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

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

    Attachment: ssvd-CDH3-or-0.21.patch.gz

Sorry for iterating too often, but this was small but important fix for a 
showstopper.
* added orthonormality assertions in local tests for V, U (they pass with 
epsilon 1e-10 or better). 
* small fixes to U, V jobs.

Now should be suitable for LSI type of work with documents having 10k-30k 
lemmas average. 

BTW when inserting dependencies for CDH3b3, additional jackson jar is required 
in order for local hadoop  test to work. 
in CDH3b2 no such change is required. 
Not sure about 0.21, but proper care should be taken as usual to integrate 
hadoop client's transitive dependencies into mahout dependencies.

> 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, Modified stochastic svd algorithm for 
> mapreduce.pdf, QR decomposition for Map.pdf, QR decomposition for Map.pdf, QR 
> decomposition for Map.pdf, sd-bib.bib, sd.pdf, sd.pdf, sd.pdf, sd.pdf, 
> sd.tex, sd.tex, sd.tex, sd.tex, SSVD working notes.pdf, SSVD working 
> notes.pdf, SSVD working notes.pdf, SSVD working notes.pdf, 
> ssvd-CDH3-or-0.21.patch.gz, ssvd-CDH3-or-0.21.patch.gz, 
> ssvd-CDH3-or-0.21.patch.gz, ssvd-CDH3-or-0.21.patch.gz, 
> ssvd-CDH3-or-0.21.patch.gz, ssvd-m1.patch.gz, ssvd-m2.patch.gz, 
> ssvd-m3.patch.gz, 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.

Reply via email to