Remove Unique value in R

[TalhaAsif]

I have this data frame (train) where I have 2314 variables and I want to drop the columns where the unique value is 0.

Example:

So, I want to remove all the columns with 0 unique values.

> unique(train$makeAm.General)
[1] 0

And, only keep the columns where the unique value is 0 and 1

> unique(train$makeAston.Martin)
[1] 0 1

[Yarnabrina]

So, basically you want to remove all columns that are all zero? Is that it?

I don't have a data to test with, but something like this will do it:

train[colSums(train)>0]

---

Edit: (in reply to #3)

@TalhaAsif that's not what your main thread says. It talks about columns with just 0's.

For your changed requirement, try something like this.

train[lapply(train, function(x) length(unique(x))) > 1]

[TalhaAsif]

No, I just want to keep all the columns with length 2. Like this:

length(unique(train$makeAcura))
[1] 2

And delete columns with:

length(unique(train$makeAm.General))
[1] 1

[TalhaAsif]

please see the next reply

[nirgrahamuk]

library(tidyverse)

#(train <- head(iris) %>% mutate(exampl0 = 0))

select(train,
       where(~length(unique(.))>1))

[TalhaAsif]

thanks a lot, it works.