Ashley Scillitoe

Adding to 0.7.0 milestone but can remove if this turns out to be a bigger job than expected.

@jklaise yeh I'm rather confused about this one...

`twine>=4.x` caused problems with `alibi-detect` notebook tests on Python 3.7 (https://github.com/SeldonIO/alibi-detect/pull/508#issuecomment-1129833673). The notebook tests should be checked for `alibi` too. The build CI is also currently failing...

The point about `pydantic` not being a validation library is a good one. I can foresee quite a few pitfalls due to this, e.g. not validating dimensions, and not catching...

p.s. my comment above is primarily related to validating `__init__` args. @jklaise your final bullet could work nicely for validating `fit` and `explain` args. But, to avoid adding numerous extra...

@mauicv @jklaise

Mmmn a tricky one. I tend to agree that `image_layout` is nicer, since it is more flexible and could be extended in the future. I also that `channel_axis` might be...

Yeh that's fair enough. Even without a need for the added flexibility of `image_layout` right now, I agree that it is easier to interpret, and might be useful for volumetric...

Hi @Simardeep27, we generally try to follow a workflow like the one described here: https://www.atlassian.com/git/tutorials/comparing-workflows/gitflow-workflow Essentially this means all new features should be implemented and PR'd into `develop`. We only...

P.s. another thought... since the implementation of discrete distributions is likely to require a bit more thought, it might be an idea to implement more continuous distributions first. For example,...