I have a dataframe which contains missing values.
# Create dataframe
df <- data.frame(Athlete = c(c(replicate(200,"Ali"), c(replicate(200,"Tyson")))),
Score = replicate(400, sample(c(1:20, NA), 1, rep = TRUE)))
My function groups factors, then counts rows which do not contain NA values.
library(dplyr)
Result <- df %>%
dplyr::group_by(Athlete, .drop = TRUE) %>%
dplyr::summarise_each(list(~sum(!is.na(.))))
I get the desired result. But there is a warning message.
`summarise_each_()` is deprecated as of dplyr 0.7.0.
Please use `across()` instead.
I'm trying to update the code base so the warning messages stop.
Note: The warning message also says;
This warning is displayed once every 8 hours.
Call `lifecycle::last_warnings()` to see where this warning was generated.
So if the warning message is absent, reboot Rstudio and rerun script to produce the warning message.
summarise_each was replaced with summarise_at/summarise_all which is now replaced with across in dplyr 1.0.0.
library(dplyr)
df %>%
group_by(Athlete) %>%
summarise(across(everything(), ~sum(!is.na(.))))
# Athlete Score
# <chr> <int>
#1 Ali 189
#2 Tyson 195
Although, if you have only one column to summarise as shown you can do this directly :
df %>%
group_by(Athlete, .drop = TRUE) %>%
summarise(Score = sum(!is.na(Score)))