I'm running Mclust 3.0 in R-2.7.0 and have run into a situation where
the optimal number of clusters, 12, seems to be greater than the number
of colors available to Mclust. The code below, which demonstrates
errors, does not if the max no of clusters is set at 10. sessionInfo,
after a restart of R and loading of packages, follows the output.
Is this a known problem with Mclust? Can I supply more colors?
As an aside, Mclust is crashing when it runs out of memory with high
dimensional datasets. R and Emacs freeze, my SystemMonitor in Ubuntu
shows that the CPU grinds to a halt but the memory is exhausted.
Shouldn't a process in R exit and return an error when memory can no
longer be allocated? Wishful thinking????
Mark
> exprs.clust <- Mclust(exprs(AOP.sig), G=9:12,
modelNames="VEI")
Warning messages:
1: In summary.mclustBIC(Bic, data, G = G, modelNames = modelNames) :
best model occurs at the min or max # of components considered
2: In Mclust(exprs(AOP.sig), G = 9:12, modelNames = "VEI") :
optimal number of clusters occurs at max choice
> plot(exprs.clust, data = exprs(AOP.sig))
Hit <Return> to see next plot:
Hit <Return> to see next plot:
Hit <Return> to see next plot:
Hit <Return> to see next plot:
Warning message:
In clPairs(data[, 1:min(5, p)], classification = x$classification, :
more colors needed
> myMclustOutput.func(exprs.clust)
#Above just outputs the classificatin and uncertainty in a data.frame.
#Usually works fine
Error: bad value
No suitable frames for recover()
> 2 + 4
Error: bad value
No suitable frames for recover()
> sessionInfo()
Error: bad value
No suitable frames for recover()
#After R-devel restart
> sessionInfo()
R version 2.7.0 Under development (unstable) (2008-03-05 r44683)
x86_64-unknown-linux-gnu
locale:
LC_CTYPE=en_US.UTF-8;LC_NUMERIC=C;LC_TIME=en_US.UTF-8;LC_COLLATE=en_US.UTF-8;LC_MONETARY=en_US.UTF-8;LC_MESSAGES=en_US.UTF-8;LC_PAPER=en_US.UTF-8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=en_US.UTF-8;LC_IDENTIFICATION=C
attached base packages:
[1] tools stats graphics grDevices datasets utils methods
[8] base
other attached packages:
[1] lme4_0.99875-9 Matrix_0.999375-5 lattice_0.17-6
[4] affy_1.17.9 preprocessCore_1.1.5 affyio_1.7.17
[7] Biobase_1.99.3 mclust_3.1-3
loaded via a namespace (and not attached):
[1] grid_2.7.0
>
--
Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry
Indiana University School of Medicine
15032 Hunter Court, Westfield, IN 46074
(317) 490-5129 Work, & Mobile & VoiceMail
(317) 204-4202 Home (no voice mail please)
mwkimpel<at>gmail<dot>com
