v0.11.1¶

September 2024

Reminder: This release will be the last 0.X minor release. Other than patches, the next release will be 1.0.0 which contains a large amount of framework changes.

Highlights¶

STOMP anomaly detector. (@wenig)
QUANT regressor (@MatthewMiddlehurst)
Series search for similarity search module (@baraline)
Piecewise Linear Approximation (PLA) collection transform (@Moonzyyy)

Anomaly Detection¶

Documentation¶

[ENH] Add performance metrics for anomaly detection (from TimeEval) (#1938) @SebastianSchmidl

Enhancements¶

[ENH] Add performance metrics for anomaly detection (from TimeEval) (#1938) @SebastianSchmidl
[ENH] Add STOMP anomaly detector. (#2031) @wenig

Benchmarking¶

Bug Fixes¶

[BUG] Typo in weasel2 alias (#2002) @TonyBagnall

Regression¶

Enhancements¶

[ENH] QUANT regressor (#1989) @MatthewMiddlehurst

Similarity Search¶

Enhancements¶

[ENH] Series search for similarity search module (#2010) @baraline

Transformations¶

Documentation¶

[DOC] fix BKFilter._transform parameters docs (#1969) @Cyril-Meyer

Enhancements¶

[ENH] Piecewise Linear Approximation (PLA) (#1807) @Moonzyyy

Visualisations¶

Bug Fixes¶

[BUG] Short term solution for shapelet feature importance (#2017) @IRKnyazev

Enhancements¶

[ENH] Adds option to use raw values instead of relative ones (#1998) @dguijo

Other¶

Maintenance¶

[MNT] Update pre-commit pyupgrade config (#1980) @MatthewMiddlehurst
[MNT] Remove keras bound (#1816) @MatthewMiddlehurst

Contributors¶

The following have contributed to this release through a collective 11 GitHub Pull Requests:

@baraline, @SebastianSchmidl, @Cyril-Meyer, @dguijo, @IRKnyazev, @MatthewMiddlehurst, @Moonzyyy, @TonyBagnall, @wenig

v0.11.0¶

August 2024

Reminder: This release will be the last 0.X minor release. Other than patches, the next release will be 1.0.0 which contains a large amount of framework changes.

Highlights¶

Added Proximity Forest classifier (@itsdivya1309)
Added a composable ensemble for classification and regression (@MatthewMiddlehurst)
New deep learning networks in AEBiGRUNetwork, AEDRNNNetwork and AEAttentionBiGRUNetwork (@aadya940)
Lots of bug and documentation fixes

Anomaly Detection¶

Enhancements¶

[ENH/DOC] Unsupervised and semi-supervised usage of PyODAdapter (#1932) @SebastianSchmidl

Classification¶

Bug Fixes¶

[BUG] Check data shape in predict-like functions (#1865) @MatthewMiddlehurst

Deprecation¶

[DEP] Move MRSQM to dictionary based (#1821) @TonyBagnall

Documentation¶

[DOC] Fix DL classification notebook (#1862) @MatthewMiddlehurst
[DOC] Update paper referenced in Arsenal docs (#1892) @Cyril-Meyer
[ENH] Remove BaseClassifier unnecessary declarations (#1887) @TonyBagnall
[DOC] Correct sast tags to shapelet (#1935) @IRKnyazev

Enhancements¶

[ENH] Change HC2 test to balance class (#1795) @TonyBagnall
[ENH] at kwargs to fit_predict to allow setting of the cv_size parameter in default (#1767) @TonyBagnall
[ENH,MNT,DOC] Tidying up deep learning modules clasification/regression (#1826) @hadifawaz1999
[ENH] Implement Proximity Forest classifier (#1729) @itsdivya1309
[ENH] Remove BaseClassifier unnecessary declarations (#1887) @TonyBagnall
[ENH] Composable ensemble for classification and regression (#1853) @MatthewMiddlehurst
[ENH] Adds the option of using callable distances for kNN regressor (#1965) @dguijo

Maintenance¶

[MNT] Unit testing revamp part 2: classification (#1770) @MatthewMiddlehurst

Clustering¶

Bug Fixes¶

[BUG] Kmeans pass through random state to averaging methods (#1850) @chrisholder
[BUG] Fixed CLARA issue (#1924) @chrisholder
[BUG] Check data shape in predict-like functions (#1865) @MatthewMiddlehurst

