0

I get the following error:

Error in model.frame.default(form = Elite ~ ., data = C_test, na.action = na.fail) : variable lengths differ (found for 'Privat1')

I check the length and not sure why it is different. How can I fix the length?

My code:

fix(colleges)
rownames(colleges)=colleges [,1]
fix(colleges)

colleges=colleges [,-1]
fix(colleges)

#create "Elite" variable

Elite=rep("No",nrow(colleges))
Elite[colleges$Top10perc >50]=" Yes"
Elite=as.factor(Elite)
colleges=data.frame(colleges ,Elite)

names(colleges)
attach(colleges)

#Change 'Elite' from factor to numeric so you can run lm

colleges$Elite<-as.numeric(colleges$Elite)

#change Private to binary to be able to use normalization

Privat1=rep(2 , length(Private))
Privat1[Private=="Yes"] <- 1
colleges=data.frame(colleges, Privat1)

#remove Private column for cleaner results

colleges=colleges [,-1]
fix(colleges)


#To tell which varibles are statistacally sig.

Linear_reg<- lm(Elite ~ ., data=colleges) 
summary(Linear_reg)

#start to prep for kNN

C_Subset<-data.frame(colleges[c('Privat1', 'Apps', 'Accept', 'Top10perc', 'Top25perc', 'Outstate',
                            'Room.Board', 'PhD', 'Terminal', 'Expend')]) 

#Normalize


normalize <- function(x){return((x - min(x)) / (max(x) - min(x)))}
C_Norm<- as.data.frame(lapply(C_Subset[1:10], normalize))



set.seed(46)

C_Dat<- sample(1:nrow(C_Subset), size = nrow(C_Subset)*.85, replace = TRUE)

C_train<- C_Subset[C_Dat,]
C_test<- C_Subset[-C_Dat,]
C_train_lab<-C_Subset[C_Dat,1]
C_test_lab<-C_Subset[-C_Dat,1]

library(class)


train_control<- trainControl(method = 'cv', number = 5)
model<- train(Elite ~., data = C_test, trControl = train_control, method = 'rpart')

# this is where the error occurs

Yuggp91
  • 1
  • We don't know what your data is hence have even less of an idea than you of why the lengths are different. Please provide a [mcve]. See [How to make a great R reproducible example?](https://stackoverflow.com/q/5963269/4996248) for what this would mean in R. – John Coleman Jul 22 '20 at 17:28
  • 1
    Hard to say without more information (especially a sample of the data), but `Privat1[Private=="Yes"] <- 1` may refer to a variable that does not exist unless it is in `colleges`. – dcarlson Jul 22 '20 at 19:17

0 Answers0