This will not work for "anonymous" R objects, that R objects not having
an associated symbol (e.g., x = rpy2.robjects.IntVector([1,2,3]) ).
For what it's worth, pickling of low-level objects (rpy2.rinterface) is
working in the tip on bitbucket (that's the dev branch for 2.1). It is
using serialize()/unserialize() with raw vectors.
L.
On Thu, 2009-04-30 at 03:09 +0200, Brandon Invergo wrote:
> FWIW, the solution I have for now is to use dump(ls(), file="") to get a
> deparsed string representation of all the objects in memory, which can
> then be pickled. I'm not having problems with it right now in any simple
> cases, but I can imagine there are situations where deparsing gets
> hairy, so being able to pickle the R globalEnv or somehow being able to
> route the file output of R's save.image command to a python object that
> can then be pickled would be the best....
> Cheers,
> -brandon invergo
>
> On Thu, 2009-04-30 at 01:01 +0200, Brandon Invergo wrote:
> > I'm trying to do something similar: I'd like to create a save file that
> > includes the current state of the R environment. I don't want to create
> > a separate file for the R environment via one of R's built-in commands,
> > so I thought I'd pickle globalEnv alongside some other objects I'm
> > pickling. When I try to pickle the globalEnv, I get this error:
> > cPickle.PicklingError: Can't pickle <type 'rinterface.SexpEnvironment'>:
> > import of module rinterface failed
> >
> >
> > How have other people gone about storing their R objects?
> >
> > Thanks!
> > brandon invergo
> >
> >
> >
> > On Mon, 2009-04-06 at 11:29 -0400, Réjean Ducharme wrote:
> > > Hi all,
> >
> > I'm using Rpy2 to train a logistic regressor (using the "glm" function
> > in R), and I need to save the model for futher use.
> > I naively try to use pickle, but I have an error message:
> >
> > import rpy2.robjects as robjects
> > fit = robjects.r.glm(formula=f, family=robjects.r.binomial(), data=data)
> > output = open('glm.pickle', 'w')
> > pickle.dump(fit, output)
> >
> > >>> pickle.PicklingError: Can't pickle <type
> > 'rinterface.SexpVector'>:
> > it's not found as rinterface.SexpVector
> >
> >
> > Is there another way to save a Rpy2 object???
> >
> > Thanks,
> > Rejean Ducharme
> >
> >
> > ------------------------------------------------------------------------------
> > Register Now & Save for Velocity, the Web Performance & Operations
> > Conference from O'Reilly Media. Velocity features a full day of
> > expert-led, hands-on workshops and two days of sessions from industry
> > leaders in dedicated Performance & Operations tracks. Use code vel09scf
> > and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
> > _______________________________________________
> > rpy-list mailing list
> > rpy-list@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/rpy-list
>
>
> ------------------------------------------------------------------------------
> Register Now & Save for Velocity, the Web Performance & Operations
> Conference from O'Reilly Media. Velocity features a full day of
> expert-led, hands-on workshops and two days of sessions from industry
> leaders in dedicated Performance & Operations tracks. Use code vel09scf
> and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
> _______________________________________________
> rpy-list mailing list
> rpy-list@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rpy-list
------------------------------------------------------------------------------
Register Now & Save for Velocity, the Web Performance & Operations
Conference from O'Reilly Media. Velocity features a full day of
expert-led, hands-on workshops and two days of sessions from industry
leaders in dedicated Performance & Operations tracks. Use code vel09scf
and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
_______________________________________________
rpy-list mailing list
rpy-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rpy-list
