mloda 0.3.3__py3-none-any.whl → 0.4.1__py3-none-any.whl

mloda_plugins/feature_group/experimental/sklearn/scaling/base.py

@@ -5,20 +5,21 @@ Base implementation for scikit-learn scaling feature groups.
 from __future__ import annotations
 
 import datetime
-from typing import Any, Dict, Optional, Set, Type, Union
-
-from mloda_core.abstract_plugins.abstract_feature_group import AbstractFeatureGroup
-from mloda_core.abstract_plugins.components.feature import Feature
-from mloda_core.abstract_plugins.components.feature_name import FeatureName
-from mloda_core.abstract_plugins.components.feature_set import FeatureSet
-from mloda_core.abstract_plugins.components.options import Options
-from mloda_core.abstract_plugins.components.feature_chainer.feature_chain_parser import FeatureChainParser
-from mloda_core.abstract_plugins.components.base_artifact import BaseArtifact
+from typing import Any, Dict, Optional, Type
+
+from mloda import FeatureGroup
+from mloda import Feature
+from mloda.provider import FeatureSet
+from mloda.provider import FeatureChainParser
+from mloda.provider import (
+    FeatureChainParserMixin,
+)
+from mloda.provider import BaseArtifact
 from mloda_plugins.feature_group.experimental.default_options_key import DefaultOptionKeys
 from mloda_plugins.feature_group.experimental.sklearn.sklearn_artifact import SklearnArtifact
 
 
-class ScalingFeatureGroup(AbstractFeatureGroup):
+class ScalingFeatureGroup(FeatureChainParserMixin, FeatureGroup):
     """
     Base class for scikit-learn scaling feature groups.
 
@@ -82,17 +83,21 @@ class ScalingFeatureGroup(AbstractFeatureGroup):
     PATTERN = "__"
     PREFIX_PATTERN = r".*__(standard|minmax|robust|normalizer)_scaled$"
 
+    # In-feature configuration for FeatureChainParserMixin
+    MIN_IN_FEATURES = 1
+    MAX_IN_FEATURES = 1
+
     # Property mapping for new configuration-based approach
     PROPERTY_MAPPING = {
         SCALER_TYPE: {
             **SUPPORTED_SCALERS,  # All supported scaler types as valid options
-            DefaultOptionKeys.mloda_context: True,  # Context parameter
-            DefaultOptionKeys.mloda_strict_validation: True,  # Enable strict validation
+            DefaultOptionKeys.context: True,  # Context parameter
+            DefaultOptionKeys.strict_validation: True,  # Enable strict validation
         },
         DefaultOptionKeys.in_features: {
             "explanation": "Source feature to scale",
-            DefaultOptionKeys.mloda_context: True,  # Context parameter
-            DefaultOptionKeys.mloda_strict_validation: False,  # Flexible validation
+            DefaultOptionKeys.context: True,  # Context parameter
+            DefaultOptionKeys.strict_validation: False,  # Flexible validation
         },
     }
 
@@ -101,22 +106,6 @@ class ScalingFeatureGroup(AbstractFeatureGroup):
         """Return the artifact class for sklearn scaler persistence."""
         return SklearnArtifact
 
-    def input_features(self, options: Options, feature_name: FeatureName) -> Optional[Set[Feature]]:
-        """Extract source feature from either configuration-based options or string parsing."""
-
-        # Try string-based parsing first
-        _, source_feature = FeatureChainParser.parse_feature_name(feature_name, [self.PREFIX_PATTERN])
-        if source_feature is not None:
-            return {Feature(source_feature)}
-
-        # Fall back to configuration-based approach
-        source_features = options.get_in_features()
-        if len(source_features) != 1:
-            raise ValueError(
-                f"Expected exactly one source feature, but found {len(source_features)}: {source_features}"
-            )
-        return set(source_features)
-
     @classmethod
     def get_scaler_type(cls, feature_name: str) -> str:
         """Extract the scaler type from the feature name."""
@@ -133,22 +122,6 @@ class ScalingFeatureGroup(AbstractFeatureGroup):
 
