corrplot
plots a correlation matrix, but it does not return a graphical object (grob)
I would like to plot several correlation matrices on a single page. For normal plots, I would use grid.arrange
from the gridExtra
package. However since corrplot only prints and does not return an object, I can't see how to do this.
Is there a workaround or a better alternative to corrplot
?
There's the old standby par(mfrow=c(x, y))
where x
is the number of rows you wish to plot and y
the numberof columns. It then posts across and then down as you call the plots.
par(mfrow = c(2, 2))
corrplot(cor(mat1))
corrplot(cor(mat2))
corrplot(cor(mat3))
corrplot(cor(mat4))
par(mfrow = c(1, 1)) #To clear layout
Will plot as
Mat1 | Mat2
-----------
Mat3 | Mat4
The recent gridGraphics
package could probably do what you asked: return the plot as a grob.
mat <- matrix(rnorm(100), ncol=10)
library(corrplot)
corrplot(cor(mat))
library(gridGraphics)
grab_grob <- function(){
grid.echo()
grid.grab()
}
g <- grab_grob()
library(gridExtra)
grid.newpage()
grid.arrange(g,g,g,g)
Not sure if I got your question right, but maybe what you are looking for is simple layout
?
mat <- matrix(rnorm(100), ncol=10)
layout(matrix(1:2))
corrplot(cor(mat))
corrplot(cor(mat))
来源:https://stackoverflow.com/questions/27929452/r-return-corrplot-as-object