Matthew Pancia

@moody-marlin Poisson, for sure; I would also include some of the other basic families, even though they are slightly less common (e.g. binomial and/or with alternate link functions e.g. probit)

After looking at it for a little bit, I think mimicing the `statsmodels` API for their families makes sense: https://github.com/statsmodels/statsmodels/blob/master/statsmodels/genmod/families/family.py They have (what amounts to) an abstract base class `Family`...

@mrocklin Yeah, sure! In, e.g. Newton's method: ```python def newton(X, y, max_steps=50, tol=1e-8, family=Logistic): '''Newtons Method for Logistic Regression.''' gradient, hessian = family.gradient, family.hessian n, p = X.shape beta =...

As we discussed earlier, I think this is a really cool idea, and I'm glad to be part of the discussion. As a novice to this (and for the purposes...