[ADD] documentation for pipelines and steps (#329)

* Add documentation for pipelines and steps

* fix flake

* Apply suggestions from code review

Co-authored-by: nabenabe0928 <47781922+nabenabe0928@users.noreply.github.com>

* accept shuhei's suggestions

Co-authored-by: nabenabe0928 <47781922+nabenabe0928@users.noreply.github.com>

Author: ravinkohli

autoPyTorch/api/tabular_classification.py

@@ -148,6 +148,7 @@ def search(
         Fit both optimizes the machine learning models and builds an ensemble out of them.
         To disable ensembling, set ensemble_size==0.
         using the optimizer.
+
         Args:
             X_train, y_train, X_test, y_test: Union[np.ndarray, List, pd.DataFrame]
                 A pair of features (X_train) and targets (y_train) used to fit a

autoPyTorch/api/tabular_regression.py

@@ -149,6 +149,7 @@ def search(
         Fit both optimizes the machine learning models and builds an ensemble out of them.
         To disable ensembling, set ensemble_size==0.
         using the optimizer.
+
         Args:
             X_train, y_train, X_test, y_test: Union[np.ndarray, List, pd.DataFrame]
                 A pair of features (X_train) and targets (y_train) used to fit a

autoPyTorch/pipeline/base_pipeline.py

@@ -29,14 +29,16 @@
 
 
 class BasePipeline(Pipeline):
-    """Base class for all pipeline objects.
+    """
+    Base class for all pipeline objects.
 
     Args:
         config (Optional[Configuration]):
             Allows to directly specify a configuration space
         steps (Optional[List[Tuple[str, PipelineStepType]]]):
-            the list of steps that build the pipeline. If provided,
-             they won't be dynamically produced.
+            The list of `autoPyTorchComponent` or `autoPyTorchChoice`
+            that build the pipeline. If provided, they won't be
+            dynamically produced.
         include (Optional[Dict[str, Any]]):
             Allows the caller to specify which configurations to honor during
             the creation of the configuration space.
@@ -46,12 +48,12 @@ class BasePipeline(Pipeline):
         random_state (np.random.RandomState):
             allows to produce reproducible results by
             setting a seed for randomized settings
-        init_params (Optional[Dict[str, Any]])
+        init_params (Optional[Dict[str, Any]]):
+            Optional initial settings for the config
         search_space_updates (Optional[HyperparameterSearchSpaceUpdates]):
             search space updates that can be used to modify the search
             space of particular components or choice modules of the pipeline
 
-
     Attributes:
         steps (List[Tuple[str, PipelineStepType]]):
             the steps of the current pipeline. Each step in an AutoPyTorch

autoPyTorch/pipeline/components/base_choice.py

@@ -16,18 +16,22 @@
 
 
 class autoPyTorchChoice(object):
-    """Allows for the dynamically generation of components as pipeline steps.
+    """
+    Allows for the dynamically generation of components as pipeline steps.
 
     Args:
-        dataset_properties (Dict[str, Union[str, BaseDatasetPropertiesType]]): Describes the dataset
-            to work on
-        random_state (Optional[np.random.RandomState]): allows to produce reproducible
-            results by setting a seed for randomized settings
+        dataset_properties (Dict[str, Union[str, BaseDatasetPropertiesType]]):
+            Describes the dataset to work on
+        random_state (Optional[np.random.RandomState]):
+            Allows to produce reproducible results by setting a
+            seed for randomized settings
 
     Attributes:
-        random_state (Optional[np.random.RandomState]): allows to produce reproducible
-            results by setting a seed for randomized settings
-        choice (autoPyTorchComponent): the choice of components for this stage
+        random_state (Optional[np.random.RandomState]):
+            Allows to produce reproducible results by setting a seed for
+            randomized settings
+        choice (autoPyTorchComponent):
+            the choice of components for this stage
     """
     def __init__(self,
                  dataset_properties: Dict[str, BaseDatasetPropertiesType],
@@ -67,11 +71,13 @@ def get_components(cls: 'autoPyTorchChoice') -> Dict[str, autoPyTorchComponent]:
         for current step.
 
         Args:
-            cls (autoPyTorchChoice): The choice object from which to query the valid
+            cls (autoPyTorchChoice):
+                The choice object from which to query the valid
                 components
 
         Returns:
-            Dict[str, autoPyTorchComponent]: The available components via a mapping
+            Dict[str, autoPyTorchComponent]:
+                The available components via a mapping
                 from the module name to the component class
 
         """
@@ -88,10 +94,13 @@ def get_available_components(
         user specification
 
         Args:
-            dataset_properties (Optional[Dict[str, BaseDatasetPropertiesType]]): Describes the dataset to work on
-            include: Optional[Dict[str, Any]]: what components to include. It is an exhaustive
+            dataset_properties (Optional[Dict[str, BaseDatasetPropertiesType]]):
+                Describes the dataset to work on
+            include: Optional[Dict[str, Any]]:
+                what components to include. It is an exhaustive
                 list, and will exclusively use this components.
-            exclude: Optional[Dict[str, Any]]: which components to skip
+            exclude: Optional[Dict[str, Any]]:
+                which components to skip. Can't be used together with include
 
         Results:
             Dict[str, autoPyTorchComponent]: A dictionary with valid components for this
@@ -137,10 +146,10 @@ def set_hyperparameters(self,
         to an actual parameter of the autoPyTorch component.
 
