optimization - CHAID error using caret in R: model fit failed for Resample01: alpha2=0.05, alpha3=-1, alpha4=0.05 Error : is.factor(x) is not TRUE

Question

CHAID error using caret in R: model fit failed for Resample01: alpha2=0.05, alpha3=-1, alpha4=0.05 Error : is.factor(x) is not TRUE

I'm getting the error above when trying to run a CHAID model in caret. The model runs fine with this data just by using the CHAID function.

Any suggestions? Code below:

model_weights <- ifelse(as.character(train_data$outcome) == "Sucess",
                        5.4,1)


model_tree_caret_cost = caret::train(outcome ~ .,  
                                data = train_data,
                                method = "chaid",
                                #tuneGrid = tunegrid,
                                #costs = costs,
                                metric = "ROC",
                                weights = model_weights,
                                trControl = trainControl(p = 0.7,
                                                         verboseIter = T,
                                                         classProbs = T#,
                                                         #costs = costs
                                                         )#,
                                #tuneLength = 10
                                )

data is 12 factor, ordered variables.

Answer 1

Waitting for answers