Merge pull request #1296 from automl/development

Development

autosklearn/__version__.py

@@ -1,4 +1,4 @@
 """Version information."""
 
 # The following line *must* be the last in the module, exactly as formatted:
-__version__ = "0.14.1"
+__version__ = "0.14.2"

autosklearn/evaluation/__init__.py

@@ -24,6 +24,7 @@
 import autosklearn.evaluation.train_evaluator
 import autosklearn.evaluation.test_evaluator
 import autosklearn.evaluation.util
+import autosklearn.pipeline.components
 from autosklearn.evaluation.train_evaluator import TYPE_ADDITIONAL_INFO
 from autosklearn.util.backend import Backend
 from autosklearn.util.logging_ import PickableLoggerAdapter, get_named_client_logger
@@ -336,6 +337,7 @@ def run(
             init_params=init_params,
             budget=budget,
             budget_type=self.budget_type,
+            additional_components=autosklearn.pipeline.components.base._addons,
         )
 
         if self.resampling_strategy != 'test':

autosklearn/evaluation/abstract_evaluator.py

@@ -16,6 +16,7 @@
 
 import autosklearn.pipeline.classification
 import autosklearn.pipeline.regression
+from autosklearn.pipeline.components.base import ThirdPartyComponents, _addons
 from autosklearn.constants import (
     CLASSIFICATION_TASKS,
     REGRESSION_TASKS,
@@ -181,6 +182,7 @@ def __init__(
         backend: Backend,
         queue: multiprocessing.Queue,
         metric: Scorer,
+        additional_components: Dict[str, ThirdPartyComponents],
         port: Optional[int],
         configuration: Optional[Union[int, Configuration]] = None,
         scoring_functions: Optional[List[Scorer]] = None,
@@ -268,6 +270,15 @@ def __init__(
         self.budget = budget
         self.budget_type = budget_type
 
+        # Add 3rd-party components to the list of 3rd-party components in case this wasn't done
+        # before (this happens if we run in parallel and the components are only passed to the
+        # AbstractEvaluator via the TAE and are not there yet because the worker is in its own
+        # process).
+        for key in additional_components:
+            for component_name, component in additional_components[key].components.items():
+                if component_name not in _addons[key].components:
+                    _addons[key].add_component(component)
+
         # Please mypy to prevent not defined attr
         self.model = self._get_model()
 

autosklearn/evaluation/test_evaluator.py

@@ -12,6 +12,7 @@
     AbstractEvaluator,
     _fit_and_suppress_warnings,
 )
+from autosklearn.pipeline.components.base import ThirdPartyComponents
 from autosklearn.metrics import calculate_loss, Scorer
 from autosklearn.util.backend import Backend
 
@@ -29,6 +30,7 @@ def __init__(
         backend: Backend,
         queue: multiprocessing.Queue,
         metric: Scorer,
+        additional_components: Dict[str, ThirdPartyComponents],
         port: Optional[int],
         configuration: Optional[Union[int, Configuration]] = None,
         scoring_functions: Optional[List[Scorer]] = None,
@@ -44,6 +46,7 @@ def __init__(
             port=port,
             configuration=configuration,
             metric=metric,
+            additional_components=additional_components,
             scoring_functions=scoring_functions,
             seed=seed,
             output_y_hat_optimization=False,
@@ -120,6 +123,7 @@ def eval_t(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = None,
     budget_type: Optional[str] = None,
@@ -131,6 +135,7 @@ def eval_t(
                               scoring_functions=scoring_functions,
                               include=include, exclude=exclude,
                               disable_file_output=disable_file_output,
-                              init_params=init_params)
+                              additional_components=additional_components,
+                              init_params=init_params,)
 
     evaluator.fit_predict_and_loss()

