/Why not modify the C code of knn? (Not knn1)/
Because I thought it would not be a good idea to apply 'knn' a 1000 times or more.
Instead I was looking for a procedure that returns a structure making it easy to find nearest neighbors for any point one wants. This way I used 'dist' for smaller data sets.
The next step could be to extent nearest neighbors to data frames with factors generalizing, for example, the 'daisy' function.
But you might be right that looking at the 'knn' implementation will be a faster road for the moment.
Hans Werner Borchers ABB Corporate Research, Germany
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