Documentation¶

[DOC][REF] K-Shape rename and add references (#1899) @Cyril-Meyer

Maintenance¶

[MNT] Make input clustering algorithm for deep leaning an estimator parameter instead of string (#1944) @hadifawaz1999
[MNT] Add metrics parameter for deep clustering module (#1945) (#1958) @ghost

Refactored¶

[DOC][REF] K-Shape rename and add references (#1899) @Cyril-Meyer

Datasets¶

Documentation¶

[DOC] Removed incomplete line from write_to_tsfile (#1808) @harshithasudhakar
[DOC] Resolved issue 1829 - corrected docstring param return_metadata to False (#1843) @IRKnyazev

Other¶

[ENH] Add support for handling file extensions in load_from_tsfile (#1953) @Sharathchenna

Distances¶

Bug Fixes¶

[BUG] Fixed incorrect bounding enforcement (#1871) @chrisholder
[BUG] Fix bug in shape dtw pairwise function (#1975) @hadifawaz1999

Documentation¶

[DOC] Fix 2 typos in distances notebook (#1873) @Datadote
[DOC, ENH] Updating distance’s notebooks + tests to ensure sklearn compatibility (#1838) @dguijo

Enhancements¶

[ENH] Plot the pairwise distance between two time series (#1977) @dguijo

Forecasting¶

Documentation¶

[DOC] Converging towards numpy doc standards V2 - issue #1540 (#1863) @IRKnyazev

Networks¶

Bug Fixes¶

[BUG] Fix config of AEBiGRUNetwork from encoder to auto-encoder (#1786) @hadifawaz1999

Deprecation¶

[DEP] Remove old class base deep network (#1817) @hadifawaz1999

Documentation¶

[DOC] Converging towards numpy doc standards V2 - issue #1540 (#1863) @IRKnyazev

Enhancements¶

[ENH] Add AEBiGRUNetwork (#1583) @aadya940
[ENH,MNT,DOC] Tidying up deep learning modules clasification/regression (#1826) @hadifawaz1999
[ENH] Improve deep learning networks test coverage for parameters which can be list (#1851) @Cyril-Meyer
[ENH] Improve deep learning networks test coverage (#1870) @Cyril-Meyer
[ENH] Add AEDRNNNetwork (#1577) @aadya940
[ENH] Implement AEAttentionBiGRUNetwork (#1608) @aadya940

Maintenance¶

[MNT] Remove tensorflow-addons dependency (#1869) @Cyril-Meyer
[MNT] Make input clustering algorithm for deep leaning an estimator parameter instead of string (#1944) @hadifawaz1999

Regression¶

Bug Fixes¶

[BUG] Check data shape in predict-like functions (#1865) @MatthewMiddlehurst

Enhancements¶

[ENH,MNT,DOC] Tidying up deep learning modules clasification/regression (#1826) @hadifawaz1999
[ENH] Composable ensemble for classification and regression (#1853) @MatthewMiddlehurst
[ENH] Adds the option of using callable distances for kNN regressor (#1965) @dguijo

Maintenance¶

[MNT] Unit testing revamp part 3: regression (#1875) @MatthewMiddlehurst

Segmentation¶

Documentation¶

[DOC] Update the docstring for BaseSegmenter (#1741) @TonyBagnall

Similarity Search¶

Documentation¶

[DOC] Update type hints for similarity search (#1939) @phershbe
[DOC] Update type hints for distance profiles (#1981) @phershbe

Transformations¶

Bug Fixes¶

[BUG] Add names to create steps_ in CollectionPipeline (#1796) @TonyBagnall
[BUG] Bug fixes for interval classes (#1988) @MatthewMiddlehurst

Deprecation¶

[DEP] Remove legacy PandasAdaptor, move TabularToSeriesAdaptor to legacy (#1789) @TonyBagnall
[DEP] Remove BKFilter BaseTransformer, redesign equivalent series transformer (#1788) @TonyBagnall
[DEP] Remove Split mixin (#1613) @TonyBagnall
[DEP] Remove redundant multivariate rockets (#1791) @TonyBagnall
[DEP] Remove DateTime and FourierFeatures (#1792) @TonyBagnall
[DEP] Move transformations.compose to legacy and isolate uses to forecasting. (#1790) @TonyBagnall
[DEP] Move summarizer to legacy (#1833) @TonyBagnall
[DEP] Remove Theta (#1832) @TonyBagnall
[DEP] Move HampelFilter to legacy (#1830) @TonyBagnall
[DEP] Move Lag to legacy (#1827) @TonyBagnall
[DEP] Remove _IndexSubset (#1831) @TonyBagnall
[DEP] Move exponent transformer to legacy (#1822) @TonyBagnall
[DEP] deprecated bootstrap to legacy (#1818) @TonyBagnall
[DEP] detrenders to legacy (#1819) @TonyBagnall
[DEP] Remove hierarchical transformers to legacy (#1820) @TonyBagnall
[DEP] Move impute to legacy (#1823) @TonyBagnall
[DEP] Remove FeatureSelection transformer (#1890) @TonyBagnall
[DEP] Remove FunctionTransformer (#1891) @TonyBagnall
[DEP] Remove update in BaseCollectionTransformer (#1894) @TonyBagnall

