mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dmitriy Lyubimov <dlie...@gmail.com>
Subject Re: Detecting rank-deficiency, or worse, via QR decomposition
Date Fri, 19 Jul 2013 05:00:56 GMT
it's fast though and doesn't require forming the whole input in memory.

btw Mahout's qr seems to garble the input argument matrix so it is unusable
after that.  It doesn't seem to  document so and intuitively that's not
what one expects. Took me quite a while to figure today where the stuff
went wrong all of a sudden (i start using it to early-test for
decomposition rank deficiency in als-wr). Should probably at least say so.




On Thu, Jul 18, 2013 at 9:11 PM, Ted Dunning <ted.dunning@gmail.com> wrote:

> On Thu, Jul 18, 2013 at 3:35 PM, Dmitriy Lyubimov <dlieu.7@gmail.com>
> wrote:
>
> > > It is hard to frame HH in a row major fashion.  I might be able to
> figure
> > > out a Given's rotation method that is row oriented.
> > >
> >
> > Mahout SSVD already has the row-wise Givens rotation solver . I guess i
> can
> > revive that as a standalone in core solver (even row-wise streaming
> solver
> > as well) if there's interest.
>
>
> I don't think that there is much interest.
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message