[MNT] Add typecheck workflow using mypy

.github/workflows/pr_typecheck.yml

@@ -0,0 +1,43 @@
+name: PR Typecheck
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+    paths:
+      - "aeon/**"
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.head_ref || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  typecheck:
+    # run the code coverage job if a PR has the '' label
+    if: ${{ github.event_name == 'pull_request' && contains(github.event.pull_request.labels.*.name, 'run typecheck test') }}
+    runs-on: ubuntu-22.04
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Python 3.10
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.10"
+
+      - name: Install aeon, dependencies and mypy
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 30
+          max_attempts: 3
+          command: python -m pip install .[all_extras,unstable_extras,dev] mypy
+
+      - name: Show dependencies
+        run: python -m pip list
+
+      - name: Run mypy typecheck
+        run: mypy aeon/

aeon/classification/distance_based/_proximity_forest.py

@@ -3,6 +3,8 @@
 The Proximity Forest is an ensemble of Proximity Trees.
 """
 
+from typing import Optional
+
 __all__ = ["ProximityForest"]
 
 from typing import Union
@@ -87,7 +89,7 @@ def __init__(
         self,
         n_trees=100,
         n_splitters: int = 5,
-        max_depth: int = None,
+        max_depth: Optional[int] = None,
         min_samples_split: int = 2,
         random_state: Union[int, np.random.RandomState, None] = None,
         n_jobs: int = 1,

aeon/classification/distance_based/_proximity_tree.py

@@ -5,7 +5,7 @@
 aeon distances.
 """
 
-from typing import Union
+from typing import Optional, Union
 
 import numpy as np
 from numba import njit
@@ -118,7 +118,7 @@ class ProximityTree(BaseClassifier):
     def __init__(
         self,
         n_splitters: int = 5,
-        max_depth: int = None,
+        max_depth: Optional[int] = None,
         min_samples_split: int = 2,
         random_state: Union[int, np.random.RandomState, None] = None,
     ) -> None:

aeon/classification/distance_based/_time_series_neighbors.py

@@ -5,6 +5,8 @@
 distances in aeon.distances.
 """
 
+from typing import Optional
+
 __maintainer__ = []
 __all__ = ["KNeighborsTimeSeriesClassifier"]
 
@@ -72,7 +74,7 @@ class KNeighborsTimeSeriesClassifier(BaseClassifier):
     def __init__(
         self,
         distance: Union[str, Callable] = "dtw",
-        distance_params: dict = None,
+        distance_params: Optional[dict] = None,
         n_neighbors: int = 1,
         weights: Union[str, Callable] = "uniform",
         n_jobs: int = 1,

aeon/classification/shapelet_based/_sast.py

@@ -3,6 +3,8 @@
 Pipeline classifier using the SAST transformer and an sklearn classifier.
 """
 
+from typing import Optional
+
 __maintainer__ = ["TonyBagnall"]
 __all__ = ["SASTClassifier"]
 
@@ -69,7 +71,7 @@ def __init__(
         length_list=None,
         stride: int = 1,
         nb_inst_per_class: int = 1,
-        seed: int = None,
+        seed: Optional[int] = None,
         classifier=None,
         n_jobs: int = -1,
     ) -> None:

aeon/clustering/_clara.py

@@ -1,5 +1,7 @@
 """Time series kmedoids."""
 
+from typing import Optional
+
 __maintainer__ = []
 
 from typing import Callable, Union
@@ -118,14 +120,14 @@ def __init__(
         n_clusters: int = 8,
         init_algorithm: Union[str, np.ndarray] = "random",
         distance: Union[str, Callable] = "msm",
-        n_samples: int = None,
+        n_samples: Optional[int] = None,
         n_sampling_iters: int = 10,
         n_init: int = 1,
         max_iter: int = 300,
         tol: float = 1e-6,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
-        distance_params: dict = None,
+        random_state: Optional[Union[int, RandomState]] = None,
+        distance_params: Optional[dict] = None,
     ):
         self.init_algorithm = init_algorithm
         self.distance = distance

aeon/clustering/_clarans.py

@@ -1,5 +1,7 @@
 """Time series kmedoids."""
 
