Colt Allen

The `lifetimes` models themselves are still able to reproduce results in research, and reliable enough to use for unit testing during development. The two most significant bugs in `lifetimes` are...

I'm confused about this Codecov report, because `pytest-cov` reports 100% code coverage in my IDE and many of the code lines in that report aren't even relevant to this PR.

Still need to add a test or two and address https://github.com/pymc-labs/pymc-marketing/issues/697, which may necessitate some refactoring to allow users to use their own segmentation methods. Might even roll https://github.com/pymc-labs/pymc-marketing/pull/698 into...

> Maybe we can leave out #698 as the PR is open and we are just missing some tests. It should not be a blocker for this one ;) Extra...

Also added a test and warning if only 2 bins are created. This is unlikely unless dealing with very small and/or invalid datasets that a model probably wouldn't converge on...

This is also impacting the CLV notebooks. I tried changing the fig size in the plotting code, but it made no difference. I'm thinking we need to specify something in...

@theorashid @zaxtax do you feel the recent merge of [PR 513](https://github.com/pymc-devs/pymc-extras/pull/513) is enough to check off the `Newton Solver` box under _Section 1. Laplace Approximation_? My thoughts are that the...

> Thank you, [@theorashid](https://github.com/theorashid) , for sending me here! I was wondering if I could pick up the sparse matrix work. I've been wanting to contribute to `pytensor` and this...

@ricardoV94 do you have a time estimate on when `numba` will became the new default backend? I'm working on the BG/BB model right now, and currently NUTS is taking over...

@ricardoV94 this sampling convention began with `ShiftedBeteGeoModelIndividual`. Is there a reason why the PMF in the paper wasn't used? See equation(3): https://faculty.wharton.upenn.edu/wp-content/uploads/2012/04/Fader_hardie_jim_07.pdf