Carter Francis

The next PR which handles the slicing is found here https://github.com/CSSFrancis/hyperspy/tree/FancySlicingRefactor if that helps understand some of the motivations for these changes.

> Sorry, I have done a review but forgot to submit it... I updated my comments with the latest changes. All good :) thanks for the review. To be honest...

@ericpre This should be in a little bit better of shape now. I added in the ability to cache the axis value for both the `FunctionalDataAxis` and the `UniformDataAxis`. Let...

> Nice job @CSSFrancis ! This is a much needed refreshment of the axes code. @francisco-dlp Thanks for the response. This was exactly the kind of idea I was hoping...

> > I think that hyperspy should try to support as much numpy as possible. > > Not sure about it: there are things that you can do with abstract...

@francisco-dlp The relevant change is here for the uniform axis. ```python if isinstance(_slice, (np.ndarray, list)): new_axis = self.convert_to_non_uniform_axis() return new_axis._slice_me(_slice) ``` In the slicing code then we just treat arrays...

@francisco-dlp did we ever fall either way on this? I can move some of the code downstream to pyxem, but I do want to get as much into hyperspy as...

@francisco-dlp No worries! Whenever you get around to it is fine, thanks for taking your time to review this.

I realize that this is still kind of messy, but the functionality should all be there now. We can remove the _masked_phase_cross_correlation.py file as well. In general most of the...

@alexdesiqueira I can get back to this sometime early next week. As for what still needs to be done I would like to: - [ ] Add in an example...