Skip to content

etna 2.6.0
Compare
Choose a tag to compare
@d-a-bunin d-a-bunin released this 11 Apr 16:02
· 62 commits to master since this release
2.6.0
755cdab
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

Highlights

  • Add ability to work with integer timestamp by using freq=None
  • Add tutorial about working with misaligned data by using integer timestamp
  • Add TSDataset.create_from_misaligned constructor
  • Add infer_alignment, apply_alignment, make_timestamp_df utils into etna.dataset
  • Add BinaryOperationTransform
  • Add TFTNativeModel
  • Add in OutliersTransform possibilities use ignore_flag_column to skip values use ignore
  • Extend base TSDataset constructor to handle long format dataframes
  • Speed up timestamp transforms
  • Update timestamp transforms to work with external timestamp by using in_column parameter

Full changelog

Added

  • Add BinaryOperationTransform to transforms (#260)
  • Add TFTNativeModel (#290)
  • Add warning on trying to pass numeric timestamp if freq is not None and add _cast_index_to_datetime (#214)
  • Add infer_alignment, apply_alignment, make_timestamp_df into etna.dataset.utils (#256)
  • Add TSDataset.create_from_misaligned constructor (#269)
  • Add tutorial about working with misaligned data (#288)
  • Add in OutliersTransform possibilities use ignore_flag_column to skip values use ignore (#291)

Changed

  • Update glossary with terms related to working with misaligned data (#288)
  • Add ignoring of integer timestamp as a feature into native DL models (#210)
  • Update pytorch_forecasting models to handle integer timestamp (#208)
  • Update datasets module to work with integer timestamp (#146)
  • Add tests for transform on data with integer timestamp (#153)
  • Add tests for models on data with integer timestamp (#188)
  • Update DateFlagsTransform, TimeFlagsTransform, HolidayTransform, SpecialDaysTransform, FourierTransform to work with external timestamp (#169)
  • Update analysis module to work with integer timestamp (#161)
  • Update StatsForecastARIMAModel, StatsForecastAutoARIMAModel, StatsForecastAutoCESModel, StatsForecastAutoETSModel, StatsForecastAutoThetaModel to handle integer timestamp (#197)
  • Update MRMRFeatureSelectionTransform to handle integer timestamp (#164)
  • Update deseasonality transforms (STLTransform, DeseasonalityTransform) to handle integer timestamp (#174)
  • Update HoltModel, HoltWintersModel, SimpleExpSmoothingModel, SARIMAXModel, AutoARIMAModel to handle integer timestamp ((#200)[https://github.com//pull/200])
  • Update detrend transforms (LinearTrendTransform, TheilSenTrendTransform) to handle integer timestamp (#163)
  • Update ResampleWithDistributionTransform to work with integer timestamp (#165)
  • Update change point transforms (ChangePointsSegmentationTransform, ChangePointsTrendTransform, ChangePointsLevelTransform, TrendTransform) to handle integer timestamp (#176)
  • Update BATSModel, TBATSModel models to work with integer timestamp (#195)
  • Update ProphetModel to handle external timestamp (#203)
  • Remove checking frequency in timestamp_column of ProphetModel (#222)
  • Update FourierTransform to handle external datetime timestamp (#223)
  • Update FoldMask to work with integer timestamp, in validate_on_dataset method add validation on presence of FoldMask parameters in ts.index, add tests for FoldMask (#226)
  • Fix FourierTransform on integer index, add inference tests (#230)
  • Update outliers transforms to handle integer timestamp (#229)
  • Update pipelines to handle integer timestamp (#241)
  • Add timestamp_range and refactor code with it (#244)
  • Update CLI to handle integer timestamp (#246)
  • Update ExogShiftTransform to handle integer timestamp (#254)
  • Extend base TSDataset constructor to handle long format dataframes, update documentation and tutorials with this change (#266)
  • Update internal datasets to work with unaligned data (#292)
  • Speed up "timestamp" transforms (#295

Fixed

  • Fix PredictionIntervalOutliersTransform fails to work with created columns (#291)
  • Prohibit empty list value and duplication of target_timestamps parameter in FoldMask (#226)
  • Fix DeseasonalityTransform fails to inverse transform short series (#174)
  • Fix indexing in stl_plot, plot_periodogram, plot_holidays, plot_backtest, plot_backtest_interactive, ResampleWithDistributionTransform (#244)
  • Fix DifferencingTransform to handle integer timestamp on test (#244)
  • Fix HolidayTransform to handle integer timestamp in days_count mode (#285)
Assets 2
Loading