0:01
Hi, my name is Brian Caffo, and
this is the Developing Data Products lecture on Plotly.
Plotly is a really nifty little web application and library for R.
I had an old video on, but it was early in their R development, and
I just went back and looked at it,
and I was really astounded at how much development has occurred since then.
And for creating interactive web graphics in R,
Plotly is certainly something that you should look into.
The nice part is it integrates with a lot of different languages, too.
So what is Plotly?
Plotly is a web application for creating and sharing visualizations.
0:41
It can work with several languages.
Off the top of my head, I know that it works with MATLAB, Python,
JavaScript, and so on.
And they also have a web interface if you like to point and
click your way to victory, just uploading a csv file, or something like that.
And you'll see, I think, in some of our first applications that we're going to do
with code demonstrations that Plotly is really easy to get up and going,
especially if you work in RStudio and in R, now.
So installing Plotly I'm not going to do code demonstrations on this.
This should be familiar to you at this point in the class.
Install.packages("plotly") and then, of course, to bring it up,
you can do Library or require ("plotly"), or something like that.
Later on, as the last part of this lecture series,
I'll talk about sharing you Plotly plot, which isn't specifically an R thing.
It's more of a how do you share HTML graphics discussion.
And one of the ways is actually on the Plotly site.
And they have free options and paid options if you want private displays and
user authentication and things like that.
So, if you want to share your visualizations through Plotly then,
of course, you can do that.
But, in addition, they just make HTML graphics that you can create
a web page and share on your own without involving Plotly at all,
other than the Plotly R package, okay?
But, also, they're web interface is pretty interesting.
It gives you a lot of nifty tools that even are not that easy to do in R,
for example, interactively changing access labels, right?
So that's something that you can't really do in R, you have to rerun the code and
that sort of thing.
And at any rate, at the Plotly web interface you can do things like that.
So we'll show that at the end, but for right now we're just going to work
in the R console and just show Plotly when running things locally.
Okay, so before I do the code demonstration,
let me go through the slides a little bit.
Here, the slides just go through creating the command, or
invoking the command, plot_ly, the ntcars dataset, and just specify the x and the y.
And then one thing I just want to show before we go to the code demonstration,
is this is actually embedded into this knitter document here.
Okay, so the R Markdown document that got then knit into an HTML document,
is actually displaying the interactive graphic as part of
the actual document in the actual slides.
So, you can zoom in, for example, and things like that,
that you can't normally do with a static graph.
And this is all embedded in the slides.
So when you look at the R Markdown document,
the commands are very simple to just do that.
So one interesting application of Plotly is embedding it in your Slidify or
R Studio Presenter or Slidy just knitter documents, okay.
So let's do some code demonstrations
where we go through a couple of different Plotly-type plots.
And then at the end, we'll talk about how we share them.
Brian Caffo, PhDProfessor, Biostatistics
Jeff Leek, PhDAssociate Professor, Biostatistics
Roger D. Peng, PhDAssociate Professor, Biostatistics