         Args:
-            configuration (Configuration): which configuration to apply to
-                the chosen component
-            init_params (Optional[Dict[str, any]]): Optional arguments to
-                initialize the chosen component
+            configuration (Configuration):
+                Which configuration to apply to the chosen component
+            init_params (Optional[Dict[str, any]]):
+                Optional arguments to initialize the chosen component
 
         Returns:
             self: returns an instance of self
@@ -177,11 +186,15 @@ def get_hyperparameter_search_space(
         """Returns the configuration space of the current chosen components
 
         Args:
-            dataset_properties (Optional[Dict[str, BaseDatasetPropertiesType]]): Describes the dataset to work on
-            default: (Optional[str]) : Default component to use in hyperparameters
-            include: Optional[Dict[str, Any]]: what components to include. It is an exhaustive
+            dataset_properties (Optional[Dict[str, BaseDatasetPropertiesType]]):
+                Describes the dataset to work on
+            default: (Optional[str]):
+                Default component to use in hyperparameters
+            include: Optional[Dict[str, Any]]:
+                what components to include. It is an exhaustive
                 list, and will exclusively use this components.
-            exclude: Optional[Dict[str, Any]]: which components to skip
+            exclude: Optional[Dict[str, Any]]:
+                which components to skip
 
         Returns:
             ConfigurationSpace: the configuration space of the hyper-parameters of the
@@ -193,8 +206,10 @@ def fit(self, X: Dict[str, Any], y: Any) -> autoPyTorchComponent:
         """Handy method to check if a component is fitted
 
         Args:
-            X (X: Dict[str, Any]): Dependencies needed by current component to perform fit
-            y (Any): not used. To comply with sklearn API
+            X (X: Dict[str, Any]):
+                Dependencies needed by current component to perform fit
+            y (Any):
+                not used. To comply with sklearn API
         """
         # Allows to use check_is_fitted on the choice object
         self.fitted_ = True
@@ -205,19 +220,23 @@ def predict(self, X: np.ndarray) -> np.ndarray:
         """Predicts the target given an input, by using the chosen component
 
         Args:
-            X (np.ndarray): input features from which to predict the target
+            X (np.ndarray):
+                input features from which to predict the target
 
         Returns:
-            np.ndarray: the predicted target
+            np.ndarray:
+                the target prediction
         """
         assert self.choice is not None, "Cannot call predict without initializing the component"
         return self.choice.predict(X)
 
     def transform(self, X: Dict[str, Any]) -> Dict[str, Any]:
         """
         Adds the current choice in the fit dictionary
+
         Args:
-            X (Dict[str, Any]): fit dictionary
+            X (Dict[str, Any]):
+                fit dictionary
 
         Returns:
             (Dict[str, Any])
@@ -233,7 +252,8 @@ def check_requirements(self, X: Dict[str, Any], y: Any = None) -> None:
         are honored before fit.
 
         Args:
-            X (Dict[str, Any]): Dictionary with fitted parameters. It is a message passing
+            X (Dict[str, Any]):
+                Dictionary with fitted parameters. It is a message passing
                 mechanism, in which during a transform, a components adds relevant information
                 so that further stages can be properly fitted
         """
@@ -246,7 +266,8 @@ def _check_dataset_properties(self, dataset_properties: Dict[str, BaseDatasetPro
         """
         A mechanism in code to ensure the correctness of the initialised dataset properties.
         Args:
-            dataset_properties:
+            dataset_properties (Dict[str, BaseDatasetPropertiesType]):
+                Describes the dataset to work on
 
         """
         assert isinstance(dataset_properties, dict), "dataset_properties must be a dictionary"

autoPyTorch/pipeline/components/base_component.py

@@ -27,9 +27,12 @@ def find_components(
     that inherit from base_class
 
     Args:
-        package (str): The associated package that contains the components
-        directory (str): The directory from which to extract the components
-        base_class (BaseEstimator): base class to filter out desired components
+        package (str):
+            The associated package that contains the components
+        directory (str):
+            The directory from which to extract the components
+        base_class (BaseEstimator):
+            base class to filter out desired components
             that don't inherit from this class
     """
     components = OrderedDict()
@@ -60,7 +63,8 @@ class ThirdPartyComponents(object):
     space to work.
 
