How can I get the data behind the curve of ggplot2::geom_smooth?

Question

take this example :

ggplot(iris,aes(Sepal.Width,Sepal.Length)) + geom_smooth()

How do I get the x and y that were used to plot this blue curve ?

This case will be computed by loess automatically, my real case is computed by gam automatically.

I tried :

reproducing it with gam function
exploring the plot object

And didn't have success with any

give an example of your gam model for the data you use, in which package gam function is included? — Mateusz1981, Oct 09 '17 at 11:38
have you looked here - https://stackoverflow.com/questions/9789871/method-to-extract-stat-smooth-line-fit — Hardik Gupta, Oct 09 '17 at 11:44

score 1 · Answer 1 · answered Oct 09 '17 at 11:46

As far as I understood the question, try this approach:

 library(gam)
    library(broom)
    mod <- gam(Sepal.Length ~ Sepal.Width,  data = iris)
    ggplot(iris, aes(Sepal.Width,Sepal.Length)) + geom_point() + 
      geom_line(data = augment(mod, type.predict = "response"), 
                aes(y = .fitted), color = "blue")

How can I get the data behind the curve of ggplot2::geom_smooth?

1 Answers1