Merge branch 'main' into aedcnn-clusterer

.all-contributorsrc

@@ -2357,10 +2357,10 @@
       ]
     },
     {
-      "login": "codelionx",
+      "login": "SebastianSchmidl",
       "name": "Sebastian Schmidl",
       "avatar_url": "https://avatars.githubusercontent.com/u/10573700?v=4",
-      "profile": "https://github.com/codelionx",
+      "profile": "https://github.com/SebastianSchmidl",
       "contributions": [
         "bug",
         "code",

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -80,7 +80,7 @@ body:
       Please run the following code snippet and paste the output here.
 
       ```python
-      from aeon.utils import show_versions; show_versions()
+      from aeon import show_versions; show_versions()
       ```
     placeholder: |
       <details>

CODEOWNERS

@@ -2,7 +2,7 @@
 # Only users with write access to the repository will be automatically added as
 # reviewers and contacted.
 
-aeon/anomaly_detection/ @CodeLionX @MatthewMiddlehurst
+aeon/anomaly_detection/ @SebastianSchmidl @MatthewMiddlehurst
 
 aeon/benchmarking/ @TonyBagnall @MatthewMiddlehurst @hadifawaz1999 @dguijo
 
@@ -17,7 +17,7 @@ aeon/distances/ @chrisholder @TonyBagnall
 
 aeon/networks/ @hadifawaz1999
 
-aeon/performance_metrics/anomaly_detection/ @codelionx @MatthewMiddlehurst
+aeon/performance_metrics/anomaly_detection/ @SebastianSchmidl @MatthewMiddlehurst
 
 aeon/regression/ @MatthewMiddlehurst @TonyBagnall @dguijo
 aeon/regression/deep_learning @hadifawaz1999 @MatthewMiddlehurst @TonyBagnall @dguijo

CONTRIBUTORS.md

@@ -247,7 +247,7 @@ Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/d
       <td align="center" valign="top" width="12.5%"><a href="https://github.com/dasgupsa"><img src="https://avatars2.githubusercontent.com/u/10398956?v=4?s=80" width="80px;" alt="Saurabh Dasgupta"/><br /><sub><b>Saurabh Dasgupta</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/commits?author=dasgupsa" title="Code">💻</a></td>
       <td align="center" valign="top" width="12.5%"><a href="https://github.com/SebasKoel"><img src="https://avatars3.githubusercontent.com/u/66252156?v=4?s=80" width="80px;" alt="Sebastiaan Koel"/><br /><sub><b>Sebastiaan Koel</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/commits?author=SebasKoel" title="Code">💻</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=SebasKoel" title="Documentation">📖</a></td>
       <td align="center" valign="top" width="12.5%"><a href="https://github.com/shagn"><img src="https://avatars.githubusercontent.com/u/16029092?v=4?s=80" width="80px;" alt="Sebastian Hagn"/><br /><sub><b>Sebastian Hagn</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/commits?author=shagn" title="Documentation">📖</a></td>
-      <td align="center" valign="top" width="12.5%"><a href="https://github.com/codelionx"><img src="https://avatars.githubusercontent.com/u/10573700?v=4?s=80" width="80px;" alt="Sebastian Schmidl"/><br /><sub><b>Sebastian Schmidl</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/issues?q=author%3Acodelionx" title="Bug reports">🐛</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=codelionx" title="Code">💻</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=codelionx" title="Documentation">📖</a> <a href="#research-codelionx" title="Research">🔬</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=codelionx" title="Tests">⚠️</a> <a href="https://github.com/aeon-toolkit/aeon/pulls?q=is%3Apr+reviewed-by%3Acodelionx" title="Reviewed Pull Requests">👀</a> <a href="#data-codelionx" title="Data">🔣</a></td>
+      <td align="center" valign="top" width="12.5%"><a href="https://github.com/SebastianSchmidl"><img src="https://avatars.githubusercontent.com/u/10573700?v=4?s=80" width="80px;" alt="Sebastian Schmidl"/><br /><sub><b>Sebastian Schmidl</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/issues?q=author%3ASebastianSchmidl" title="Bug reports">🐛</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=SebastianSchmidl" title="Code">💻</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=SebastianSchmidl" title="Documentation">📖</a> <a href="#research-SebastianSchmidl" title="Research">🔬</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=SebastianSchmidl" title="Tests">⚠️</a> <a href="https://github.com/aeon-toolkit/aeon/pulls?q=is%3Apr+reviewed-by%3ASebastianSchmidl" title="Reviewed Pull Requests">👀</a> <a href="#data-SebastianSchmidl" title="Data">🔣</a></td>
       <td align="center" valign="top" width="12.5%"><a href="https://github.com/Sharathchenna"><img src="https://avatars.githubusercontent.com/u/50939688?v=4?s=80" width="80px;" alt="Sharathchenna"/><br /><sub><b>Sharathchenna</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/commits?author=Sharathchenna" title="Code">💻</a></td>
       <td align="center" valign="top" width="12.5%"><a href="https://github.com/AurumnPegasus"><img src="https://avatars.githubusercontent.com/u/54315149?v=4?s=80" width="80px;" alt="Shivansh Subramanian"/><br /><sub><b>Shivansh Subramanian</b></sub></a><br /><a href="https://github.com/aeon-toolkit/aeon/commits?author=AurumnPegasus" title="Documentation">📖</a> <a href="https://github.com/aeon-toolkit/aeon/commits?author=AurumnPegasus" title="Code">💻</a></td>
       <td align="center" valign="top" width="12.5%"><a href="https://www.linkedin.com/in/solomon-botchway-a1383821b/"><img src="https://avatars.githubusercontent.com/u/62394255?v=4?s=80" width="80px;" alt="Solomon Botchway"/><br /><sub><b>Solomon Botchway</b></sub></a><br /><a href="#maintenance-snnbotchway" title="Maintenance">🚧</a></td>

