Search Archives
Search tips
from:Name
Search by author name, e.g. from:Duncan Murdoch
"exact phrase"
Match an exact phrase
word1 word2
Match messages containing both words
Date range
Use the date pickers to filter results to a time period
Use the list dropdown to narrow results to a specific mailing list. Combine from: with other terms to filter by author and content.
This first message is just to make sure the archiving works properly. Martin
This is now fixed in R-patched (and R-devel). Martin
It looks to me that R-forge has been down for at least two hours now...
Did you try pam() from package "cluster"? I'd recommend it as an improvement compared to kmeans(). Martin Maechler, ETH Zurich
Thank you, Ivan, for the documentation update; Yes, such small "fixes"/patches are welcome as well. Martin
Please stop posting your homework problems to R-help! The posting guide quite explictly does not allow it. M.Maechler, ETH Zurich
Thank you, Ben! I've added your suggestion to R-devel. Martin
Thank you, Ross. This was still in R-devel and is now fixed. Regards, Martin Mächler
I'm replying to this on the R-devel mailing list and the Subject "On modes, types and R documentation" Martin
Thank you, John. This has been fixed in "R-patched" to become R 1.7.1 early June. Martin
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-devel/attachments/20100518/e10d0ddf/attachment.pl>
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20100518/8fba2602/attachment.pl>
Thank you, Arni, for your bug report! However, this has been reported before, as PR#2295, and fixed for "R-patched" quite a while. Martin
Hi Jim, are you sure you haven't confused *source* packages with *installed* (aka binary) packages ?? Source packages' ./data/ subdirectories have no "funny files" in them. Martin
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20130423/57eb7217/attachment.pl>
Without going into details, did you remember that the intercept depends very much on the **contrasts** you use (for your factors)? Regards, Martin
Thank you David, the report and the patch look perfectly valid to me and I will commit a patch shortly {Brian is still traveling currently}. Martin
Well, your example can be simplified to > -1^(1/3) [1] -1 > (-1)^(1/3) [1] NaN Do you see the light? [think "precedence rules"] Regards, Martin
Can't find what you're looking for? Try searching with Google .