Documentation¶

[DOC] DFT instead of PAA in SFA docstring (#1849) @hadifawaz1999
[DOC] Correct sast tags to shapelet (#1935) @IRKnyazev
[DOC] Added Type Hints for _dilated_shapelet_transform (#1949) @aryanpola

Enhancements¶

[ENH] channel selector test (#1783) @TonyBagnall
[ENH] Fixes in Catch22 (#1809) @Moonzyyy
[ENH] Import RandomChannelSelector through init (#1933) @TonyBagnall
[ENH] Added useful attributes to extracted shapelets for RDST (#1959) @IRKnyazev
[ENH] Discrete Fourier Approximation Transformer (#1967) @Cyril-Meyer
[ENH] add type hints to _shapelet_transform.py init method (#1926) @Datadote

Other¶

Added type hints for rsast and sast files (#1994) @aryanpola

Unit Testing¶

Bug Fixes¶

[BUG] Fixed incorrect bounding enforcement (#1871) @chrisholder

Enhancements¶

[ENH] Add minimum number of class labels for numpy data generators (#1974) @MatthewMiddlehurst

Maintenance¶

[MNT] Add Regressor Results For Comparison (#1787) @harshithasudhakar
[MNT] Fix sporadic PF breakage and matplotlib version bound (#1906) @MatthewMiddlehurst
[MNT] Unit testing revamp part 2: classification (#1770) @MatthewMiddlehurst
[MNT] Unit testing revamp part 3: regression (#1875) @MatthewMiddlehurst

Visualisations¶

Bug Fixes¶

[BUG] Mapped input index to class index for shapelet viz (#1943) @IRKnyazev
[BUG] Corrected shapelet plot for RDST (#1971) @IRKnyazev

Documentation¶

[DOC] Update shapelet notebook (#1824) @baraline

Enhancements¶

[ENH] Plot the pairwise distance between two time series (#1977) @dguijo

Other¶

Bug Fixes¶

[BUG] get_n_channels check channels consistent (#1895) @Cyril-Meyer

Documentation¶

[DOC] Typo correction in getting started guide (#1842) @IRKnyazev
[DOC] Correct info for issue assign bot (#1860) @MatthewMiddlehurst
[Doc] Corrected mis-rendered double references. (#1883) @IRKnyazev
[DOC] Add ddtw distance to the API references (#1897) @Cyril-Meyer
[DOC] Change order of build and install instructions for clarity (#1915) @phershbe
[DOC] Grammar corrections to developer installation and installation (#1918) @phershbe
[DOC] Shapelet classifier notebook extensions & minor docstring corrections (#1930) @IRKnyazev

Enhancements¶

[ENH] Documentation fixes and function changes for BaseCollectionEstimator (#1864) @MatthewMiddlehurst

Maintenance¶

[MNT] Adding Scorecard.yml Workflow and Scorecard Badge GOSST (#1619) @PatriceJada
[MNT] mrsqm and pycatch22 bounds (#1753) @MatthewMiddlehurst
[MNT] Scorecard badge fix (#1804) @MatthewMiddlehurst
[MNT] Update scikit-learn version (#1803) @MatthewMiddlehurst
[MNT] Raise numba version bound (#1814) @MatthewMiddlehurst
[MNT] Try shifting to Ubuntu 22.04 (#1929) @hadifawaz1999

Contributors¶

The following have contributed to this release through a collective 91 GitHub Pull Requests:

@aadya940, @aryanpola, @baraline, @chrisholder, @SebastianSchmidl, @Cyril-Meyer, @Datadote, @dguijo, @ghost, @hadifawaz1999, @harshithasudhakar, @IRKnyazev, @itsdivya1309, @MatthewMiddlehurst, @Moonzyyy, @PatriceJada, @phershbe, @Sharathchenna, @TonyBagnall