CODEWITHSUNDEEP
rdplyr
Fiona
Fiona

Reputation: 53

Filter_all(any_vars()) Where am I going wrong?

I have a dataframe with 51 obs. of 20 variables called test. Like below:

X1 P1 P2 P3
A  1  0  5
B  3  9  0
C  5  0  0
D  1  0  0

I want to remove all rows where there are values under or equal to 1. So the resulting dataframe would only have A,B,C rows. I thought I cracked it using the below command:

test2 <- test %>% filter_all(any_vars(. > 2))

I also tried

test2 <- test %>% select(X1, P1:P3) %>% filter_all(any_vars(. > 2).

This doesn't remove any of the rows even though I can see rows have values under or equal to 1.

Any help would be appreciated

Upvotes: 1

Views: 414

Answers (2)

Ronak Shah
Ronak Shah

Reputation: 389135

In base R, you can take help of rowSums -

res <- test[rowSums(test[-1] > 2) > 0, ]
res

#  X1 P1 P2 P3
#1  A  1  0  5
#2  B  3  9  2
#3  C  5  0  0

Upvotes: 0

Anoushiravan R
Anoushiravan R

Reputation: 21938

You can use the following solution:

library(dplyr)

df %>%
  filter(if_all(everything(), ~ .x > 1))

With a modified data set it returns:

  X1 P1 P2 P3
1  B  3  9  2

Data

structure(list(X1 = c("A", "B", "C", "D"), P1 = c(1L, 3L, 5L, 
1L), P2 = c(0L, 9L, 0L, 0L), P3 = c(5L, 2L, 0L, 0L)), class = "data.frame", row.names = c(NA, 
-4L))

Here is another way of doing this with purrr::pmap:

  • pmap is used for row-wise operations, here we capture each row of our data set in every iteration with c(...) while excluding the first variable X1
  • Then I check if all of them are greater than 1
df %>%
  filter(pmap_lgl(df[-1], ~ all(c(...) > 1)))

  X1 P1 P2 P3
1  B  3  9  2

Upvotes: 1

Related Questions