         return scaler_type
 
-    @classmethod
-    def match_feature_group_criteria(
-        cls,
-        feature_name: Union[FeatureName, str],
-        options: Options,
-        data_access_collection: Optional[Any] = None,
-    ) -> bool:
-        """Check if feature name matches the expected pattern using unified parser."""
-        # Use the unified parser with property mapping for full configuration support
-        return FeatureChainParser.match_configuration_feature_chain_parser(
-            feature_name,
-            options,
-            property_mapping=cls.PROPERTY_MAPPING,
-            prefix_patterns=[cls.PREFIX_PATTERN],
-        )
-
     @classmethod
     def calculate_feature(cls, data: Any, features: FeatureSet) -> Any:
         """
@@ -213,33 +186,44 @@ class ScalingFeatureGroup(AbstractFeatureGroup):
         Raises:
             ValueError: If parameters cannot be extracted
         """
-        scaler_type = None
-        source_feature_name: str | None = None
+        source_features = cls._extract_source_features(feature)
+        scaler_type = cls._extract_scaler_type(feature)
+        if scaler_type is None:
+            raise ValueError(f"Could not extract scaler type from: {feature.name}")
+        return scaler_type, source_features[0]
 
-        # Try string-based parsing first
+    @classmethod
+    def _extract_scaler_type(cls, feature: Feature) -> Optional[str]:
+        """
+        Extract scaler type from a feature.
+
+        Tries string-based parsing first, falls back to configuration-based approach.
+
+        Args:
+            feature: The feature to extract scaler type from
+
+        Returns:
+            The scaler type string
+
+        Raises:
+            ValueError: If scaler type is unsupported
+        """
         feature_name_str = feature.name.name if hasattr(feature.name, "name") else str(feature.name)
 
+        # Try string-based parsing first
         if FeatureChainParser.is_chained_feature(feature_name_str):
             scaler_type = cls.get_scaler_type(feature_name_str)
-            source_feature_name = FeatureChainParser.extract_source_feature(feature_name_str, cls.PREFIX_PATTERN)
-            return scaler_type, source_feature_name
+            return scaler_type
 
         # Fall back to configuration-based approach
-        source_features = feature.options.get_in_features()
-        source_feature = next(iter(source_features))
-        source_feature_name = source_feature.get_name()
-
         scaler_type = feature.options.get(cls.SCALER_TYPE)
 
-        if scaler_type is None or source_feature_name is None:
-            raise ValueError(f"Could not extract scaler type and source feature from: {feature.name}")
-
-        if scaler_type not in cls.SUPPORTED_SCALERS:
+        if scaler_type is not None and scaler_type not in cls.SUPPORTED_SCALERS:
             raise ValueError(
                 f"Unsupported scaler type: {scaler_type}. Supported types: {', '.join(cls.SUPPORTED_SCALERS.keys())}"
             )
 
-        return scaler_type, source_feature_name
+        return str(scaler_type) if scaler_type is not None else None
 
     @classmethod
     def _import_sklearn_components(cls) -> Dict[str, Any]:

mloda_plugins/feature_group/experimental/sklearn/scaling/pandas.py

@@ -6,7 +6,7 @@ from __future__ import annotations
 
 from typing import Any, Set, Type, Union
 
-from mloda_core.abstract_plugins.compute_frame_work import ComputeFrameWork
+from mloda import ComputeFramework
 
 from mloda_plugins.compute_framework.base_implementations.pandas.dataframe import PandasDataFrame
 from mloda_plugins.feature_group.experimental.sklearn.scaling.base import ScalingFeatureGroup
@@ -21,7 +21,7 @@ class PandasScalingFeatureGroup(ScalingFeatureGroup):
     """
 
     @classmethod
-    def compute_framework_rule(cls) -> Union[bool, Set[Type[ComputeFrameWork]]]:
+    def compute_framework_rule(cls) -> Union[bool, Set[Type[ComputeFramework]]]:
         """Specify that this feature group works with Pandas."""
         return {PandasDataFrame}
 

mloda_plugins/feature_group/experimental/sklearn/sklearn_artifact.py

@@ -4,14 +4,13 @@ Artifact for storing fitted scikit-learn transformers and estimators.
 
