I am trying to use k-fold cross-validation on a WESBROOK dataset. It uses the train function from the caret package to do this. So far this function has worked for me with methods such as svm, knn and rpart, however with the nb (naive bayes) method, I get the following error:
Error in { : task 1 failed - "Not all variable names used in object found in newdata" This is how my train function look like:
k_folds <- 5 train_control <- trainControl(method = "cv", number = k_folds, classProbs = TRUE, summaryFunction = twoClassSummary) nb_model <- train( TOTLGIVE ~ ., data = train_data, method = "nb", trControl = train_control ) I checked, there are no missing data, the column names and their types are the same in the training and test set.