Make stacking available in SKLL

[desilinguist]

One way to to this is via https://github.com/civisanalytics/civisml-extensions

I am not sure what the best approach for incorporation is: (a) include their code as a dependency and use it as is (b) fork their repository and modify it for use with SKLL and give them proper credit obviously.

Another option is to basically build stacking into SKLL using the SKLL API itself. I am currently leaning towards this option. However, I am not sure whether this should live in SKLL itself or perhaps another package that's built on top of SKLL.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.

[desilinguist]

Keep it open please.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.

[desilinguist]

Keep it open.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.

[aoifecahill]

keep it open

[desilinguist]

Just FYI: scikit-learn will have StackingClassifier and StackingRegressor available starting with 0.22.

So, our work should now be much simpler. Basically we need to figure out how to pass in multiple estimators as inputs in the configuration file. This solution would also apply to #488 @bndgyawali.