Bokeh

Bokeh is an interactive visualisation library for modern web browsers. Its goal is to provide versatile graphics and extend this capability to very large and streaming datasets through high-performance interactivity. Bokeh is useful for quickly and easily creating interactive charts, dashboards and data applications.

To provide both simple and powerful and flexible features required for extensible customisation, Bokeh provides two interfaces:

bokeh.models

Low-level interface that offers application developers the greatest possible flexibility

bokeh.plotting

High-level interface for the creation of visual glyphs

Installation

With Spack you can deploy Bokeh in your kernel, for example with:

$ spack env activate python-311
$ spack install   py-bokeh

Alternatively, you can also install Bokeh with other package managers, for example

$ pipenv install bokeh

Optional extensions

There are extensions for Bokeh for the following functions:

NodeJS

Necessary to extend Bokeh or to define CustomJS implementations in CoffeeScript or TypeScript

pandas

Necessary for the Hexbin function. Some applications are simplified by using pandas, for example pandas DataFrames are automatically converted to Bokeh data sources by Glyph functions

Psutil

Required to enable detailed memory logging in the Bokeh server

NetworkX

With from_networkx, the bokeh diagram renderer can be applied directly to NetworkX data

Selenium, PhantomJS

Necessary for exporting plots to PNG and SVG images

Examples

When installing with pip, the examples are not installed. However, you can clone the Git repository and look at the examples/ directory to see the examples.

Most of these examples use sample data, which must also be provided separately. To download these files, simply enter the following:

$ pipenv run bokeh sampledata