Skip to content

Latest commit

 

History

History
972 lines (666 loc) · 41.1 KB

CHANGELOG.md

File metadata and controls

972 lines (666 loc) · 41.1 KB

Release Notes

v1.6.1

Patch release focusing on:

  • Bug fix to solve the persisting bug of passing weights_only in load_from_checkpoint for lightning <2.6.
  • Bug fix to non-writeable encoder issue caused by pandas copy-on-write behavior.

Fixes

  • [BUG] Torch doesn't support the conversion of non-writeable numpyFix non-writeable encoder issue and update tests (#1989) @cngmid
  • [BUG] Solve the persisting bug of passing weights_only in load_from_checkpoint (#2027) @phoeenniixx

Maintenance

  • [MNT] Add CI step with pinned dependencies as of Nov 2025 (#2029) @phoeenniixx

All Contributors

@cngmid, @phoeenniixx

v1.6.0

Release focusing on:

  • python 3.14 support
  • Solving the unpickling error in weight loading
  • Deduplicating utilities with scikit-base and adding it as a core dependency
  • Addition of new predict interface for Beta v2
  • Improvements to model backends

Highlights

pytorch-forecasting v1.6.0

  • Refactor N-BEATS blocks to separate KAN logic by @khenm in #2012
  • Efficient Attention Backend for TimeXer @anasashbin #1997

pytorch-forecasting Beta v2

  • New predict interface for v2 models by @phoeenniixx in #1984
  • Efficient Attention Backend for TimeXer @anasashbin #1997

API Changes

  • Tuner import change due to a Lightning breaking change. Lightning v2.6 introduced a breaking change in its checkpoint loading behavior, which caused unpickling errors during weight loading in pytorch-forecasting (see #2000). To address this, pytorch-forecasting now provides its own Tuner wrapper that exposes the required weights_only argument when calling lr_find().

    • When using pytorch-forecasting > 1.5.0 with lightning > 2.5, please use pytorch_forecasting.tuning.Tuner in place of lightning.pytorch.tuner.Tuner. See #2000 for details.

Maintenance

  • [MNT] Dependabot: Bump actions/upload-artifact from 4 to 5 (#1986) @dependabot[bot]
  • [MNT] Dependabot: Bump actions/download-artifact from 5 to 6 (#1985) @dependabot[bot]
  • [MNT] Fix typos (#1988) @szepeviktor
  • [MNT] Dependabot: Bump actions/checkout from 5 to 6 (#1991) @dependabot[bot]
  • [MNT] Add version bound for lightning (#2001) @phoeenniixx
  • [MNT] Dependabot: Bump actions/upload-artifact from 5 to 6 (#2005) @dependabot[bot]
  • [MNT] Dependabot: Bump actions/download-artifact from 6 to 7 (#2006) @dependabot[bot]
  • [MNT] Dependabot: Update sphinx requirement from <8.2.4,>3.2 to >3.2,<9.1.1 (#2013) @dependabot[bot]
  • [MNT] Dependabot: Update lightning requirement from <2.6.0,>=2.0.0 to >=2.0.0,<2.7.0 (#2002) @dependabot[bot]
  • [MNT] Add python 3.14 support (#2015) @phoeenniixx
  • [MNT] Update changelog generator script to return markdown files (#2016) @phoeenniixx
  • [MNT] deduplicating utilities with scikit-base (#1929) @fkiraly
  • [MNT] Update ruff linting target version to python 3.10 (#2017) @phoeenniixx

Enhancements

  • [ENH] Consistent 3D output for single-target point predictions in TimeXer v1. (#1936) @PranavBhatP
  • [ENH] Efficient Attention Backend for TimeXer (#1997) @anasashb
  • [ENH] Add predict to v2 models (#1984) @phoeenniixx
  • [ENH] Refactor N-BEATS blocks to separate KAN logic (#2012) @khenm
  • [MNT] deduplicating utilities with scikit-base (#1929) @fkiraly

Fixes

  • [BUG] Align TimeXer v2 endogenous/exogenous usage with tslib metadata (#2009) @ahmedkansulum
  • [BUG] Solve the unpickling error in weight Loading (#2000) @phoeenniixx

Documentation

  • [DOC] add CODE_OF_CONDUCT.md and GOVERNANCE.md (#2014) @phoeenniixx

All Contributors

@ahmedkansulum, @anasashb, @dependabot[bot], @fkiraly, @khenm, @phoeenniixx, @PranavBhatP, @szepeviktor, @agobbifbk

v1.5.0

Release focusing on:

  • python 3.9 end-of-life
  • changes to testing framework.
  • New estimators in pytorch-forecasting v1 and beta v2.

