Lukas Strack: Fixing hps remain active & meta hp configuration (#1536)

Author: Github Actions

development/_downloads/23ae4950352edc8dd9ea5443ba77886b/example_extending_regression.py

@@ -6,6 +6,7 @@
 The following example demonstrates how to create a new regression
 component for using in auto-sklearn.
 """
+from typing import Optional
 from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
@@ -17,6 +18,8 @@
 from ConfigSpace.conditions import EqualsCondition
 
 import sklearn.metrics
+
+from autosklearn.askl_typing import FEAT_TYPE_TYPE
 import autosklearn.regression
 import autosklearn.pipeline.components.regression
 from autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm
@@ -86,7 +89,9 @@ def get_properties(dataset_properties=None):
         }
 
     @staticmethod
-    def get_hyperparameter_search_space(dataset_properties=None):
+    def get_hyperparameter_search_space(
+        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None
+    ):
         cs = ConfigurationSpace()
         alpha = UniformFloatHyperparameter(
             name="alpha", lower=10**-5, upper=1, log=True, default_value=1.0

development/_downloads/3188e6424499bf3242cdcdfc032bf614/example_restrict_number_of_hyperparameters.py

@@ -7,6 +7,7 @@
 component with a new component, implementing the same classifier,
 but with different hyperparameters .
 """
+from typing import Optional
 
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import (
@@ -17,6 +18,7 @@
 from sklearn.datasets import load_breast_cancer
 from sklearn.model_selection import train_test_split
 
+from autosklearn.askl_typing import FEAT_TYPE_TYPE
 import autosklearn.classification
 import autosklearn.pipeline.components.classification
 from autosklearn.pipeline.components.classification import (
@@ -84,7 +86,9 @@ def get_properties(dataset_properties=None):
         }
 
     @staticmethod
-    def get_hyperparameter_search_space(dataset_properties=None):
+    def get_hyperparameter_search_space(
+        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None
+    ):
         cs = ConfigurationSpace()
 
         # The maximum number of features used in the forest is calculated as m^max_features, where

development/_downloads/4f9b78e1d6464520c85232e30bf19d2b/example_text_preprocessing.ipynb

@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "# create an autosklearn Classifier or Regressor depending on your task at hand.\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=30,\n    tmp_folder=\"/tmp/autosklearn_text_example_tmp\",\n)\n\nautoml.fit(X_train, y_train, dataset_name=\"20_Newsgroups\")  # fit the automl model"
+        "# create an autosklearn Classifier or Regressor depending on your task at hand.\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=30,\n)\n\nautoml.fit(X_train, y_train, dataset_name=\"20_Newsgroups\")  # fit the automl model"
       ]
     },
     {

development/_downloads/515ab036d01801cb08e4878be1aef556/example_extending_classification.py

@@ -6,6 +6,7 @@
 The following example demonstrates how to create a new classification
 component for using in auto-sklearn.
 """
+from typing import Optional
 from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
@@ -16,6 +17,8 @@
 )
 
 import sklearn.metrics
+
+from autosklearn.askl_typing import FEAT_TYPE_TYPE
 import autosklearn.classification
 import autosklearn.pipeline.components.classification
 from autosklearn.pipeline.components.base import AutoSklearnClassificationAlgorithm
@@ -100,7 +103,9 @@ def get_properties(dataset_properties=None):
         }
 
     @staticmethod
-    def get_hyperparameter_search_space(dataset_properties=None):
+    def get_hyperparameter_search_space(
+        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None
+    ):
         cs = ConfigurationSpace()
         hidden_layer_depth = UniformIntegerHyperparameter(
             name="hidden_layer_depth", lower=1, upper=3, default_value=1

development/_downloads/60abb5e0c4b0861f5ecbe4ae9c2e51dd/example_extending_data_preprocessor.py

@@ -5,12 +5,15 @@
 
 The following example demonstrates how to turn off data preprocessing step in auto-skearn.
 """
+from typing import Optional
 from pprint import pprint
 
 import autosklearn.classification
 import autosklearn.pipeline.components.data_preprocessing
 import sklearn.metrics
 from ConfigSpace.configuration_space import ConfigurationSpace
+
+from autosklearn.askl_typing import FEAT_TYPE_TYPE
 from autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm
 from autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT
 from sklearn.datasets import load_breast_cancer
@@ -49,7 +52,9 @@ def get_properties(dataset_properties=None):
         }
 
