MethodMapping

classsklearn.utils.metadata_routing.MethodMapping[source]

Stores the mapping between caller and callee methods for a router.

This class is primarily used in a get_metadata_routing() of a router object when defining the mapping between the router’s methods and a sub-object (a sub-estimator or a scorer).

Iterating through an instance of this class yields MethodPair(caller, callee) instances.

Read more on developing custom estimators that can route metadata in the Metadata Routing Developing Guide.

Added in version 1.3.

Examples

>>> from sklearn.utils.metadata_routing import MethodMapping
>>> MethodMapping().add(caller="fit", callee="split")
[{'caller': 'fit', 'callee': 'split'}]

add(*, caller, callee)[source]

Add a method mapping.

Parameters:

callerstr

Parent estimator’s method name in which the callee is called.

calleestr

Child object’s method name. This method is called in caller.

Returns:

selfMethodMapping

Returns self.

Gallery examples

Developing Estimators Compliant with Metadata Routing