autosklearn/evaluation/train_evaluator.py

@@ -35,7 +35,7 @@
      SUPPORTED_TARGET_TYPES,
  )
 from autosklearn.pipeline.base import PIPELINE_DATA_DTYPE
-from autosklearn.pipeline.components.base import IterativeComponent
+from autosklearn.pipeline.components.base import IterativeComponent, ThirdPartyComponents
 from autosklearn.metrics import Scorer
 from autosklearn.util.backend import Backend
 from autosklearn.util.logging_ import PicklableClientLogger
@@ -160,6 +160,7 @@ def __init__(
         backend: Backend,
         queue: multiprocessing.Queue,
         metric: Scorer,
+        additional_components: Dict[str, ThirdPartyComponents],
         port: Optional[int],
         configuration: Optional[Union[int, Configuration]] = None,
         scoring_functions: Optional[List[Scorer]] = None,
@@ -184,6 +185,7 @@ def __init__(
             port=port,
             configuration=configuration,
             metric=metric,
+            additional_components=additional_components,
             scoring_functions=scoring_functions,
             seed=seed,
             output_y_hat_optimization=output_y_hat_optimization,
@@ -1163,6 +1165,7 @@ def eval_holdout(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = 100.0,
     budget_type: Optional[str] = None,
@@ -1183,6 +1186,7 @@ def eval_holdout(
         include=include,
         exclude=exclude,
         disable_file_output=disable_file_output,
+        additional_components=additional_components,
         init_params=init_params,
         budget=budget,
         budget_type=budget_type,
@@ -1206,6 +1210,7 @@ def eval_iterative_holdout(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = 100.0,
     budget_type: Optional[str] = None,
@@ -1227,6 +1232,7 @@ def eval_iterative_holdout(
         instance=instance,
         disable_file_output=disable_file_output,
         iterative=True,
+        additional_components=additional_components,
         init_params=init_params,
         budget=budget,
         budget_type=budget_type
@@ -1249,6 +1255,7 @@ def eval_partial_cv(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = None,
     budget_type: Optional[str] = None,
@@ -1274,6 +1281,7 @@ def eval_partial_cv(
         include=include,
         exclude=exclude,
         disable_file_output=disable_file_output,
+        additional_components=additional_components,
         init_params=init_params,
         budget=budget,
         budget_type=budget_type,
@@ -1298,6 +1306,7 @@ def eval_partial_cv_iterative(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = None,
     budget_type: Optional[str] = None,
@@ -1321,6 +1330,7 @@ def eval_partial_cv_iterative(
         exclude=exclude,
         disable_file_output=disable_file_output,
         iterative=True,
+        additional_components=additional_components,
         init_params=init_params,
     )
 
@@ -1342,6 +1352,7 @@ def eval_cv(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = None,
     budget_type: Optional[str] = None,
@@ -1362,6 +1373,7 @@ def eval_cv(
         include=include,
         exclude=exclude,
         disable_file_output=disable_file_output,
+        additional_components=additional_components,
         init_params=init_params,
         budget=budget,
         budget_type=budget_type,
@@ -1386,6 +1398,7 @@ def eval_iterative_cv(
     exclude: Optional[List[str]],
     disable_file_output: bool,
     port: Optional[int],
+    additional_components: Dict[str, ThirdPartyComponents],
     init_params: Optional[Dict[str, Any]] = None,
     budget: Optional[float] = None,
     budget_type: Optional[str] = None,
@@ -1406,6 +1419,7 @@ def eval_iterative_cv(
         exclude=exclude,
         disable_file_output=disable_file_output,
         port=port,
+        additional_components=additional_components,
         init_params=init_params,
         budget=budget,
         budget_type=budget_type,

autosklearn/pipeline/components/base.py

@@ -3,11 +3,14 @@
 import inspect
 import pkgutil
 import sys
+from typing import Dict
 
 from sklearn.base import BaseEstimator, TransformerMixin
 
 from autosklearn.pipeline.constants import SPARSE
 
+_addons = dict()  # type: Dict[str, 'ThirdPartyComponents']
+
 
 def find_components(package, directory, base_class):
     components = OrderedDict()

autosklearn/pipeline/components/classification/__init__.py

@@ -5,19 +5,20 @@
 import os
 
 from ..base import AutoSklearnClassificationAlgorithm, find_components, \
-    ThirdPartyComponents, AutoSklearnChoice
+    ThirdPartyComponents, AutoSklearnChoice, _addons
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import CategoricalHyperparameter
 
 classifier_directory = os.path.split(__file__)[0]
 _classifiers = find_components(__package__,
                                classifier_directory,
                                AutoSklearnClassificationAlgorithm)
-_addons = ThirdPartyComponents(AutoSklearnClassificationAlgorithm)
+additional_components = ThirdPartyComponents(AutoSklearnClassificationAlgorithm)
+_addons['classification'] = additional_components
 
 
 def add_classifier(classifier: Type[AutoSklearnClassificationAlgorithm]) -> None:
-    _addons.add_component(classifier)
+    additional_components.add_component(classifier)
 
 
 class ClassifierChoice(AutoSklearnChoice):
@@ -26,7 +27,7 @@ class ClassifierChoice(AutoSklearnChoice):
     def get_components(cls):
         components = OrderedDict()
         components.update(_classifiers)
-        components.update(_addons.components)
+        components.update(additional_components.components)
         return components
 
     def get_available_components(cls, dataset_properties=None,

autosklearn/pipeline/components/data_preprocessing/categorical_encoding/__init__.py

