Hi, Dear Greg, I have one question, if I boosting decision tree, the distribution "bernoulli", after that, I use predict.gbm to predict the fitted value for new data set. predict(gbm1, newdata, n.trees=best..iteration, type="response") If type="response" then gbm converts back to the same scale as the outcome. My outcome is 0 and 1. Will it be 0 and 1? Also on the help document, " Currently the only effect this will have is returning probabilities for bernoulli" how to change to 1 and 0, which is what I want? Thanks so much! -- Sincerely, Changbin -- [[alternative HTML version deleted]]