CODEWITHSUNDEEP
rfor-loopdplyrsubset
jungpooh
jungpooh

Reputation: 1

How to change values between columns which have same colnames

I would like to replace certain columns of a data frame with columns from another data frame if the column names match:

I extracted some numeric columns from a data.frame named train. Then I made a second data.frame named all_box which contain column-wise boxcox transformations of train. After that, I want to replace those imputed data(all_box) with part of original data(train) finding by same colnames. My code was this..

I was trying to change value if the colnames are same or did not want to change any of it. 

for(i in 1:length(train)){

  train[,i]<-ifelse(colnames(train)[i]%in%colnames(all_box),

                    dplyr::select(all_box,colnames(train[i])),
                    dplyr::select(train,colnames(train[i])))
  }

but I get the error:

Error: Unknown column id Call rlang::last_error() to see a backtrace

(column 'id' is the name of first column of original data which I do not want to make any change)

Upvotes: 0

Views: 54

Answers (1)

Cettt
Cettt

Reputation: 11981

the ifelse function should not be used in this case. Try this:

train2 <- train
my_cols <- colnames(train)[colnames(train) %in% colnames(all_box)]
train2[, my_cols] <- all_box[, my_cols]

Upvotes: 0

Related Questions