Highlights

pytorch-forecasting v1.5.0

  • Kolmogorov Arnold Block for NBeats by @Sohaib-Ahmed21 in sktime#1751
  • xLSTMTime implementation by @phoeenniixx in sktime#1709

pytorch-forecasting Beta v2

  • Implementing D2 data module, tests and TimeXer model from tslib for PTF v2 by @PranavBhatP in sktime#1836
  • Add DLinear model from tslib for PTF v2 by @PranavBhatP in sktime#1874
  • Add Samformer model for PTF v2 from DSIPTS by @PranavBhatP in sktime#1952
  • Tide model in PTF v2 interface from dsipts by @phoeenniixx in sktime#1889

Enhancements

  • [ENH] Test framework for ptf-v2 by @phoeenniixx in sktime#1841
  • [ENH] Implementing D2 data module, tests and TimeXer model from tslib for v2 by @PranavBhatP in sktime#1836
  • [ENH] DLinear model from tslib by @PranavBhatP in sktime#1874
  • [ENH] Enable DeprecationWarning , PendingDeprecationWarning and FutureWarning when running pytest by @fnhirwa in sktime#1912
  • [ENH] Suppress __array_wrap__ warning in numpy 2 for torch and pandas by @fnhirwa in sktime#1911
  • [ENH] Suppress PyTorch deprecation warning: UserWarning: nn.init.constant is now deprecated in favor of nn.init.constant_ by @fnhirwa in sktime#1915
  • [ENH] two-way linkage of model package classes and neural network classes by @fkiraly in sktime#1888
  • [ENH] Add a copy of BaseFixtureGenerator to pytorch-forecasting/tests/_base as a true base class by @PranavBhatP in sktime#1919
  • [ENH] Remove references to model from the BaseFixtureGenerator by @phoeenniixx in sktime#1923
  • [ENH] Improve test framework for v1 models by @phoeenniixx in sktime#1908
  • [ENH] xLSTMTime implementation by @phoeenniixx in sktime#1709
  • [ENH] Improve test framework for v1 metrics by @PranavBhatP in sktime#1907
  • [ENH] Tide model in v2 interface by @phoeenniixx in sktime#1889
  • [ENH] docstring test suite for functions by @fkiraly in sktime#1955
  • [ENH] Add missing test for forward output of TimeXer as proposed in #1936 by @PranavBhatP in sktime#1951
  • [ENH] Add Samformer model for PTF v2 from DSIPTS by @PranavBhatP in sktime#1952
  • [ENH] Kolmogorov Arnold Block for NBeats by @Sohaib-Ahmed21 in sktime#1751
  • [ENH] Standardize output format for tslib v2 models by @phoeenniixx in sktime#1965
  • [ENH] Add Metrics support to ptf-v2 by @phoeenniixx in sktime#1960
  • [ENH] check_estimator utility for checking new estimators against unified API contract by @fkiraly in sktime#1954
  • [ENH] Standardize testing of estimator outputs and skip tests for non-conformant estimators by @PranavBhatP in sktime#1971

Fixes

  • [BUG] Fix issue with EncodeNormalizer(method='standard', center=False) for scale value by @fnhirwa in sktime#1902
  • [BUG] fixed memory leak in TimeSeriesDataset by using @cached_property and clean-up of index construction by @Vishnu-Rangiah in sktime#1905
  • [BUG] Fix issue with plot_prediction_actual_by_variable unsupported operand type(s) for *: 'numpy.ndarray' and 'Tensor' by @fnhirwa in sktime#1903
  • [BUG] Correctly set lagged variables to known when lag >= horizon by @hubkrieb in sktime#1910
  • [BUG] Updated base_model.py to account for importing error by @Himanshu-Verma-ds in sktime#1488
  • [BUG][DOC] Fix documentation: pass loss argument to BaseModel in custom models tutorial example by @PranavBhatP in sktime#1931
  • [BUG] fix broken version inspection if package distribution has None name by @lohraspco in sktime#1926
  • [BUG] fix sporadic tkinter failures in CI by @fkiraly in sktime#1937
  • [BUG] Device inconsistency in MQF2DistributionLoss raising: RuntimeError: Expected all tensors to be on the same device by @fnhirwa in sktime#1916
  • [BUG] fixed memory leak in BaseModel by detach some tensor by @zju-ys in sktime#1924
  • [BUG] Fix TimeSeriesDataSet wrong inferred tensor dtype when time_idx is included in features by @cngmid in sktime#1950
  • [BUG] standardize output format of xLSTMTime estimator for point predictions by @sanskarmodi8 in sktime#1978
  • [BUG] Standardize output format of NBeats and NBeatsKAN estimators by @sanskarmodi8 in sktime#1977

Documentation

  • [DOC] Correct documentation for N-BEATS by @Pinaka07 in sktime#1914
  • [DOC] 1.1.0 changelog - missing entries by @jdb78 in sktime#1512
  • [DOC] fix minor typo in changelog by @fkiraly in sktime#1917
  • [DOC] Missing parenthesis in docstring of MASE by @caph1993 in sktime#1944