+from typing import Optional
+
 __maintainer__ = []
 
 import math
@@ -104,11 +106,11 @@ def __init__(
         n_clusters: int = 8,
         init_algorithm: Union[str, np.ndarray] = "random",
         distance: Union[str, Callable] = "msm",
-        max_neighbours: int = None,
+        max_neighbours: Optional[int] = None,
         n_init: int = 10,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
-        distance_params: dict = None,
+        random_state: Optional[Union[int, RandomState]] = None,
+        distance_params: Optional[dict] = None,
     ):
         self.max_neighbours = max_neighbours
 

aeon/clustering/_k_means.py

@@ -1,5 +1,7 @@
 """Time series kmeans."""
 
+from typing import Optional
+
 __maintainer__ = []
 
 from typing import Callable, Union
@@ -163,11 +165,11 @@ def __init__(
         max_iter: int = 300,
         tol: float = 1e-6,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
+        random_state: Optional[Union[int, RandomState]] = None,
         averaging_method: Union[str, Callable[[np.ndarray], np.ndarray]] = "ba",
-        distance_params: dict = None,
-        average_params: dict = None,
-        init_algorithm: Union[str, np.ndarray] = None,
+        distance_params: Optional[dict] = None,
+        average_params: Optional[dict] = None,
+        init_algorithm: Optional[Union[str, np.ndarray]] = None,
     ):
         self.init = init
         self.init_algorithm = init_algorithm

aeon/clustering/_k_medoids.py

@@ -1,5 +1,7 @@
 """Time series kmedoids."""
 
+from typing import Optional
+
 __maintainer__ = []
 
 import warnings
@@ -157,8 +159,8 @@ def __init__(
         max_iter: int = 300,
         tol: float = 1e-6,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
-        distance_params: dict = None,
+        random_state: Optional[Union[int, RandomState]] = None,
+        distance_params: Optional[dict] = None,
     ):
         self.init_algorithm = init_algorithm
         self.distance = distance

aeon/clustering/_k_shape.py

@@ -1,6 +1,6 @@
 """Time series kshapes."""
 
-from typing import Union
+from typing import Optional, Union
 
 import numpy as np
 from numpy.random import RandomState
@@ -82,7 +82,7 @@ def __init__(
         max_iter: int = 300,
         tol: float = 1e-4,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
+        random_state: Optional[Union[int, RandomState]] = None,
     ):
         self.init_algorithm = init_algorithm
         self.n_init = n_init

aeon/clustering/_k_shapes.py

@@ -1,6 +1,6 @@
 """Time series kshapes."""
 
-from typing import Union
+from typing import Optional, Union
 
 import numpy as np
 from deprecated.sphinx import deprecated
@@ -83,7 +83,7 @@ def __init__(
         max_iter: int = 300,
         tol: float = 1e-4,
         verbose: bool = False,
-        random_state: Union[int, RandomState] = None,
+        random_state: Optional[Union[int, RandomState]] = None,
     ):
         self.init_algorithm = init_algorithm
         self.n_init = n_init

aeon/clustering/_kernel_k_means.py

@@ -1,6 +1,6 @@
 """Time series kernel kmeans."""
 
-from typing import Union
+from typing import Optional, Union
 
 import numpy as np
 from numpy.random import RandomState
@@ -99,7 +99,7 @@ def __init__(
         kernel_params: Union[dict, None] = None,
         verbose: bool = False,
         n_jobs: Union[int, None] = None,
-        random_state: Union[int, RandomState] = None,
+        random_state: Optional[Union[int, RandomState]] = None,
     ):
         self.kernel = kernel
         self.n_init = n_init

aeon/clustering/base.py

@@ -1,5 +1,7 @@
 """Base class for clustering."""
 
+from typing import Optional
+
 __maintainer__ = []
 __all__ = ["BaseClusterer"]
 
@@ -22,7 +24,7 @@ class BaseClusterer(BaseCollectionEstimator, ABC):
         Number of clusters for model.
     """
 
-    def __init__(self, n_clusters: int = None):
+    def __init__(self, n_clusters: Optional[int] = None):
         self.n_clusters = n_clusters
         # required for compatibility with some sklearn interfaces e.g.
         # CalibratedClassifierCV

aeon/datasets/_data_loaders.py

@@ -1,5 +1,7 @@
 """Dataset loading functions."""
 
