Bokeh can create any type of custom graph or visualization. For example, here is a screenshot of a bar chart created with the figure plot:
For more references, including interactive live demonstrations, check out these sites:
The official Bokeh gallery has many example Bokeh visual formats.
Bokeh Applications hosts numerous data visualizations built with Bokeh.
Bokeh is under heavy development ahead of the upcoming 1.0 release. Note that while all of the following tutorials are useful, it is possible some of the basic syntax will change as the library's API is not yet stable.
Responsive Bar Charts with Bokeh, Flask and Python 3 is my recommended tutorial for those new to Bokeh who want to try out the library and get an example project running quickly with Flask.
Fun with NFL Stats, Bokeh, and Pandas takes an NFL play-by-play data set, shows how to wrangle the data into an appropriate format then explains the code that uses Bokeh to visualize it.
Data is beautiful: Visualizing Roman imperial dynasties provides a walkthrough for creating a gorgeous visualization based on historical Roman data. The post is about more than just the visual, it also goes into the ideation, data wrangling and analysis phases that came before using Bokeh to show the results.
Interactive Data Visualization in Python With Bokeh is a great beginners tutorial that shows you how to structure your data, draw your first figures and add interactivity to the visualizations.
Creating Bar Chart Visuals with Bokeh, Bottle and Python 3 is a tutorial that combines the Bottle web framework
Building Bullet Graphs and Waterfall Charts with Bokeh covers buildings two types of useful visualizations into your applications using Bokeh.
Interactive Visualization of Australian Wine Ratings builds a non-trivial visualization with a nice sample set of data based on wine ratings.
Drawing a Brain with Bokeh is a fun example of a chord diagram that represents neural connections in the brain.
Bryan Van de Ven on Bokeh is a podcast episode by one of the main Bokeh maintainers.
The Python Visualization Landscape by Jake VanderPlas at PyCon 2017 covers many Python data visualization tools, including Bokeh.
Bokeh vs Dash — Which is the Best Dashboard Framework for Python? contains a single project that was written in both Dash and Bokeh. The author gives his subjective view on the implementation difficulty although the web application only contained a single type of data visualization so it is hard to drawn any real conclusions from his opinion.
Fix errors in your Python code before your users see them by monitoring with Rollbar.