complete restructure of ml-test-functions

src/surfaces/test_functions/__init__.py

@@ -2,9 +2,8 @@
 # Email: [email protected]
 # License: MIT License
 
-from .mathematical import (
-    mathematical_functions,
-)
+from .mathematical import mathematical_functions
+
 from .machine_learning import machine_learning_functions
 
 

src/surfaces/test_functions/machine_learning/__init__.py

@@ -3,18 +3,20 @@
 # License: MIT License
 
 
-from .tabular_classifiers import KNeighborsClassifierFunction
-from .tabular_regressors import (
-    GradientBoostingRegressorFunction,
+from .tabular import (
+    KNeighborsClassifierFunction,
     KNeighborsRegressorFunction,
+    GradientBoostingRegressorFunction,
 )
 
+
 __all__ = [
     "KNeighborsClassifierFunction",
-    "GradientBoostingRegressorFunction",
     "KNeighborsRegressorFunction",
+    "GradientBoostingRegressorFunction",
 ]
 
+
 machine_learning_functions = [
     KNeighborsClassifierFunction,
     GradientBoostingRegressorFunction,

src/surfaces/test_functions/machine_learning/tabular/__init__.py

@@ -0,0 +1,13 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+from .classification import KNeighborsClassifierFunction
+from .regression import KNeighborsRegressorFunction, GradientBoostingRegressorFunction
+
+
+__all__ = [
+    "KNeighborsClassifierFunction",
+    "KNeighborsRegressorFunction",
+    "GradientBoostingRegressorFunction",
+]

src/surfaces/test_functions/machine_learning/tabular/_base_tabular.py

@@ -0,0 +1,11 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+from .._base_machine_learning import MachineLearningFunction
+
+
+class BaseTabular(MachineLearningFunction):
+    def __init__(self, metric, sleep, evaluate_from_data):
+        super().__init__(metric, sleep, evaluate_from_data)

src/surfaces/test_functions/machine_learning/tabular/classification/__init__.py

@@ -0,0 +1,12 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+from .test_functions import (
+    KNeighborsClassifierFunction,
+)
+
+
+__all__ = [
+    "KNeighborsClassifierFunction",
+]

src/surfaces/test_functions/machine_learning/tabular/classification/_base_classification.py

@@ -0,0 +1,13 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+from .._base_tabular import BaseTabular
+
+
+class BaseClassification(BaseTabular):
+    metric = "accuracy"  # called 'scoring' in sklearn
+
+    def __init__(self, metric, sleep, evaluate_from_data):
+        super().__init__(metric, sleep, evaluate_from_data)

src/surfaces/test_functions/machine_learning/tabular/classification/test_functions/__init__.py

@@ -0,0 +1,11 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+from .k_neighbors_classifier import KNeighborsClassifierFunction
+
+
+__all__ = [
+    "KNeighborsClassifierFunction",
+]

src/surfaces/test_functions/machine_learning/tabular_classifiers.py → src/surfaces/test_functions/machine_learning/tabular/classification/test_functions/k_neighbors_classifier.py

@@ -3,12 +3,12 @@
 from sklearn.neighbors import KNeighborsClassifier
 
 from sklearn.model_selection import cross_val_score
-from .datasets import digits_data, wine_data, iris_data
+from ..datasets import digits_data, wine_data, iris_data
 
-from .base_machine_learning_function import MachineLearningFunction
+from .._base_classification import BaseClassification
 
 
-class KNeighborsClassifierFunction(MachineLearningFunction):
+class KNeighborsClassifierFunction(BaseClassification):
     name = "KNeighbors Classifier Function"
     _name_ = "k_neighbors_classifier"
     __name__ = "KNeighborsClassifierFunction"

src/surfaces/test_functions/machine_learning/tabular/regression/__init__.py

@@ -0,0 +1,14 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+from .test_functions import (
+    KNeighborsRegressorFunction,
+    GradientBoostingRegressorFunction,
+)
+
+
+__all__ = [
+    "KNeighborsRegressorFunction",
+    "GradientBoostingRegressorFunction",
+]

src/surfaces/test_functions/machine_learning/tabular/regression/_base_regression.py

@@ -0,0 +1,13 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+from .._base_tabular import BaseTabular
+
+
+class BaseRegression(BaseTabular):
+    metric = "r2"  # called 'scoring' in sklearn
+
+    def __init__(self, metric, sleep, evaluate_from_data):
+        super().__init__(metric, sleep, evaluate_from_data)

src/surfaces/test_functions/machine_learning/tabular/regression/test_functions/__init__.py

