Oscar Esteban
Oscar Esteban
As brought up by Chris in the last bit of https://github.com/nipy/nitransforms/pull/166#issuecomment-1189631982, the right implementation should be carefully considered.
Ensure we have tests of `.h5` composites to make sure that #81 was properly addressed (see https://github.com/nipy/nitransforms/pull/165#issuecomment-1189201198 for further details).
This problem has not been addressed with the AFNI-obliques which only addressed linear transforms.
It would be beneficial to have this integrated, so that composite B-Spline can naturally be handled internally.
My plan is to merge #138 knowingly that the `BSplineFieldTransform.map()` method is broken, as reflected by the doctest (the outputs of this doctest should be very close to those of...
The LTA format allows reference and moving to be defined per entry in the affine list. However, our current implementation interprets these as a list of affines all referenced to...
Complete the implementation of mappings with nonlinear transforms.