| Re: [eigen] SVD with singular matrices |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/eigen Archives
]
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] SVD with singular matrices
- From: Benoit Jacob <jacob.benoit.1@xxxxxxxxx>
- Date: Wed, 29 Sep 2010 10:41:39 -0400
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=ZngPcji05h5i9Yos6PtLUqv16pWig23jF4EgTcJCfTM=; b=XO0jmZJI9VlvNuJPDC/4CfMHueYIiyqXCN5xyMHQ4Ac2zlNOWphpdf5Pi430zKTG8C cjHU8vms8PseLybjHbx5W+Be4uE6GTdwzz2xdpjwvJIivmZHscoRqsSwTVdKxhkGMoEv rhi3J2bM4DZlDyBQrH4lRnVSVhXbeZzTFULzI=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; b=Y5rEgXV1smwzUq0URh5G2PvPurgFkoM7/BgN6NR2gVMdMzb768p1/UuG9iBOIadkBT RIxA85xVY+PyRNfssdtRhnI6Np3qasnxk207VZJ8PsI2u/bbGe98vtKqOZLddUP7a0h4 +qOGruCNummtyjANHjdYDucJqKDYmcBB6qyEQ=
2010/9/29 <hamelin.philippe@xxxxxxx>:
> After some more investigation, I found that the problem is effectively the
> SVD. It fails with the singular matrix :
>
> [1 1; 0 0]
>
> while it works fine with JacobiSVD. Is that something to do with the
> implemented method or this is really a bug?
This is a bug. Actually this matrix is already bidiagonal, so the SVD
should skip directly to the iterative Givens process, which should
have no trouble at all since the singular values are distinct.
Again, please make a compilable testcase. It saves us time.
Benoit
>
> Thank you,
>
> Philippe
> ________________________________
> De : Listengine [mailto:listengine@xxxxxxxxxxxxxxxxx] De la part de
> hamelin.philippe@xxxxxxx
> Envoyé : 29 septembre 2010 10:27
> À : eigen@xxxxxxxxxxxxxxxxxxx
> Objet : [eigen] SVD with singular matrices
>
> Hello,
>
> when updating from eigen2 to eigen3, I found that my pseudo-inverse was not
> working correctly for singular matrices. However, just replacing SVD with
> JacobiSVD makes it work. Before looking further, is there any limitation
> with SVD with singular matrices such as:
>
> [1 1;0 0]
>
> Thank you,
>
> ------------------------------------
> Philippe Hamelin, ing. jr, M. Ing
> Chercheur / Researcher
>
> T: 450-652-8499 x2198
> F: 450-652-1316
>
> Expertise robotique et civil
> Institut de recherche d'Hydro-Québec (IREQ)
> 1740, boul. Lionel-Boulet
> Varennes (QC) J3X 1S1, Canada
>