Is it possible to use reframe in dplyr to get the former behaviour of summarize?

Question

We've just upgraded from dplyr to version 1.1.0 and now we have a problem with summarize and unique. We're trying to use reframe instead.

the problem is that one variable is sometimes NA and sometimes a value, e.g.

With the old dplyr summarize behaviour, it would pick the non-nan value if there was one and otherwise it would keep it as NAN.

With reframe, it always keeps both the value and the NAN.

output how it should look like

Example 'old' code:

t <- buttons %>%
group_by(envIndex,predIndex) %>%
dplyr::summarize(predatorLocation= unique(predatorLocation)

Is it possible to use reframe to reproduce the former behaviour of summarize?

Welcome to SO! It would be easier to help you if you provide [a minimal reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) including a snippet of your data or some fake data. Please do not post an image of code/data/errors [for these reasons](https://meta.stackoverflow.com/questions/285551/why-not-upload-images-of-code-errors-when-asking-a-question/285557#285557). Just include the code, console output, or data (e.g. dput(head(x)) or data.frame(...)) directly. — stefan, Apr 24 '23 at 09:34
Please provide enough code so others can better understand or reproduce the problem. — Community, Apr 24 '23 at 15:07

jared_mamrot · Answer 1 · 2023-04-24T11:42:29.743

Interesting question; one potential option would be to take the first non-NA value for each group, i.e.

Example data:

library(tidyverse)

data(ChickWeight)

df <- ChickWeight %>%
  slice_head(n = 36) %>%
  mutate(Diet = ifelse(Chick == 1, NA, Diet),
         Diet = ifelse(Chick == 2 & Time < 16, NA, Diet))
df
#>    weight Time Chick Diet
#> 1      42    0     1   NA
#> 2      51    2     1   NA
#> 3      59    4     1   NA
#> 4      64    6     1   NA
#> 5      76    8     1   NA
#> 6      93   10     1   NA
#> 7     106   12     1   NA
#> 8     125   14     1   NA
#> 9     149   16     1   NA
#> 10    171   18     1   NA
#> 11    199   20     1   NA
#> 12    205   21     1   NA
#> 13     40    0     2   NA
#> 14     49    2     2   NA
#> 15     58    4     2   NA
#> 16     72    6     2   NA
#> 17     84    8     2   NA
#> 18    103   10     2   NA
#> 19    122   12     2   NA
#> 20    138   14     2   NA
#> 21    162   16     2    1
#> 22    187   18     2    1
#> 23    209   20     2    1
#> 24    215   21     2    1
#> 25     43    0     3    1
#> 26     39    2     3    1
#> 27     55    4     3    1
#> 28     67    6     3    1
#> 29     84    8     3    1
#> 30     99   10     3    1
#> 31    115   12     3    1
#> 32    138   14     3    1
#> 33    163   16     3    1
#> 34    187   18     3    1
#> 35    198   20     3    1
#> 36    202   21     3    1

Answer:

# Chick 2 has both value and NA in reframed df
df %>%
  reframe(Diet = unique(Diet), .by = Chick)
#>   Chick Diet
#> 1     1   NA
#> 2     2   NA
#> 3     2    1
#> 4     3    1

# Potential solution
df %>%
  reframe(Diet = first(unique(Diet), na_rm = TRUE), .by = Chick)
#>   Chick Diet
#> 1     1   NA
#> 2     2    1
#> 3     3    1

^{Created on 2023-04-24 with reprex v2.0.2}

Does this approach solve your problem?

Is it possible to use reframe in dplyr to get the former behaviour of summarize?

1 Answers1