Test + docs: classical registration toy examples for integration testing purposes

[tvercaut]

Such a feature can be a nice simple integration test to ensure for example that the gradients are useable.

A basic toy example could involve translating an image and registering it with itself.

If useful, this quick and dirty notebook could be a source of inspiration:
https://colab.research.google.com/drive/1OmU-jTvWmRRVerc23nsyNz_p8jU3S7wX?usp=sharing

[YipengHu]

@tvercaut see: https://github.com/ucl-candi/DeepReg/tree/28-classical-registration/test/sit
It looks like working fine. ;)

[YipengHu]

@zacbaum can you help with converting the two examples to sit please? let me know if you want them to be merged first, though the branch has just been rebased - so should be no big diff.

[YipengHu]

@zacbaum can you help with converting the two examples to sit please? let me know if you want them to be merged first, though the branch has just been rebased - so should be no big diff.

I've PR anyway ;) please review/approve and you can start a new branch for it.

[YipengHu]

@zacbaum @NMontanaBrown if adding SITs is too much for R1, I can convert this to a "special demo". What do you think?

[NMontanaBrown]

@YipengHu Let me review the test and get back to you

[YipengHu]

@YipengHu Let me review the test and get back to you

these two are just examples, there are a few others you can add to cover other functions/modules, such as affine registration, registration on labels/features.

[NMontanaBrown]

@YipengHu This is cool, but the test works by looking at the plotted output? I think this is easily modifiable. We can incorporate this after unit tests, I will add to timeline if that makes sense?

[YipengHu]

Good idea. We have to streamline, if not required by joss, not required for R1.
And it is a cool idea for sit!

[YipengHu]

@NMontanaBrown @zacbaum what's our plan for this? if not before R1, please add post-r1 label.

[NMontanaBrown]

Still on track to refactor for R1 @YipengHu !

[YipengHu]

@zacbaum @NMontanaBrown
see #193 we will move some of the code for purpose of one or two demos. we will discuss it again how to make this sit again

• potentially post-r1

[YipengHu]

closing this as they have been moved to demos.