ggplotly demos

In this RPub I will explain how to use the new ggplotly function in library(plotly), which allows you to upload ggplots to the https://plot.ly web app, for sharing and interactive analysis.

## Loading required package: RCurl
## Loading required package: bitops
## Loading required package: RJSONIO
## Loading required package: ggplot2

As a useR and reader of RPubs, you are probably already familiar with the ggplot2 package, which can be used to produce a wide variety of plots, such as this classic visualization of Fisher’s iris data:

library(ggplot2)
ggiris <- qplot(Petal.Width, Sepal.Length, data=iris, color=Species)
print(ggiris)

Wouldn’t it be nice to be able to directly upload this ggplot to https://plot.ly for interactive annotation, editing, and sharing with your colleagues and the public? Well, now you can! Read on.

If you don’t already have the newest version of the plotly package on your computer, install it using the devtools package:

install.packages("devtools")
library(devtools)
install_github("plotly", "ropensci")

Then load the plotly package:

library(plotly)

If you don’t already have a plotly account, get one!

signup("new_username", "your_email@domain.com")

That should have responded with your new key. Use that to create a plotly interface object:

py <- plotly("new_username", "YOUR_KEY")

Now, the moment you have been waiting for: ggplots can be uploaded to plotly using the new py$ggplotly function. Give it a ggplot, and it will upload it to https://plot.ly – it is that simple!

py$ggplotly(ggiris)

Note that if you are on the R command line, this will open a new web browser tab with the uploaded plotly. And if you write this in an Rmd document, the plotly will be embedded, as long as you specify the plotly=TRUE chunk option (see https://github.com/tdhock/plotly-demos/ggplotly-iris-canada.Rmd).

data(canada.cities, package="maps")
viz <- ggplot(canada.cities, aes(long, lat))+
  borders(regions="canada", name="borders")+
  coord_equal()+
  geom_point(aes(text=name, size=pop), colour="red",
               alpha=1/2, name="cities")
print(viz)

Note how we used name=“borders” and name=“cities” – this is not standard ggplot2, so is just ignored when rendering the plot above. However, it allows us to name the traces that we upload to plotly:

py$ggplotly(viz)

As of writing (14 March 2014) only a few simple ggplot types are supported by ggplotly, but please don’t hesitate to post issues: https://github.com/ropensci/plotly/issues.

Thanks for reading and happy plotly-ing!