Paul Talbot

It's possible we could directly use optimization such as scipy and scikitlearn offer and still use the sampling, code interfaces, and job distribution in RAVEN; I think Andrea looked at...

I intend to work on the GD exercises this week. I'm not sure how likely I am to fulfill that intent! 😄

# 2022-02-02 Design Meeting Notes We quickly concluded that including the `PostProcessor` in the `EnsembleModel` is challenging and does not present sufficient use cases currently to justify the effort involved;...

I don't see an updated list of where it's not in place yet; does that information exist? It will be challenging to carve out other work to include that without...

I provided a few comments, but not a full review.

We do have a mechanic for this, using the `writeXML` method, but I don't think it's used for all the ROMs yet. To use it you have an IOStep with...

I believe the base method is defined in SupervisedLearning, but is extendable in each inheritor.

Yeah, the real challenge is that each ROM representation is dramatically different. We initially attempted to do a full roll-out of this meta-printing strategy across all of SupervisedLearning but quickly...

It looks like in PR #1607 (ROM API) there's more modularity introduced to the ScikitLearn ROM, allowing each strategy to be independent. Maybe that will facilitate rolling out attribute exposure.

LGTM. Do you want to move this into a PR?