IdentifierLeakage

class IdentifierLeakage

Check if identifiers (Index/Date) can be used to predict the label.

Parameters

ppscore_paramsany , default: None

dictionary containing params to pass to ppscore predictor

__init__(ppscore_params=None)

__new__(*args, **kwargs)

Methods

IdentifierLeakage.add_condition(name, ...)	Add new condition function to the check.
IdentifierLeakage.add_condition_pps_not_greater_than([...])	Add condition - require columns not to have a greater pps than given max.
IdentifierLeakage.clean_conditions()	Remove all conditions from this check instance.
IdentifierLeakage.conditions_decision(result)	Run conditions on given result.
IdentifierLeakage.name()	Name of class in split camel case.
IdentifierLeakage.params([show_defaults])	Return parameters to show when printing the check.
IdentifierLeakage.remove_condition(index)	Remove given condition by index.
IdentifierLeakage.run(dataset[, model])	Run check.
IdentifierLeakage.run_logic(context[, ...])	Run check.

Example

IdentifierLeakage Example Notebook

previous

DateTrainTestLeakageOverlap.run_logic

next

IdentifierLeakage.add_condition