Applications built to run with Bokeh server require a host to run the server. Fortunately, the same code can run either on a remote webserver (and displaying back to your local browser) or on your local machine (using a localhost address as a URL within your browser). We will discuss these issues in more detail in the next topic. On this page, we will point you to some of the demo applications that are hosted on the Bokeh website, to provide a sense of what is possible.

In another browser window or tab, open this link to demo.bokeh.org and feel free to explore the different applications hosted there. Each image on that page is a link to an application hosted on the bokeh.org webserver. You can click on an image to activate an application, and then interact with the application through whatever widgets that application provides, such as sliders, drop-down menus, etc.

We will examine here in a little more detail one of these hosted applications. In another browser window or tab, open the crossfilter application that investigates the autompg dataset that we have been examining. In our previous examples, we loaded the autompg data into a ColumnDataSource, but then we had to manually code which columns of the data we wanted to plot against each other, and had to rerun the code if we wanted to plot different columns. In the crossfilter application, one can select which columns to plot against each by choosing them from the drop-down menus for the "X-Axis" and the "Y-Axis", and the figure is replotted once a change is made using one of the selector widgets. In addition, one can add additional dimensions of data to the figure by making selections from the "Color" and "Size" menus, which change the color and size of the scatter plot markers based upon the data in some other columns. The image below is an example of this usage, plotting vehicle weight vs. engine displacement, with color used to represent acceleration and size used to represent mpg (miles per gallon).

We will revisit the crossfilter application in the subsequent topic where we take a deeper dive into Bokeh server applications and their use of callback functions to support data exploration of the sort seen here.