Hello, I would like to have any help with the function Kmeans of R.. I use this to do a classification of my data...I have chosen 12 classes but, I have always an error message: Error: empty cluster: try a better set of initial centers So, I don't understand the probleme with this function.. Thank you to help me!! All the Best Clothilde Clothilde Kussener CNRS - CEBC 79360 Villiers en bois France --------------------------------- [[alternative HTML version deleted]]
Hello, When clustering with kmeans, your data should have more than one variable. Matthias -----Urspr??ngliche Nachricht----- Von: clothilde kussener [mailto:clothilde13 at yahoo.fr] Gesendet: Dienstag, 11. Mai 2004 12:01 An: r-help at stat.math.ethz.ch Betreff: [R] Probleme with Kmeans... Hello, I would like to have any help with the function Kmeans of R.. I use this to do a classification of my data...I have chosen 12 classes but, I have always an error message: Error: empty cluster: try a better set of initial centers So, I don't understand the probleme with this function.. Thank you to help me!! All the Best Clothilde Clothilde Kussener CNRS - CEBC 79360 Villiers en bois France --------------------------------- [[alternative HTML version deleted]] ______________________________________________ R-help at stat.math.ethz.ch 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
clothilde kussener wrote:> Hello, > I would like to have any help with the function Kmeans of R.. > I use this to do a classification of my data...I have chosen 12 classes but, I have always an error message: > Error: empty cluster: try a better set of initial centers > > So, I don't understand the probleme with this function.. > Thank you to help me!!Well, after running, there are less than 12 clusters with observations found by kmeans(). The error message suggests to specify initial cluster centers instead of randomly chosen ones, if you have got a rough guess where centers will be. BTW: If you have got known "classes", don't you want to perform classification rather than clustering??? Uwe Ligges> All the Best > Clothilde > > Clothilde Kussener > CNRS - CEBC > 79360 Villiers en bois > France > > > --------------------------------- > > > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch 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