refactor: name property in providers

Author: iraedeus

pysatl_cpd/analysis/labeled_data.py

@@ -8,12 +8,10 @@
 __license__ = "SPDX-License-Identifier: MIT"
 
 from collections.abc import Collection, Iterator, Sequence
-from dataclasses import dataclass
 
 from pysatl_cpd.core.data_providers import DataProvider
 
 
-@dataclass
 class LabeledData[T](DataProvider[T]):
     """
     Container for labeled time series data with known change point locations.
@@ -51,37 +49,23 @@ class LabeledData[T](DataProvider[T]):
     6
     """
 
-    raw_data: Collection[T]
-    change_points: Sequence[int]
-    name: str | None = None
+    def __init__(self, raw_data: Collection[T], change_points: Sequence[int], name: str | None = None):
+        super().__init__(name)
 
-    def __post_init__(self) -> None:
-        """
-        Validate change point indices after initialization.
-
-        Verifies that all change point indices are positive and within
-        the bounds of the raw data. Change points must be >= 1 because
-        they represent the first observation index after a regime change,
-        and index 0 would imply a change before any data exists.
-
-        Raises
-        ------
-        ValueError
-            If any change point index is <= 0.
-        ValueError
-            If any change point index exceeds the length of raw_data.
-        """
         # Validate that change points are positive
-        if self.change_points and min(self.change_points) <= 0:
-            raise ValueError(f"Change point indices must be positive (>= 1). Found index: {min(self.change_points)}")
+        if change_points and min(change_points) <= 0:
+            raise ValueError(f"Change point indices must be positive (>= 1). Found index: {min(change_points)}")
 
         # Validate that change points are within data bounds
-        max_index = max(self.change_points) if self.change_points else 0
-        if max_index > len(self.raw_data):
+        max_index = max(change_points) if change_points else 0
+        if max_index > len(raw_data):
             raise ValueError(
-                f"Change point index exceeds data length. Max index: {max_index}, data length: {len(self.raw_data)}"
+                f"Change point index exceeds data length. Max index: {max_index}, data length: {len(raw_data)}"
             )
 
+        self.__raw_data = raw_data
+        self.__change_points = change_points
+
     def __iter__(self) -> Iterator[T]:
         """
         Return an iterator over the raw observations.
@@ -91,7 +75,7 @@ def __iter__(self) -> Iterator[T]:
         Iterator[T]
             Iterator yielding each observation in sequence.
         """
-        return iter(self.raw_data)
+        return iter(self.__raw_data)
 
     def __len__(self) -> int:
         """
@@ -102,7 +86,7 @@ def __len__(self) -> int:
         int
             Total number of observations.
         """
-        return len(self.raw_data)
+        return len(self.__raw_data)
 
     def __str__(self) -> str:
         """
@@ -113,6 +97,12 @@ def __str__(self) -> str:
         str
             String representation with dataset name (if provided) and length.
         """
-        if self.name is not None:
-            return f"{self.name} (len = {len(self)})"
-        return f"Labeled Data (len = {len(self)})"
+        return f"{self.name} (len = {len(self)})"
+
+    @property
+    def raw_data(self) -> Collection[T]:
+        return self.__raw_data
+
+    @property
+    def change_points(self) -> Sequence[int]:
+        return self.__change_points

pysatl_cpd/core/data_providers/idata_provider.py

@@ -34,6 +34,9 @@ class DataProvider[T](ABC):
         For multivariate data, T is typically a one-dimensional array.
     """
 
+    def __init__(self, name: str | None) -> None:
+        self._name = name if name is not None else type(self).__name__
+
     @abstractmethod
     def __iter__(self) -> Iterator[T]:
         """
@@ -59,7 +62,6 @@ def __len__(self) -> int:
         raise NotImplementedError  # pragma: no cover
 
     @property
-    @abstractmethod
     def name(self) -> str:
         """Return the name of the DataProvider"""
-        raise NotImplementedError  # pragma: no cover
+        return self._name

pysatl_cpd/core/data_providers/numpy_data_provider.py

@@ -49,7 +49,7 @@ class NDArrayUnivariateProvider(DataProvider[NumPyNumber]):
     [1.0, 2.0, 3.0, 4.0, 5.0]
     """
 
-    def __init__(self, data: NumericArray) -> None:
+    def __init__(self, data: NumericArray, name: str | None = None) -> None:
         """
         Initialize the univariate provider with a NumPy array.
 