     Args:
-        base_class (BaseEstimator) component type desired to be created
+        base_class (BaseEstimator):
+            Component type desired to be created
     """
 
     def __init__(self, base_class: BaseEstimator):
@@ -96,6 +100,16 @@ def add_component(self, obj: BaseEstimator) -> None:
 
 
 class autoPyTorchComponent(BaseEstimator):
+    """
+    Provides an abstract interface which can be used to
+    create steps of a pipeline in AutoPyTorch.
+
+    Args:
+        random_state (Optional[np.random.RandomState]):
+            Allows to produce reproducible results by setting a
+            seed for randomized settings
+
+    """
     _required_properties: Optional[List[str]] = None
 
     def __init__(self, random_state: Optional[np.random.RandomState] = None) -> None:
@@ -115,7 +129,8 @@ def get_required_properties(cls) -> Optional[List[str]]:
         Usually defined in the base class of the component
 
         Returns:
-            List[str]: list of properties autopytorch component must have for proper functioning of the pipeline
+            List[str]:
+                list of properties autopytorch component must have for proper functioning of the pipeline
         """
         return cls._required_properties
 
@@ -125,8 +140,8 @@ def get_fit_requirements(self) -> Optional[List[FitRequirement]]:
         that need to be in the fit dictionary
 
         Returns:
-            List[FitRequirement]: a list containing required keys
-                            in a named tuple (name: str, type: object)
+            List[FitRequirement]:
+                a list containing required keys in a named tuple (name: str, type: object)
         """
         return self._fit_requirements
 
@@ -139,11 +154,12 @@ def get_properties(dataset_properties: Optional[Dict[str, BaseDatasetPropertiesT
         """Get the properties of the underlying algorithm.
 
         Args:
-            dataset_properties (Optional[Dict[str, Union[str, int]]): Describes the dataset
-               to work on
+            dataset_properties (Optional[Dict[str, Union[str, int]]):
+                Describes the dataset to work on
 
         Returns:
-            Dict[str, Any]: Properties of the algorithm
+            Dict[str, Any]:
+                Properties of the algorithm
         """
         raise NotImplementedError()
 
@@ -154,11 +170,12 @@ def get_hyperparameter_search_space(
         """Return the configuration space of this classification algorithm.
 
         Args:
-            dataset_properties (Optional[Dict[str, Union[str, int]]): Describes the dataset
-               to work on
+            dataset_properties (Optional[Dict[str, Union[str, int]]):
+                Describes the dataset to work on
 
         Returns:
-            ConfigurationSpace: The configuration space of this algorithm.
+            ConfigurationSpace:
+                The configuration space of this algorithm.
         """
         raise NotImplementedError()
 
@@ -167,13 +184,16 @@ def fit(self, X: Dict[str, Any], y: Any = None) -> "autoPyTorchComponent":
         model and returns `self`.
 
         Args:
-            X (Dict[str, Any]): Dictionary with fitted parameters. It is a message passing
+            X (Dict[str, Any]):
+                Dictionary with fitted parameters. It is a message passing
                 mechanism, in which during a transform, a components adds relevant information
                 so that further stages can be properly fitted
-            y (Any): Not Used -- to comply with API
+            y (Any):
+                Not Used -- to comply with API
 
         Returns:
-            self : returns an instance of self.
+            self:
+                returns an instance of self.
 
         Notes:
             Please see the `scikit-learn API documentation
@@ -192,10 +212,10 @@ def set_hyperparameters(self,
         to an actual parameter of the autoPyTorch component.
 
         Args:
-            configuration (Configuration): which configuration to apply to
-                the chosen component
-            init_params (Optional[Dict[str, any]]): Optional arguments to
-                initialize the chosen component
+            configuration (Configuration):
+                Which configuration to apply to the chosen component
+            init_params (Optional[Dict[str, any]]):
+                Optional arguments to initialize the chosen component
 
         Returns:
             An instance of self
@@ -226,7 +246,8 @@ def check_requirements(self, X: Dict[str, Any], y: Any = None) -> None:
         are honored before fit.
 
         Args:
-            X (Dict[str, Any]): Dictionary with fitted parameters. It is a message passing
+            X (Dict[str, Any]):
+                Dictionary with fitted parameters. It is a message passing
                 mechanism, in which during a transform, a components adds relevant information
                 so that further stages can be properly fitted
         """
@@ -267,10 +288,12 @@ def _apply_search_space_update(self, hyperparameter_search_space_update: Hyperpa
         """Allows the user to update a hyperparameter
 
         Args:
-            name (str): name of hyperparameter
+            name (str):
+                name of hyperparameter
             new_value_range (List[Union[int, str, float]]):
                 value range can be either lower, upper or a list of possible candidates
-            log (bool): Whether to use log scale
+            log (bool):
+                Whether to use log scale
         """
 
         self._cs_updates[hyperparameter_search_space_update.hyperparameter] = hyperparameter_search_space_update