     @staticmethod
-    def get_hyperparameter_search_space(dataset_properties=None):
+    def get_hyperparameter_search_space(
+        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None
+    ):
         return ConfigurationSpace()  # Return an empty configuration as there is None
 
 

development/_downloads/6dea5849db1f35abbefe123cd0eb49fd/example_extending_data_preprocessor.ipynb

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from pprint import pprint\n\nimport autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from typing import Optional\nfrom pprint import pprint\n\nimport autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\n\nfrom autosklearn.askl_typing import FEAT_TYPE_TYPE\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "class NoPreprocessing(AutoSklearnPreprocessingAlgorithm):\n    def __init__(self, **kwargs):\n        \"\"\"This preprocessors does not change the data\"\"\"\n        # Some internal checks makes sure parameters are set\n        for key, val in kwargs.items():\n            setattr(self, key, val)\n\n    def fit(self, X, Y=None):\n        return self\n\n    def transform(self, X):\n        return X\n\n    @staticmethod\n    def get_properties(dataset_properties=None):\n        return {\n            \"shortname\": \"NoPreprocessing\",\n            \"name\": \"NoPreprocessing\",\n            \"handles_regression\": True,\n            \"handles_classification\": True,\n            \"handles_multiclass\": True,\n            \"handles_multilabel\": True,\n            \"handles_multioutput\": True,\n            \"is_deterministic\": True,\n            \"input\": (SPARSE, DENSE, UNSIGNED_DATA),\n            \"output\": (INPUT,),\n        }\n\n    @staticmethod\n    def get_hyperparameter_search_space(dataset_properties=None):\n        return ConfigurationSpace()  # Return an empty configuration as there is None\n\n\n# Add NoPreprocessing component to auto-sklearn.\nautosklearn.pipeline.components.data_preprocessing.add_preprocessor(NoPreprocessing)"
+        "class NoPreprocessing(AutoSklearnPreprocessingAlgorithm):\n    def __init__(self, **kwargs):\n        \"\"\"This preprocessors does not change the data\"\"\"\n        # Some internal checks makes sure parameters are set\n        for key, val in kwargs.items():\n            setattr(self, key, val)\n\n    def fit(self, X, Y=None):\n        return self\n\n    def transform(self, X):\n        return X\n\n    @staticmethod\n    def get_properties(dataset_properties=None):\n        return {\n            \"shortname\": \"NoPreprocessing\",\n            \"name\": \"NoPreprocessing\",\n            \"handles_regression\": True,\n            \"handles_classification\": True,\n            \"handles_multiclass\": True,\n            \"handles_multilabel\": True,\n            \"handles_multioutput\": True,\n            \"is_deterministic\": True,\n            \"input\": (SPARSE, DENSE, UNSIGNED_DATA),\n            \"output\": (INPUT,),\n        }\n\n    @staticmethod\n    def get_hyperparameter_search_space(\n        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None\n    ):\n        return ConfigurationSpace()  # Return an empty configuration as there is None\n\n\n# Add NoPreprocessing component to auto-sklearn.\nautosklearn.pipeline.components.data_preprocessing.add_preprocessor(NoPreprocessing)"
       ]
     },
     {

development/_downloads/89acefb6af0174645412e5af4eafade1/example_text_preprocessing.py

@@ -59,7 +59,6 @@
 automl = autosklearn.classification.AutoSklearnClassifier(
     time_left_for_this_task=60,
     per_run_time_limit=30,
-    tmp_folder="/tmp/autosklearn_text_example_tmp",
 )
 
 automl.fit(X_train, y_train, dataset_name="20_Newsgroups")  # fit the automl model

development/_downloads/a23bc40b83b60b7e97a3eb188a82ad24/example_extending_preprocessor.py

@@ -7,6 +7,7 @@
 discriminant analysis (LDA) algorithm from sklearn and use it as a preprocessor
 in auto-sklearn.
 """
+from typing import Optional
 from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
@@ -17,6 +18,8 @@
 from ConfigSpace.conditions import InCondition
 
 import sklearn.metrics
+
+from autosklearn.askl_typing import FEAT_TYPE_TYPE
 import autosklearn.classification
 import autosklearn.pipeline.components.feature_preprocessing
 from autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm
@@ -76,7 +79,9 @@ def get_properties(dataset_properties=None):
         }
 
