CODEWITHSUNDEEP
pythonrdata-visualizationcorrelationscatter-plot
user3115933
user3115933

Reputation: 4453

What is the R equivalent package to achieve this scatterplot visualization between variables?

I am using R to perform a machine learning analysis and I came across this beautiful data visualization done with Python.

I would like to replicate this in R. Is there a specific package in R that can achieve this? Or may be some online examples of the same visualizations with R.

scatterplot visualization

Upvotes: 2

Views: 543

Answers (2)

milan
milan

Reputation: 5000

You can use a package but can also just do this using base R. You should decide where each single plot (e.g., histogram, scatter plot) goes in the panel. This is an example, also using the R iris dataset. I did not format the plots much as that info can be found elsewhere and it is beyond the scope of the question.

v1 <- iris$Sepal.Length
v2 <- iris$Sepal.Width
v3 <- iris$Petal.Length

png('panel.png', units="in", width=9, height=9, res=150)
par(mfrow=c(3,3))
hist(v1, main='')
plot(v2, v1)
plot(v3, v1)
 plot(v1, v2)
 hist(v2, main='')
 plot(v3, v2)
  plot(v1, v3)
  plot(v2, v3)
  hist(v3, main='')
dev.off()

enter image description here

Upvotes: 0

G5W
G5W

Reputation: 37661

You can do this with the pairs function in R. Here is an example that is only a modest modification of an example given in the help page ?pairs. You need to define a function that plots the histogram on the diagonal - but the help page provides that code. 

panel.hist <- function(x, ...) {
    usr <- par("usr"); on.exit(par(usr))
    par(usr = c(usr[1:2], 0, 1.5) )
    h <- hist(x, plot = FALSE)
    breaks <- h$breaks; nB <- length(breaks)
    y <- h$counts; y <- y/max(y)
    rect(breaks[-nB], 0, breaks[-1], y, ...)
}

pairs(iris[,1:4], diag.panel=panel.hist, pch=16, col="steelblue")

Pairs plot

Upvotes: 5

Related Questions