Maintenance

  • [MNT] remove import conditionals for python 3.6 by @fkiraly in sktime#1928
  • [MNT] Dependabot: bump actions/download-artifact from 4 to 5 by @dependabot[bot] in sktime#1939
  • [MNT] Dependabot: Bump actions/checkout from 4 to 5 by @dependabot[bot] in sktime#1942
  • [MNT] Check versions in wheels workflow by @szepeviktor in sktime#1948
  • [MNT] Dependabot: Bump actions/setup-python from 5 to 6 by @dependabot[bot] in sktime#1963
  • [MNT] Update CODEOWNERS with current core dev state by @fkiraly in sktime#1972
  • [MNT] python 3.9 end-of-life by @phoeenniixx in sktime#1980

All Contributors

@agobbifbk, @caph1993, @cngmid, @fkiraly, @fnhirwa, @Himanshu-Verma-ds, @hubkrieb, @jdb78, @lohraspco, @phoeenniixx, @Pinaka07, @PranavBhatP, @sanskarmodi8, @Sohaib-Ahmed21, @szepeviktor @Vishnu-Rangiah, @zju-ys

v1.4.0

Feature and maintenance update.

Highlights

Enhancements

  • [ENH] Add Type hints to TimeSeriesDataSet to align with pep 585 by @fnhirwa in sktime#1819
  • [ENH] Allow multiple instances from multiple mock classes in _safe_import by @fnhirwa in sktime#1818
  • [ENH] EXPERIMENTAL PR: D1 and D2 layer for v2 refactor by @phoeenniixx in sktime#1811
  • [ENH] EXPERIMENTAL PR: make the data_module dataclass-like by @phoeenniixx in sktime#1832
  • [ENH] EXPERIMENTAL: TFT model based on the new data pipeline by @phoeenniixx in sktime#1812
  • [ENH] test suite for pytorch-forecasting forecasters by @fkiraly in sktime#1780
  • [ENH] TemporalFusionTransformer - allow mixed precision training by @Marcrb2 in sktime#1518
  • [ENH] move model base classes into models.base module - part 1 by @fkiraly in sktime#1773
  • [ENH] move model base classes into models.base module - part 2 by @fkiraly in sktime#1774
  • [ENH] move model base classes into models.base module - part 3 by @fkiraly in sktime#1776
  • [ENH] tests for TiDE Model by @PranavBhatP in sktime#1843
  • [ENH] refactor test metadata container to include data loader configs by @fkiraly in sktime#1861
  • [ENH] DecoderMLP metadata container for v1 tests by @fkiraly in sktime#1859
  • [ENH] TimeXer model from thuml by @PranavBhatP in sktime#1797
  • [ENH] EXPERIMENTAL: Example notebook based on the new data pipeline by @phoeenniixx in sktime#1813
  • [ENH] refactor test data scenario generation to tests._data_scenarios by @fkiraly in sktime#1877

Fixes

  • [BUG] fix absolute errorbar by @MartinoMensio in sktime#1579
  • [BUG] EXPERIMENTAL PR: Solve the bug in data_module by @phoeenniixx in sktime#1834
  • [BUG] fix incorrect concatenation dimension in concat_sequences by @cngmid in sktime#1827
  • [BUG] Fix for the case when reduction is set to none by @fnhirwa in sktime#1872
  • [BUG] enable silenced TFT v2 tests by @fkiraly in sktime#1878

Documentation

  • [DOC] fix gradient_clip value in tutorials to ensure reproducible outputs similar to the committed cell output by @gbilleyPeco in sktime#1750
  • [DOC] Fix typos in getting started section of the documentation by @pietsjoh in sktime#1399
  • [DOC] improved pull request template by @fkiraly in sktime#1866
  • [DOC] add project badges to README: sponsoring and downloads by @fkiraly in sktime#1891

Maintenance

  • [MNT] Isolate cpflow package, towards fixing readthedocs build by @fkiraly in sktime#1775
  • [MNT] fix readthedocs build by @fkiraly in sktime#1777
  • [MNT] move release to trusted publishers by @fkiraly in sktime#1800
  • [MNT] standardize dependabot.yml by @fkiraly in sktime#1799
  • [MNT] remove tj-actions by @fkiraly in sktime#1798
  • [MNT] Dependabot: bump codecov/codecov-action from 1 to 5 by @dependabot in sktime#1803
  • [MNT] disable automated merge and approve actions by @fkiraly in sktime#1804
  • build(deps): update sphinx requirement from <7.2.6,>3.2 to >3.2,<8.2.4 by @dependabot in sktime#1787
  • [MNT] Move config from setup.cfg to pyproject.toml by @Borda in sktime#1852
  • [MNT] Move pytest configuration to pyproject.toml by @Borda in sktime#1851
  • [MNT] Add 'UP' to extend-select for pyupgrade python syntax by @Borda in sktime#1856
  • [MNT] Replace Black with Ruff formatting and update configuration by @Borda in sktime#1853
  • [MNT] issue templates by @fkiraly in sktime#1867
  • [MNT] Clearly define the MLP as a class/nn.model by @jobs-git in sktime#1864

All Contributors

@agobbifbk, @Borda, @cngmid, @fkiraly, @fnhirwa, @gbilleyPeco, @jobs-git, @Marcrb2, @MartinoMensio, @phoeenniixx, @pietsjoh, @PranavBhatP

v1.3.0

Feature and maintenance update.

