v1.3.0

September 2025

Highlights

• A massive update to the forecasting module and its framework. New forecasters include ARIMA/AutoARIMA, SETAR-Tree/Forest, Theta, TAR/SETAR, and deep learners. Thank you to all contributors including our GSoC 2024 students!

• The shapelet transform and extending algorithms now accept unequal length series and have had a massive speed-up

• KNN and KMeans algorithms now have n_jobs support for multithreading

• A download function for the RehabPile dataset collection has been implemented

• An implementation of ESMOTE for imbalanced classification problems has been added to the collection transformers module

• The Time Mixup Contrastive Learning (TimeMCL) algorithm has been added to the self supervised module housed in the transformations package

Anomaly Detection

Documentation

• [DOC] Fixing typos in docstrings/comments (#3036) @OutragedOctopus

Benchmarking

Deprecation

• [DEP] Remove deprecated stats (#2963) @MatthewMiddlehurst

Classification

Bug Fixes

• [BUG] Fix redcomets bug when using only one sample (#2952) @baraline

Documentation

• [DOC] Classification notebook links (#2942) @TonyBagnall

Enhancements

• [ENH] Improvements to ST transformer and classifier (#2968) @MatthewMiddlehurst

• [ENH] KNN n_jobs and updated kneighbours method (#2578) @chrisholder

• [ENH] Refactor signature code (#2943) @TonyBagnall

• [ENH] Change seed to random_state (#3031) @TonyBagnall

Clustering

Documentation

• [DOC] Added missing references in deep learning clustering #3021 (#3029) @PipaFlores

Enhancements

• [ENH] Averaging module refactor (#3053) @chrisholder

• [ENH] K-means n-job support (#3058) @chrisholder

Datasets

Deprecation

• [DEP] Deprecate PLAID dataset (#2928) @TonyBagnall

• [ENH,DEP] Update .ts writer and deprecate old writers (#3042) @MatthewMiddlehurst

Enhancements

• [ENH,DEP] Update .ts writer and deprecate old writers (#3042) @MatthewMiddlehurst

Maintenance

• [MNT] Adding rehab-pile loaders in init (#3035) @hadifawaz1999

Other

• [ENH] Add RehabPile dataset collection (#2993) @sarthakuwar

Distances

Bug Fixes

• [BUG] MSM cost matrix initialisation bug fix. (#3010) @chrisholder

Enhancements

• [ENH] Distance module n_jobs support (#2545) @chrisholder

• [ENH] KNN n_jobs and updated kneighbours method (#2578) @chrisholder

Forecasting

Bug Fixes

• [BUG] Fix Nelder-Mead (#2992) @TonyBagnall

• [BUG] Fix incorrect undifferencing in ARIMA (#2987) @alexbanwell1

• [BUG] Add mixin for setar (#3009) @TonyBagnall

• [BUG] Change forecast() method in BaseForecaster (#3043) @TonyBagnall

Documentation

• [DOC] Base forecaster docstring (#2998) @TonyBagnall

• [DOC] Update api references for Deep Forecasters (#3060) @hadifawaz1999

Enhancements

• [ENH] Refactor to set forecast_ in fit and to require y in predict (#2927) @TonyBagnall

• [ENH] Clone estimator in direct forecast (#2936) @TonyBagnall

• [ENH] Fix typo in forecaster docstring (#2944) @TonyBagnall

• [ENH] Rework ETS Forecaster (#2939) @TonyBagnall

• [ENH] Add Basic ARIMA model (#2860) @alexbanwell1

• [ENH] Forecasting refactor (#2962) @TonyBagnall

• [ENH] forecasting coverage (#2948) @TonyBagnall

• [ENH] refactor and tidy (#2988) @TonyBagnall

• [ENH] Add axis to base forecaster (#3000) @TonyBagnall

• [ENH] Forecasing mixins and base class checking (#3006) @MatthewMiddlehurst

• [ENH] basic setar-tree module and tests (#2890) @TinaJin0228

• [ENH] Convert ETS to automatically optimise for the alpha, beta, gamma and phi parameters (#3011) @alexbanwell1

• [ENH] Theta forecaster (#2978) @TonyBagnall

• [ENH] Coverage for setar forecasting (#3022) @TonyBagnall

• [ENH] TAR forecaster (#3012) @TonyBagnall

• [ENH] Added BaseDeepForecaster in forecasting/deep_learning (#2905) @lucifer4073

• [ENH] Added TCN forecaster in aeon/forecasting/deep_learning (#2938) @lucifer4073

• [ENH] Add AutoARIMA algorithm in (#2861) @alexbanwell1

Other

• [EHN] Allow exogenous variables in regression forecasters (#2915) @TinaJin0228

• [ENH] Implement Time-Varying Parameter (TVP) regression forecaster using Kalman filter (#2920) @TonyBagnall

Networks

Bug Fixes

• [BUG] Issue with axis in TCNNetwork (#3007) @hadifawaz1999

Documentation

• [DOC] Update api references for Deep Forecasters (#3060) @hadifawaz1999

Enhancements

• [ENH] Added RecurrentRegressor for time series regression (#2894) @lucifer4073

• [ENH] Added TemporalConvolutionalNetwork in aeon/networks (#2933) @lucifer4073

• [ENH] Avoid error in calculating kernel sizes in InceptionNetwork (#3034) @hadifawaz1999

• [ENH] Added DeepARNetwork to aeon/networks (#2956) @lucifer4073

• [ENH] Added TCN forecaster in aeon/forecasting/deep_learning (#2938) @lucifer4073

Regression

Enhancements

• [ENH] Added RecurrentRegressor for time series regression (#2894) @lucifer4073

Documentation

• [DOC] Add RecurrentRegressor and RecurrentNetwork to api_reference (#2935) @lucifer4073