@@ -63,6 +63,8 @@ def __init__(self, data: NumericArray) -> None:
         ValueError
             If the array is not one-dimensional.
         """
+        super().__init__(name)
+
         if data.ndim != 1:
             raise ValueError(f"Expected 1-dimensional array, got {data.ndim} dimensions")
         self.__data = cast(UnivariateNumericArray, data)
@@ -89,10 +91,6 @@ def __len__(self) -> int:
         """
         return self.__data.shape[0]
 
-    @property
-    def name(self) -> str:
-        return "NDArrayUnivariateProvider"
-
 
 class NDArrayMultivariateProvider(DataProvider[UnivariateNumericArray]):
     """
@@ -126,7 +124,7 @@ class NDArrayMultivariateProvider(DataProvider[UnivariateNumericArray]):
     [array([1., 2.]), array([3., 4.]), array([5., 6.])]
     """
 
-    def __init__(self, data: NumericArray) -> None:
+    def __init__(self, data: NumericArray, name: str | None = None) -> None:
         """
         Initialize the multivariate provider with a NumPy array.
 
@@ -140,6 +138,8 @@ def __init__(self, data: NumericArray) -> None:
         ValueError
             If the array is not two-dimensional.
         """
+        super().__init__(name)
+
         if data.ndim != 2:
             raise ValueError(f"Expected 2 dimensions, got {data.ndim}")
         self.__data = cast(MultivariateNumericArray, data)
@@ -166,7 +166,3 @@ def __len__(self) -> int:
             The number of rows in the underlying array (number of observations).
         """
         return self.__data.shape[0]
-
-    @property
-    def name(self) -> str:
-        return "NDArrayMultivariateProvider"

tests/mocks/core/data_providers/dirty.py

@@ -40,7 +40,10 @@ def __init__(
         source: DataProvider[Number],
         nan_indices: list[int] | None = None,
         inf_indices: list[int] | None = None,
+        name: str | None = None,
     ) -> None:
+        super().__init__(name)
+
         self._source = source
         self._nan_indices = set(nan_indices or [])
         self._inf_indices = set(inf_indices or [])
@@ -82,10 +85,6 @@ def __getitem__(self, index: int) -> Number:
             return float("inf")
         return self._data[index]
 
-    @property
-    def name(self) -> str:
-        return "MockUnivariateDirtyDataProvider"
-
     def get_call_count(self) -> int:
         """Return number of times __iter__ was called."""
         return self._call_count
@@ -126,7 +125,10 @@ def __init__(
         source: DataProvider[list[Number]],
         nan_positions: list[tuple[int, int]] | None = None,
         inf_positions: list[tuple[int, int]] | None = None,
+        name: str | None = None,
     ) -> None:
+        super().__init__(name)
+
         self._source = source
         self._nan_positions = set(nan_positions or [])
         self._inf_positions = set(inf_positions or [])
@@ -199,10 +201,6 @@ def dimensions(self) -> int:
         """Return number of dimensions (variables)."""
         return self._dimensions
 
-    @property
-    def name(self) -> str:
-        return "MockMultivariateDirtyDataProvider"
-
     def __repr__(self) -> str:
         """Return string representation."""
         return (

tests/mocks/core/data_providers/edge.py

@@ -17,7 +17,9 @@
 class MockEmptyDataProvider[T](DataProvider[T]):
     """Mock data provider that yields no observations."""
 
-    def __init__(self) -> None:
+    def __init__(self, name: str | None = None) -> None:
+        super().__init__(name)
+
         self._call_count = 0
 
     def __iter__(self) -> Iterator[T]:
@@ -37,10 +39,6 @@ def reset_call_count(self) -> None:
         """Reset the call counter."""
         self._call_count = 0
 
-    @property
-    def name(self) -> str:
-        return "MockEmptyDataProvider"
-
     def __repr__(self) -> str:
         """Return string representation."""
         return "MockEmptyDataProvider()"
@@ -49,7 +47,9 @@ def __repr__(self) -> str:
 class MockSingleObservationProvider[T](DataProvider[T]):
     """Mock data provider with a single observation."""
 
