[vtkusers] PCA statistics change

[Sean McBride]

On Wed, 31 Jan 2018 08:45:11 -0500, David Thompson said:

>Recently, Ben switched VTK to use Eigen's singular value decomposition.
>It reports some eigenvectors flipped. Notably, it seems to prefer a
>right-handed coordinate system as its output basis in 3D. This is
>great! ... but note that image tests that plot assessed values (i.e.,
>the inputs projected into the new basis) may now return different
>results with the y axis assessed-value flipped.
>
>Does anyone know if the VTKExamples repo does image-based tests? If so,
>I expect that may be causing a failure. However, it is unclear from the
>test output[1] whether an OpenGL/machine configuration error is masking
>an image test failure.

David, Ben,

We are seeing this too when running our company's own tests built against VTK master.

Is this change considered a breakage of vtkPCAStatistics?  Will you be changing things to make vtkPCAStatistics behave as it used to?

Cheers,

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montréal, Québec, Canada