How can I flatten the output of R’s base::table function (applied to columns of a dataframe) in order to return an integer vector with names according to the convention specified below?
set.seed(123); N <- 100
df <- data.frame(A = sample(c("GOOD", "BAD"), N, TRUE, c(0.4, 0.6)),
B = sample(c("GOOD", "BAD"), N, TRUE, c(0.7, 0.3)))
table(df)
# B
# A BAD GOOD
# BAD 16 44
# GOOD 10 30
# Desired output (naming convention: A-B)
# BAD-BAD GOOD-BAD BAD-GOOD GOOD-GOOD
# 16 10 44 30
Using as.vector(table(df)) drops names, which I’d like to retain in one step for use downstream for example, in case the ordering of factors changes or to enable "toggling" of useNA = "always" when calling table whilst not having to manually track the positions of the respective cells in the output vector.
>Solution :
From memory (currently no access to R):
table(interaction(df$A, df$B, sep = "-"))
Could be that you need to specify the correct separator to interaction