Datashader (https://github.com/bokeh/datashader) is another library with which Bokeh output can be extended. It is a Python library that pre-renders large datasets as a large-sized raster image. This ability overcomes limitation of browser when it comes to very large data. Datashader includes tools to build interactive Bokeh plots that dynamically re-render these images when zooming and panning in Bokeh, making it practical to work with arbitrarily large datasets in a web browser.
Another library is Holoviews ((http://holoviews.org/) that provides a concise declarative interface for building Bokeh plots, especially in Jupyter notebook. It facilitates quick prototyping of figures for data analysis.