The rgl package is used to produce interactive 3-D plots. It contains high-level graphics commands modelled loosely after classic R graphics, but working in three dimensions. It also contains low level structure inspired by (but incompatible with) the grid package.
This document gives an overview. See the help pages for details.
This document was written in R Markdown, using the knitr package for production. It corresponds to rgl version 0.96.0.
Most of the highlighted function names are HTML links. The internal links should work in any browser; the links to help topics should work if you view the vignette from within the R help system.
The document includes WebGL figures. To view these, you must have Javascript and WebGL enabled in your browser. Some older browsers may not support this – see http://get.webgl.org for tests and links to a discussion.
The plot3d function plots points within an rgl window. It is similar to the classic plot function, but works in 3 dimensions.
For example
with(iris, plot3d(Sepal.Length, Sepal.Width, Petal.Length,
type="s", col=as.numeric(Species)))iris data."p", "l", "h", "s", meaning points, lines, segments from z=0, and spheres. There’s a lot of flexibility in specifying the coordinates; the xyz.coords function from the grDevices package is used for this.
You can use your mouse to manipulate the plot. The default is that if you click and hold with the left mouse button, you can rotate the plot by dragging it. The right mouse button is used to resize it, and the middle button changes the perspective in the point of view.
If you call plot3d again, it will overwrite the current plot. To open a new graphics window, use open3d.
The other high level function is persp3d to draw surfaces. It is similar to the classic persp function, but with greater flexibility. First, any of x, y or z can be specified using matrices, not just z. This allows parametric surfaces to be plotted. An even simpler specification is possible: x may be a function, in which case persp3d will work out the grid itself. See ?persp3d.function for details. For example, the MASS package estimates Gamma parameters using maximum likelihood in a ?MASS::fitdistr example. Here we show the log likelihood surface.
library(MASS)
# from the fitdistr example
set.seed(123)
x <- rgamma(100, shape = 5, rate = 0.1)
fit <- fitdistr(x, dgamma, list(shape = 1, rate = 0.1), lower = 0.001)
loglik <- function(shape, rate) sum(dgamma(x, shape=shape, rate=rate,
log=TRUE))
loglik <- Vectorize(loglik)
xlim <- fit$estimate[1]+4*fit$sd[1]*c(-1,1)
ylim <- fit$estimate[2]+4*fit$sd[2]*c(-1,1)
mfrow3d(1, 2, sharedMouse = TRUE)
persp3d(loglik,
xlim = xlim, ylim = ylim,
n = 30)
zlim <- fit$loglik + c(-qchisq(0.99, 2)/2, 0)
next3d()
persp3d(loglik,
xlim = xlim, ylim = ylim, zlim = zlim,
n = 30)