Dear Paul,
On Sat, Aug 20, 2011 at 4:35 PM, <[email protected]> wrote:
>
> after having trained a model
> "
> cmp = Composite()
> cmp.Grow
> (pts,attrs=attrs,nPossibleVals=nPossible,nTries=1,buildDriver=CrossValidate.CrossValidationDriver,treeBuilder=QuantTreeBoot,
> Â needsQuantization=False,nQuantBounds=boundsPerVar, maxDepth=3)
> "
>
> How can this model be stored on disk?
> pickle will not help me, because this only works for molecules, or am I
> wrong?
You should be able to store it to disk by pickling without problems,
something like:
cPickle.dump(cmp,file('cmp.pkl','wb+'))
> I would like to store this model on disk, because I would like predict
> molecules in a later stage without training evry time running the
> prediction.
> For the prediction part, I will be using "cmp.ClassifyExample()" - or is
> there any other possibility?
Pickling the models and then using ClassifyExample is correct.
Best,
-greg
------------------------------------------------------------------------------
uberSVN's rich system and user administration capabilities and model
configuration take the hassle out of deploying and managing Subversion and
the tools developers use with it. Learn more about uberSVN and get a free
download at: http://p.sf.net/sfu/wandisco-dev2dev
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
