[ENH] Add primitive type hints to segmentation module

aeon/segmentation/_binseg.py

@@ -58,7 +58,9 @@ class BinSegmenter(BaseSegmenter):
         "python_dependencies": "ruptures",
     }
 
-    def __init__(self, n_cps=1, model="l2", min_size=2, jump=5):
+    def __init__(
+        self, n_cps: int = 1, model: str = "l2", min_size: int = 2, jump: int = 5
+    ):
         self.n_cps = n_cps
         self.model = model
         self.min_size = min_size
@@ -122,7 +124,7 @@ def _get_interval_series(self, X, found_cps):
         return pd.IntervalIndex.from_arrays(start, end)
 
     @classmethod
-    def _get_test_params(cls, parameter_set="default"):
+    def _get_test_params(cls, parameter_set: str = "default"):
         """Return testing parameter settings for the estimator.
 
         Parameters

aeon/segmentation/_clasp.py

@@ -4,7 +4,6 @@
 
 __maintainer__ = []
 __all__ = ["ClaSPSegmenter", "find_dominant_window_sizes"]
-
 from queue import PriorityQueue
 
 import numpy as np
@@ -15,7 +14,7 @@
 from aeon.utils.validation import check_n_jobs
 
 
-def find_dominant_window_sizes(X, offset=0.05):
+def find_dominant_window_sizes(X, offset: float = 0.05):
     """Determine the Window-Size using dominant FFT-frequencies.
 
     Parameters
@@ -55,7 +54,9 @@ def find_dominant_window_sizes(X, offset=0.05):
     )
 
 
-def _is_trivial_match(candidate, change_points, n_timepoints, exclusion_radius=0.05):
+def _is_trivial_match(
+    candidate: int, change_points, n_timepoints: int, exclusion_radius: float = 0.05
+):
     """Check if a candidate change point is in close proximity to other change points.
 
     Parameters
@@ -87,7 +88,9 @@ def _is_trivial_match(candidate, change_points, n_timepoints, exclusion_radius=0
     return False
 
 
-def _segmentation(X, clasp, n_change_points=None, exclusion_radius=0.05):
+def _segmentation(
+    X, clasp, n_change_points: int | None = None, exclusion_radius: float = 0.05
+):
     """Segments the time series by extracting change points.
 
     Parameters
@@ -182,7 +185,7 @@ class ClaSPSegmenter(BaseSegmenter):
         Size of window for sliding, based on the period length of the data.
     n_cps : int, default = 1
         The number of change points to search.
-    exclusion_radius : int
+    exclusion_radius : float
         Exclusion Radius for change points to be non-trivial matches.
     n_jobs : int, default=1
         Number of jobs to be used.
@@ -209,7 +212,13 @@ class ClaSPSegmenter(BaseSegmenter):
 
     _tags = {"capability:multithreading": True, "fit_is_empty": True}
 
-    def __init__(self, period_length=10, n_cps=1, exclusion_radius=0.05, n_jobs=1):
+    def __init__(
+        self,
+        period_length: int = 10,
+        n_cps: int = 1,
+        exclusion_radius: float = 0.05,
+        n_jobs: int = 1,
+    ):
         self.period_length = int(period_length)
         self.n_cps = n_cps
         self.exclusion_radius = exclusion_radius
@@ -289,7 +298,7 @@ def _get_interval_series(self, X, found_cps):
         Parameters
         ----------
         X :         array-like, shape = [n]
-           Univariate time-series data to be segmented.
+        Univariate time-series data to be segmented.
         found_cps : array-like, shape = [n_cps] The found change points found
 
         Returns
@@ -304,7 +313,7 @@ def _get_interval_series(self, X, found_cps):
         return pd.IntervalIndex.from_arrays(start, end)
 
     @classmethod
-    def _get_test_params(cls, parameter_set="default"):
+    def _get_test_params(cls, parameter_set: str = "default"):
         """Return testing parameter settings for the estimator.
 
         Parameters

aeon/segmentation/_eagglo.py

@@ -89,7 +89,7 @@ class EAggloSegmenter(BaseSegmenter):
     def __init__(
         self,
         member=None,
-        alpha=1.0,
+        alpha: float = 1.0,
         penalty=None,
     ):
         self.member = member

aeon/segmentation/_fluss.py

@@ -48,7 +48,9 @@ class FLUSSSegmenter(BaseSegmenter):
         "python_dependencies": "stumpy",
     }
 
-    def __init__(self, period_length=10, n_regimes=2, exclusion_factor=5):
+    def __init__(
+        self, period_length: int = 10, n_regimes: int = 2, exclusion_factor: int = 5
+    ):
         self.period_length = int(period_length)
         self.n_regimes = n_regimes
         self.exclusion_factor = exclusion_factor
@@ -137,7 +139,7 @@ def _get_interval_series(self, X, found_cps):
         return pd.IntervalIndex.from_arrays(start, end)
 
     @classmethod
-    def _get_test_params(cls, parameter_set="default"):
+    def _get_test_params(cls, parameter_set: str = "default"):
         """Return testing parameter settings for the estimator.
 
         Parameters

aeon/segmentation/_random.py

@@ -10,7 +10,23 @@
 class RandomSegmenter(BaseSegmenter):
     """Random Segmenter.
 
-    Randomly segments a time series.
+    Randomly segments a time series into a specified number of segments
+    at random cut points.
+
+    Parameters
+    ----------
+    n_segments : int, default=2
+        Number of segments to split the time series into  (must be >= 2).
+    random_state : int, RandomState instance or None, default=None
+        Random seed for reproducibility.
+
+    Examples
+    --------
+        >>> from aeon.segmentation import RandomSegmenter
+        >>> segmenter = RandomSegmenter(n_segments=3)
+        >>> from aeon.testing.data_generation import make_example_1d_numpy
+        >>> X = make_example_1d_numpy(n_timepoints=100, random_state=0)
+        >>> found_cps = segmenter.fit_predict(X)
     """
 
     _tags = {
@@ -20,7 +36,7 @@ class RandomSegmenter(BaseSegmenter):
         "returns_dense": True,
     }
 
-    def __init__(self, random_state=None, n_segments=2):
+    def __init__(self, random_state=None, n_segments: int = 2):
         self.random_state = random_state
         self.breakpoints_ = []
         super().__init__(axis=1, n_segments=n_segments)