@@ -0,0 +1,13 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+from .k_neighbors_regressor import KNeighborsRegressorFunction
+from .gradient_boosting_regressor import GradientBoostingRegressorFunction
+
+
+__all__ = [
+    "KNeighborsRegressorFunction",
+    "GradientBoostingRegressorFunction",
+]

src/surfaces/test_functions/machine_learning/tabular/regression/test_functions/gradient_boosting_regressor.py

@@ -0,0 +1,66 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+import numpy as np
+
+from sklearn.ensemble import GradientBoostingRegressor
+
+from sklearn.model_selection import cross_val_score
+from ..datasets import diabetes_data
+
+from .._base_regression import BaseRegression
+
+
+class GradientBoostingRegressorFunction(BaseRegression):
+    name = "Gradient Boosting Regressor Function"
+    _name_ = "gradient_boosting_regressor"
+    __name__ = "GradientBoostingRegressorFunction"
+
+    para_names = ["n_estimators", "max_depth", "cv", "dataset"]
+
+    n_estimators_default = list(np.arange(3, 150, 5))
+    max_depth_default = list(np.arange(2, 25))
+    cv_default = [2, 3, 4, 5, 8, 10]
+    dataset_default = [diabetes_data]
+
+    def __init__(
+        self,
+        metric=None,
+        sleep=0,
+        evaluate_from_data=False,
+    ):
+        super().__init__(metric, sleep, evaluate_from_data)
+
+    def search_space(
+        self,
+        n_estimators: list = None,
+        max_depth: list = None,
+        cv: list = None,
+        dataset: list = None,
+    ):
+        search_space: dict = {}
+
+        search_space["n_estimators"] = (
+            self.n_estimators_default if n_estimators is None else n_estimators
+        )
+        search_space["max_depth"] = (
+            self.max_depth_default if max_depth is None else max_depth
+        )
+        search_space["cv"] = self.cv_default if cv is None else cv
+        search_space["dataset"] = self.dataset_default if dataset is None else dataset
+
+        return search_space
+
+    def create_objective_function(self):
+        def gradient_boosting_regressor(params):
+            knc = GradientBoostingRegressor(
+                n_estimators=params["n_estimators"],
+                max_depth=params["max_depth"],
+            )
+            X, y = params["dataset"]()
+            scores = cross_val_score(knc, X, y, cv=params["cv"], scoring=self.metric)
+            return scores.mean()
+
+        self.pure_objective_function = gradient_boosting_regressor

src/surfaces/test_functions/machine_learning/tabular/regression/test_functions/k_neighbors_regressor.py

@@ -0,0 +1,66 @@
+# Author: Simon Blanke
+# Email: [email protected]
+# License: MIT License
+
+
+import numpy as np
+
+from sklearn.neighbors import KNeighborsRegressor
+
+from sklearn.model_selection import cross_val_score
+from ..datasets import diabetes_data
+
+from .._base_regression import BaseRegression
+
+
+class KNeighborsRegressorFunction(BaseRegression):
+    name = "KNeighbors Regressor Function"
+    _name_ = "k_neighbors_regressor"
+    __name__ = "KNeighborsRegressorFunction"
+
+    para_names = ["n_neighbors", "algorithm", "cv", "dataset"]
+
+    n_neighbors_default = list(np.arange(3, 150, 5))
+    algorithm_default = ["auto", "ball_tree", "kd_tree", "brute"]
+    cv_default = [2, 3, 4, 5, 8, 10]
+    dataset_default = [diabetes_data]
+
+    def __init__(
+        self,
+        metric=None,
+        sleep=0,
+        evaluate_from_data=False,
+    ):
+        super().__init__(metric, sleep, evaluate_from_data)
+
+    def search_space(
+        self,
+        n_neighbors: list = None,
+        algorithm: list = None,
+        cv: list = None,
+        dataset: list = None,
+    ):
+        search_space: dict = {}
+
+        search_space["n_neighbors"] = (
+            self.n_neighbors_default if n_neighbors is None else n_neighbors
+        )
+        search_space["algorithm"] = (
+            self.algorithm_default if algorithm is None else algorithm
+        )
+        search_space["cv"] = self.cv_default if cv is None else cv
+        search_space["dataset"] = self.dataset_default if dataset is None else dataset
+
+        return search_space
+
+    def create_objective_function(self):
+        def k_neighbors_regressor(params):
+            knc = KNeighborsRegressor(
+                n_neighbors=params["n_neighbors"],
+                algorithm=params["algorithm"],
+            )
+            X, y = params["dataset"]()
+            scores = cross_val_score(knc, X, y, cv=params["cv"], scoring=self.metric)
+            return scores.mean()
+
+        self.pure_objective_function = k_neighbors_regressor