How to export gbm model to R?

Is there a standard (or available) way to export a gbm model to R? PMML will work, but when I try to use the pmml library, probably wrong, I get the error:

For example, my code looks something like this:


  model <- gbm(
      data =,
      distribution = "adaboost",
      n.trees = 450,
      n.minobsinnode = 10,
      interaction.depth = 4, shrinkage=0.05, verbose=TRUE)
  export <- pmml(model)
  # and then export to xml


And I am getting the error:

Error in UseMethod("pmml") : no applicable method for 'pmml' applied to an object of class "gbm"


I have also tried traversal on dataset. Anyway, I could live with a different format that I can parse programmatically (I will be evaluating the JVM), but PMML would be great if there was a way to make this work.


source to share

1 answer

You can get the job done using r2pmml

. It currently supports regression (i.e. distribution = "gaussian"

) and binary classification (i.e. distribution = "adaboost"

or distribution = "bernoulli"

) model types .

Below is the sample Auto MPG

code :


auto = read.csv(file = "AutoNA.csv", header = TRUE)

auto.formula = gbm(mpg ~ ., data = auto, interaction.depth = 3, shrinkage = 0.1, n.trees = 100, = "mpg")

r2pmml(auto.formula, "/tmp/gbm.pmml")




All Articles