You can probably get around it with casting, but I ended up using wrapRDD -- which is not a static method -- from another JavaRDD in scope to address this more directly without casting or warnings. It's not ideal but both should work, just a matter of which you think is less hacky.
On Fri, Dec 5, 2014 at 5:51 PM, Xingwei Yang <happy...@gmail.com> wrote: > I use Spark in Java. > > I want to access the vectors of RowMatrix M, thus I use M.rows(), which is a > RDD<Vector> > > I want to transform it to JavaRDD<Vector>, I used the following command; > > JavaRDD<Vector> data = JavaRDD.fromRDD(M.rows(), > scala.reflect.ClassTag$.MODULE$.apply(Vector.class); > > However, it shows a error like this: > > The method fromRDD(RDD<T>, ClassTag<T>) in the type JavaRDD is not > applicable for the arguments (RDD<Vector>, ClassTag<Object>) > > Is there anything wrong with the method? Thanks a lot. > > > -- > Sincerely Yours > Xingwei Yang > https://sites.google.com/site/xingweiyang1223/ > --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
