Update from unit8co/darts. #11
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…2218) * add support for more lr scheduler config parameters to torch models * update changelog
* update dev requirements with new pre commit hook lint dependency versions * black reformatting * fix flake8 checks
* fix: append/prepend correctul retain components names and hierarchy * updated changelog * fix: revert unecessary change * update changelog --------- Co-authored-by: dennisbader <dennis.bader@gmx.ch>
* fix: update hierarchy for single transform window_transform * update changelog * update changelog * fix: using set to check overlap * fix: corrected logic to update the hierarchy after window_transform * fix: hierarchy can be conserved when applying non-overlapping transforms * feat: add new argument, improve logic * feat: adding tests * fix: expected argument match docstring in resample() * fix: addressing review comments * fix: linting issue * fix: linting * linting * update changelog and remane keep_old_names to keep_names --------- Co-authored-by: dennisbader <dennis.bader@gmx.ch>
* remove pytorch lightning upper version cap * fix failing unit test and update changelog
…rror (#2222) * fix: arguments must be provided to model cls in order to check presence of the fitted_values attribute * fix: added a check that parameters is indeed a dict * updated changelog * fix: update test to pass the new sanity checks * fix: addressing review comments --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
* Change default kwarg * Update CHANGELOG.md --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
* add output chunk shift to lightning modeuls * training torch model with shifted output * first shifted output inference works for mixed covariates models * full covariateds support for shifted mixed covariates dataset * add shift support to all torch models * update torch model extreme lags with shift * update torch model encoder settings with shift * update torch model encoder settings with shift * add unit test for shifted torch mmodel with encoders * add unit tests for tft model * add unit tests for all torch models * update output_chunk_shift description * apply suggestions from PR review * add output chunk shift to extreme lags * udpate historical forecasts to work with shifted output * update historical forecasts start description for shifted output * apply suggestions from PR review * prepare regression models for output chunk shift * fix failing unit tests * prepare regression models for output chunk shift part 2 * update hist fc for regression models with output shift * update tabularization * add test for comparing results between output shift and normal multi models * historical forecasts for shifted regression models * update tabularization training tests * update tabulirazion get feature times tests * update tabularization get shared times tests * update tabularization get shared bounds tests * update tabularization get lagged prediction data tests * add tests for tabularization without target lags but only covariate lags * update n_steps_between docs * update changelog * add unit tests for inference datasets * add unit tests for sequential training datasts * update changelog * make ocs property non optional * skip output_chunk_shift checks when loading weights since not relevant for parameter shape * apply suggestions from PR review
…#2242) * fix: datetime_attribute account for 0 or 1-indexing of the attributes * feat: 1-indexed date attribute are shifted to enforce 0-indexing for all the generated encodings * updated changelog * fix: remove commented lines * fix: typo in comment * make ONE_INDEXED_FREQS a constant * fix: simplified test by using year 2001 * feat: better handling of years with 53 weeks or 366 days * fix: properly take the index length when adding the extra week * fix: simplifying test * fix: update tests to account for the forced 0-indexing of the datetime attributes encoding * fix: passing lmbda parameter as BoxCox doesn't converge when encodings contains a 0 --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
* feat: adding docstring and check to get_multioutput_estimator * fix: added lowbound check * fix: update docstring, indexing account for multi_models param * feat: added corresponding test * feat: added tests for estimator getter * feat: store and expose the lagged label names (for each model estimator) * fix: rephrasing docstring * update changelog * fix: linting * fix: replaced ocl with hrz in naming of the lagged label * fix: update error messages * feat: simplify test, overfit XGB on only one training example * feat: added a method to get estimator for models supporting multi-output natively * feat: added corresponding test * update changelog * fix: linting * Update CHANGELOG.md --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
* update changelog * bump u8darts 0.27.2 to 0.28.0 * update changelog
* update code owners * udpated PR template
* Remove unnessesary `pass` statements * Rename ForecastingModel_is_probabilistic to supports_probabilistic_prediction, rearrange some documentation * Remove redundant overrides * Reformat * Add CHANGELOG entry --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
* fix type hinting for _with_sanity_checks * update changelog
