neurodata / SPORF

This is the implementation of Sparse Projection Oblique Randomer Forest
https://neurodata.io/forests/
97 stars 46 forks source link

make python bindings sklearn compatible/compliant #148

Open jovo opened 5 years ago

falkben commented 5 years ago

Some notes on this:

We'll want our class to inherit from sklearn's ForestClassifier (or maybe RandomForestClassifier) and have our methods named the same as sklearn's methods and return data the same way.

Methods that the RandomForestClassifier has: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html#sklearn.ensemble.RandomForestClassifier.__init__

This then allows us to use other sklearn functionality on our randomERforest object.