Sam Lendle

I suppose the easiest way would be to have something like ``` julia abstract AbstractLoss abstract Loss

It might be better for `tloss_and_gradient` to calculate the mean loss instead of sum, so that the loss is O(1) instead of O(n), and the amount of regularization does not...

That's true that you'd want the relative amount of regularization to decrease as your total sample size increases, and typically, you'd probably want to calculate the loss/gradient on the full...

I like the `DiscriminativeLoss`/potential `GenerativeLoss` names. If it turns out that we don't want to support generative models it will be easy to remove.

A general ensemble package would be great to have under the `RegressionBase.jl` umbrella. A consistent interface makes developing a such a package very easy. @svs14 has done a lot of...

I have [OnlineLearning.jl](https://github.com/lendle/OnlineLearning.jl) which fits GLMs (linear regression, logistic, and quantile regression for now) (optionally with L1 and/or L2 regularization) with SGD. Standard SGD and some variants (ADAGRAD, ADADELTA, and...

(I'm really interested in documenting methods and types, not really integers as in my example above.) How hard would it be to do something like this: ``` julia module M...

Regarding the deprecation warnings, I see them in IJulia but not in the REPL. I'm on `Julia Version 0.3.7-pre+19 Commit e3fbe7e (2015-03-10 07:59 UTC)`

I'm facing a similar issue as @stoney95 where we're using artifactory with every team member & CI pipeline having unique username & password credentials. An additional wrinkle is that each...

I'm also seeing this on svelte 5 with vidstack 1.12.12. 1.11.30 is okay.