No it was actually all over the place. It was strange. I ended up using the
eigen solver from the apache Commons math 2.1. Btw mahout core doesn't have
the dependency on it.
On Oct 10, 2010 10:33 AM, "Jake Mannix" <[email protected]> wrote:
> I'm pretty sure that the COLT code does it in reverse order (see
> LanczosSolver for some usage there).
>
> -jake
>
> On Sun, Oct 10, 2010 at 7:00 AM, Ted Dunning <[email protected]>
wrote:
>
>> Cool.
>>
>> I previously potted and tested the qr decomposition. Looks I have a
similar
>> task ahead for the svd.
>>
>> It us the convention that the singular values come out in order but there
>> is usually no guarantee that the will so the svd codes usually have a
sort
>> step to do this.
>>
>> Sent from my iPhone
>>
>> On Oct 9, 2010, at 10:09 PM, "Dmitriy Lyubimov (JIRA)" <[email protected]>
>> wrote:
>>
>> >
>> > [
>>
https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12919567#action_12919567
]
>> >
>> > Dmitriy Lyubimov edited comment on MAHOUT-376 at 10/10/10 1:08 AM:
>> > -------------------------------------------------------------------
>> >
>> > ok, i got to BBt normalized for rank deficiences in Q blocks.
>> >
>> > What eigensolver should i use in context of Mahout? I know there's
>> something in that Cern library but it is all deprecateed. There's also a
>> solver in apache commons, but it is not readily imported into the mahout
>> core.
>> >
>> > Thanks.
>> >
>> > PS. i tried EigenvalueDecomposition from cern library. I know i am
quite
>> likely providing an ill-conditioned matrix in my unit test but strange
thing
>> is that eigen values are not always in descending order. i thought they
were
>> supposed to come out in descending order?
>> >
>> > was (Author: dlyubimov2):
>> > ok, i got to BBt normalized for rank deficiences in Q blocks.
>> >
>> > What eigensolver should i use in context of Mahout? I know there's
>> something in that Cern library but it is all deprecateed. There's also a
>> solver in apache commons, but it is not readily imported into the mahout
>> core.
>> >
>> > Thanks.
>> >
>> >> Implement Map-reduce version of stochastic SVD
>> >> ----------------------------------------------
>> >>
>> >> Key: MAHOUT-376
>> >> URL: https://issues.apache.org/jira/browse/MAHOUT-376
>> >> Project: Mahout
>> >> Issue Type: Improvement
>> >> Components: Math
>> >> Reporter: Ted Dunning
>> >> Assignee: Ted Dunning
>> >> Fix For: 0.5
>> >>
>> >> Attachments: MAHOUT-376.patch, sd-bib.bib, sd.pdf, sd.tex,
>> Stochastic SVD using eigensolver trick.pdf
>> >>
>> >>
>> >> See attached pdf for outline of proposed method.
>> >> All comments are welcome.
>> >
>> > --
>> > This message is automatically generated by JIRA.
>> > -
>> > You can reply to this email to add a comment to the issue online.
>> >
>>
