What is Bokeh?

A data visualization package that enables the development of interactive, web-based data visualizations.
- Excellent to leverage when you want to display and interact visually with complex data sets or serve up such visualizations through the web.
- Declarative, not procedural: you describe the visualization (figure, glyphs, properties) rather than writing step-by-step drawing instructions.
  - Matplotlib is more procedural: plt.plot(x, y), plt.xlabel(‘X-axis’), plt.ylabel(‘Y-axis’), plt.title(‘Title’), etc. Each function call modifies the plot in a specific way.

Building Graphs in Bokeh:

Start out by creating an empty figure using the figure object. This unlocks all the available tools for visualizing data.
Glyphs are the building blocks of Bokeh. A glyph is a graphical shape or marker that is used to represent your data, like a circle or square. They can also represent lines, bars, etc.
- Glyphs can be combined as needed to fit your data visualization needs.

Bokeh code example showing a figure object with circle glyph rendering data points on a plot

Labeled diagram of a Bokeh figure showing title, axes, plot area, toolbar, and legend components

bokeh.models: Low-level interface offering flexibility and control. It provides access to the underlying components of Bokeh, allowing for more customization and fine-tuning of visualizations. (e.g., ColumnDataSource, HoverTool, etc.)
bokeh.plotting: High-level interface designed to make plotting quick and simple. It represents the most commonly used interface for creating visualizations with Bokeh.

Code example using bokeh.plotting interface to create a figure and add glyphs with show() call

Bokeh is available in the public repository called PyPI (Python Package Index)
- https://pypi.org/
The tool can be installed from the command line using the Python Management System, pip.
For Windows users, from the command line, running as an administrator, simply issue the following command:


pip install bokeh


pip3 install bokeh