[
https://issues.apache.org/jira/browse/HADOOP-2515?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Edward Yoon updated HADOOP-2515:
--------------------------------
Description:
After discuss with gary bradski, i decided to try working for matrix processing
on Hadoop.
If you have any advice, please let me know.
* A matrix of features
** Each row is one input vector.
** Each column coreesponds to a separate feature.
*** indicates the i-th row vector
* Splitting the input matrix
** The input matrix is split into n submatrix, where n is the number of mapper.
** Computation is are on input vectors; only 1-D decomposition is done.
was:
After talk with gary bradski, i decided to try working for matrix.
If you have any advice, please let me know.
* A matrix of features
** Each row is one input vector.
** Each column coreesponds to a separate feature.
*** indicates the i-th row vector
* Splitting the input matrix
** The input matrix is split into n submatrix, where n is the number of mapper.
** Computation is are on input vectors; only 1-D decomposition is done.
> Addition of Matrix Input-Output format
> --------------------------------------
>
> Key: HADOOP-2515
> URL: https://issues.apache.org/jira/browse/HADOOP-2515
> Project: Hadoop
> Issue Type: New Feature
> Components: mapred
> Reporter: Edward Yoon
> Assignee: Edward Yoon
> Priority: Minor
>
> After discuss with gary bradski, i decided to try working for matrix
> processing on Hadoop.
> If you have any advice, please let me know.
> * A matrix of features
> ** Each row is one input vector.
> ** Each column coreesponds to a separate feature.
> *** indicates the i-th row vector
> * Splitting the input matrix
> ** The input matrix is split into n submatrix, where n is the number of
> mapper.
> ** Computation is are on input vectors; only 1-D decomposition is done.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
