC0 license / public domain dedication. Full license info available here.

using Makie

 showlibrary(:Plots)

CMOCEAN



 showlibrary(:cmocean)

COLORCET

Released under The MIT License (MIT) Copyright (c) 2015 Peter Kovesi. These are the perceptually correct color maps designed by Peter Kovesi and described in Peter Kovesi. Good Colour Maps: How to Design Them. arXiv:1509.03700 [cs.GR] 2015



 showlibrary(:colorcet)

COLORBREWER

Created by Cynthia Brewer, Mark Harrower, and The Pennsylvania State University. Released under the Apache License, Version 2.0. Full license info available here.



 showlibrary(:colorbrewer)

Note

Due to the font in the image above, it may be difficult to tell that the yellow-orange-brown gradient, YlOrBr, is not YIOrBr - the second character is an l, not an I.

MISC



 showlibrary(:misc)

Color legends

To show the colormap and its scaling, you can use a color legend. Color legends can be automatically produced by the colorlegend function, to which a Plot object must be passed. Its range and the colormap it shows can also be manually altered, as can many of its attributes.

To simply produce a color legend and plot it to the left of the original plot, you can produce a colorlegend and vbox it. In the example below, p1 is the initial Scene, with only one plot.

scene = vbox(
  p1,
  colorlegend(
    p1[end],            # get Plot object from Scene
    camera = campixel!, # let vbox decide scene limits
    raw = true          # no axes, other things as well
  )
)

Be warned that in its current form, this will look quite small compared to the size of the plot!

To fix that, you can theme it, as shown below:

using Makie
 using ColorSchemes      # colormaps galore


 t = range(0, stop=1, length=500) # time steps

 θ = (6π) .* t    # angles

 x = t .* cos.(θ) # x coords of spiral
 y = t .* sin.(θ) # y coords of spiral

 p1 = lines(
     x,
     y,
     color = t,
     colormap = ColorSchemes.magma.colors,
     linewidth=8)

 cm = colorlegend(
     p1[end],             # access the plot of Scene p1
     raw = true,          # without axes or grid
     camera = campixel!,  # gives a concrete bounding box in pixels
                          # so that the `vbox` gives you the right size
     width = (            # make the colorlegend longer so it looks nicer
         30,              # the width
         540              # the height
     )
     )

 scene_final = vbox(p1, cm) # put the colorlegend and the plot together in a `vbox`