On Thu, Mar 4, 2010 at 9:10 AM, Ted Dunning <ted.dunn...@gmail.com> wrote:
> To be conformable, the dictionaries must be the identical object. > > At that point, you do what we do now. The labels are irrelevant to the dot > product and are only used during input and output. > > a.times(b) // this checks that a.d == b.d, then does > a.rawMatrix.times(b.rawMatrix) > * > make sense? > Ok, you're just saying that you can *have* labels without loss of performance, but the calculations are all still done with the integers? What do the labels actually serve as a use here? Just for mapping back to user-space at the end of the day? If so, this can be done distributed to, as long as row and column dictionaries can be given some sense of a "UID" to identify themselves well enough (even just doing string equality on the Path URI to the dictionary should be fine). -jake > > On Thu, Mar 4, 2010 at 8:59 AM, Jake Mannix <jake.man...@gmail.com> wrote: > > > On Thu, Mar 4, 2010 at 8:54 AM, Ted Dunning <ted.dunn...@gmail.com> > wrote: > > > > > I haven't examined the out-of-core scenarios at all, but in-memory, it > is > > > possible to have labels with no performance cost if you assume add the > > > constraint that labeled matrices are only conformable if they share the > > > identical label dictionary. That implies that you can use the internal > > row > > > and column indexes for all internal operations. > > > > > > Care to elaborate? If you're multiplying two a matrix by a vector, both > > labeled by Map<Integer,String> and reverse Map<String,Integer> for both > > the rows and columns (and they match in the right way), what is the fast > > way to do the individual dot products, which performs comparably to > > walking the sparse int[] / double[] parallel arrays? > > > > -jake > > > > > > -- > Ted Dunning, CTO > DeepDyve >