 import json
 import base64
-import os
 import hashlib
 import tempfile
 from pathlib import Path
 from typing import Any, Dict, Optional
 
-from mloda_core.abstract_plugins.components.base_artifact import BaseArtifact
-from mloda_core.abstract_plugins.components.feature_set import FeatureSet
+from mloda.provider import BaseArtifact
+from mloda.provider import FeatureSet
 
 
 class SklearnArtifact(BaseArtifact):

mloda_plugins/feature_group/experimental/source_input_feature.py

@@ -39,16 +39,16 @@ Further, it allows defining:
 """
 
 from typing import Any, Dict, NamedTuple, Optional, Set, Tuple, Type, Union
-from mloda_core.abstract_plugins.abstract_feature_group import AbstractFeatureGroup
-from mloda_core.abstract_plugins.components.feature import Feature
-from mloda_core.abstract_plugins.components.feature_name import FeatureName
-from mloda_core.abstract_plugins.components.index.index import Index
-from mloda_core.abstract_plugins.components.link import JoinType, Link, JoinSpec
-from mloda_core.abstract_plugins.components.options import Options
+from mloda import FeatureGroup
+from mloda import Feature
+from mloda.user import FeatureName
+from mloda.user import Index
+from mloda.user import JoinType, Link, JoinSpec
+from mloda import Options
 from mloda_plugins.feature_group.experimental.default_options_key import DefaultOptionKeys
 
 
-class SourceInputFeature(AbstractFeatureGroup):
+class SourceInputFeature(FeatureGroup):
     """
     This feature group focuses on defining input features, especially when they originate
     from other sources or require joins/merges.
@@ -100,19 +100,19 @@ class SourceTuple(NamedTuple):
 
     Attributes:
         feature_name: The name of the feature.
-        source_class: (Optional) The source class of the feature, can be an `AbstractFeatureGroup` class or a `str` representing a scope.
+        source_class: (Optional) The source class of the feature, can be an `FeatureGroup` class or a `str` representing a scope.
         source_value: (Optional) The value associated with the source class, if applicable.
-        left_link: (Optional)  A tuple containing the left-side `AbstractFeatureGroup` class and index for join operations.
-        right_link: (Optional) A tuple containing the right-side `AbstractFeatureGroup` class and index for join operations.
+        left_link: (Optional)  A tuple containing the left-side `FeatureGroup` class and index for join operations.
+        right_link: (Optional) A tuple containing the right-side `FeatureGroup` class and index for join operations.
         join_type: (Optional) The type of join operation (`JoinType`).
         merge_index: (Optional) The index to use for merge operations.
     """
 
     feature_name: str
-    source_class: Optional[Type[Union[AbstractFeatureGroup, str]]] = None
+    source_class: Optional[Type[Union[FeatureGroup, str]]] = None
     source_value: Optional[str] = None
-    left_link: Optional[Tuple[Type[AbstractFeatureGroup], Union[str, Index]]] = None
-    right_link: Optional[Tuple[Type[AbstractFeatureGroup], Union[str, Index]]] = None
+    left_link: Optional[Tuple[Type[FeatureGroup], Union[str, Index]]] = None
+    right_link: Optional[Tuple[Type[FeatureGroup], Union[str, Index]]] = None
     join_type: Optional[JoinType] = None
     merge_index: Optional[Union[str, Index]] = None
 
@@ -207,8 +207,8 @@ class SourceInputFeatureComposite:
     @classmethod
     def _handle_link(
         cls,
-        left_link: Tuple[Type[AbstractFeatureGroup], Union[str, Index]],
-        right_link: Tuple[Type[AbstractFeatureGroup], Union[str, Index]],
+        left_link: Tuple[Type[FeatureGroup], Union[str, Index]],
+        right_link: Tuple[Type[FeatureGroup], Union[str, Index]],
         join_type: Any,
     ) -> Link:
         """

mloda_plugins/feature_group/experimental/text_cleaning/base.py

@@ -4,18 +4,19 @@ Base implementation for text cleaning feature groups.
 
