CODEWITHSUNDEEP
rdplyrset-difference
Navya
Navya

Reputation: 287

How to print observations existing in one dataframe and missing from the other r?

I have a one dataframe with 3719 (actual data) rows and another with 3721 (from coding) rows. I got 2 extra observations.

I have tried with setdiff but it is giving zero rows

dplyr::setdiff(d1,d2)

o/p: [1] col1 col2 col3 col4       
     [5] col5 col6                
<0 rows> (or 0-length row.names)

I have tried vise versa also,i.e, 

dplyr::setdiff(d2,d1)

o/p: [1] col1 col2 col3 col4       
     [5] col5 col6                
<0 rows> (or 0-length row.names)

How to identify those 2 extra observation in R?

Upvotes: 0

Views: 69

Answers (2)

r.user.05apr
r.user.05apr

Reputation: 5456

Anti-join would be the "tidy" option:

library(tidyverse)

d1 <- tribble(~a, ~b,
              "a", 3,
              "f", 9,
              "g", 10)

d2 <- tribble(~a, ~b,
              "a", 333,
              "b", 999,
              "f", 444,
              "g", 111)

d2 %>%
  anti_join(d1, by = "a")

# A tibble: 1 x 2
# a         b
# <chr> <dbl>
#   1 b       999

Upvotes: 2

MatthewR
MatthewR

Reputation: 2770

Option 1 You can use the %in% operator

#Make Fake Data
a <- mtcars
b <- mtcars[ 3:nrow(mtcars) , ] 

a$id <- rownames( a )
b$id <- rownames( b )

#In A not B
a[ !(a$id %in% b$id) , ]
#In B not A
b[ !(b$id %in% a$id) , ]

Option 2 - use merge with all=T

a$flaga <- 1
b$flagb <- 1

d <- merge( a[ ,c("id","flaga")] , b[ ,c("id","flagb")], by= "id" , all=T)

d[ is.na(d$flaga) | is.na(d$flagb) , "id" ]

Upvotes: 0

Related Questions