Owner

Nixtla

10 indexed repositories · View on GitHub

statsforecast
Lightning ⚡️ fast forecasting with statistical and econometric models.

Python · 4909 stars
neuralforecast
Scalable and user friendly neural :brain: forecasting algorithms.

Python · 4275 stars
nixtla
TimeGPT-2.1: production ready pre-trained Time Series Foundation Model for forecasting and anomaly detection. Generative pretrained tran...

Jupyter Notebook · 4009 stars
mlforecast
Scalable machine 🤖 learning for time series forecasting.

Python · 1278 stars
hierarchicalforecast
Probabilistic Hierarchical forecasting 👑 with statistical and econometric methods.

Python · 760 stars
datasetsforecast
Datasets for time series forecasting

Python · 129 stars
utilsforecast
Helper functions to plot, evaluate, preprocess and engineer features for forecasting

Python · 109 stars
coreforecast
Fast implementations of common forecasting routines

Python · 43 stars
nixtlar
R SDK for TimeGPT

R · 40 stars
docs

JavaScript · 4 stars

Beginner friendly Everything

Open All

[Models] Support historical-only exogenous variables in DeepAR Open

Difficulty 4/5 3-5 days Newbie friendliness 38/100

Nixtla/neuralforecast#1622 · Opened Sep 19, 2026
Tech debt: pre-existing duplication/coupling issues surfaced by _payloads.py / _preprocessing.py split Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

Nixtla/nixtla#899 · Opened Sep 18, 2026
xLSTM: decoder_activation is neither validated nor warned about when decoder_layers=1 Open

Difficulty 2/5 1-3 hours Newbie friendliness 86/100

Nixtla/neuralforecast#1616 · Opened Sep 14, 2026
AutoSOFTSSharp with Ray backend fails with CUDNN_STATUS_SUBLIBRARY_VERSION_MISMATCH on NVIDIA A800 CUDA 13.0 Open

awaiting response bug

Difficulty 4/5 3-5 days Newbie friendliness 42/100

Nixtla/neuralforecast#1614 · 1 comment · Opened Sep 14, 2026
[scalers] Box-Cox lambda is the optimizer's starting point, not an estimate, when the data has a NaN Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

Nixtla/coreforecast#146 · Opened Sep 9, 2026
AutoARIMA: drop collinear exogenous regressors instead of fitting them Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Nixtla/statsforecast#1235 · 1 comment · Opened Sep 9, 2026
polars date features diverge from the pandas convention: 5 raise AttributeError, weekday silently returns 1-7 Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 75/100

Nixtla/mlforecast#738 · 1 comment · Opened Sep 9, 2026
Target transforms corrupt conformal calibration with refit=False cross-validation Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Nixtla/mlforecast#733 · Opened Sep 7, 2026
[Flaky] test_cv_forecast_consistency fails Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 45/100

Nixtla/nixtla#863 · 4 comments · Opened Aug 21, 2026
prediction interval in transfer learning with exogenous inputs Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Nixtla/mlforecast#721 · Opened Aug 21, 2026
Seasonality and trend extraction in the presence of level shifts Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

Nixtla/statsforecast#1214 · Opened Aug 21, 2026
[Models] ConformalSeasonalPool: documented n_samples sufficiency rule covers only the lower bound and its example is off by one; no validation or warning when the pool is too small Open

Difficulty 3/5 1-2 days Newbie friendliness 72/100

Nixtla/statsforecast#1202 · 1 comment · Opened Aug 4, 2026
[lag_transforms] Honor `skipna` in the stateful `update()` path (ExpandingMean/Std/Min/Max, ExponentiallyWeightedMean) Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

Nixtla/coreforecast#142 · Opened Jul 30, 2026
[Core] Allow NaN target values at `fit`/`preprocess`, dropping them only before model fitting Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 38/100

Nixtla/mlforecast#704 · Opened Jul 29, 2026
Accept a one-row-per-id DataFrame for `static_features` Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 35/100

Nixtla/mlforecast#702 · 3 comments · Opened Jul 24, 2026
[Enhancement] Support empirical bootstrap prediction intervals in `forecast_arima` and `forecast_ets` Open

Difficulty 5/5 Over a week Newbie friendliness 45/100

Nixtla/statsforecast#1191 · Opened Jul 17, 2026
Support `models_fit_kwargs` in LightGBMCV.fit Open

enhancement feature

Difficulty 3/5 1-2 days Newbie friendliness 68/100

Nixtla/mlforecast#691 · 1 comment · Opened Jul 17, 2026
Add support for pandas>=3.0 Open

Difficulty 3/5 1-2 days Newbie friendliness 65/100

Nixtla/statsforecast#1182 · 3 comments · 3 reactions · Opened Jul 14, 2026
[FEAT - evaluation/losses/plotting] Integrate forecast stability evaluation in utilsforecast Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 28/100

