v0.9.0¶

May 2024

Highlights¶

New anomaly detection base class and MERLIN anomaly detector
Implementation for the RSAST classifier
Fixes for pipelines and new clustering and collection transformer pipeline classes
Improved model loading ability for deep learning modules
Large number of deprecations as we port transformations over to a new interface

Anomaly Detection¶

Enhancements¶

[ENH] Anomaly detection base class and MERLIN detector (#1343) @MatthewMiddlehurst

Benchmarking¶

Enhancements¶

[ENH] Results lists and loading (#1549) @TonyBagnall

Classification¶

Bug Fixes¶

[BUG] Set dynamic tag after __init__ for classifiers and regressors. (#1483) @MatthewMiddlehurst

Deprecation¶

[DEP] Remove shapeDTW (#1470) @TonyBagnall

Documentation¶

[DOC] Fixing random_state in docstrings (#1417) @AnonymousCodes911
[DOC] Add reference to paper in FreshPRINCEClassifier (#1515) @jasonmokk
[DOC] Update the ElasticEnsemble documentation (#1544) @itsdivya1309
[DOC] Typos in shapelet notebook (#1572) @TonyBagnall

Enhancements¶

[ENH] Type hints for primitives and string arguments:Part1 (#1472) @nileenagp
[ENH] Add control over the score method of classification and regression (#1443) @aadya940
[ENH] Type hints for primitives and string arguments:Part2 (#1487) @nileenagp
[ENH] Included RSAST as a alternative to SAST (2.0) (#1383) @nirojasva
[ENH] Loading ability in deep learning modules for classification/regression/clustering (#1374) @hadifawaz1999
[ENH] Type hints for primitives and string arguments:Part3 (#1525) @nileenagp

Maintenance¶

[MNT] Update ElasticEnsemble test to work with all distances (#1532) @itsdivya1309

Clustering¶

Bug Fixes¶

[ENH,BUG] Fixes for pipeline utilities and additional pipeline classes for collection based estimators (#1533) @MatthewMiddlehurst
[ENH,BUG] added weighted elastic barycenter and fix cache issue in subgradient method (#1563) @maxwell1503

Enhancements¶

[ENH,BUG] Fixes for pipeline utilities and additional pipeline classes for collection based estimators (#1533) @MatthewMiddlehurst
[ENH,BUG] added weighted elastic barycenter and fix cache issue in subgradient method (#1563) @maxwell1503
[ENH] Loading ability in deep learning modules for classification/regression/clustering (#1374) @hadifawaz1999

Maintenance¶

[MNT] Add tests for BA weight parameters (#1570) @chrisholder

Datasets¶

Enhancements¶

[ENH] Load tsc problems from zenodo as a back up to tsc.com (#1531) @TonyBagnall
[ENH] Results lists and loading (#1549) @TonyBagnall

Distances¶

Bug Fixes¶

[BUG] mpdist window length (#1513) @TonyBagnall

Enhancements¶

[ENH] Update MPDist function (#1415) @itsdivya1309

Regression¶

Bug Fixes¶

[BUG] Set dynamic tag after __init__ for classifiers and regressors. (#1483) @MatthewMiddlehurst

Deprecation¶

[DEP] deprecation V9 actions (#1464) @TonyBagnall

Documentation¶

[MNT] Add note in project README (#1391) @aadya940

Enhancements¶

[ENH] Add control over the score method of classification and regression (#1443) @aadya940
[ENH] Loading ability in deep learning modules for classification/regression/clustering (#1374) @hadifawaz1999

Transformations¶

Bug Fixes¶

[ENH,BUG] Fixes for pipeline utilities and additional pipeline classes for collection based estimators (#1533) @MatthewMiddlehurst

Deprecation¶

[DEP] Move ElbowClass channel selection algorithms (#1466) @TonyBagnall
[DEP] Move BoxCox and Log transformer to series transformers (#1493) @TonyBagnall
[DEP] Deprecate the ACF and PACF BaseTransformer wrappers (#1473) @TonyBagnall
[DEP] Deprecate simple wrapper transformers (#1494) @TonyBagnall
[DEP] Deprecate Differencer transformer (#1535) @TonyBagnall
[DEP] Convert DOBIN to a BaseSeriesTransformer (#1539) @TonyBagnall
[DEP] convert Scaled Logit (#1552) @TonyBagnall
[ENH] Convert Theta transform to series transformer (#1550) @TonyBagnall
[DEP] Move and deprecate pca (#1558) @TonyBagnall
[DEP] Deprecate TimeSince (#1559) @TonyBagnall
[DEP] Move and deprecate BKFilter (#1560) @TonyBagnall
[DEP] Deprecate kalman filter (#1561) @TonyBagnall

Enhancements¶

[ENH] BaseSeriesEstimator and BaseSeriesTransformer (#1490) @TonyBagnall
[ENH] Changes to series base classes from #1490 (#1506) @MatthewMiddlehurst
[ENH] Channel selection (#1486) @TonyBagnall
[ENH,BUG] Fixes for pipeline utilities and additional pipeline classes for collection based estimators (#1533) @MatthewMiddlehurst
[ENH] Included RSAST as a alternative to SAST (2.0) (#1383) @nirojasva
[ENH] Convert Theta transform to series transformer (#1550) @TonyBagnall

Unit Testing¶

Maintenance¶

[MNT] Changes to no soft dependency tests (#1476) @MatthewMiddlehurst

Other¶

Bug Fixes¶

[BUG] Dependencies version checking not handling python release candidate correctly (#1450) @baraline

Deprecation¶

[DEP] Remove annotation module (#1441) @TonyBagnall
[DEP] remove deprecation for datatypes adapter functions (#1489) @TonyBagnall
[DEP] deprecate SKLearnRegression pipeline (#1488) @TonyBagnall
[DEP] Removal of various annotation mentions and functions (#1505) @MatthewMiddlehurst
[DEP] Deprecate exceptions.py (#1517) @MatthewMiddlehurst
[ENH] Remove duplicate check estimator (#1545) @TonyBagnall

Documentation¶

[DOC] Change about text, add a link from docs to repo (#1485) @TonyBagnall
[DOC] add distance and channel selection projects (#1468) @TonyBagnall
[DOC] Update classification API docs (#1510) @itsdivya1309

Enhancements¶

[ENH] Exclude notebooks from PR testing that load from tsc.com (#1501) @TonyBagnall
[ENH] Remove duplicate check estimator (#1545) @TonyBagnall
[ENH] Lower bound scipy to 1.9 (#1546) @TonyBagnall

Maintenance¶

[MNT] Adds an Issue/PR Assigning Bot (#1459) @AnonymousCodes911
[MNT] Fix issue assign action and general maintenance (#1509) @MatthewMiddlehurst
[MNT] Update CODEOWNERS (#1507) @MatthewMiddlehurst
[MNT] Fix BaseSeriesEstimator test (#1512) @MatthewMiddlehurst

Contributors¶

The following have contributed to this release through a collective 55 GitHub Pull Requests:

@aadya940, @AnonymousCodes911, @baraline, @chrisholder, @CodeLionX, @hadifawaz1999, @itsdivya1309, @jasonmokk, @MatthewMiddlehurst, @maxwell1503, @nileenagp, @nirojasva, @TonyBagnall