> 1. matrices are stored columnwise so R is better at column-wise operations > than row-wise.
I am seeing this by my code which contains more t() than what seems healthy. However, the summaries are patient-wise over repeated measurements. Out of convention, I am storing patients in rows and measurements in columns. > > 2. Here is one way to do it (although I am not sure its better than the > index approach): > > row.apply <- function(f, a, b) > t(mapply(f, as.data.frame(t(a)), as.data.frame(t(b)))) > Ah, thank you so much. I'll take the generalization to N arguments à la mapply() as an exercise for the reader. > 3. The code for the example in this post could be simplified to: > > first.1 <- apply(cbind(goldstandard, 1), 1, which.max) > ifelse(col(newtest) > first.1, NA, newtest) > Ouch! Consider this scholar slapped. > 4. given that both examples did not inherently need row by row operations > I wonder if that is the wrong generalization in the first place? > Given that you managed to squeeze my 20 lines of code into 2 lines AND that row.apply() does not exist in base without many people missing it, I'll have to concede this point and eliminate the craving for row.apply() in favour of the whole-object approach. ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
