Error: $ operator is invalid for atomic vectors while using fscaret to do feature selection (wrapper approach)

Question

While I'm trying to do feature selection (Using Wrapper Approach) by fscaret function I got this error:

*Elapsed time:  0.81 0 0.88 NA NA 

[1] "Building model has failed or reached timelimit!"

Error: $ operator is invalid for atomic vectors"*

Note: I want to get relevant features for neural network and SVM algorithm as wrapper approach gets relevant features based on the used algorithm.

Here is my code:

library(caret)
library(fscaret)
library(R.methodsS3)
#load data:
mov = read.csv("bank.csv",header=TRUE,sep = ";")
mov$y = as.factor(mov$y)
mov$y = as.numeric(mov$y)

#to convert data(factors) into binary format
dumdmov <- dummyVars("~.", data=mov , fullRank=F)
mov <- as.data.frame(predict(dumdmov,mov))
head(mov)

#train and test set

trainIndex <- createDataPartition(mov$y, p = .70,  list = FALSE, times = 1)
head(trainIndex)
mov_train <- mov[ trainIndex,]
mov_test  <- mov[-trainIndex,]

#apply fscaret function
fsmodel2<- c("neuralnet","svmRadial")

results <- fscaret(mov_train, mov_test,installReqPckg = TRUE ,
 myTimeLimit = 40, preprocessData=TRUE, classPred = TRUE, regPred = FALSE,
                  Used.funcClassPred = fsmodel2, with.labels=TRUE,
                  supress.output=FALSE,no.cores = NULL )

Answer 1

You should use [ not $.

See here: R Error in x$ed : $ operator is invalid for atomic vectors

For example here: trainIndex <- createDataPartition(mov$y, p = .70, list = FALSE, times = 1) head(trainIndex)

try: trainIndex <- createDataPartition(mov[y], p = .70, list = FALSE, times = 1) head(trainIndex)