aeon/anomaly_detection/_dwt_mlead.py

@@ -1,6 +1,6 @@
 """DWT-MLEAD anomaly detector."""
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 __all__ = ["DWT_MLEAD"]
 
 import warnings

aeon/anomaly_detection/_kmeans.py

@@ -1,6 +1,6 @@
 """k-Means anomaly detector."""
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 __all__ = ["KMeansAD"]
 
 from typing import Optional

aeon/anomaly_detection/_pyodadapter.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 __all__ = ["PyODAdapter"]
 
 from typing import TYPE_CHECKING, Any

aeon/anomaly_detection/_stomp.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 __all__ = ["STOMP"]
 
 import numpy as np

aeon/anomaly_detection/tests/test_base.py

@@ -7,18 +7,20 @@
 import pytest
 from numpy.testing import assert_almost_equal
 
-from aeon.testing.data_generation._legacy import make_series
+from aeon.testing.data_generation import (
+    make_example_1d_numpy,
+    make_example_2d_numpy_series,
+)
 from aeon.testing.mock_estimators._mock_anomaly_detectors import (
     MockAnomalyDetector,
     MockAnomalyDetectorRequiresFit,
     MockAnomalyDetectorRequiresY,
 )
 
-test_series = make_series(n_timepoints=10, return_numpy=True)
-test_series_2d = make_series(n_timepoints=10, n_columns=2, return_numpy=True).T
-test_series_pd = make_series(n_timepoints=10)
-test_series_pd_2d = make_series(n_timepoints=10, n_columns=2).T
-
+test_series = make_example_1d_numpy(n_timepoints=10)
+test_series_2d = make_example_2d_numpy_series(n_timepoints=10, n_channels=2)
+test_series_pd = pd.Series(test_series)
+test_series_pd_2d = pd.DataFrame(test_series_2d)
 test_y = np.array([0, 0, 0, 1, 1, 0, 0, 0, 1, 1])
 
