[
https://issues.apache.org/jira/browse/SYSTEMML-2364?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
LI Guobao updated SYSTEMML-2364:
--------------------------------
Description:
It aims to extend the scheme of data partition.
* Disjoint_Round_Robin: for each worker, use a permutation multiply or simpler
a removeEmpty such as {{removeEmpty(target=X, margin=rows,
select=(seq(1,nrow(X))%%k)==id)}}.
was:
It aims to extend the three other schemes of data partition.
* Disjoint_Round_Robin: for each worker, use a permutation multiply or simpler
a removeEmpty such as {{removeEmpty(target=X, margin=rows,
select=(seq(1,nrow(X))%%k)==id)}}.
* Disjoint_Random: for each worker, use a permutation multiply {{P[beg:end,]
%*% X}}, where P is constructed for example with
{{P=table(seq(1,nrow(X),sample(nrow(X), nrow(X))))}}, i.e., sampling without
replacement to ensure disjointness.
* Overlap_Reshuffle: Similar to the above, except you create a new permutation
matrix for each worker and without the indexing on P.
Summary: Extend scheme Disjoint_Round_Robin (was: Extend the data
partition scheme)
> Extend scheme Disjoint_Round_Robin
> ----------------------------------
>
> Key: SYSTEMML-2364
> URL: https://issues.apache.org/jira/browse/SYSTEMML-2364
> Project: SystemML
> Issue Type: Sub-task
> Reporter: LI Guobao
> Assignee: LI Guobao
> Priority: Major
>
> It aims to extend the scheme of data partition.
> * Disjoint_Round_Robin: for each worker, use a permutation multiply or
> simpler a removeEmpty such as {{removeEmpty(target=X, margin=rows,
> select=(seq(1,nrow(X))%%k)==id)}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