Nixtla/utilsforecast#255 · 7 comments · Opened Jun 25, 2026
Allow `predict(ids=...)` with safe pooled/grouped lag-transform subsets Open

enhancement feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Nixtla/mlforecast#649 · Opened May 20, 2026
[Core] Support exogenous transforms inside `MLForecast` and allow raw exogenous inputs to be dropped from the model feature set Open

enhancement feature

Difficulty 4/5 3-5 days Newbie friendliness 48/100

Nixtla/mlforecast#635 · Opened Apr 28, 2026
Conformal intervals become extreme/null after inverse log transform (expm1) — what's the recommended approach? Open

bug

Difficulty 5/5 Over a week Newbie friendliness 35/100

Nixtla/statsforecast#1119 · 6 comments · Opened Mar 23, 2026
Add ensemble method Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 25/100

Nixtla/utilsforecast#220 · Opened Mar 14, 2026
[FEAT] Scalable reconciliation Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

Nixtla/hierarchicalforecast#454 · 1 comment · 1 reaction · Opened Mar 3, 2026
[FEAT] Cross-temporal reconciliation via projection approach Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

Nixtla/hierarchicalforecast#453 · Opened Mar 3, 2026
docs (notebooks): list items not rendered Open

documentation

Difficulty 2/5 1-3 hours Newbie friendliness 54/100

Nixtla/statsforecast#1089 · Opened Feb 10, 2026
docs: equations of exponential smoothing models don't render correctly Open

documentation

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

Nixtla/statsforecast#1088 · Opened Feb 10, 2026
Error using Ray with StatsForecast Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 35/100

Nixtla/nixtla#705 · Opened Nov 12, 2025
OverflowError / OutOfBoundsDatetime when loading Favorita Dataset Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

Nixtla/datasetsforecast#73 · 1 reaction · Opened Oct 23, 2025
Unordered results when finetune_steps have values greater than 1 Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

Nixtla/nixtlar#59 · Opened Oct 20, 2025
TimeGEN-1 Azure API key setup and forecast 422 Error Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Nixtla/nixtlar#58 · 1 comment · Opened Oct 16, 2025
[Test] Failure of test_autonlinear_longer_horizon test in .circleci/config.yml Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Nixtla/neuralforecast#1385 · Opened Sep 16, 2025
[docs] Broken image urls Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 35/100

Nixtla/nixtla#676 · 1 comment · Opened Sep 12, 2025
Collaboration with Tempus Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 15/100

Nixtla/nixtla#674 · Opened Sep 8, 2025
Issue on docs Open

Difficulty 2/5 1-3 hours Newbie friendliness 52/100

Nixtla/docs#19 · Opened Sep 8, 2025
[<Library component: Models|Core|etc...>] How to align TimeGPT evaluation with rolling window strategies used in other time series models (e.g., TimeLLM)? Open

enhancement feature

Difficulty 5/5 Over a week Newbie friendliness 25/100

Nixtla/nixtla#668 · Opened Aug 28, 2025
Issue on docs: deprecated links Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

Nixtla/docs#16 · Opened Aug 19, 2025
Missing Source Links in Documentation Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

Nixtla/docs#15 · Opened Aug 18, 2025
PatchTST: hist_exog_list Parameter does not work Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

Nixtla/docs#14 · Opened Aug 18, 2025
[ARIMA] C++ max_size() error when using long `p` or `seasonalP` terms Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Nixtla/statsforecast#1034 · Opened Aug 8, 2025
Access Denied when downloading data Open

Difficulty 2/5 1-3 hours Newbie friendliness 30/100

Nixtla/datasetsforecast#64 · 1 comment · Opened Aug 5, 2025
Website: How to Guides 🚧 Open

documentation

Difficulty 1/5 Under an hour Newbie friendliness 25/100

Nixtla/statsforecast#1033 · Opened Jul 25, 2025
forecast_fitted_values() only works with h = 1 Open

Difficulty 4/5 3-5 days Newbie friendliness 42/100

Nixtla/statsforecast#1032 · Opened Jul 21, 2025
docs: Some broken links in the documentation Open

documentation

Nixtla/statsforecast#1031 · 2 comments · 1 assignee · Opened Jul 14, 2025
Errors on direct installation from a forked repository Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 25/100

Nixtla/statsforecast#1017 · Opened Jul 3, 2025
Forecast on non-datetime temporal indicator Open

documentation

Difficulty 4/5 3-5 days Newbie friendliness 35/100

Nixtla/nixtla#655 · 1 comment · Opened Jun 27, 2025
Issue on ray library Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

Nixtla/docs#11 · Opened Jun 26, 2025
Cross-validation with refit=True failing on particular datasets using Pandas Open

bug

Nixtla/neuralforecast#1343 · 1 assignee · Opened Jun 18, 2025
[<Library component: Models|Core|etc...>] TBATS parameters labeling Open

enhancement feature

Difficulty 4/5 3-5 days Newbie friendliness 38/100

Nixtla/statsforecast#1026 · Opened Jun 10, 2025
[docs] contribution instructions need to be updated. Open

documentation

Difficulty 1/5 Under an hour Newbie friendliness 45/100

Nixtla/nixtla#650 · Opened Jun 10, 2025

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

Nixtla

Showing the newest 100

Get new issues in your inbox