@@ -10,19 +10,20 @@
 from sklearn.base import BaseEstimator
 
 from ...base import AutoSklearnPreprocessingAlgorithm, find_components, \
-    ThirdPartyComponents, AutoSklearnChoice
+    ThirdPartyComponents, AutoSklearnChoice, _addons
 
 from autosklearn.pipeline.base import DATASET_PROPERTIES_TYPE, PIPELINE_DATA_DTYPE
 
 ohe_directory = os.path.split(__file__)[0]
 _ohes = find_components(__package__,
                         ohe_directory,
                         AutoSklearnPreprocessingAlgorithm)
-_addons = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+additional_components = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+_addons['data_preprocessing.categorical_encoding'] = additional_components
 
 
 def add_ohe(ohe: 'OHEChoice') -> None:
-    _addons.add_component(ohe)
+    additional_components.add_component(ohe)
 
 
 class OHEChoice(AutoSklearnChoice):
@@ -31,7 +32,7 @@ class OHEChoice(AutoSklearnChoice):
     def get_components(cls: BaseEstimator) -> Dict[str, BaseEstimator]:
         components: Dict[str, BaseEstimator] = OrderedDict()
         components.update(_ohes)
-        components.update(_addons.components)
+        components.update(additional_components.components)
         return components
 
     def get_hyperparameter_search_space(

autosklearn/pipeline/components/data_preprocessing/minority_coalescense/__init__.py

@@ -8,7 +8,7 @@
 from ConfigSpace.hyperparameters import CategoricalHyperparameter
 
 from ...base import AutoSklearnPreprocessingAlgorithm, find_components, \
-    ThirdPartyComponents, AutoSklearnChoice
+    ThirdPartyComponents, AutoSklearnChoice, _addons
 
 from sklearn.base import BaseEstimator
 
@@ -17,11 +17,12 @@
 mc_directory = os.path.split(__file__)[0]
 _mcs = find_components(
     __package__, mc_directory, AutoSklearnPreprocessingAlgorithm)
-_addons = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+additional_components = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+_addons['data_preprocessing.minority_coalescense'] = additional_components
 
 
 def add_mc(mc: BaseEstimator) -> None:
-    _addons.add_component(mc)
+    additional_components.add_component(mc)
 
 
 class CoalescenseChoice(AutoSklearnChoice):
@@ -30,7 +31,7 @@ class CoalescenseChoice(AutoSklearnChoice):
     def get_components(cls: BaseEstimator) -> Dict[str, BaseEstimator]:
         components: Dict[str, BaseEstimator] = OrderedDict()
         components.update(_mcs)
-        components.update(_addons.components)
+        components.update(additional_components.components)
         return components
 
     def get_hyperparameter_search_space(

autosklearn/pipeline/components/data_preprocessing/rescaling/__init__.py

@@ -9,7 +9,7 @@
 from sklearn.base import BaseEstimator
 
 from ...base import AutoSklearnPreprocessingAlgorithm, find_components, \
-    ThirdPartyComponents, AutoSklearnChoice
+    ThirdPartyComponents, AutoSklearnChoice, _addons
 from autosklearn.pipeline.base import DATASET_PROPERTIES_TYPE, PIPELINE_DATA_DTYPE
 from autosklearn.pipeline.components.data_preprocessing.rescaling.abstract_rescaling import (
     Rescaling
@@ -19,11 +19,12 @@
 _rescalers = find_components(__package__,
                              rescaling_directory,
                              AutoSklearnPreprocessingAlgorithm)
-_addons = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+additional_components = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+_addons['data_preprocessing.rescaling'] = additional_components
 
 
 def add_rescaler(rescaler: Rescaling) -> None:
-    _addons.add_component(rescaler)
+    additional_components.add_component(rescaler)
 
 
 class RescalingChoice(AutoSklearnChoice):
@@ -32,7 +33,7 @@ class RescalingChoice(AutoSklearnChoice):
     def get_components(cls: BaseEstimator) -> Dict[str, BaseEstimator]:
         components: Dict[str, BaseEstimator] = OrderedDict()
         components.update(_rescalers)
-        components.update(_addons.components)
+        components.update(additional_components.components)
         return components
 
     def get_hyperparameter_search_space(

autosklearn/pipeline/components/feature_preprocessing/__init__.py

@@ -3,19 +3,20 @@
 from typing import Type
 
 from ..base import AutoSklearnPreprocessingAlgorithm, find_components, \
-    ThirdPartyComponents, AutoSklearnChoice
+    ThirdPartyComponents, AutoSklearnChoice, _addons
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import CategoricalHyperparameter
 
 classifier_directory = os.path.split(__file__)[0]
 _preprocessors = find_components(__package__,
                                  classifier_directory,
                                  AutoSklearnPreprocessingAlgorithm)
-_addons = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+additional_components = ThirdPartyComponents(AutoSklearnPreprocessingAlgorithm)
+_addons['feature_preprocessing'] = additional_components
 
 
 def add_preprocessor(preprocessor: Type[AutoSklearnPreprocessingAlgorithm]) -> None:
-    _addons.add_component(preprocessor)
+    additional_components.add_component(preprocessor)
 
 
 class FeaturePreprocessorChoice(AutoSklearnChoice):
@@ -24,7 +25,7 @@ class FeaturePreprocessorChoice(AutoSklearnChoice):
     def get_components(cls):
         components = OrderedDict()
         components.update(_preprocessors)
-        components.update(_addons.components)
+        components.update(additional_components.components)
         return components
 
     def get_available_components(self, dataset_properties=None,