Efficient row-wise operations on a data.table

After some discussion around row-wise first/last occurrences from column series in data.table, which suggested that melting first would be faster than a row-wise calculation, I decided to benchmark:

• pmin (Matt Dowle's answer above), below as tm1
• apply (Andrie's answer above), below as tm2
• melting first, then min by group, below as tm3

so:

library(microbenchmark); library(data.table)
set.seed(1000)
b <- data.table(m=integer(), n=integer(), tm1 = numeric(), tm2 = numeric(), tm3 = numeric())

for (m in c(2.5,100)*1e5){

  for (n in c(3,50)){
    my.df <- sample(1:5, m*n, replace=TRUE)
    dim(my.df) <- c(m,n)    
    my.df <- as.data.frame(my.df)
    names(my.df) <- c(LETTERS,letters)[1:n]   
    my.dt <- as.data.table(my.df)

    tm1 <- mean(microbenchmark(my.dt[, foo := do.call(pmin, .SD)], times=30L)$time)/1e6
    my.dt <- as.data.table(my.df)
    tm2 <- mean(microbenchmark(apply(my.dt, 1, min), times=30L)$time)/1e6
    my.dt <- as.data.table(my.df)sv
    tm3 <- mean(microbenchmark(
                melt(my.dt[, id:=1:nrow(my.dt)], id.vars='id')[, min(value), by=id], 
                times=30L
               )$time)/1e6
    b <- rbind(b, data.table(m, n, tm1, tm2, tm3) ) 
  }
}

(I ran out of time to try more combinations) gives us:

b
#          m  n        tm1       tm2         tm3
# 1: 2.5e+05  3   16.20598  1000.345    39.36171
# 2: 2.5e+05 50  166.60470  1452.239   588.49519
# 3: 1.0e+07  3  662.60692 31122.386  1668.83134
# 4: 1.0e+07 50 6594.63368 50915.079 17098.96169
c <- melt(b, id.vars=c('m','n'))

library(ggplot2)
ggplot(c, aes(x=m, linetype=as.factor(n), col=variable, y=value)) + geom_line() +
  ylab('Runtime (millisec)') + xlab('# of rows') +  
  guides(linetype=guide_legend(title='Number of columns'))

Although I knew apply (tm2) would scale poorly, I am surprised that pmin (tm1) scales so well if R is not really designed for row-wise operations. I couldn't identify a case where pmin shouldn't be used over melt-min-by-group (tm3).

Or, just pmin.

my.dt <- as.data.table(my.df)
system.time(my.dt[,row.min:=pmin(A,B,C)])
# user  system elapsed 
# 0.02    0.00    0.01 
head(my.dt)
#      A B C row.min
# [1,] 2 5 2       2
# [2,] 4 5 5       4
# [3,] 1 5 3       1
# [4,] 4 4 3       3
# [5,] 3 5 5       3
# [6,] 1 5 3       1

The classical way of doing row-wise operations in R is to use apply:

apply(my.df, 1, min)
> head(my.df)
  A B C min
1 2 5 4   2
2 4 3 1   1
3 1 1 5   1
4 4 1 5   1
5 3 3 4   3
6 1 1 1   1

On my machine, this operation takes about 0.25 of a second.