I have a dataframe in R with lots of columns and lots of rows with numeric values. I have multiple colums with the same name (e.g. A, A, B, B, C, C...) that should be combined into a single colum with the maximum of the rows (A,B,C)

So this

Should be combined to

Can somebody help me with that?

I found an example online that is used to summarize row values of columns with the same name

df<-
data.frame(x=rpois(25,1),y=rpois(25,2),x=rpois(25,5),z=rpois(25,2),y=rpois(25,1),z=rpoi
s(25,5),check.names=FALSE)
Merged_df<-as.data.frame(do.call(cbind,
by(t(df),INDICES=names(df),FUN=colSums)))
Merged_df

and replaced FUN=colSums it with

 FUN = function(x) max(x,na.rm = TRUE)

but that resulted in an error

Error in tapply(seq_len(3635L), list(INDICES = NULL), function (x)  : 
  arguments must have same length

Now i am not sure if this is a problem in my data or in the function i was using as i am not experienced with R.

CodePudding user response：

You can use split.default to split the columns according to their names, and then get the rowwise max with do.call and pmax, before binding the rows.

dat <- data.frame(A = 1:2, A = c(0, 3), B = 3:4, B = 0:1, check.names = F)
#   A A B B
# 1 1 0 3 0
# 2 2 3 4 1

split.default(dat, colnames(dat)) |>
  lapply(do.call, what = "pmax") |>
  rbind.data.frame()
#   A B
# 1 1 3
# 2 3 4