I tried using Caret with no cross validation, training on the whole data set. Here is my code Fit <- train(Result ~ ., data = FitData, method = method, #trControl = FitControl, trControl=trainControl(method="none"), verbose = FALSE) #tuneGrid = Grid) Error in train.default(x, y, weights = w, ...) : Only one model should be specified in tuneGrid with no resampling Wonder if anyone can help. Thanks Tjun Kiat [[alternative HTML version deleted]]