Should MMM preprocessing methods be defined in mixin class that aren't derived from `BaseMMM`?

pymc-labs / pymc-marketing

Bayesian marketing toolbox in PyMC. Media Mix (MMM), customer lifetime value (CLV), buy-till-you-die (BTYD) models and more.

https://www.pymc-marketing.io/

Apache License 2.0

614 stars 148 forks source link

Should MMM preprocessing methods be defined in mixin class that aren't derived from `BaseMMM`? #89

Open lucianopaz opened 1 year ago

wd60622 commented 2 months ago

Overall, I am against mixins used in this way as it makes switching out the processing steps a wayharder. I would be in favor of composition / pulling out processing outside of the class entirely

This would go for regardless of how this steps are written (sklearn, pytensor, pandas, etc)