Highlights

  • python 3.13 support
  • tide model
  • bugfixes for TFT

Enhancements

  • [ENH] Tide model. by @Sohaib-Ahmed21 in sktime#1734
  • [ENH] refactor __init__ modules to no longer contain classes - preparatory commit by @fkiraly in sktime#1739
  • [ENH] refactor __init__ modules to no longer contain classes by @fkiraly in sktime#1738
  • [ENH] extend package author attribution requirement in license to present by @fkiraly in sktime#1737
  • [ENH] linting tide model by @fkiraly in sktime#1742
  • [ENH] move tide model - part 1 by @fkiraly in sktime#1743
  • [ENH] move tide model - part 2 by @fkiraly in sktime#1744
  • [ENH] clean-up refactor of TimeSeriesDataSet by @fkiraly in sktime#1746

Fixes

  • [BUG] Bugfix when no exogenous variable is passed to TFT by @XinyuWuu in sktime#1667
  • [BUG] Fix issue when training TFT model on mac M1 mps device. element 0 of tensors does not require grad and does not have a grad_fn by @fnhirwa in sktime#1725

Documentation

  • [DOC] Fix the spelling error of holding by @xiaokongkong in sktime#1719
  • [DOC] Updated documentation on TimeSeriesDataSet.predict_mode by @madprogramer in sktime#1720
  • [DOC] General PR to improve docs by @julian-fong in sktime#1705
  • [DOC] Correct argument for optimizer ranger in Temporal Fusion Transformer tutorial by @fnhirwa in sktime#1724
  • [DOC] Fixed typo "monotone_constaints" by @Luke-Chesley in sktime#1516
  • [DOC] minor fixes in documentation by @fkiraly in sktime#1763
  • [DOC] improve and add tide model to docs by @PranavBhatP in sktime#1762

Maintenance

  • [MNT] update linting: limit line length to 88, add isort by @fkiraly in sktime#1740
  • [MNT] update nbeats/sub_modules.py to remove overhead in tensor creation by @d-schmitt in sktime#1580
  • [MNT] Temporary fix for lint errors to conform to the recent changes in linting rules see #1749 by @fnhirwa in sktime#1748
  • [MNT] python 3.13 support by @fkiraly in sktime#1691

All Contributors

@d-schmitt, @fkiraly, @fnhirwa, @julian-fong, @Luke-Chesley, @madprogramer, @PranavBhatP, @Sohaib-Ahmed21, @xiaokongkong, @XinyuWuu

v1.2.0

Maintenance update, minor feature additions and bugfixes.

  • support for numpy 2.X
  • end of life for python 3.8
  • fixed documentation build
  • bugfixes

Dependency changes

  • pytorch-forecasting is now compatible with numpy 2.X (core dependency)
  • optuna (tuning soft dependency) bounds have been update to >=3.1.0,<5.0.0

Fixes

  • [BUG] fix AttributeError: 'ExperimentWriter' object has no attribute 'add_figure' by @ewth in sktime#1694

Documentation

  • [DOC] typo fixes in changelog by @fkiraly in sktime#1660
  • [DOC] update URLs to sktime org by @fkiraly in sktime#1674

Maintenance

  • [MNT] handle mps backend for lower versions of pytorch and fix mps failure on macOS-latest runner by @fnhirwa in sktime#1648
  • [MNT] updates the actions in the doc build CI by @fkiraly in sktime#1673
  • [MNT] fixes to readthedocs.yml by @fkiraly in sktime#1676
  • [MNT] updates references in CI and doc locations to main by @fkiraly in sktime#1677
  • [MNT] show_versions utility by @fkiraly in sktime#1688
  • [MNT] Relax numpy bound to numpy<3.0.0 by @XinyuWuu in sktime#1624
  • [MNT] fix pre-commit failures on main by @ewth in sktime#1696
  • [MNT] Move linting to ruff by @airookie17 in sktime#1692 1693
  • [MNT] ruff linting - allow use of assert (S101) by @fkiraly in sktime#1701
  • [MNT] ruff - fix list related linting failures C416 and C419 by @fkiraly in sktime#1702
  • [MNT] Delete poetry.lock by @benHeid in sktime#1704
  • [MNT] fix black doesn't have extras dependency by @fnhirwa in sktime#1697
  • [MNT] Remove mutable objects from defaults by @eugenio-mercuriali in sktime#1699
  • [MNT] remove docs build in ci for all pr by @yarnabrina in sktime#1712
  • [MNT] EOL for python 3.8 by @fkiraly in sktime#1661
  • [MNT] remove poetry.lock by @fkiraly in sktime#1651
  • [MNT] update pre-commit requirement from <4.0.0,>=3.2.0 to >=3.2.0,<5.0.0 by @dependabot in https://github.com/sktime/pytorch-forecasting/pull/
  • [MNT] update optuna requirement from <4.0.0,>=3.1.0 to >=3.1.0,<5.0.0 by @dependabot in sktime#1715
  • [MNT] CODEOWNERS file by @fkiraly in sktime#1710