-    def __init__(self, observation: T) -> None:
+    def __init__(self, observation: T, name: str | None = None) -> None:
+        super().__init__(name)
+
         self._observation = observation
         self._call_count = 0
 
@@ -70,10 +70,6 @@ def reset_call_count(self) -> None:
         """Reset the call counter."""
         self._call_count = 0
 
-    @property
-    def name(self) -> str:
-        return "MockSingleObservationProvider"
-
     def __repr__(self) -> str:
         """Return string representation."""
         return f"MockSingleObservationProvider(value={self._observation})"

tests/mocks/core/data_providers/multivariate.py

@@ -29,7 +29,9 @@ class MockMultivariateDataProvider(DataProvider[list[Number]]):
         All observations must have the same length.
     """
 
-    def __init__(self, data: Sequence[Sequence[Number]]) -> None:
+    def __init__(self, data: Sequence[Sequence[Number]], name: str | None = None) -> None:
+        super().__init__(name)
+
         if not data:
             self._data: list[list[Number]] = []
             self._dimensions = 0
@@ -76,10 +78,6 @@ def dimensions(self) -> int:
         """Return number of dimensions (variables)."""
         return self._dimensions
 
-    @property
-    def name(self) -> str:
-        return "MockMultivariateDataProvider"
-
     def __repr__(self) -> str:
         """Return string representation."""
         return f"MockMultivariateDataProvider(observations={len(self)}, dimensions={self.dimensions})"
@@ -99,14 +97,10 @@ class MockMultivariateConstantDataProvider(MockMultivariateDataProvider):
         Number of observations to yield.
     """
 
-    def __init__(self, value: list[Number], length: int) -> None:
+    def __init__(self, value: list[Number], length: int, name: str | None = None) -> None:
         self._value = value
         self._length = length
-        super().__init__([value] * length)
-
-    @property
-    def name(self) -> str:
-        return "MockMultivariateConstantDataProvider"
+        super().__init__([value] * length, name)
 
 
 class MockMultivariateZeroDataProvider(MockMultivariateConstantDataProvider):
@@ -123,13 +117,9 @@ class MockMultivariateZeroDataProvider(MockMultivariateConstantDataProvider):
         Number of observations to yield.
     """
 
-    def __init__(self, dimensions: int, length: int) -> None:
+    def __init__(self, dimensions: int, length: int, name: str | None = None) -> None:
         value = [0.0] * dimensions
-        super().__init__(value, length)
-
-    @property
-    def name(self) -> str:
-        return "MockMultivariateZeroDataProvider"
+        super().__init__(value, length, name)
 
 
 class MockMultivariateNaNDataProvider(MockMultivariateConstantDataProvider):
@@ -146,13 +136,9 @@ class MockMultivariateNaNDataProvider(MockMultivariateConstantDataProvider):
         Number of observations to yield.
     """
 
-    def __init__(self, dimensions: int, length: int) -> None:
+    def __init__(self, dimensions: int, length: int, name: str | None = None) -> None:
         value = [float("nan")] * dimensions
-        super().__init__(value, length)
-
-    @property
-    def name(self) -> str:
-        return "MockMultivariateNaNDataProvider"
+        super().__init__(value, length, name)
 
 
 class MockMultivariateInfDataProvider(MockMultivariateConstantDataProvider):
@@ -169,10 +155,6 @@ class MockMultivariateInfDataProvider(MockMultivariateConstantDataProvider):
         Number of observations to yield.
     """
 
-    def __init__(self, dimensions: int, length: int) -> None:
+    def __init__(self, dimensions: int, length: int, name: str | None = None) -> None:
         value = [float("inf")] * dimensions
-        super().__init__(value, length)
-
-    @property
-    def name(self) -> str:
-        return "MockMultivariateInfDataProvider"
+        super().__init__(value, length, name)