Python bindings for GraphX are still in development - see https://issues.apache.org/jira/browse/SPARK-3789?jql=project%20%3D%20SPARK%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20GraphX%20ORDER%20BY%20updated%20DESC <https://issues.apache.org/jira/browse/SPARK-3789?jql=project%20=%20SPARK%20AND%20resolution%20=%20Unresolved%20AND%20component%20=%20GraphX%20ORDER%20BY%20updated%20DESC> for the latest.
------------------------------------------------------------------------------- Robin East Spark GraphX in Action Michael Malak and Robin East Manning Publications Co. http://www.manning.com/books/spark-graphx-in-action <http://www.manning.com/books/spark-graphx-in-action> > On 7 Dec 2015, at 02:16, 张志强(旺轩) <zzq98...@alibaba-inc.com> wrote: > > Yes, > > We already used ALS in our production environment, we also want to try SVD++ > but it has no python interface. > > Any idea? Thanks > > -Allen > > 发件人: Yanbo Liang [mailto:yblia...@gmail.com <mailto:yblia...@gmail.com>] > 发送时间: 2015年12月3日 10:30 > 收件人: 张志强(旺轩) > 抄送: dev@spark.apache.org <mailto:dev@spark.apache.org> > 主题: Re: query on SVD++ > > You means the SVDPlusPlus in GraphX? If you want to use SVD++ to train CF > model, I recommend you to use ALS which is more efficiency and has python > interface. > > 2015-12-02 11:21 GMT+08:00 张志强(旺轩) <zzq98...@alibaba-inc.com > <mailto:zzq98...@alibaba-inc.com>>: > Hi All, > > I came across the SVD++ algorithm implementation in Spark code base, but I > was wondering why we didn’t expose the scala api interface to python? > Any plan to do this? > > BR, > -Allen Zhang
