I have looked at the help (?gbm) and documentation for gbm.object, it only says that
fit: a vector containing the fitted values on the scale of regression function (e.g. log-odds scale for bernoulli, log scale for poisson)
Is it the OOB estimates like randomForest? That would seem unlikely to me since best_iter is not supplied to get it unless gbm has secretly decided which test I prefer.
Is there an efficient way to get the OOB estimates at all since I've already ran 5-fold cv to get best_iter?