skforecast
https://github.com/joaquinamatrodrigo/skforecast
Jupyter Notebook
Time series forecasting with scikit-learn models
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Jupyter Notebook not yet supported1 Subscribers
Add a CodeTriage badge to skforecast
Help out
- Issues
- Avoid copying the public training matrix with pandas 3 and the exog before encoding categorical features
- [BUG]: `RollingFeatures` gives NaN for a window with a missing value when training, but a statistic of the remaining values when predicting
- Add implementation plan for initial_trials in the Bayesian searches
- docs: add user guide on forecasted exogenous variables
- Foundation backends compatibility check is failing
- Fix failing tests: Pandas Datetime Resolution Mismatch
- [FEATURE]: Support for "As-Of" timestamp filtering for exogenous covariates in backtesting
- roadmap towards `skforecast` as primary reduction engine used by `sktime`
- Best way to use optuna pruning in skforecast search_forecaster?
- [FEATURE]: suport for cudf.pandas
- Docs
- Jupyter Notebook not yet supported