+from typing import Optional
+
 __all__ = [  # Load functions
     "load_from_tsfile",
     "load_from_tsf_file",
@@ -680,7 +682,7 @@ def load_from_tsv_file(full_file_path_and_name):
 def _convert_tsf_to_hierarchical(
     data: pd.DataFrame,
     metadata,
-    freq: str = None,
+    freq: Optional[str] = None,
     value_column_name: str = "series_value",
 ) -> pd.DataFrame:
     """Convert the data from default_tsf to pd_multiindex_hier.

aeon/regression/distance_based/_time_series_neighbors.py

@@ -5,6 +5,8 @@
 It can also be used with callables, or aeon (pairwise transformer) estimators.
 """
 
+from typing import Optional
+
 __maintainer__ = []
 __all__ = ["KNeighborsTimeSeriesRegressor"]
 
@@ -72,7 +74,7 @@ class KNeighborsTimeSeriesRegressor(BaseRegressor):
     def __init__(
         self,
         distance: Union[str, Callable] = "dtw",
-        distance_params: dict = None,
+        distance_params: Optional[dict] = None,
         n_neighbors: int = 1,
         weights: Union[str, Callable] = "uniform",
         n_jobs: int = 1,

aeon/segmentation/_ggs.py

@@ -36,6 +36,7 @@
 import logging
 import math
 from dataclasses import dataclass, field
+from typing import Optional
 
 import numpy as np
 import numpy.typing as npt
@@ -457,7 +458,7 @@ def __init__(
         lamb: float = 1.0,
         max_shuffles: int = 250,
         verbose: bool = False,
-        random_state: int = None,
+        random_state: Optional[int] = None,
     ):
         self.k_max = k_max
         self.lamb = lamb

aeon/segmentation/_hmm.py

@@ -7,6 +7,7 @@
 """
 
 import warnings
+from typing import Optional
 
 import numpy as np
 from scipy.stats import norm
@@ -137,7 +138,7 @@ def __init__(
         self,
         emission_funcs: list,
         transition_prob_mat: np.ndarray,
-        initial_probs: np.ndarray = None,
+        initial_probs: Optional[np.ndarray] = None,
     ):
         self.initial_probs = initial_probs
         self.emission_funcs = emission_funcs

aeon/similarity_search/matrix_profiles/stomp.py

@@ -1,5 +1,7 @@
 """Implementation of stomp for euclidean and squared euclidean distance profile."""
 
+from typing import Optional
+
 __maintainer__ = ["baraline"]
 
 
@@ -29,7 +31,7 @@ def stomp_euclidean_matrix_profile(
     k: int = 1,
     threshold: float = np.inf,
     inverse_distance: bool = False,
-    exclusion_size: int = None,
+    exclusion_size: Optional[int] = None,
 ):
     """
     Compute a euclidean euclidean matrix profile using STOMP [1]_.
@@ -109,7 +111,7 @@ def stomp_squared_matrix_profile(
     k: int = 1,
     threshold: float = np.inf,
     inverse_distance: bool = False,
-    exclusion_size: int = None,
+    exclusion_size: Optional[int] = None,
 ):
     """
     Compute a squared euclidean matrix profile using STOMP [1]_.
@@ -198,7 +200,7 @@ def stomp_normalized_euclidean_matrix_profile(
     k: int = 1,
     threshold: float = np.inf,
     inverse_distance: bool = False,
-    exclusion_size: int = None,
+    exclusion_size: Optional[int] = None,
 ):
     """
     Compute a euclidean matrix profile using STOMP [1]_.
@@ -296,7 +298,7 @@ def stomp_normalized_squared_matrix_profile(
     k: int = 1,
     threshold: float = np.inf,
     inverse_distance: bool = False,
-    exclusion_size: int = None,
+    exclusion_size: Optional[int] = None,
 ):
     """
     Compute a squared euclidean matrix profile using STOMP [1]_.