All Contributors

@airookie17, @benHeid, @eugenio-mercuriali, @ewth, @fkiraly, @fnhirwa, @XinyuWuu, @yarnabrina

v1.1.1

Hotfix for accidental package name change in pyproject.toml.

The package name is now corrected to pytorch-forecasting.

v1.1.0

Maintenance update widening compatibility ranges and consolidating dependencies:

  • TSMixer model, see TSMixer: An All-MLP Architecture for Time Series Forecasting.
  • support for python 3.11 and 3.12, added CI testing
  • support for MacOS, added CI testing
  • core dependencies have been minimized to numpy, torch, lightning, scipy, pandas, and scikit-learn.
  • soft dependencies are available in soft dependency sets: all_extras for all soft dependencies, and tuning for optuna based optimization.

Dependency changes

  • the following are no longer core dependencies and have been changed to optional dependencies : optuna, statsmodels, pytorch-optimize, matplotlib. Environments relying on functionality requiring these dependencies need to be updated to install these explicitly.
  • optuna bounds have been updated to optuna >=3.1.0,<4.0.0
  • optuna-integrate is now an additional soft dependency, in case of optuna >=3.3.0

Deprecations and removals

  • from 1.2.0, the default optimizer will be changed from "ranger" to "adam" to avoid non-torch dependencies in defaults. pytorch-optimize optimizers can still be used. Users should set the optimizer explicitly to continue using "ranger".
  • from 1.1.0, the loggers do not log figures if soft dependency matplotlib is not present, but will raise no exceptions in this case. To log figures, ensure that matplotlib is installed.

All Contributors

@andre-marcos-perez, @avirsaha, @bendavidsteel, @benHeid, @bohdan-safoniuk, @Borda, @CahidArda, @fkiraly, @fnhirwa, @germanKoch, @jacktang, @jdb78, @jurgispods, @maartensukel, @MBelniak, @orangehe, @pavelzw, @sfalkena, @tmct, @XinyuWuu, @yarnabrina

v1.0.0 Update to pytorch 2.0 (10/04/2023)

Breaking Changes

  • Upgraded to pytorch 2.0 and lightning 2.0. This brings a couple of changes, such as configuration of trainers. See the lightning upgrade guide. For PyTorch Forecasting, this particularly means if you are developing own models, the class method epoch_end has been renamed to on_epoch_end and replacing model.summarize() with ModelSummary(model, max_depth=-1) and Tuner(trainer) is its own class, so trainer.tuner needs replacing. (#1280)
  • Changed the predict() interface returning named tuple - see tutorials.

Changes

  • The predict method is now using the lightning predict functionality and allows writing results to disk (#1280).

Fixed

  • Fixed robust scaler when quantiles are 0.0, and 1.0, i.e. minimum and maximum (#1142)

v0.10.3 Poetry update (07/09/2022)

Fixed

  • Removed pandoc from dependencies as issue with poetry install (#1126)
  • Added metric attributes for torchmetric resulting in better multi-GPU performance (#1126)

Added

  • "robust" encoder method can be customized by setting "center", "lower" and "upper" quantiles (#1126)

v0.10.2 Multivariate networks (23/05/2022)

Added

  • DeepVar network (#923)
  • Enable quantile loss for N-HiTS (#926)
  • MQF2 loss (multivariate quantile loss) (#949)
  • Non-causal attention for TFT (#949)
  • Tweedie loss (#949)
  • ImplicitQuantileNetworkDistributionLoss (#995)

Fixed

  • Fix learning scale schedule (#912)
  • Fix TFT list/tuple issue at interpretation (#924)
  • Allowed encoder length down to zero for EncoderNormalizer if transformation is not needed (#949)
  • Fix Aggregation and CompositeMetric resets (#949)

Changed

  • Dropping Python 3.6 support, adding 3.10 support (#479)
  • Refactored dataloader sampling - moved samplers to pytorch_forecasting.data.samplers module (#479)
  • Changed transformation format for Encoders to dict from tuple (#949)