 

aeon/anomaly_detection/tests/test_dwt_mlead.py

@@ -1,25 +1,46 @@
 """Tests for the DWT_MLEAD class."""
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 
 import numpy as np
 import pytest
+from sklearn.utils import check_random_state
 
 from aeon.anomaly_detection import DWT_MLEAD
-from aeon.testing.data_generation._legacy import make_series
 
 
 def test_dwt_mlead_output():
     """Test DWT_MLEAD output."""
-    series = make_series(n_timepoints=100, return_numpy=True, random_state=42)
+    rng = check_random_state(seed=42)
+    series = rng.normal(size=(100,))
     series[50:58] -= 5
 
     ad = DWT_MLEAD(start_level=2)
     pred = ad.predict(series)
 
     assert pred.shape == (100,)
     assert pred.dtype == np.float64
-    assert 50 <= np.argmax(pred) <= 58
+    assert 45 <= np.argmax(pred) <= 60
+
+
+def test_dwt_mlead_incorrect_input():
+    """Test DWT_MLEAD with incorrect input."""
+    rng = check_random_state(seed=42)
+    series = rng.normal(size=(100,))
+    with pytest.raises(ValueError, match="start_level must be >= 0"):
+        ad = DWT_MLEAD(start_level=-1)
+        ad.predict(series)
+    with pytest.raises(
+        ValueError, match="quantile_boundary_type must be 'percentile' or 'monte-carlo'"
+    ):
+        ad = DWT_MLEAD(quantile_boundary_type="Arsenal")
+        ad.predict(series)
+    with pytest.raises(ValueError, match="epsilon must be in"):
+        ad = DWT_MLEAD(quantile_epsilon=-1.0)
+        ad.predict(series)
+    with pytest.raises(ValueError):
+        ad = DWT_MLEAD(start_level=100)
+        ad.predict(series)
 
 
 def test_dwt_mlead_monte_carlo_unimplemented():

aeon/anomaly_detection/tests/test_iforest.py

@@ -2,9 +2,9 @@
 
 import numpy as np
 import pytest
+from sklearn.utils import check_random_state
 
 from aeon.anomaly_detection import IsolationForest
-from aeon.testing.data_generation._legacy import make_series
 from aeon.utils.validation._dependencies import _check_soft_dependencies
 
 
@@ -14,7 +14,8 @@
 )
 def test_iforest_default():
     """Test IsolationForest."""
-    series = make_series(n_timepoints=80, return_numpy=True, random_state=0)
+    rng = check_random_state(0)
+    series = rng.normal(size=(80,))
     series[50:58] -= 2
 
     iforest = IsolationForest(window_size=10, stride=1, random_state=0)
@@ -31,9 +32,8 @@ def test_iforest_default():
 )
 def test_iforest_multivariate():
     """Test IsolationForest multivariate."""
-    series = make_series(
-        n_timepoints=80, n_columns=2, return_numpy=True, random_state=0
-    )
+    rng = check_random_state(0)
+    series = rng.normal(size=(80, 2))
     series[50:58, 0] -= 2
 
     iforest = IsolationForest(window_size=10, stride=1, random_state=0)
@@ -50,7 +50,8 @@ def test_iforest_multivariate():
 )
 def test_iforest_no_window_univariate():
     """Test IsolationForest without windows univariate."""
-    series = make_series(n_timepoints=80, return_numpy=True, random_state=0)
+    rng = check_random_state(0)
+    series = rng.normal(size=(80,))
     series[50:58] -= 2
     iforest = IsolationForest(window_size=1, stride=1, random_state=0)
     pred = iforest.fit_predict(series, axis=0)
@@ -66,7 +67,9 @@ def test_iforest_no_window_univariate():
 )
 def test_iforest_stride():
     """Test IsolationForest with stride."""
-    series = make_series(n_timepoints=80, return_numpy=True, random_state=0)
+    rng = check_random_state(0)
+    series = rng.normal(size=(80,))
+
     series[50:58] -= 2
 
