You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Ted Dunning (JIRA)" <ji...@apache.org> on 2010/04/11 15:42:41 UTC

[jira] Created: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Implement Map-reduce version of stochastic SVD
----------------------------------------------

                 Key: MAHOUT-376
                 URL: https://issues.apache.org/jira/browse/MAHOUT-376
             Project: Mahout
          Issue Type: Bug
            Reporter: Ted Dunning


See attached pdf for outline of proposed method.

All comments are welcome.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Re: [jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by Ted Dunning <te...@gmail.com>.
Please do take a look.  I think that the details are somewhat different from
our previous discussions.  The differences have largely to do with making
sure that all phases of the problem involve scalable computations.  You may
well be ahead of me on this.

On Sun, Apr 11, 2010 at 6:46 PM, Jake Mannix <ja...@gmail.com> wrote:

> I haven't had a chance to read your attached pdf, but I *have* had a chance
> to code up an impl of this jira.  Patch coming soon.
>

Re: [jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by Jake Mannix <ja...@gmail.com>.
I haven't had a chance to read your attached pdf, but I *have* had a chance
to code up an impl of this jira.  Patch coming soon.

On Apr 11, 2010 6:50 AM, "Ted Dunning (JIRA)" <ji...@apache.org> wrote:


    [
https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel]

Ted Dunning updated MAHOUT-376:
-------------------------------

   Attachment: sd.tex
               sd-bib.bib
               sd.pdf

Algorithm details.

> Implement Map-reduce version of stochastic SVD
> ----------------------------------------------
>
>                 Key: MAHOUT-376
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-376
>             Project: Mahout
>          Issue Type: Bug
>            Reporter: Ted Dunning
>         Attachments: sd-bib.bib, sd.pdf, sd.tex
>
>
> See attached pdf for outline of proposed method.
> All comments are welcome.

--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by "Ted Dunning (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ted Dunning updated MAHOUT-376:
-------------------------------

    Attachment: MAHOUT-376.patch

Here is a work-in-progress patch that illustrates how I plan to do the stochastic multiplication.

For moderate sized problems, this will be the major step required since all of the dense intermediate products will fit in memory.  For larger problems, additional tricks will be necessary.

> Implement Map-reduce version of stochastic SVD
> ----------------------------------------------
>
>                 Key: MAHOUT-376
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-376
>             Project: Mahout
>          Issue Type: Bug
>            Reporter: Ted Dunning
>         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.


[jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by "Dmitriy Lyubimov (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Dmitriy Lyubimov updated MAHOUT-376:
------------------------------------

    Attachment: Stochastic SVD using eigensolver trick.pdf

Per Ted's request, i am attaching a conspectus of our previous discussion of Ted's suggested mods to Tropp's stochastic svd. It doesn't include Q orthonormalization.

> Implement Map-reduce version of stochastic SVD
> ----------------------------------------------
>
>                 Key: MAHOUT-376
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-376
>             Project: Mahout
>          Issue Type: Bug
>            Reporter: Ted Dunning
>         Attachments: 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.
-
If you think it was sent incorrectly contact one of the administrators: https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by "Ted Dunning (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ted Dunning updated MAHOUT-376:
-------------------------------

    Attachment: sd.tex
                sd-bib.bib
                sd.pdf

Algorithm details.

> Implement Map-reduce version of stochastic SVD
> ----------------------------------------------
>
>                 Key: MAHOUT-376
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-376
>             Project: Mahout
>          Issue Type: Bug
>            Reporter: Ted Dunning
>         Attachments: sd-bib.bib, sd.pdf, sd.tex
>
>
> See attached pdf for outline of proposed method.
> All comments are welcome.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Updated: (MAHOUT-376) Implement Map-reduce version of stochastic SVD

Posted by "Sean Owen (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAHOUT-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sean Owen updated MAHOUT-376:
-----------------------------

       Issue Type: Improvement  (was: Bug)
         Assignee: Ted Dunning
    Fix Version/s: 0.4
      Component/s: Math

More minor comments:

- Don't forget a copyright header for new files
- murmurInt should inline those constants or declare them constants
- More VirtualRandomMatrix fields ought be final, conceptually
- VirtualRandomVector doesn't need "size"
- "Utils" classes ought to have a private constructor IMHO

> 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: 0.4
>
>         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.