Contributors

  • jdb78

v0.10.1 Bugfixes (24/03/2022)

Fixed

  • Fix with creating tensors on correct devices (#908)
  • Fix with MultiLoss when calculating gradient (#908)

Contributors

  • jdb78

v0.10.0 Adding N-HiTS network (N-BEATS successor) (23/03/2022)

Added

  • Added new N-HiTS network that has consistently beaten N-BEATS (#890)
  • Allow using torchmetrics as loss metrics (#776)
  • Enable fitting EncoderNormalizer() with limited data history using max_length argument (#782)
  • More flexible MultiEmbedding() with convenience output_size and input_size properties (#829)
  • Fix concatenation of attention (#902)

Fixed

  • Fix pip install via github (#798)

Contributors

  • jdb78
  • christy
  • lukemerrick
  • Seon82

v0.9.2 Maintenance Release (30/11/2021)

Added

  • Added support for running lightning.trainer.test (#759)

Fixed

  • Fix inattention mutation to x_cont (#732).
  • Compatibility with pytorch-lightning 1.5 (#758)

Contributors

  • eavae
  • danielgafni
  • jdb78

v0.9.1 Maintenance Release (26/09/2021)

Added

  • Use target name instead of target number for logging metrics (#588)
  • Optimizer can be initialized by passing string, class or function (#602)
  • Add support for multiple outputs in Baseline model (#603)
  • Added Optuna pruner as optional parameter in TemporalFusionTransformer.optimize_hyperparameters (#619)
  • Dropping support for Python 3.6 and starting support for Python 3.9 (#639)

Fixed

  • Initialization of TemporalFusionTransformer with multiple targets but loss for only one target (#550)
  • Added missing transformation of prediction for MLP (#602)
  • Fixed logging hyperparameters (#688)
  • Ensure MultiNormalizer fit state is detected (#681)
  • Fix infinite loop in TimeDistributedEmbeddingBag (#672)

Contributors

  • jdb78
  • TKlerx
  • chefPony
  • eavae
  • L0Z1K

v0.9.0 Simplified API (04/06/2021)

Breaking changes

  • Removed dropout_categoricals parameter from TimeSeriesDataSet. Use categorical_encoders=dict(<variable_name>=NaNLabelEncoder(add_nan=True)) instead (#518)

  • Rename parameter allow_missings for TimeSeriesDataSet to allow_missing_timesteps (#518)

  • Transparent handling of transformations. Forward methods should now call two new methods (#518):

    • transform_output to explicitly rescale the network outputs into the de-normalized space
    • to_network_output to create a dict-like named tuple. This allows tracing the modules with PyTorch's JIT. Only prediction is still required which is the main network output.

    Example:

    def forward(self, x):
    normalized_prediction = self.module(x)
    prediction = self.transform_output(prediction=normalized_prediction, target_scale=x["target_scale"])
    return self.to_network_output(prediction=prediction)

Fixed

  • Fix quantile prediction for tensors on GPUs for distribution losses (#491)
  • Fix hyperparameter update for RecurrentNetwork.from_dataset method (#497)

Added

  • Improved validation of input parameters of TimeSeriesDataSet (#518)

v0.8.5 Generic distribution loss(es) (27/04/2021)

Added

  • Allow lists for multiple losses and normalizers (#405)
  • Warn if normalization is with scale < 1e-7 (#429)
  • Allow usage of distribution losses in all settings (#434)

Fixed

  • Fix issue when predicting and data is on different devices (#402)
  • Fix non-iterable output (#404)
  • Fix problem with moving data to CPU for multiple targets (#434)

Contributors

  • jdb78
  • domplexity

v0.8.4 Simple models (07/03/2021)

Added

  • Adding a filter functionality to the timeseries dataset (#329)
  • Add simple models such as LSTM, GRU and a MLP on the decoder (#380)
  • Allow usage of any torch optimizer such as SGD (#380)

Fixed

  • Moving predictions to CPU to avoid running out of memory (#329)
  • Correct determination of output_size for multi-target forecasting with the TemporalFusionTransformer (#328)
  • Tqdm autonotebook fix to work outside of Jupyter (#338)
  • Fix issue with yaml serialization for TensorboardLogger (#379)

Contributors

  • jdb78
  • JakeForsey
  • vakker

v0.8.3 Bugfix release (31/01/2021)

Added

  • Make tuning trainer kwargs overwritable (#300)
  • Allow adding categories to NaNEncoder (#303)

Fixed

  • Underlying data is copied if modified. Original data is not modified inplace (#263)
  • Allow plotting of interpretation on passed figure for NBEATS (#280)
  • Fix memory leak for plotting and logging interpretation (#311)
  • Correct shape of predict() method output for multi-targets (#268)
  • Remove cloudpickle to allow GPU trained models to be loaded on CPU devices from checkpoints (#314)

