Cluto assertion error


I'm tring to use cluto to compute the Gap Statistic and i'm running into an odd assertion error.

When running the following command

vcluster -clmethod=direct -clustfile=/tmp/gap-clustering-output7524246358192229718.matrix -crfun=h2 /tmp/cluto-sparse-matrix6495249055097069589.dat 6

with a sparse matrix file that has 103 rows, 5000 columns, and 1071 non zero values, i get the following output:

vcluster (CLUTO 2.1.2) Copyright 2001-06, Regents of the University of Minnesota

Matrix Information -----------------------------------------------------------
Name: /tmp/cluto-sparse-matrix3460122025714674563.dat, #Rows: 103, #Columns: 5000, #NonZeros: 1071

Options ----------------------------------------------------------------------
CLMethod=Direct, CRfun=H2, SimFun=Cosine, #Clusters: 5
RowModel=None, ColModel=IDF, GrModel=SY-DIR, NNbrs=40
Colprune=1.00, EdgePrune=-1.00, VtxPrune=-1.00, MinComponent=5
CSType=Best, AggloFrom=0, AggloCRFun=H2, NTrials=10, NIter=10

Solution ---------------------------------------------------------------------

WARNING: Cluto exited with error status. 134 stderr:
vcluster: kwaydirect.c:116: cluto_PartitionMatKWayDirect: Assertion `cnorm >= 0.0' failed.

I'm unsure of what this error implies. I've run the same command with a smaller number of clusters, 1 up to 5, and everything has worked well. Is there any reason 6 clusters would cause cluto to fail an assertion when fewer clusters don't?

Thank you

RE: This looks like a bug. Can

This looks like a bug. Can you send me the input file.