commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gilles (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (MATH-1430) Singularity check in QRDecomposition
Date Fri, 25 Aug 2017 10:46:01 GMT

     [ https://issues.apache.org/jira/browse/MATH-1430?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Gilles updated MATH-1430:
-------------------------
    Attachment: MATH-1430.unit_test.diff

Unit tests (see attached patch) suggest that
* the default value of the threshold is dangerous (yielding garbage to an unsuspecting user),
* the threshold is absolute whereas singularity is independent of the "global scale".


> Singularity check in QRDecomposition
> ------------------------------------
>
>                 Key: MATH-1430
>                 URL: https://issues.apache.org/jira/browse/MATH-1430
>             Project: Commons Math
>          Issue Type: Improvement
>    Affects Versions: 3.6.1
>            Reporter: Gilles
>            Priority: Minor
>             Fix For: 4.0
>
>         Attachments: MATH-1430.unit_test.diff
>
>
> In class {{o.a.c.m.linear.QRDecomposition}}, the singularity check seems not to be robust
enough to catch some (numerically) singular matrices.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message