Contributors

  • jdb78
  • kigawas
  • snumumrik

v0.8.2 Fix for output transformer (12/01/2021)

  • Added missing output transformation which was switched off by default (#260)

v0.8.1 Adding support for lag variables (10/01/2021)

Added

  • Add "Release Notes" section to docs (#237)
  • Enable usage of lag variables for any model (#252)

Changed

  • Require PyTorch>=1.7 (#245)

Fixed

  • Fix issue for multi-target forecasting when decoder length varies in single batch (#249)
  • Enable longer subsequences for min_prediction_idx that were previously wrongfully excluded (#250)

Contributors

  • jdb78

v0.8.0 Adding multi-target support (03/01/2021)

Added

  • Adding support for multiple targets in the TimeSeriesDataSet (#199) and amended tutorials.
  • Temporal fusion transformer and DeepAR with support for multiple targets (#199)
  • Check for non-finite values in TimeSeriesDataSet and better validate scaler argument (#220)
  • LSTM and GRU implementations that can handle zero-length sequences (#235)
  • Helpers for implementing auto-regressive models (#236)

Changed

  • TimeSeriesDataSet's y of the dataloader is a tuple of (target(s), weight) - potentially breaking for model or metrics implementation Most implementations will not be affected as hooks in BaseModel and MultiHorizonMetric were modified. (#199)

Fixed

  • Fixed autocorrelation for pytorch 1.7 (#220)
  • Ensure reproducibility by replacing python set() with dict.fromkeys() (mostly TimeSeriesDataSet) (#221)
  • Ensures BetaDistributionLoss does not lead to infinite loss if actuals are 0 or 1 (#233)
  • Fix for GroupNormalizer if scaling by group (#223)
  • Fix for TimeSeriesDataSet when using min_prediction_idx (#226)

Contributors

  • jdb78
  • JustinNeumann
  • reumar
  • rustyconover

v0.7.1 Tutorial on how to implement a new architecture (07/12/2020)

Added

  • Tutorial on how to implement a new architecture covering basic and advanced use cases (#188)
  • Additional and improved documentation - particularly of implementation details (#188)

Changed (breaking for new model implementations)

  • Moved multiple private methods to public methods (particularly logging) (#188)
  • Moved get_mask method from BaseModel into utils module (#188)
  • Instead of using label to communicate if model is training or validating, using self.training attribute (#188)
  • Using sample((n,)) of pytorch distributions instead of deprecated sample_n(n) method (#188)

v0.7.0 New API for transforming inputs and outputs with encoders (03/12/2020)

Added

  • Beta distribution loss for probabilistic models such as DeepAR (#160)

Changed

  • BREAKING: Simplifying how to apply transforms (such as logit or log) before and after applying encoder. Some transformations are included by default but a tuple of a forward and reverse transform function can be passed for arbitrary transformations. This requires to use a transformation keyword in target normalizers instead of, e.g. log_scale (#185)

Fixed

  • Incorrect target position if len(static_reals) > 0 leading to leakage (#184)
  • Fixing predicting completely unseen series (#172)

Contributors

  • jdb78
  • JakeForsey

v0.6.1 Bugfixes and DeepAR improvements (24/11/2020)

Added

  • Using GRU cells with DeepAR (#153)

Fixed

  • GPU fix for variable sequence length (#169)
  • Fix incorrect syntax for warning when removing series (#167)
  • Fix issue when using unknown group ids in validation or test dataset (#172)
  • Run non-failing CI on PRs from forks (#166, #156)

Docs

  • Improved model selection guidance and explanations on how TimeSeriesDataSet works (#148)
  • Clarify how to use with conda (#168)

Contributors

  • jdb78
  • JakeForsey

v0.6.0 Adding DeepAR (10/11/2020)

Added

  • DeepAR by Amazon (#115)
    • First autoregressive model in PyTorch Forecasting
    • Distribution loss: normal, negative binomial and log-normal distributions
    • Currently missing: handling lag variables and tutorial (planned for 0.6.1)
  • Improved documentation on TimeSeriesDataSet and how to implement a new network (#145)

Changed

  • Internals of encoders and how they store center and scale (#115)

Fixed

  • Update to PyTorch 1.7 and PyTorch Lightning 1.0.5 which came with breaking changes for CUDA handling and with optimizers (PyTorch Forecasting Ranger version) (#143, #137, #115)

Contributors

  • jdb78
  • JakeForesey

v0.5.3 Bug fixes (31/10/2020)

Fixes

  • Fix issue where hyperparameter verbosity controlled only part of output (#118)
  • Fix occasional error when .get_parameters() from TimeSeriesDataSet failed (#117)
  • Remove redundant double pass through LSTM for temporal fusion transformer (#125)
  • Prevent installation of pytorch-lightning 1.0.4 as it breaks the code (#127)
  • Prevent modification of model defaults in-place (#112)

v0.5.2 Fixes to interpretation and more control over hyperparameter verbosity (18/10/2020)

