fix(KDP): added fixes for the distribution estimator and tests

Author: Gandalfdore

.ruff.toml

@@ -123,7 +123,7 @@ encoder = DistributionAwareEncoder(
 |-----------|------|---------|-------------|
 | num_bins | int | 1000 | Number of bins for quantile encoding |
 | epsilon | float | 1e-6 | Small value for numerical stability |
-| detect_periodicity | bool | True | Enable periodic pattern detection |
+| detect_periodicity | bool | True | Enable periodic pattern detection | Remove this parameter when having multimodal functions/distributions
 | handle_sparsity | bool | True | Enable special handling for sparse data |
 | adaptive_binning | bool | True | Enable adaptive bin boundaries |
 | mixture_components | int | 3 | Number of components for mixture models |

docs/distribution_aware_encoder.md

@@ -53,7 +53,11 @@ def __init__(
             **kwargs: Additional keyword arguments for the feature.
         """
         self.name = name
-        self.feature_type = FeatureType.from_string(feature_type) if isinstance(feature_type, str) else feature_type
+        self.feature_type = (
+            FeatureType.from_string(feature_type)
+            if isinstance(feature_type, str)
+            else feature_type
+        )
         self.preprocessors = preprocessors or []
         self.kwargs = kwargs
 
@@ -93,7 +97,9 @@ def from_string(type_str: str) -> "FeatureType":
 class NumericalFeature(Feature):
     """NumericalFeature with dynamic kwargs passing."""
 
-    def __init__(self, name: str, feature_type: FeatureType = FeatureType.FLOAT, **kwargs) -> None:
+    def __init__(
+        self, name: str, feature_type: FeatureType = FeatureType.FLOAT, **kwargs
+    ) -> None:
         """Initializes a NumericalFeature instance.
 
         Args:
@@ -126,7 +132,9 @@ def __init__(
         """
         super().__init__(name, feature_type, **kwargs)
         self.category_encoding = category_encoding
-        self.dtype = tf.int32 if feature_type == FeatureType.INTEGER_CATEGORICAL else tf.string
+        self.dtype = (
+            tf.int32 if feature_type == FeatureType.INTEGER_CATEGORICAL else tf.string
+        )
         self.kwargs = kwargs
 
     def _embedding_size_rule(self, nr_categories: int) -> int:
@@ -144,7 +152,9 @@ def _embedding_size_rule(self, nr_categories: int) -> int:
 class TextFeature(Feature):
     """TextFeature with dynamic kwargs passing."""
 
-    def __init__(self, name: str, feature_type: FeatureType = FeatureType.TEXT, **kwargs) -> None:
+    def __init__(
+        self, name: str, feature_type: FeatureType = FeatureType.TEXT, **kwargs
+    ) -> None:
         """Initializes a TextFeature instance.
 
         Args:
@@ -160,7 +170,9 @@ def __init__(self, name: str, feature_type: FeatureType = FeatureType.TEXT, **kw
 class DateFeature(Feature):
     """TextFeature with dynamic kwargs passing."""
 
-    def __init__(self, name: str, feature_type: FeatureType = FeatureType.DATE, **kwargs) -> None:
+    def __init__(
+        self, name: str, feature_type: FeatureType = FeatureType.DATE, **kwargs
+    ) -> None:
         """Initializes a DateFeature instance.
 
         Args:

kdp/custom_layers.py

@@ -18,7 +18,9 @@
 
 class PreprocessorLayerFactory:
     @staticmethod
-    def create_layer(layer_class: str | object, name: str = None, **kwargs) -> tf.keras.layers.Layer:
+    def create_layer(
+        layer_class: str | object, name: str = None, **kwargs
+    ) -> tf.keras.layers.Layer:
         """Create a layer using the layer class name, automatically filtering kwargs based on the layer class.
 
