Using ggvis¶
ggvis is a data visualization package for R
For more information, see http://ggvis.rstudio.com/
Installing ggvis¶
Installing the ggvis package¶
The ggvis package is not installed by default. The recommended way to install it is to use a code environment
For a regular R environment, you need to install the
ggvis
packageIf you are using a Conda environment, you can also choose instead to install in the Conda section the
r-ggvis
package.
Installing the frontend dependencies¶
To work, ggvis first needs some frontend libraries, that need to be preinstalled once.
To install the dependencies, open a R notebook and run
library(dataiku)
dkuInstallGgvisDependenciesOnce()
Warning
It is not possible to run this if your DSS instance has the User Isolation Framework enabled.
In that case, your DSS administrator needs to run this from a command-line ./bin/R prompt, after setting the DIP_HOME env variable to the location of the DSS data directory
Displaying charts in a Jupyter notebook¶
ggvis charts will not work properly if you only enter it in a Jupyter notebook
Instead, use the dkuDisplayGgvis
method.
For example; to display the first example in the ggvis documentation:
library(dataiku)
library(ggvis)
# Prepare the chart
chart <- mtcars %>% ggvis(~wt, ~mpg) %>% layer_points()
# And display it
dkuDisplayGgvis(Line)
Displaying charts on a dashboard¶
googleVis charts generated using R code can be shared on a DSS dashboard using the “static insights” system.
Each chart can become a single insight in the dashboard.
To do so, create static insights
# Prepare the chart
chart <- mtcars %>% ggvis(~wt, ~mpg) %>% layer_points()
# Save it as an insight
dkuSaveGgvisInsight("my-ggvis-plot", chart)
From the Dashboard, you can then add a new “Static” insight, select the my-ggvis-plot
insight
Plots can be donwloaded in SVG or PNG format
Refreshing charts on a dashboard¶
You can refresh the charts automatically on a dashboard by using a scenario to re-run the above piece of code.
This call to dkuSaveGgvisInsight
code can be:
In a DSS recipe (use a regular “Build” scenario step)
In a Jupyter notebook (use a “Export notebook” scenario step)