Hi,
In python the results for a specific input look like this:
['modelName', 'n', 'd', 'G', 'BIC', 'bic', 'loglik', 'parameters', 'z',
'classification', 'uncertainty']
Mean:
[[ 0.20609646 -0.40144849]
[ 0.86991148 0.82791771]]
BIC:
[[-8035.90938198]
[-7677.29330621]]
1
0
Variance$sigma:
[[[ 1.23021647 0.75045067]
[ 0.37992429 0.21127281]]
[[ 0.37992429 0.21127281]
[ 0.18390347 0.07193298]]]
But in R for the same inputs the outputs are:
> colorMclust$parameters
$Vinv
NULL
$pro
[1] 0.479828 0.520172
$mean
[,1] [,2]
[1,] 0.84819158 -0.2718649
[2,] -0.01039735 0.9009248
$variance
$variance$modelName
[1] "VVV"
$variance$d
[1] 2
$variance$G
[1] 2
$variance$sigma
, , 1
[,1] [,2]
[1,] 0.10121396 -0.08937613
[2,] -0.08937613 1.06053090
, , 2
[,1] [,2]
[1,] 0.95559538 -0.09661115
[2,] -0.09661115 0.11975590
Do you think whether there is a practical solution for my problem or not?
Zahra.
On Tue, Mar 2, 2010 at 6:53 PM, Peter <rpy-l...@maubp.freeserve.co.uk>wrote:
> On Tue, Mar 2, 2010 at 5:48 PM, zahra sheikhbahaee
> <sheikhbah...@gmail.com> wrote:
> > Hi all,
> >
> > I am using mclust package of R. I checked the results which I have
> already
> > covert by rpy2 to python by R's one. It calculates the covariance matrix
> of
> > my data but I got surprised because it seems that the numbers changed.
> They
> > are totally different. I wonder whether there is any way to solve my
> problem
> > or it is impossible??
> >
> > Zahra.
>
> Can you construct a (small) self contained example to show the issue?
>
> Peter
>
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> rpy-list mailing list
> rpy-list@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rpy-list
>
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
rpy-list mailing list
rpy-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rpy-list
