Add to_sklearn and from_sklearn APIs to serialize to CPU sklearn-estimators for supported models

[dantegd]

No description provided.

[betatim]

Why have the methods do both the conversion cuml<>sklearn and the serialisation? Having a way to convert to and from scikit-learn seems like a useful thing by itself. Maybe because you have your own way of serialising the model, or because you need a particular type of model or who-knows-what.

So to serialise it you'd do something like pickle.dumps(cuml_est.to_sklearn()) (or dill, joblib, ...)

How hard would it be to have cuml.from_sklearn(estimator)? As in one top level function that takes a scikit-learn estimator and converts it to the cuml equivalent? It seems like it should be easy to figure out the estimator's class name: "just look at .__class__.__name__" but I wonder if there is a trap here?

---

Name bike shedding: if we don't save things to a file, how about as_sklearn? A bit like other functions that do type conversion like astype.

If we do save to a file, then save_sklearn and load_sklearn? Basically getting words like "save" and "load" in there to make it clear that this is about storing things (to a file).