search

ZELLMECHANIK-DRESDEN / ShapeOut

Shape-Out has been superseded by Shape-Out 2.
https://shapeout.readthedocs.io
GNU General Public License v2.0
5 stars 5 forks source link

Implementing Generalized linear mixed models #135

Closed maikherbig closed 7 years ago

maikherbig commented 7 years ago

I have added the variable "model" to the function linmixmod, which allows to perfom either a linear mixed model (model=0) or a generalized linear mixed model (GLMM) with a log-link function (model=1). GLMM might be better suited for skewed distributions like deformation. The user should be able to decide which model should be applied. Therefore I suggest a dropdown menu for LMM and GLMM

paulmueller commented 7 years ago

Thanks for the pull request. The checks fail. I think it is because of the new argument model which should be a keyword argument. Instead of the possible values "0" and "1", I would prefer "linear" and "generalized" or some such kind.

maikherbig commented 7 years ago

very goof idea. I changed it to 'glmm' and 'lmm'

paulmueller commented 7 years ago

The tests still fail... Please make sure to set 'lmm' as default

maikherbig commented 7 years ago

I have added 'lmm' as default argument, but the tests still fail. On my machine this script runs coorectly and without any error. Paul, do you have an idea what could be the problem?

paulmueller commented 7 years ago

It's a coveralls issue. The tests pass.

katpla commented 7 years ago

Hi, what is the status of this one? there seems to be no artifacts in this release

paulmueller commented 7 years ago

There are no artifacts, because Maiks branch is not allowed to submit to coveralls (I will fix this in the future). Either way, you would not benefit from a build, because the user interface needs to be updated (by me) to allow generalized lmm.

paulmueller commented 7 years ago

I am merging this into a separate branch for testing.

paulmueller commented 7 years ago

I am not merging into develop yet. Further discussions in issue #126.