0

I want to access the spatial distribution of errors in a map based on random forest predictions on 10-fold Cross validation.

I run:

model <- train(predictors,response,method="rf",metric="Rsquared",
                 trControl=ctrl,ntree=1500)

predictions <- predict(preds, model)

I thought I could use the predictions interval of each pixel (max pred - min pred) as a measure of uncertainty. But I can´t find how to obtain the max and min predictions per pixes in the predictions output. How could I obtain that?

jpsmith
  • 11,023
  • 5
  • 15
  • 36
Gabriela Zuquim
  • 1
  • 2
  • Please read these guidelines and edit your question to create a minimal, reproducible example. https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example – John Polo Feb 23 '23 at 22:48

0 Answers0