v0.2.2

What's Changed

• Naive models by @Sburyachenko in #159
• TimeSeriesForecastingModel by @deatinor in #160
• Hierarchycal models by @deatinor in #161
• Horizon as list by @deatinor in #164
• Exogenous feature by @deatinor in #166
• Time Series Split by @Srinidhi-Patil in #163
• moving custom function applied to acoustics by @Hella in #145
• Multi feature multi output regressor by @deatinor in #169
• Explainability by @deatinor in #170
• Explainability by @deatinor in #173
• Refactoring by @deatinor in #176
• [WIP] CV model selection by @Sburyachenko in #178
• kwargs in TimeSeriesForecastingModel by @deatinor in #182
• Cv model selection by @Sburyachenko in #181
• CV model selection by @Sburyachenko in #183
• Added modification on granger_causality to boost the code by @niko992 in #184
• added 2 of the three algorithms for the top down approach+tests by @niko992 in #179
• Cv by @deatinor in #185
• Fixed granger causality by @deatinor in #186
• Added support for calendar feature and datetime_index by @deatinor in #187
• fixing tests and moving to actions by @matteocao in #188

New Contributors

• @Hella made their first contribution in #145
• @niko992 made their first contribution in #184

Full Changelog: v0.2.1...v0.2.2

Pandas v1.0 compatibility

This release introduces compatibility with Pandas v1.

New API

• New API interface
• Bug fixes
• New features

Please check the release notes for more details.

p_values for causality test.

The causality test also computes the p_values to check show at which confidence level the test is significant.

Minor fixes

The purpose of this minor release is to improve the documentation by fixing some errors and typos in the README

First Public Release

v0.1.0

Fixed README and chenged version for release

First upload on PyPi

v0.1a.0

Fix in check_is_fitted method