* Add optional inverse transform in historical forecast * Update variables names and docstrings * Move the inverse transform to InvertibleDataTransformer * Fix single element list * Update docstrings * Move the inverse transform of list of lists to inverse_transform method * make invertible transformers act on list of lists of series * add tests * update changelog --------- Co-authored-by: dennisbader <dennis.bader@gmx.ch>
* lxml_html_clean for nbshinx * update changelog
* fix lighgbm segmentation fualt * update changelog * parameterize unit tests
* fix lighgbm segmentation fualt * update changelog * parameterize unit tests * make metric_kwargs metric specific rather than infereing which kwarg belongs to which metric * update hierarchical reconciliation notebook * fix failing residuals tests
* use pytest to skip torch tests * fix some mistakes in tsmixer notebook
* add TimesSeries.from_group_dataframe parallel mode * remove code mess * add doc string for new parameters * update CHANGELOG.md * add miss dtype * fix static covariates * make parallel function as local and fix tests * fix parallel utils imports * update changelog * Update CHANGELOG.md --------- Co-authored-by: Bohdan Bilonoh <bilonogo@rozetka.ua> Co-authored-by: dennisbader <dennis.bader@gmx.ch>
* bump black[jupyter] 24.1.1 to 24.3.0 * update changeloig
* improve hist fc start point * add tests * update documentation * update changelog * clean up code * fix tests * fix missed lines * improve codecov
* reduce tests to failing macos * retry * retry * retry * add back cache file * add comment * try different caching * fix bug * remove venv * add back codecov upload and ubuntu os * add linting, docs, and example checks * fix papermill missing * change to exmaples workdir for example checks * improve caching, update merge workflow * update merge workflow to use cache * add update cache step whenever a branch is merged to master
* update comment * fix failing merge workflow
* fix failing merge workflow * generate cache hash using uv for latest requirements
* update deploy github pages * update release workflow * run docker build test * fix bug * add bump type * test * use python image * update release workflow
* remove gradle * update changelog
* bump minimum python version from 3.8 to 3.9 * remove last gradle file * update changelog * update changelog
* ruff update start * ruff * changlog * lint * fix: remove unused dependencies in notebooks * feat: upgrading ruff to 0.7.2 * feat: update target version from python38 to python39 * update changelog * fix: typing imports * fix: issue of the test --------- Co-authored-by: Greg DeVosNouri <gdevos010@gmail.com>
* fix: check that model is probabilistic when num samples is greater than 1 for optimized historical forecasts * feat: update the tests accordingly * update changelog * fix: simplify the test * fix: remove typo * fix: ignoring a linting commit for git blame
* Fix deprecated usage of torch.nn.utils.weight_norm The previous implementation in darts.darts.models.forecasting.tcn_mode was using `torch.nn.utils.weight_norm`, which is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`. This commit replaces two occurrences of `torch.nn.utils.weight_norm` with the recommended `torch.nn.utils.parametrizations.weight_norm` to resolve the deprecation warning. * Update torch_forecasting_model.py Corrected file saving process for checkpoint files (ckpt) to filter out occurrences of the string '.pt' from the previous file path." * fix: revert changes * update changelog --------- Co-authored-by: Saeed Foroutan <foroutansaeed@gmail.com>
Co-authored-by: madtoinou <32447896+madtoinou@users.noreply.github.com> Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
Bumps [lxml-html-clean](https://github.com/fedora-python/lxml_html_clean) from 0.1.1 to 0.4.0. - [Changelog](https://github.com/fedora-python/lxml_html_clean/blob/main/CHANGES.rst) - [Commits](fedora-python/lxml_html_clean@0.1.1...0.4.0) --- updated-dependencies: - dependency-name: lxml-html-clean dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
…historical forecasts (#2604) * fix backtest and residuals for mutliple series with different number of forecasts * update changelog
* add basic scalar window support * change scaler to a more robust generalisation Co-authored-by: Dennis Bader <dennis.bader@gmx.ch> * delete unused functions and add util function * delete unused functions and add util function * add transformers to optimized historical forecasts * add covariate transformers, refactor + docstring update * delete util to avoid circular import * delete unused param and add transforms to torch models * fix param name * move all series and covariates fitting into one place, allow data transform without model retrain * update readme and data types * optimized forecasts only support invertible data transform * feat: harmonize application of scaler in hf, support for Pipeline * feat: adding basic test for regression models * fix: using an util method to reduce code duplication * fix: simplify the tests * fix: makes things faster if no data transformer are passed * feat: add test for the optimized hf * fix: using util method in gridsearch as well * fix: reverting some changes * fix: make sure the series have