Comparison of operations on normal and immutable (see package plyr) data frames

idata_frame.R

# load packages
require(plyr); require(reshape2)

# Make immutable data frame
baseball_i <- idata.frame(baseball)

# Example 1 - idata.frame more than twice as fast
system.time( replicate(50, ddply( baseball, "year", summarise, mean(rbi))) )
system.time( replicate(50, ddply( baseball_i, "year", summarise, mean(rbi))) )

# Example 2 - Bummer, this does not work with idata.frame's
colwise(max, is.numeric) ( baseball ) # works
colwise(max, is.numeric) ( baseball_i ) # doesn't work

# Example 3 - idata.frame twice as fast
system.time( replicate(100, baseball[baseball$year == "1884", ] ) )
system.time( replicate(100, baseball_i[baseball_i$year == "1884", ] ) )

# Example 4 - idata.frame faster
system.time( replicate(50, melt(baseball[, 1:4], id = 1) ) )
system.time( replicate(50, melt(baseball_i[, 1:4], id = 1) ) )

# And you can go back to a data frame by 
d <- as.data.frame(baseball_i)
str(d)


# idata.frame doesn't work with the doBy package
require(doBy)
summaryBy(rbi ~ year, baseball_i, FUN=c(mean), na.rm=T)

# But idata.frame works with aggregate in base (but with minimal speed gains)
# and aggregate is faster than ddply 
system.time( replicate(100, aggregate(rbi ~ year, baseball, mean) ) )
system.time( replicate(100, aggregate(rbi ~ year, baseball_i, mean) ) )
system.time( replicate(100, ddply( baseball_i, "year", summarise, mean(rbi)) ) )