User Guide

Installation

In Julia, press ] to enter the Pkg REPL and do

pkg> add Bokeh

Create your first plot

You will need to use Pluto, Jupyter, or any other environment capable of displaying HTML.

Alternatively (e.g. if you are using the Julia REPL) you may activate a display backend.

Creating a plot generally consists of three steps:

1. Create an empty figure using figure.
2. Add glyphs to the figure using plot!.
3. Display the figure.

using Bokeh
p = figure()
plot!(p, Scatter, x=randn(1000), y=randn(1000))
p

The final line causes the REPL or notebook you are using to display the plot p. You may explicitly call display(p) instead, e.g. if you are plotting in a script or a loop.

If you are in the REPL, you may like to put ; at the end of each intermediate command to supress displaying it immediately.

Getting help

Aside from this documentation, you can access extensive documentation from your Julia session.

If you are at the REPL, you can use the help mode to find out about Bokeh functions, models and properties. For example, to learn more about the Scatter model used above, press ? then do

help?> Scatter
search: Scatter AbstractPattern BasicTickFormatter ScientificFormatter

  Bokeh model type Scatter.

  Render scatter markers selected from a predefined list of designs.

[...]

This mentions the property marker, which we can find out about with

help?> Scatter.marker
  Bokeh model property Scatter.marker.

  Which marker to render. This can be the name of any built in marker, e.g. "circle", or a
  reference to a data column containing such names.

If you are not at the REPL, the Docs.doc function can be used to access model documentation:

Docs.doc(Scatter)
Docs.doc(Scatter, :size)