 from __future__ import annotations
 
-from typing import Any, Optional, Set, Union
-
-from mloda_core.abstract_plugins.abstract_feature_group import AbstractFeatureGroup
-from mloda_core.abstract_plugins.components.feature import Feature
-from mloda_core.abstract_plugins.components.feature_chainer.feature_chain_parser import FeatureChainParser
-from mloda_core.abstract_plugins.components.feature_name import FeatureName
-from mloda_core.abstract_plugins.components.feature_set import FeatureSet
-from mloda_core.abstract_plugins.components.options import Options
+from typing import Any, Optional
+
+from mloda import FeatureGroup
+from mloda import Feature
+from mloda.provider import FeatureChainParser
+from mloda.provider import (
+    FeatureChainParserMixin,
+)
+from mloda.provider import FeatureSet
 from mloda_plugins.feature_group.experimental.default_options_key import DefaultOptionKeys
 
 
-class TextCleaningFeatureGroup(AbstractFeatureGroup):
+class TextCleaningFeatureGroup(FeatureChainParserMixin, FeatureGroup):
     # Option key for the list of operations
     CLEANING_OPERATIONS = "cleaning_operations"
 
@@ -33,13 +34,17 @@ class TextCleaningFeatureGroup(AbstractFeatureGroup):
     PATTERN = "__"
     PREFIX_PATTERN = r".*__cleaned_text$"
 
+    # In-feature configuration for FeatureChainParserMixin
+    MIN_IN_FEATURES = 1
+    MAX_IN_FEATURES = 1
+
     # Property mapping for configuration-based features
     PROPERTY_MAPPING = {
         CLEANING_OPERATIONS: {
             **SUPPORTED_OPERATIONS,  # All supported operations as valid options
-            DefaultOptionKeys.mloda_context: True,  # Mark as context parameter
-            DefaultOptionKeys.mloda_strict_validation: True,  # Enable strict validation
-            DefaultOptionKeys.mloda_validation_function: lambda operations: (
+            DefaultOptionKeys.context: True,  # Mark as context parameter
+            DefaultOptionKeys.strict_validation: True,  # Enable strict validation
+            DefaultOptionKeys.validation_function: lambda operations: (
                 # Handle both actual tuples/lists and string representations
                 (
                     isinstance(operations, (tuple, list))
@@ -59,7 +64,7 @@ class TextCleaningFeatureGroup(AbstractFeatureGroup):
         },
         DefaultOptionKeys.in_features: {
             "explanation": "Source feature to apply text cleaning operations to",
-            DefaultOptionKeys.mloda_context: True,
+            DefaultOptionKeys.context: True,
         },
     }
 
@@ -115,41 +120,6 @@ class TextCleaningFeatureGroup(AbstractFeatureGroup):
     - The source feature must contain text data
     """
 
-    def input_features(self, options: Options, feature_name: FeatureName) -> Optional[Set[Feature]]:
-        """Extract source feature from either configuration-based options or string parsing."""
-
-        source_feature: str | None = None
-
-        # Try string-based parsing first
-        _, source_feature = FeatureChainParser.parse_feature_name(feature_name, [self.PREFIX_PATTERN])
-        if source_feature is not None:
-            return {Feature(source_feature)}
-
-        # Fall back to configuration-based approach
-        source_features = options.get_in_features()
-        if len(source_features) != 1:
-            raise ValueError(
-                f"Expected exactly one source feature, but found {len(source_features)}: {source_features}"
-            )
-        return set(source_features)
-
-    @classmethod
-    def match_feature_group_criteria(
-        cls,
-        feature_name: Union[FeatureName, str],
-        options: Options,
-        data_access_collection: Optional[Any] = None,
-    ) -> bool:
-        """Check if feature name matches the expected pattern for text cleaning features."""
-
-        # Use the unified parser with property mapping for full configuration support
-        return FeatureChainParser.match_configuration_feature_chain_parser(
-            feature_name,
-            options,
-            property_mapping=cls.PROPERTY_MAPPING,
-            prefix_patterns=[cls.PREFIX_PATTERN],
-        )
-
     @classmethod
     def _extract_operations_and_source_feature(cls, feature: Feature) -> tuple[tuple[Any, Any], str]:
         """
@@ -166,31 +136,36 @@ class TextCleaningFeatureGroup(AbstractFeatureGroup):
         Raises:
             ValueError: If parameters cannot be extracted
         """
-        operations = None
-        source_feature_name: str | None = None
+        source_features = cls._extract_source_features(feature)
+        operations = cls._extract_cleaning_operations(feature)
+        if operations is None:
+            raise ValueError(f"Could not extract operations from: {feature.name}")
+        return operations, source_features[0]
 
