Zernike refactor

[brianthelion]

1. Legacy code should be moved into the test suite.
2. Test suite should be updated.
3. File proliferation should be reduced or eliminated.
4. Code should be PEP8 compliant.
5. Remaining MATLAB idioms should be replaced with Numpy idioms.

Anything else?

[binarybottle]

Those are all good. I would add documentation of the code so that it will be easier to maintain in the future.

[brianthelion]

Making incremental progress here, a few hours a day. Expect a push this week.

[binarybottle]

Great!

Cheers,
@rno

On Mon, Apr 28, 2014 at 4:05 PM, brianthelion [email protected]:

Making incremental progress here, a few hours a day. Expect a push this
week.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/42#issuecomment-41625205
.

[brianthelion]

Didn't get quite as much done as I had hoped last week. Picking it up again tomorrow and will update on progress then. Cheers!

[binarybottle]

Great -- Looking forward to your update!

Cheers,
@rno

On Mon, May 5, 2014 at 12:54 PM, brianthelion [email protected]:

Didn't get quite as much done as I had hoped last week. Picking it up
again tomorrow and will update on progress then. Cheers!

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/42#issuecomment-42230861
.

[brianthelion]

Documentation is still to-do.

[binarybottle]

Good show, Brian!
After you implemented Koehl's exact method, it became significantly faster. It took only 10 minutes for order 10 for two cortical hemispheres!

[brianthelion]

The Koehl implementation is covered in #45. There are still a few things to do to wrap this ticket, like pulling together the documentation.

[brianthelion]

@binarybottle: I finally have some time to put on this cleanup task. Unfortunately I haven't tracked our priorities w.r.t tasks 1-5 listed above. Is documentation the only thing that remains to be done at this time? Cheers!

[binarybottle]

I believe documentation was the only thing left to do, but after Satra found that a 2GB memory limit was exceeded by computing Zernike moments, this strikes me as extremely important to address!

From Satra:
150215-15:17:45,874 workflow INFO:
Executing node Zernike_sulci.a1 in dir: /om/scratch/Tue/ps/MB_work/734db8e05f6be469df79c1419f253ad7/Mindboggle/Surface_feature_shapes/_hemi_rh/Zernike_sulci
Load "sulci" scalars from sulci.vtk
8329 vertices for label 1
Reduced 160076 to 15921 triangular faces
srun: Exceeded job memory limit

[brianthelion]

Ok, see my comments on #52.

[brianthelion]

@binarybottle Is there a particular documentation standard that you're sticking with on Mindboggle?

[binarybottle]

If you take a look at any of the other scripts, you'll see that I use restructured text. I use Sphinx to create docs from the code. Does that answer your question?

[brianthelion]

@binarybottle Yep! Thanks.

[binarybottle]

@brianthelion -- If you are unable to devote any more time to documentation of the code, shall I close this issue?

[satra]

@binarybottle - was this fixed?

[binarybottle]

Since the only lingering part of the discussion was the question of documentation, I didn't think it helpful to keep this issue open.