     @staticmethod
-    def get_hyperparameter_search_space(dataset_properties=None):
+    def get_hyperparameter_search_space(
+        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None
+    ):
         cs = ConfigurationSpace()
         solver = CategoricalHyperparameter(
             name="solver", choices=["svd", "lsqr", "eigen"], default_value="svd"

development/_downloads/bc82bea3a5dd7bdba60b65220891d9e5/examples_python.zip

@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import (\n    CategoricalHyperparameter,\n    UniformIntegerHyperparameter,\n    UniformFloatHyperparameter,\n)\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import (\n    DENSE,\n    SIGNED_DATA,\n    UNSIGNED_DATA,\n    PREDICTIONS,\n)\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from typing import Optional\nfrom pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import (\n    CategoricalHyperparameter,\n    UniformIntegerHyperparameter,\n    UniformFloatHyperparameter,\n)\n\nimport sklearn.metrics\n\nfrom autosklearn.askl_typing import FEAT_TYPE_TYPE\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import (\n    DENSE,\n    SIGNED_DATA,\n    UNSIGNED_DATA,\n    PREDICTIONS,\n)\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "class MLPClassifier(AutoSklearnClassificationAlgorithm):\n    def __init__(\n        self,\n        hidden_layer_depth,\n        num_nodes_per_layer,\n        activation,\n        alpha,\n        solver,\n        random_state=None,\n    ):\n        self.hidden_layer_depth = hidden_layer_depth\n        self.num_nodes_per_layer = num_nodes_per_layer\n        self.activation = activation\n        self.alpha = alpha\n        self.solver = solver\n        self.random_state = random_state\n\n    def fit(self, X, y):\n        self.num_nodes_per_layer = int(self.num_nodes_per_layer)\n        self.hidden_layer_depth = int(self.hidden_layer_depth)\n        self.alpha = float(self.alpha)\n\n        from sklearn.neural_network import MLPClassifier\n\n        hidden_layer_sizes = tuple(\n            self.num_nodes_per_layer for i in range(self.hidden_layer_depth)\n        )\n\n        self.estimator = MLPClassifier(\n            hidden_layer_sizes=hidden_layer_sizes,\n            activation=self.activation,\n            alpha=self.alpha,\n            solver=self.solver,\n            random_state=self.random_state,\n        )\n        self.estimator.fit(X, y)\n        return self\n\n    def predict(self, X):\n        if self.estimator is None:\n            raise NotImplementedError()\n        return self.estimator.predict(X)\n\n    def predict_proba(self, X):\n        if self.estimator is None:\n            raise NotImplementedError()\n        return self.estimator.predict_proba(X)\n\n    @staticmethod\n    def get_properties(dataset_properties=None):\n        return {\n            \"shortname\": \"MLP Classifier\",\n            \"name\": \"MLP CLassifier\",\n            \"handles_regression\": False,\n            \"handles_classification\": True,\n            \"handles_multiclass\": True,\n            \"handles_multilabel\": False,\n            \"handles_multioutput\": False,\n            \"is_deterministic\": False,\n            # Both input and output must be tuple(iterable)\n            \"input\": [DENSE, SIGNED_DATA, UNSIGNED_DATA],\n            \"output\": [PREDICTIONS],\n        }\n\n    @staticmethod\n    def get_hyperparameter_search_space(dataset_properties=None):\n        cs = ConfigurationSpace()\n        hidden_layer_depth = UniformIntegerHyperparameter(\n            name=\"hidden_layer_depth\", lower=1, upper=3, default_value=1\n        )\n        num_nodes_per_layer = UniformIntegerHyperparameter(\n            name=\"num_nodes_per_layer\", lower=16, upper=216, default_value=32\n        )\n        activation = CategoricalHyperparameter(\n            name=\"activation\",\n            choices=[\"identity\", \"logistic\", \"tanh\", \"relu\"],\n            default_value=\"relu\",\n        )\n        alpha = UniformFloatHyperparameter(\n            name=\"alpha\", lower=0.0001, upper=1.0, default_value=0.0001\n        )\n        solver = CategoricalHyperparameter(\n            name=\"solver\", choices=[\"lbfgs\", \"sgd\", \"adam\"], default_value=\"adam\"\n        )\n        cs.add_hyperparameters(\n            [\n                hidden_layer_depth,\n                num_nodes_per_layer,\n                activation,\n                alpha,\n                solver,\n            ]\n        )\n        return cs\n\n\n# Add MLP classifier component to auto-sklearn.\nautosklearn.pipeline.components.classification.add_classifier(MLPClassifier)\ncs = MLPClassifier.get_hyperparameter_search_space()\nprint(cs)"
