This is a port of Mike Bostock's D3 scatter plot matrix code to the htmlwidgets framework. There have been some minor adjustments, including the addition of tooltips.
You could also consider the pairedVis() function in the healthvis package.
Take it for a test run here (you can upload your own data).
The pairsD3
package is available on CRAN:
install.packages("pairsD3")
Alternatively, you can install the development version of pairsD3
from Github using the devtools
package as follows:
devtools::install_github("garthtarr/pairsD3")
A canonical example with the iris data:
data(iris)
require(pairsD3)
pairsD3(iris[,1:4],group=iris[,5])
Use savePairs
to save a pairs plot as a stand alone HTML file:
library(magrittr)
pairsD3(iris[,1:4],group=iris[,5]) %>% savePairs(file = 'iris.html')
You can view an interactive scatterplot matrix using the shinypairs
function:
shinypairs(iris)
You can include interactive scatterplot matrices in rmarkdown documents in the usual way:
```{r}
require(pairsD3)
pairsD3(iris)
```
HTML widgets are not (yet) supported in slidify. A workaround is to do save the widget as a webpage then include that webpage in slidify using an iframe:
```{r, results='asis',echo=FALSE}
require(pairsD3)
pd3 = pairsD3(iris)
savePairs(pd3, 'pD3.html')
cat('<iframe src="pD3.html"> </iframe>')
```