Asaf Manela

In that sense, I expect it to be fine. Lasso reuses a lot of GLM code and mostly adheres to the StatsBase interface.

I would love to see this one merged for two downstream packages. thanks!

There is currently no dev doc, but feel free to ask question and I can try to answer. In any case, the contribution is welcome.

The packages may be generating a different path of regularization lambdas. You can get them from the lasso path as `l.λ`, and from GLMNet as `g.lambda`. Also, in your example...

Sure, a PR would be welcome!

This line cannot be removed entirely because then the jupyter notebooks that include sotu.jl won't get the variables they expect (covarsdf, counts, terms). If you mean you want to move...

Gave this a go, but it resulted in a small improvement for dmr and large slowdown for hdmr. Will try again after 1.2 is out