a range that require scaling * update changelog * feat: adding small example about how to use scaler in historical forecasts * fix: adress review comments * fix: adapting the tests * fix: moved the historical forecasts test to dedicated folder * feat: make sure the already fitted data transformer process series correctly * feat: adding tests for historical forecasts with scaler * fix: remove duplicated test, also test tfm historical forecast with scaler * fix: typo * fix: adressing review comments * fix: adjust the example of historical forecasts with auto-scaling according to revire comments * fix: typo * fix: adress review comments * feat: possibility to select transformer idx * fix: adding tests, fixing logic * feat: added tests for the new data transformer features, fixed logic * feat: add tests, fix logic * fix: adding virtual env to gitignore * fix: renamed the idx_params argument idx_series * apply minor changes * apply minor changes part 2 * add additional info to docs * fix: replaced exception with warning when multiple series, retrain=True and data transformer defined with global_fit=True --------- Co-authored-by: Jan Fidor <jan.p.fidor@gmail.com> Co-authored-by: Dennis Bader <dennis.bader@gmx.ch> Co-authored-by: JanFidor <66260538+JanFidor@users.noreply.github.com>
* Update sf_auto_arima.py * Update sf_auto_ets.py * Update sf_auto_ces.py * Update sf_auto_theta.py * Update CHANGELOG.md
* naive conformal prediction * first hist fc version works * add component names * add support for train length * support for last points only * add hist fc unit tests * add first conformal unit tests * overlap end checkpoint * overlap end checkpoint 2 * ignore start * finalize hist fc test * start, train length tests * finalize start train length tests * fix residuals with overlap end * refactor calibration for predict and hist fc * base and child conformal * checks for calibration set * rename conformal naive model * add additional forecasting model logic * add more unit tests * add output chunk shift support * support train length with cal input * support train lenght part 2 * restructure hist fc logic * test with shorter covariates * add checks for min lengths * corrections for minimum input * improve hist fc tests * make naive conformal model accept quantiles * add winkler score quantile interval metric * update tests for quantile instead of alpha * add coverage metric and improve residuals and backtest * add save load as in ensemble mode * quantile tests * remove checks * add non conformity scores for cqr * add conformalized quantile regression * allow all global prob models for ConformalQR * add asymmetric naive model * remove old code * add tests for asymetric naive mdoel * add tests for cqr * add progress bars * add quantile sampler * add predict lkl params and num samples * add random method for handling randomness of non-torch models * fix all tests * code cleanup * add probabilistic test * add conformal models to readme and covariates user guide * fix failing tests * improve docs * add sketch of cp example notebook * small update * improve docs * attempt to fix failing test on linux * update start logic * upgrade python target version * improve stride handling * remove optional input calibration set * use cal stride * make predict work with cal_stride * add cal stride to historical forecasts * hist fc optimized cal set selection * add hist fc start test with different strides * improve comments * add more tests * stridden conformal model tests * apply suggestions from pr review * update docs * cleanup * update changelog * update changelog * update example notebook * add conformal prediction notebook * apply suggestions from PR review * update notebook * update changelog
* bump u8darts 0.31.0 to 0.32.0 * update changelog
* Add new model StatsForecastAutoTBATS * Update darts/models/forecasting/sf_auto_tbats.py Co-authored-by: Dennis Bader <dennis.bader@gmx.ch> * Update CHANGELOG * Update README.md * Update covariates.md * Update test_probabilistic_models.py * Update test_probabilistic_models.py * update changelog and readme --------- Co-authored-by: Dennis Bader <dennis.bader@gmx.ch>
…with output_chunk_shift in regression models (#2634) * FIX-2633: Use correct time indices when running historical forecasts on regression models with 'output_chunk_shift > 0' and 'output_chunk_length == 1'. Extended unit tests to cover this * fix tests * update changelog * fix type * check expected time index for historical forecasts --------- Co-authored-by: dennisbader <dennis.bader@gmx.ch>
* Add new feature to plot each series's component separately * Update the docstring * refactor the show_anomalies_from_scores() * Update changelog * Improve code in utils.py * Improve code in utils.py * make check_multivariate depend on multivariate_plot * update utils * improve the spacing between suptitle and axes * update utils * improve the spacing between suptitle and axes * Fix the height_ratios of each subplot * Change "multivariate_plot" parameter name to "component_wise" * make title fit better --------- Co-authored-by: madtoinou <32447896+madtoinou@users.noreply.github.com> Co-authored-by: dennisbader <dennis.bader@gmx.ch>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #.
Summary
Other Information