+    @classmethod
+    def _extract_cleaning_operations(cls, feature: Feature) -> Optional[tuple[Any, Any]]:
+        """
+        Extract cleaning operations from a feature.
+
+        Tries string-based parsing first, falls back to configuration-based approach.
+
+        Args:
+            feature: The feature to extract operations from
+
+        Returns:
+            Tuple of cleaning operations, or None if not found
+        """
         # Try string-based parsing first
         feature_name_str = feature.name.name if hasattr(feature.name, "name") else str(feature.name)
 
         if FeatureChainParser.is_chained_feature(feature_name_str):
-            _, source_feature_name = FeatureChainParser.parse_feature_name(feature_name_str, [cls.PREFIX_PATTERN])
             # For string-based features, get operations from options
             operations = feature.options.get(cls.CLEANING_OPERATIONS) or ()
-            if source_feature_name is None:
-                raise ValueError(f"Could not extract source feature from string-based feature: {feature.name}")
-            return operations, source_feature_name  # type: ignore
+            return operations  # type: ignore
 
         # Fall back to configuration-based approach
-        source_features = feature.options.get_in_features()
-        source_feature = next(iter(source_features))
-        source_feature_name = source_feature.get_name()
-
         operations = feature.options.get(cls.CLEANING_OPERATIONS)
-
-        if operations is None or source_feature_name is None:
-            raise ValueError(f"Could not extract cleaning operations and source feature from: {feature.name}")
-
-        return operations, source_feature_name
+        return operations if operations is not None else None
 
     @classmethod
     def calculate_feature(cls, data: Any, features: FeatureSet) -> Any:

mloda_plugins/feature_group/experimental/text_cleaning/pandas.py

@@ -25,7 +25,7 @@ except ImportError:
     pd = None
 
 
-from mloda_core.abstract_plugins.compute_frame_work import ComputeFrameWork
+from mloda import ComputeFramework
 from mloda_plugins.compute_framework.base_implementations.pandas.dataframe import PandasDataFrame
 from mloda_plugins.feature_group.experimental.text_cleaning.base import TextCleaningFeatureGroup
 
@@ -40,7 +40,7 @@ class PandasTextCleaningFeatureGroup(TextCleaningFeatureGroup):
     """
 
     @classmethod
-    def compute_framework_rule(cls) -> set[type[ComputeFrameWork]]:
+    def compute_framework_rule(cls) -> set[type[ComputeFramework]]:
         """Define the compute framework for this feature group."""
         return {PandasDataFrame}
 

mloda_plugins/feature_group/experimental/text_cleaning/python_dict.py

@@ -9,7 +9,7 @@ import string
 import unicodedata
 from typing import Any, Dict, List, Set, Type, Union
 
-from mloda_core.abstract_plugins.compute_frame_work import ComputeFrameWork
+from mloda import ComputeFramework
 
 from mloda_plugins.compute_framework.base_implementations.python_dict.python_dict_framework import PythonDictFramework
 from mloda_plugins.feature_group.experimental.text_cleaning.base import TextCleaningFeatureGroup
@@ -35,7 +35,7 @@ class PythonDictTextCleaningFeatureGroup(TextCleaningFeatureGroup):
     """
 
     @classmethod
-    def compute_framework_rule(cls) -> Union[bool, Set[Type[ComputeFrameWork]]]:
+    def compute_framework_rule(cls) -> Union[bool, Set[Type[ComputeFramework]]]:
         return {PythonDictFramework}
 
     @classmethod