+        "class MLPClassifier(AutoSklearnClassificationAlgorithm):\n    def __init__(\n        self,\n        hidden_layer_depth,\n        num_nodes_per_layer,\n        activation,\n        alpha,\n        solver,\n        random_state=None,\n    ):\n        self.hidden_layer_depth = hidden_layer_depth\n        self.num_nodes_per_layer = num_nodes_per_layer\n        self.activation = activation\n        self.alpha = alpha\n        self.solver = solver\n        self.random_state = random_state\n\n    def fit(self, X, y):\n        self.num_nodes_per_layer = int(self.num_nodes_per_layer)\n        self.hidden_layer_depth = int(self.hidden_layer_depth)\n        self.alpha = float(self.alpha)\n\n        from sklearn.neural_network import MLPClassifier\n\n        hidden_layer_sizes = tuple(\n            self.num_nodes_per_layer for i in range(self.hidden_layer_depth)\n        )\n\n        self.estimator = MLPClassifier(\n            hidden_layer_sizes=hidden_layer_sizes,\n            activation=self.activation,\n            alpha=self.alpha,\n            solver=self.solver,\n            random_state=self.random_state,\n        )\n        self.estimator.fit(X, y)\n        return self\n\n    def predict(self, X):\n        if self.estimator is None:\n            raise NotImplementedError()\n        return self.estimator.predict(X)\n\n    def predict_proba(self, X):\n        if self.estimator is None:\n            raise NotImplementedError()\n        return self.estimator.predict_proba(X)\n\n    @staticmethod\n    def get_properties(dataset_properties=None):\n        return {\n            \"shortname\": \"MLP Classifier\",\n            \"name\": \"MLP CLassifier\",\n            \"handles_regression\": False,\n            \"handles_classification\": True,\n            \"handles_multiclass\": True,\n            \"handles_multilabel\": False,\n            \"handles_multioutput\": False,\n            \"is_deterministic\": False,\n            # Both input and output must be tuple(iterable)\n            \"input\": [DENSE, SIGNED_DATA, UNSIGNED_DATA],\n            \"output\": [PREDICTIONS],\n        }\n\n    @staticmethod\n    def get_hyperparameter_search_space(\n        feat_type: Optional[FEAT_TYPE_TYPE] = None, dataset_properties=None\n    ):\n        cs = ConfigurationSpace()\n        hidden_layer_depth = UniformIntegerHyperparameter(\n            name=\"hidden_layer_depth\", lower=1, upper=3, default_value=1\n        )\n        num_nodes_per_layer = UniformIntegerHyperparameter(\n            name=\"num_nodes_per_layer\", lower=16, upper=216, default_value=32\n        )\n        activation = CategoricalHyperparameter(\n            name=\"activation\",\n            choices=[\"identity\", \"logistic\", \"tanh\", \"relu\"],\n            default_value=\"relu\",\n        )\n        alpha = UniformFloatHyperparameter(\n            name=\"alpha\", lower=0.0001, upper=1.0, default_value=0.0001\n        )\n        solver = CategoricalHyperparameter(\n            name=\"solver\", choices=[\"lbfgs\", \"sgd\", \"adam\"], default_value=\"adam\"\n        )\n        cs.add_hyperparameters(\n            [\n                hidden_layer_depth,\n                num_nodes_per_layer,\n                activation,\n                alpha,\n                solver,\n            ]\n        )\n        return cs\n\n\n# Add MLP classifier component to auto-sklearn.\nautosklearn.pipeline.components.classification.add_classifier(MLPClassifier)\ncs = MLPClassifier.get_hyperparameter_search_space()\nprint(cs)"
       ]
     },
     {