ENH: Add reverse-ordered transform lists to ants.Registration outputs

[sulantha2006]

… ANTS Registration outputs.

Summary

Fixes # .

List of changes proposed in this PR (pull-request)

Acknowledgment

• (Mandatory) I acknowledge that this contribution will be available under the Apache 2 license.

[sulantha2006]

Any reason why the doctests are failing here. ?

[codecov]

Codecov Report

Merging #3192 into master will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #3192   +/-   ##
=======================================
  Coverage   65.07%   65.08%           
=======================================
  Files         301      301           
  Lines       39624    39628    +4     
  Branches     5247     5247           
=======================================
+ Hits        25786    25790    +4     
  Misses      12771    12771           
  Partials     1067     1067           

Flag	Coverage Δ
#unittests	65.08% <100.00%> (+<0.01%)	⬆️

Impacted Files	Coverage Δ
nipype/interfaces/ants/registration.py	73.04% <100.00%> (+0.21%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c37fcfb...f6ba809. Read the comment docs.

[SulanthaM]

Do you think we can incorporate this into pypi soon?
This output reverse_forward_transforms is mentioned in the documentation of NIPYPE antsRegistration but was not implemented to be used. So, if anyone tries to access this output based on the documentation, it fails.

[effigies]

Can you link to the documentation? It's difficult to review a pull request without context.

[SulanthaM]

Here is the documentation.

nipype/nipype/interfaces/ants/registration.py

Line 646 in c37fcfb

``reverse_forward_transforms`` outputs ``forward_transforms`` in reverse order and can be used for

[effigies]

Thanks. Judging by the discussion in #2187, it looks like reverse_transforms was deemed to be what was needed, but the docs didn't change...

@Gilles86 Can you confirm?

[SulanthaM]

reverse_transform is actually a collection of inverse transformations. This corresponds to the reverse_invert_flags as well.
The suggested modification is only a list reverse (reverse the elements in the list) of the forward_transforms list and forward_invert_flags list such that you can feed the output directly to a resampling node without a list reverse function node.

[sulantha2006]

Any way we can integrate this sooner to pypi?

[AlanKuurstra]

I also would appreciate reverse_forward_transforms which is mentioned in the documentation but not implemented. I understand reverse_forward_transforms to be different from reverse_transform.

[effigies]

Seems fine...

@Gilles86 @satra Is this consistent with your understanding?

[satra]

@effigies - seems reasonable from a conceptual standpoint and provides a handy helper to ease inverting lists. i think the docstring still says to use composite transforms when possible.