Added

  • Hyperparameter tuning with optuna to tutorial
  • Control over verbosity of hyper parameter tuning

Fixes

  • Interpretation error when different batches had different maximum decoder lengths
  • Fix some typos (no changes to user API)

v0.5.1 PyTorch Lightning 1.0 compatibility (14/10/2020)

This release has only one purpose: Allow usage of PyTorch Lightning 1.0 - all tests have passed.

v0.5.0 PyTorch Lightning 0.10 compatibility and classification (12/10/2020)

Added

  • Additional checks for TimeSeriesDataSet inputs - now flagging if series are lost due to high min_encoder_length and ensure parameters are integers
  • Enable classification - simply change the target in the TimeSeriesDataSet to a non-float variable, use the CrossEntropy metric to optimize and output as many classes as you want to predict

Changed

  • Ensured PyTorch Lightning 0.10 compatibility
    • Using LearningRateMonitor instead of LearningRateLogger
    • Use EarlyStopping callback in trainer callbacks instead of early_stopping argument
    • Update metric system update() and compute() methods
    • Use Tuner(trainer).lr_find() instead of trainer.lr_find() in tutorials and examples
  • Update poetry to 1.1.0

v0.4.1 Various fixes models and data (01/10/2020)

Fixes

Model

  • Removed attention to current datapoint in TFT decoder to generalise better over various sequence lengths
  • Allow resuming optuna hyperparamter tuning study

Data

  • Fixed inconsistent naming and calculation of encoder_lengthin TimeSeriesDataSet when added as feature

Contributors

  • jdb78

v0.4.0 Metrics, performance, and subsequence detection (28/09/2020)

Added

Models

  • Backcast loss for N-BEATS network for better regularisation
  • logging_metrics as explicit arguments to models

Metrics

  • MASE (Mean absolute scaled error) metric for training and reporting
  • Metrics can be composed, e.g. 0.3* metric1 + 0.7 * metric2
  • Aggregation metric that is computed on mean prediction over all samples to reduce mean-bias

Data

  • Increased speed of parsing data with missing datapoints. About 2s for 1M data points. If numba is installed, 0.2s for 1M data points
  • Time-synchronize samples in batches: ensure that all samples in each batch have with same time index in decoder

Breaking changes

  • Improved subsequence detection in TimeSeriesDataSet ensures that there exists a subsequence starting and ending on each point in time.
  • Fix min_encoder_length = 0 being ignored and processed as min_encoder_length = max_encoder_length

Contributors

  • jdb78
  • dehoyosb

v0.3.1 More tests and better docs (13/09/2020)

  • More tests driving coverage to ~90%
  • Performance tweaks for temporal fusion transformer
  • Reformatting with sort
  • Improve documentation - particularly expand on hyper parameter tuning

Fixed

  • Fix PoissonLoss quantiles calculation
  • Fix N-Beats visualisations

v0.3.0 More testing and interpretation features (02/09/2020)

Added

  • Calculating partial dependency for a variable
  • Improved documentation - in particular added FAQ section and improved tutorial
  • Data for examples and tutorials can now be downloaded. Cloning the repo is not a requirement anymore
  • Added Ranger Optimizer from pytorch_ranger package and fixed its warnings (part of preparations for conda package release)
  • Use GPU for tests if available as part of preparation for GPU tests in CI

Changes

  • BREAKING: Fix typo "add_decoder_length" to "add_encoder_length" in TimeSeriesDataSet

Bugfixes

  • Fixing plotting predictions vs actuals by slicing variables

v0.2.4 Fix edge case in prediction logging (26/08/2020)

Fixed

Fix bug where predictions were not correctly logged in case of decoder_length == 1.

Added

  • Add favicon to docs page

v0.2.3 Make pip installable from master branch (23/08/2020)

Update build system requirements to be parsed correctly when installing with pip install git+https://github.com/jdb78/pytorch-forecasting

v0.2.2 Improving tests (23/08/2020)

  • Add tests for MacOS
  • Automatic releases
  • Coverage reporting

v0.2.1 Patch release (23/08/2020)

This release improves robustness of the code.

  • Fixing bug across code, in particularly

    • Ensuring that code works on GPUs
    • Adding tests for models, dataset and normalisers
    • Test using GitHub Actions (tests on GPU are still missing)

  • Extend documentation by improving docstrings and adding two tutorials.

  • Improving default arguments for TimeSeriesDataSet to avoid surprises

v0.2.0 Minor release (16/08/2020)

Added

  • Basic tests for data and model (mostly integration tests)
  • Automatic target normalization
  • Improved visualization and logging of temporal fusion transformer
  • Model bugfixes and performance improvements for temporal fusion transformer

Modified

  • Metrics are reduced to calculating loss. Target transformations are done by new target transformer