apply rowwise max is very slow

[thorstenwagner]

The following command needs for a 65700x2 matrix more than 30 seconds in renjin.

apply(fprob[1:nrow(fprob),],1,max)

Here is the fprob object: maxslowRData.zip

In R it is done in a couple of seconds. I'm using Renjin 0.8.2356

What is the reason?

Best, Thorsten

[akbertram]

Good question! The indexing on fprob might be getting deferred... in which case it's possible something is going wrong. Will check it out. Thanks for the benchmark!

[thorstenwagner]

Any progress here? At the moment this is really a bottle neck for me :-( Or are there any alternative to do it faster?

[mjkallen]

Try pmax(fprob[,1], fprob[,2]) since it looks like your trying to find the largest value in each row.

If the number of columns is variable, you can do

do.call(pmax, lapply(seq(ncol(fprob)), function(col) fprob[,col]))

[thorstenwagner]

Thanks! This works perfectly!!!