Hi Jacob, Thank you for clarification.
My problem however is the size of data in terms of number of samples. The features are engineered and are only 80. I wanted to try training on bigger set of data for improvement. Thanks & Best, Maryam > > ------------------------------ > > Message: 3 > Date: Wed, 30 Sep 2015 12:11:00 -0700 > From: Jacob Schreiber <jmschreibe...@gmail.com> > Subject: Re: [Scikit-learn-general] Scalability of Gradient Boosting > Classifier > To: scikit-learn-general@lists.sourceforge.net > Message-ID: > < > ca+ad8etyev331pfafp60jctxj5sbtbsf+yj19kg9tzodee-...@mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Hi Maryam > > Currently, no tree based methods have a partial fit method. We are > currently working on expanding the tree module, you can see our checklist > here; https://github.com/scikit-learn/scikit-learn/issues/5212 > > There are many methods to reduce the dimensionality of data, if you are > using high dimensional data, including Gaussian random projections or LSH > for continuous data, and collapsing samples into higher weight samples for > discrete data. If you provide more information about your use case, I may > be able to be of more help. > > Jacob > > On Wed, Sep 30, 2015 at 5:12 AM, Maryam Tavakol < > maryam.tavakol...@gmail.com > > wrote: > > > Dear all, > > > > I am using Gradient Boosting Classifier from scikit-learn for a huge set > > of data. Unfortunately, the method loads the whole data into memory > (around > > 45 GBs!). As it is not very easy to modify the code to stream data, is > > there any other way to make it scalable? > > > > Best Regards, > > Maryam Tavakol > > > > > > > ------------------------------------------------------------------------------ > > > > _______________________________________________ > > Scikit-learn-general mailing list > > Scikit-learn-general@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general > > > > >
------------------------------------------------------------------------------
_______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
