Error : Error in grouped_df_impl(data, unname(vars), drop) : Column col1 is unknown
pivot <- function(df, col1, col2){
df %>%
group_by(col1, col2) %>%
summarise(n = n()) %>%
spread(col2, n, fill = 0)
}
Running the command pivot (iris, Species, Petal.Width)
Can anyone tell why this error message ? and how to solve it
You need to use group_by_ for programming.
library(tidyverse)
pivot <- function(df, col1, col2){
df %>%
group_by_(col1, col2) %>%
summarise(n = n()) %>%
spread(col2, n, fill = 0)
}
pivot (iris, "Species", "Petal.Width")
This should work too. Don't need to quote the columns here.
pivot <- function(df, col1, col2) {
group_var1 <- enquo(col1)
group_var2 <- enquo(col2)
df %>%
group_by(!!group_var1, !!group_var2) %>%
summarise(n = n())
}
pivot (iris, Species, Petal.Width)
The output on the second is easier to read, I think.
For more information, see dplyr: How to use group_by inside a function? and https://dplyr.tidyverse.org/articles/programming.html.
Another answer is:
pivot <- function(df, col1, col2){
col1<-deparse(substitute(col1))
col2<-deparse(substitute(col2))
df %>%
group_by_(col1, col2) %>%
summarise(n = n()) %>%
spread(col2, n, fill = 0)
}
pivot (iris, Species, Petal.Width)