     iforest = IsolationForest(window_size=10, stride=2, random_state=0)
@@ -83,9 +86,8 @@ def test_iforest_stride():
 )
 def test_iforest_multivariate_stride():
     """Test IsolationForest multivariate with stride."""
-    series = make_series(
-        n_timepoints=80, n_columns=2, return_numpy=True, random_state=0
-    )
+    rng = check_random_state(0)
+    series = rng.normal(size=(80, 2))
     series[50:58, 0] -= 2
 
     iforest = IsolationForest(window_size=10, stride=2, random_state=0)
@@ -102,9 +104,10 @@ def test_iforest_multivariate_stride():
 )
 def test_iforest_semi_supervised_univariate():
     """Test IsolationForest semi-supervised univariate."""
-    series = make_series(n_timepoints=80, return_numpy=True, random_state=0)
+    rng = check_random_state(0)
+    series = rng.normal(size=(80,))
     series[50:58] -= 2
-    train_series = make_series(n_timepoints=100, return_numpy=True, random_state=0)
+    train_series = rng.normal(size=(80,))
 
     iforest = IsolationForest(window_size=10, stride=1, random_state=0)
     iforest.fit(train_series, axis=0)
@@ -121,13 +124,10 @@ def test_iforest_semi_supervised_univariate():
 )
 def test_iforest_semi_supervised_multivariate():
     """Test IsolationForest semi-supervised multivariate."""
-    series = make_series(
-        n_timepoints=80, n_columns=2, return_numpy=True, random_state=0
-    )
+    rng = check_random_state(0)
+    series = rng.normal(size=(80, 2))
     series[50:58, 0] -= 2
-    train_series = make_series(
-        n_timepoints=100, n_columns=2, return_numpy=True, random_state=0
-    )
+    train_series = rng.normal(size=(80, 2))
 
     iforest = IsolationForest(window_size=10, stride=1, random_state=0)
     iforest.fit(train_series, axis=0)

aeon/anomaly_detection/tests/test_kmeans.py

@@ -1,16 +1,18 @@
 """Tests for the KMeansAD class."""
 
-__maintainer__ = ["CodeLionX"]
+__maintainer__ = ["SebastianSchmidl"]
 
 import numpy as np
+import pytest
+from sklearn.utils import check_random_state
 
 from aeon.anomaly_detection import KMeansAD
-from aeon.testing.data_generation._legacy import make_series
 
 
 def test_kmeansad_univariate():
     """Test KMeansAD univariate output."""
-    series = make_series(n_timepoints=100, return_numpy=True, random_state=42)
+    rng = check_random_state(seed=2)
+    series = rng.normal(size=(100,))
     series[50:58] -= 5
 
     ad = KMeansAD(n_clusters=2, window_size=10)
@@ -23,9 +25,8 @@ def test_kmeansad_univariate():
 
 def test_kmeansad_multivariate():
     """Test KMeansAD multivariate output."""
-    series = make_series(
-        n_timepoints=100, n_columns=3, return_numpy=True, random_state=42
-    )
+    rng = check_random_state(seed=2)
+    series = rng.normal(size=(100, 3))
     series[50:58, 0] -= 5
     series[87:90, 1] += 0.1
 
@@ -35,3 +36,19 @@ def test_kmeansad_multivariate():
     assert pred.shape == (100,)
     assert pred.dtype == np.float64
     assert 50 <= np.argmax(pred) <= 58
+
+
+def test_kmeansad_incorrect_input():
+    """Test KMeansAD univariate output."""
+    rng = check_random_state(seed=2)
+    series = rng.normal(size=(100,))
+
+    with pytest.raises(ValueError, match="The window size must be at least 1"):
+        ad = KMeansAD(window_size=0)
+        ad.fit_predict(series)
+    with pytest.raises(ValueError, match="The stride must be at least 1"):
+        ad = KMeansAD(stride=0)
+        ad.fit_predict(series)
+    with pytest.raises(ValueError, match="The number of clusters must be at least 1"):
+        ad = KMeansAD(n_clusters=0)
+        ad.fit_predict(series)