Refactor credit risk to use wave-ml

h2oai / wave-apps

Sample AI Apps built with H2O Wave.

MIT License

144 stars 51 forks source link

Refactor credit risk to use wave-ml #54

Closed vopani closed 3 years ago

vopani commented 3 years ago

Using Wave ML

Bare minimum refactoring to use wave-ml for modelling. There is more that can be done but this is a good place to start. Part of #41

Have tested this locally ✅

cc: @mtanco @geomodular

vopani commented 3 years ago

@mtanco This is ready to be merged. Things to note:

Wave ML uses H2O-3 AutoML for modelling. Currently runtime set to 30 seconds - You can change this here
Doesn't use validation split nor does cross-validation (didn't find it necessary) but I could add it if required.
Models enabled are DRF, XGBoost and GBM. This is to ensure the local shapley contributions are supported.

mtanco commented 3 years ago

@vopani : @g-eoj has refactored this in another branch to use native wave plots @g-eoj : @vopani has refracted this here to use wave-ml instead of native h2o-3

Can you two make sure both of your PRs merge nicely, as both of these are needed for our users but there is some overlap. Thank you!!

vopani commented 3 years ago

Please merge @g-eoj 's code first and I will redo the WaveML part, it should be simple.

g-eoj commented 3 years ago

@vopani my PR is merged

vopani commented 3 years ago

Ooopps, deleted by mistake. Opened a new PR here: https://github.com/h2oai/wave-apps/pull/65