Segmentation

Bug Fixes

• [BUG] stop overwriting y in BaseSegmenter (#2969) @TonyBagnall

Similarity Search

Enhancements

• [ENH] Use np.argpartition for efficient top-k selection instead of np.argsort (#2805) @Kaustbh

Transformations

Bug Fixes

• [BUG] Improve signature docs (#2929) @TonyBagnall

• [BUG] Fix redcomets bug when using only one sample (#2952) @baraline

• [BUG] Fix incorrect undifferencing in ARIMA (#2987) @alexbanwell1

• [BUG] Fix multirocket bug variables not updated in loop (#3003) @angus924

• [BUG] Fix RDST admissible sampling point method (#2763) @baraline

• [BUG] patch esig function and remove test exclusion for signatures (#3030) @MatthewMiddlehurst

Deprecation

• [DEP] Remove deprecated transforms (#2961) @MatthewMiddlehurst

• [DEP] Remove update function from base transformer and small tidying (#3047) @MatthewMiddlehurst

Documentation

• [DOC] fix typos in MatrixProfileSeriesTransformer (#2940) (#2941) @CodeFor2001

• [DOC] Add Imbalance transformers to API (#2973) @TonyBagnall

• [DOC] Simple imbalance notebook (#2976) @TonyBagnall

• [DOC] Update self supervised api references (#3061) @hadifawaz1999

Enhancements

• [ENH] New experimental module: imbalance in collection transformers (#2498) @TonyBagnall

• [ENH] Improvements to ST transformer and classifier (#2968) @MatthewMiddlehurst

• [ENH] Deprecate MatrixProfile collection transformer and MPDist (#3002) @TonyBagnall

• [ENH] Refactor signature code (#2943) @TonyBagnall

• [ENH] Implement of ESMOTE for imbalanced classification problems (#2971) @LinGinQiu

• [ENH] Change seed to random_state (#3031) @TonyBagnall

• [ENH] Adding Time Mixingup Contrastive Learning to Self Supervised module (#3015) @hadifawaz1999

Maintenance

• [MNT] add n_jobs parameter for sax inversion (#2953) @baraline

Unit Testing

Bug Fixes

• [BUG] patch esig function and remove test exclusion for signatures (#3030) @MatthewMiddlehurst

Enhancements

• [ENH] Forecasing mixins and base class checking (#3006) @MatthewMiddlehurst

• [ENH] Update testing data dictionary with series testing data (#2975) @MatthewMiddlehurst

• [ENH] Change seed to random_state (#3031) @TonyBagnall

Maintenance

• [MNT] Comments on numba caching and release CI fix (#2916) @MatthewMiddlehurst

• [MNT] Add testing workflow for multithreading and make threaded estimators use check_n_jobs (#2913) @MatthewMiddlehurst

• [MNT] Update expected results generation (#3017) @MatthewMiddlehurst

Visualisations

Documentation

• [DOC] Fixing typos in docstrings/comments (#3036) @OutragedOctopus

• [DOC] Improve visualization module documentation (#2602) @Kaustbh

Enhancements

• [ENH] Improves create_multi_comparison_matrix parameters and saving (#2657) @aryanpola

Other

Deprecation

• [ENH] Series data validation improvements and data validation deprecations (#2979) @MatthewMiddlehurst

Documentation

• [DOC,MNT] Documentation configuration clean-up (#2911) @MatthewMiddlehurst

• [DOC] fix incorrect references of classes in getting started page (#2762) @inclinedadarsh

• [DOC] improve base classes example (#2966) @baraline

• [DOC] Update experimental modules and social media links (#3014) @MatthewMiddlehurst

• [DOC] Time series machine learning (#3025) @TonyBagnall

Enhancements

• [ENH] Series data validation improvements and data validation deprecations (#2979) @MatthewMiddlehurst

Maintenance

• [MNT] Prevents assignment on Restricted Labels (#2706) @kavya-r30

• [MNT] issue-assign-bot - only users with write access should be able to assign others (#2739) @shinymack

• [MNT] Created workflow for closing “AI Spam” pull requests (#2750) @Ramana-Raja

• [DOC,MNT] Documentation configuration clean-up (#2911) @MatthewMiddlehurst

• [MNT] Add codespell support (config, workflow to detect/not fix) and make it fix a “few” typos (#2653) @yarikoptic

• [MNT] Correct checkout for codespell (#2945) @MatthewMiddlehurst

• [MNT] Workflow maintenance (#2919) @MatthewMiddlehurst

• [MNT] Clean disk space for CI to try stop sporadic failure (#2965) @MatthewMiddlehurst

• [MNT] Workflow maintenance and dev table update (#2984) @MatthewMiddlehurst

• [MNT] Fixes the label buttons in the PR bot comment (#3024) @MatthewMiddlehurst

• [MNT] Update pyupgrade for Python 10+ (#3023) @MatthewMiddlehurst

• [MNT] Remove ruptures install to fix CI (#3040) @MatthewMiddlehurst

• [MNT] Fixes from missed merges (#3045) @MatthewMiddlehurst

• [MNT] Fix CI failures (#3059) @MatthewMiddlehurst

Contributors

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

@alexbanwell1, @angus924, @aryanpola, @baraline, @chrisholder, @CodeFor2001, @hadifawaz1999, @inclinedadarsh, @Kaustbh, @kavya-r30, @LinGinQiu, @lucifer4073, @MatthewMiddlehurst, @OutragedOctopus, @PipaFlores, @Ramana-Raja, @sarthakuwar, @shinymack, @TinaJin0228, @TonyBagnall, @yarikoptic