All pull requests

Pull requests

• Add ZeroInflatedLikelihood

  #3245·AbhayJuloori opened Oct 10, 2026Contributor·3 tasks done·

  ·
  1

  2

• Fix NegativeBinomialLikelihood parameter order for multivariate series

  #3243·AbhayJuloori opened Oct 10, 2026Contributor·3 tasks done·

  ·
  1

  7

• fix: correct parameter ordering in NegativeBinomialLikelihood for multivariate series

  #3242·shyamsharmas124-commits opened Oct 9, 2026Contributor·

  ·
  1

  1

• Feature: add TiRex-2 foundation model

  #3222·daidahao opened Sep 28, 2026Contributor·3 tasks done·

  ·
  1

  4

• Add step-wise future covariates lags to SKLearnModel

  #3213·JuanCruzC97 opened Sep 24, 2026Contributor·3 tasks done·

  ·
  1

  5

• Fix two error messages that printed their placeholders

  #3211·David-Wu1119 opened Sep 23, 2026·3 tasks done·

  ·
  1

• Expose last hidden state on RNNModel and BlockRNNModel

  #3190·AmjadAAYD opened Aug 22, 2026·

  ·
  1

  2

• Add TSPulse filter for reconstruction and imputation

  #3187·ishitta-iyer opened Aug 21, 2026·3 tasks done·

  ·
  2

• Support multi-batch inputs in TFTExplainer

  #3182·mahimn01 opened Aug 18, 2026Contributor·3 tasks done·

  ·
  1

  3

• Add @overload type signatures for series2seq and seq2series

  #3178·authierj opened Aug 14, 2026Contributor·1 of 3 tasks·

  ·
  1

• Add effective_sample_size() and document that overlapping backtest windows are not independent

  #3176·ipezygj opened Aug 9, 2026·

  ·
  4

• Enable RINorm for past and future covariates

  #3175·Kurokabe opened Aug 4, 2026Collaborator·3 tasks done·

  ·
  1

  1

• Add per-timestep TFT variable importance plot

  #3174·exactml opened Aug 1, 2026Contributor·3 tasks done·

  ·
  1

• [Feature] Efficient DTW in C++

  #3141·daidahao opened Jun 20, 2026Contributor·2 of 3 tasks·

  ·
  1

  5

• feat/3037 phase 1 refactor return types

  #3131·ozink-u8 opened Jun 4, 2026Contributor·3 tasks done·

  ·
  19

• Add save/load methods to anomaly detection classes

  #3130·robrui opened Jun 4, 2026Contributor·3 tasks done·

  ·
  1

  2

• fix(examples): handle NaNs in 14-transfer-learning air dataset

  #3116·jbbqqf opened May 9, 2026Contributor·

  ·
  1

  4

• feat(timeseries): add idxmin / idxmax to TimeSeries

  #3115·jbbqqf opened May 9, 2026Contributor·

  ·
  1

  9

• Docs/fit dataset error

  #3086·jakubchlapek opened Apr 24, 2026Collaborator·3 tasks done·

  ·
  1

  7

• Add Reverso foundation model for zero-shot time series forecasting

  #3061·shinfxh opened Apr 6, 2026·3 tasks done·

  ·
  1

  27

• xarray dependency migration

  #2974·jakubchlapek opened Dec 19, 2025Collaborator·3 tasks done·

  ·
  5

• add os.Path support for load/save torch models

  #2947·turbotimon opened Nov 14, 2025Contributor·3 tasks done·

  ·
  7

• Direct Recursive Strategy for prediction models

  #2941·authierj opened Nov 4, 2025Contributor·2 of 6 tasks·

  ·
  1

  1

• feat: Added frequency aware one-hot and relative cyclic encoding.

  #2893·konsram opened Sep 7, 2025·2 of 3 tasks·

  ·
  1

  3

• Bugfix: Recommend pytorch-lightning>=2.5.3 for Google Colab notebook

  #2885·daidahao opened Aug 27, 2025Contributor·3 tasks done·

  ·
  1

  8