remove legacy detector

Author: alelavml3

src/ml3_drift/sklearn/base.py

@@ -1,103 +1,13 @@
 import numpy as np
 
-from collections.abc import Callable
 from sklearn.base import BaseEstimator, TransformerMixin, check_is_fitted
-from abc import ABC, abstractmethod
 from sklearn.utils.validation import validate_data
-from ml3_drift.callbacks.models import DriftInfo
 from ml3_drift.enums.monitoring import DataDimension, DataType
 from ml3_drift.monitoring.base import MonitoringAlgorithm
 
 from copy import deepcopy
 
 
-class BaseDriftDetector(TransformerMixin, BaseEstimator, ABC):
-    """
-    Base class for drift detector.
-    Base Drift Detectors are neither transformers nor predictors, they
-    just observe the data and detects drift, executing specified
-    actions when necessary.
-    For this reason, they implement both the transform and predict methods.
-
-    Parameters
-    ----------
-    callbacks: list[Callable[[DriftInfo], None]
-        list of callbacks function used to act when drift are detected
-    """
-
-    def __init__(self, callbacks: list[Callable[[DriftInfo], None]] | None = None):
-        super().__init__()
-        self.callbacks = callbacks
-
-    @abstractmethod
-    def _fit(self, X, y=None):
-        """
-        Fit method that should be implemented in child classes
-        """
-
-    @abstractmethod
-    def _detect(self, X) -> list[DriftInfo]:
-        """
-        Core method for detecting drift that is implemented in child classes.
-        """
-
-    def fit(self, X, y=None):
-        """
-        Fit method. Calls _inner_fit and returns self
-        """
-
-        X = self._validate_data(X, y, reset=True)
-        self._fit(X)
-        self.is_fitted_ = True
-        return self
-
-    def transform(self, X):
-        """
-        Transform method. Calls _detect method and return X.
-        This step does not change the data but only performs drift detection.
-        """
-        X = self._validate_data(X, reset=False)
-        check_is_fitted(self)
-        if (drift_info_list := self._detect(X)) and (self.callbacks is not None):
-            for drift_info in drift_info_list:
-                for callback in self.callbacks:
-                    callback(drift_info)
-        return X
-
-    def predict(self, X):
-        """
-        Predict method. Calls _detect method and return X.
-        This step does not change the data but only performs drift detection.
-        """
-        X = self._validate_data(X, reset=False)
-        check_is_fitted(self)
-        if (drift_info_list := self._detect(X)) and (self.callbacks is not None):
-            for drift_info in drift_info_list:
-                for callback in self._callbacks:
-                    callback(drift_info)
-        return X
-
-    def _validate_data(self, X, y=None, reset=False):
-        """
-        Validate data method. This calls validate_data sklearn method with
-        provided parameters and returns the validated X.
-        Child classes can override with their own validation methods if needed
-        or just call the base class method with the custom parameters.
-        """
-
-        # Workaround since validate data doesn't return y if it is None
-        if y is None:
-            X = validate_data(self, X, reset=reset, accept_sparse=False)
-        else:
-            X, _ = validate_data(self, X, y, reset=reset, accept_sparse=False)
-        return X
-
-    def __sklearn_tags__(self):
-        tags = super().__sklearn_tags__()
-        # Currently empty, but can be used to add tags to the estimator
-        return tags
-
-
 class SklearnDriftDetector(TransformerMixin, BaseEstimator):
     """Adapter class for sklearn library.