scikit-learn-intelex 2025.7.0__py311-none-manylinux_2_28_x86_64.whl → 2025.9.0__py311-none-manylinux_2_28_x86_64.whl

daal4py/mb/__init__.py

@@ -53,15 +53,26 @@ def convert_model(model) -> "GBTDAALModel | LogisticDAALModel":
         offers faster prediction methods.
     """
     if isinstance(model, LogisticRegression):
+        # The multi_class keyword is removed in scikit-learn 1.8, and OvR functionality
+        # has been replaced by other estimators. Therefore checking for linear classifiers
+        # only dependent on the solver.
         if model.classes_.shape[0] > 2:
-            if (model.multi_class == "ovr") or (
-                model.multi_class == "auto" and model.solver == "liblinear"
-            ):
-                raise TypeError(
-                    "Supplied 'model' object is a linear classifier, but not multinomial logistic"
-                    " (hint: pass multi_class='multinomial' to 'LogisticRegression')."
-                )
-        elif (model.classes_.shape[0] == 2) and (model.multi_class == "multinomial"):
+            if not hasattr(model, "multi_class"):
+                if model.solver == "liblinear":
+                    raise TypeError(
+                        "Supplied 'model' object is a linear classifier, but not multinomial logistic"
+                    )
+            else:
+                if (model.multi_class == "ovr") or (
+                    model.multi_class == "auto" and model.solver == "liblinear"
+                ):
+                    raise TypeError(
+                        "Supplied 'model' object is a linear classifier, but not multinomial logistic"
+                        " (hint: pass multi_class='multinomial' to 'LogisticRegression')."
+                    )
+        elif (model.classes_.shape[0] == 2) and (
+            getattr(model, "multi_class", "auto") == "multinomial"
+        ):
             raise TypeError(
                 "Supplied 'model' object is not a logistic regressor "
                 "(hint: pass multi_class='auto' to 'LogisticRegression')."

daal4py/mb/logistic_regression_builders.py

@@ -97,21 +97,6 @@ class LogisticDAALModel:
         )
         builder.set_beta(coefs, intercepts)
         self._model = builder.model
-        self._alg_pred_class = logistic_regression_prediction(
-            nClasses=self.n_classes_,
-            fptype=self._fptype,
-            resultsToEvaluate="computeClassLabels",
-        )
-        self._alg_pred_prob = logistic_regression_prediction(
-            nClasses=self.n_classes_,
-            fptype=self._fptype,
-            resultsToEvaluate="computeClassProbabilities",
-        )
-        self._alg_pred_logprob = logistic_regression_prediction(
-            nClasses=self.n_classes_,
-            fptype=self._fptype,
-            resultsToEvaluate="computeClassLogProbabilities",
-        )
 
     @property
     def coef_(self):
@@ -121,6 +106,15 @@ class LogisticDAALModel:
     def intercept_(self):
         return self._model.Beta[:, 0]
 
+    def _logistic_regression_prediction(
+        self, X: np.ndarray, resultsToEvaluate: str
+    ) -> classifier_prediction_result:
+        return logistic_regression_prediction(
+            nClasses=self.n_classes_,
+            fptype=self._fptype,
+            resultsToEvaluate=resultsToEvaluate,
+        ).compute(X, self._model)
+
     def predict(self, X) -> np.ndarray:
         """
         Predict most probable class
@@ -133,7 +127,7 @@ class LogisticDAALModel:
             The most probable class, as integer indexes
         """
         return (
-            self._alg_pred_class.compute(X, self._model)
+            self._logistic_regression_prediction(X, "computeClassLabels")
             .prediction.reshape(-1)
             .astype(int)
         )
@@ -151,7 +145,9 @@ class LogisticDAALModel:
         proba : array(n_samples, n_classes)
             The predicted probabilities for each class.
         """
-        return self._alg_pred_prob.compute(X, self._model).probabilities
+        return self._logistic_regression_prediction(
+            X, "computeClassProbabilities"
+        ).probabilities
 
     predict_proba.__doc__ = predict_proba.__doc__.replace(r"%docstring_X%", _docstring_X)
 
@@ -166,7 +162,9 @@ class LogisticDAALModel:
         log_proba : array(n_samples, n_classes)
             The logarithms of the predicted probabilities for each class.
         """
-        return self._alg_pred_logprob.compute(X, self._model).logProbabilities
+        return self._logistic_regression_prediction(
+            X, "computeClassLogProbabilities"
+        ).logProbabilities
 
     predict_log_proba.__doc__ = predict_log_proba.__doc__.replace(
         r"%docstring_X%", _docstring_X
@@ -206,11 +204,7 @@ class LogisticDAALModel:
             raise ValueError(
                 "Must request at least one of 'classes', 'proba', 'log_proba'."
             )
-        return logistic_regression_prediction(
-            nClasses=self.n_classes_,
-            fptype=self._fptype,
-            resultsToEvaluate=pred_request,
-        ).compute(X, self._model)
+        return self._logistic_regression_prediction(X, pred_request)
 
     predict_multiple.__doc__ = predict_multiple.__doc__.replace(
         r"%docstring_X%", _docstring_X

daal4py/sklearn/linear_model/logistic_loss.py

@@ -61,8 +61,8 @@ def _daal4py_logistic_loss_extra_args(
         fptype=getFPType(X),
         method="defaultDense",
         interceptFlag=fit_intercept,
-        penaltyL1=l1 / n,
-        penaltyL2=l2 / n,
+        penaltyL1=l1,
+        penaltyL2=l2,
         resultsToCompute=results_to_compute,
     )
     objective_function_algorithm_instance.setup(X, y, beta)
@@ -99,8 +99,8 @@ def _daal4py_cross_entropy_loss_extra_args(
             fptype=getFPType(X),
             method="defaultDense",
             interceptFlag=fit_intercept,
-            penaltyL1=l1 / n,
-            penaltyL2=l2 / n,
+            penaltyL1=l1,
+            penaltyL2=l2,
             resultsToCompute=results_to_compute,
         )
     )