         Args:
@@ -39,7 +41,9 @@ def create_layer(layer_class: str | object, name: str = None, **kwargs) -> tf.ke
         constructor_params = inspect.signature(layer_class.__init__).parameters
 
         # Filter kwargs to only include those that the constructor can accept
-        filtered_kwargs = {key: value for key, value in kwargs.items() if key in constructor_params}
+        filtered_kwargs = {
+            key: value for key, value in kwargs.items() if key in constructor_params
+        }
 
         # Add the 'name' argument if provided else default the class name lowercase option
         if name:
@@ -86,7 +90,9 @@ def distribution_aware_encoder(
         )
 
     @staticmethod
-    def text_preprocessing_layer(name: str = "text_preprocessing", **kwargs: dict) -> tf.keras.layers.Layer:
+    def text_preprocessing_layer(
+        name: str = "text_preprocessing", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a TextPreprocessingLayer layer.
 
         Args:
@@ -103,7 +109,9 @@ def text_preprocessing_layer(name: str = "text_preprocessing", **kwargs: dict) -
         )
 
     @staticmethod
-    def cast_to_float32_layer(name: str = "cast_to_float32", **kwargs: dict) -> tf.keras.layers.Layer:
+    def cast_to_float32_layer(
+        name: str = "cast_to_float32", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a CastToFloat32Layer layer.
 
         Args:
@@ -120,7 +128,9 @@ def cast_to_float32_layer(name: str = "cast_to_float32", **kwargs: dict) -> tf.k
         )
 
     @staticmethod
-    def date_parsing_layer(name: str = "date_parsing_layer", **kwargs: dict) -> tf.keras.layers.Layer:
+    def date_parsing_layer(
+        name: str = "date_parsing_layer", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a DateParsingLayer layer.
 
         Args:
@@ -137,7 +147,9 @@ def date_parsing_layer(name: str = "date_parsing_layer", **kwargs: dict) -> tf.k
         )
 
     @staticmethod
-    def date_encoding_layer(name: str = "date_encoding_layer", **kwargs: dict) -> tf.keras.layers.Layer:
+    def date_encoding_layer(
+        name: str = "date_encoding_layer", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a DateEncodingLayer layer.
 
         Args:
@@ -154,7 +166,9 @@ def date_encoding_layer(name: str = "date_encoding_layer", **kwargs: dict) -> tf
         )
 
     @staticmethod
-    def date_season_layer(name: str = "date_season_layer", **kwargs: dict) -> tf.keras.layers.Layer:
+    def date_season_layer(
+        name: str = "date_season_layer", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a SeasonLayer layer.
 
         Args:
@@ -171,7 +185,9 @@ def date_season_layer(name: str = "date_season_layer", **kwargs: dict) -> tf.ker
         )
 
     @staticmethod
-    def transformer_block_layer(name: str = "transformer", **kwargs: dict) -> tf.keras.layers.Layer:
+    def transformer_block_layer(
+        name: str = "transformer", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a TransformerBlock layer.
 
         Args:
@@ -241,7 +257,9 @@ def multi_resolution_attention_layer(
         )
 
     @staticmethod
-    def variable_selection_layer(name: str = "variable_selection", **kwargs: dict) -> tf.keras.layers.Layer:
+    def variable_selection_layer(
+        name: str = "variable_selection", **kwargs: dict
+    ) -> tf.keras.layers.Layer:
         """Create a VariableSelection layer.
 
         Args:

kdp/features.py

@@ -7,7 +7,9 @@
 
 
 class ProcessingStep:
-    def __init__(self, layer_creator: Callable[..., tf.keras.layers.Layer], **layer_kwargs) -> None:
+    def __init__(
+        self, layer_creator: Callable[..., tf.keras.layers.Layer], **layer_kwargs
+    ) -> None:
         """Initialize a processing step.
 
         Args:
@@ -94,7 +96,9 @@ def __init__(self, name: str) -> None:
         self.name = name
         self.pipeline = Pipeline(name=name)
 
-    def add_processing_step(self, layer_creator: Callable[..., tf.keras.layers.Layer] = None, **layer_kwargs) -> None:
+    def add_processing_step(
+        self, layer_creator: Callable[..., tf.keras.layers.Layer] = None, **layer_kwargs
+    ) -> None:
         """Add a processing step to the feature preprocessor.
 
         Args: