[BioC] Problems with clValid: "fanny unable to find clusters"
Hervé Pagès
hpages at fhcrc.org
Thu Mar 26 02:38:36 CET 2009
Hi Jörg,
Even if it suggests a few BioC packages, clValid is not a BioC package
(it's hosted on CRAN). You will get better results if you ask directly
to its authors/maintainer (see DESCRIPTION file).
Cheers,
H.
Jörg Linde wrote:
> Hello,
>
> I have been working a lot with the clValid package lately. I find it a really helpful, well written and documented package. However, I ran into a problem and hope someone can help me:
>
> Using this command:
> "
> clust=clValid(obj=matrix,nClust=2:8,validation=c("internal", "stability"),metric ="euclidean", method ="average",neighbSize = 10,maxitems =dim(matrix)[1], clMethods = c("fanny"))"
>
> I receive the following messages:
>
> Warning messages:
> 1: In vClusters(mat, clMethods[i], nClust, validation = validation, :
> fanny unable to find 4 clusters, returning NA for these validation measures
> 2: In vClusters(mat, clMethods[i], nClust, validation = validation, :
> fanny unable to find 5 clusters, returning NA for these validation measures
>
> ...(repeats until 8 clusters)
> There are no fuzzy clusters computed for k=4..8
> However, using fanny directly there is no warning message at all (the same for k=4..8):
> fanny(x=matrix,k=5,metric ="euclidean",maxit = 500)
>
> Note that maxit=500 is the standard and not reached for any of the clusters.
> Interestingly using metric ="correlation" reports no warnings for clValid.
>
> Any idea what the problem might be?
>
> Thanks a lot
> Jörg
>
> --
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
--
Hervé Pagès
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M2-B876
P.O. Box 19024
Seattle, WA 98109-1024
E-mail: hpages at fhcrc.org
Phone: (206) 667-5791
Fax: (206) 667-1319
More information about the Bioconductor
mailing list