implement sample weights to allow use on weighted data

[ghost]

Proof-of concept is in https://github.com/peterfoley605/pyglmnet/tree/implement_sample_weight

[jasmainak]

Feel free to open a WIP pull request so that it's not too much to review in one go

[ghost]

tricky thing there is that the branch is built as a "fork of pyglmnet that does things the way I need them done" rather than a "prepping for a PR", and it includes additional logging along with the sample weights, so it probably wouldn't be possible to merge directly. It does have tests and updated documentation and everything, though.

[jasmainak]

okay I see, whenever you are ready.

It's usually a good idea to open PR earlier than later so that it's possible to discuss and change things while there isn't too much code.

[arkottke]

It's unfortunate that the original user account on this request was deleted. I would also be interested in this feature.

[jasmainak]

it's here: https://github.com/glm-tools/pyglmnet/pull/324

I am happy to review/help if you want to try and finish this work ... however it's unlikely to be on the top of my priority list at the moment

[arkottke]

Oh great! I will see if I can carve out some time to contribute to this.

[jasmainak]

keep us posted!