Ziqi Zhang
2015-Aug-07 14:40 UTC
[R] fpc package - computing silhouette value returns error
Hi I am using R to do clustering and compute a number of cluster statistics. Below are my code. I do not understand why I get the error: "Error in summary(silhouette(clustering[ss[[i]]], dx)) : error in evaluating the argument 'object' in selecting a method for function 'summary': Error in round(x) : non-numeric argument to mathematical function" Any suggestsion highly appreciated! Code: -------------- library(fpc) require(graphics) distance <- dist(USArrests, "euclidean") hc <-hclust(distance, "average") distcritmulti(USArrests, hc, criterion="asw", fun="dist", metric="euclidean") ------------- -- Ziqi Zhang Research Associate Department of Computer Science University of Sheffield --- This email has been checked for viruses by Avast antivirus software.
William Dunlap
2015-Aug-07 15:40 UTC
[R] fpc package - computing silhouette value returns error
help(distcritmulti) says that its 2nd argument is "clustering: vector of integers indicating the clustering" and you supplied the output of hclust, which is not a vector of group identifiers. You can make a group identifier vector from hc with cutree(hc, k=4), where k is the number of groups. Bill Dunlap TIBCO Software wdunlap tibco.com On Fri, Aug 7, 2015 at 7:40 AM, Ziqi Zhang <ziqi.zhang at sheffield.ac.uk> wrote:> Hi > I am using R to do clustering and compute a number of cluster statistics. > Below are my code. I do not understand why I get the error: > > "Error in summary(silhouette(clustering[ss[[i]]], dx)) : error in > evaluating the argument 'object' in selecting a method for function > 'summary': Error in round(x) : non-numeric argument to mathematical > function" > > Any suggestsion highly appreciated! > Code: > -------------- > library(fpc) > require(graphics) > distance <- dist(USArrests, "euclidean") > hc <-hclust(distance, "average") > distcritmulti(USArrests, hc, criterion="asw", fun="dist", > metric="euclidean") > ------------- > > > > -- > Ziqi Zhang > Research Associate > Department of Computer Science > University of Sheffield > > > --- > This email has been checked for viruses by Avast antivirus software. > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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. >[[alternative HTML version deleted]]