Improvement
- Added support for async emhass code (@gieljnssns)
- Added support for websocket data retrieval (@gieljnssns)
- Add feature to cache pv regression model (@sokorn)
- Added support for a more comprehensive list of
scikit-learnmodels for both ML Forecaster and Regressor classes - Improved documentation re-arranged sections
- Thermal model improvements:
- Implemented a Min/Max temperature control, proposed by @werdnum
- Implemented needed unit tests
- Improved documentation
Fix
- Fixed warning message because pickle files saved with old Numpy version
- A buch of code refactor following SonarCloud Code Analysis
- Solved runtime param optimization_time_step not recognized, issue #563
- Attempt at solving data publish problem when containing both battery sensor definitions and a def_load_config, issue #587
- Fix timezone comparison issue with InfluxDB (@sokorn)
- Update skforecast requirement from <0.19.0,>=0.18.0 to >=0.18.0,<0.20.0
- Implement inverter AC input limit constraint
What's Changed
- Thermal model improvements by @davidusb-geek in #637
- Emhass async + websocket data retrieval by @gieljnssns in #626
- Code fix sonar cube by @davidusb-geek in #640
- Add feature to cache pv regression model by @sokorn in #638
- Implement inverter AC input limit constraint by @davidusb-geek in #643
- Fix timezone comparison issue with InfluxDB by @sokorn in #642
- Update skforecast requirement from <0.19.0,>=0.18.0 to >=0.18.0,<0.20.0 by @dependabot[bot] in #641
- Add new inverter and module data entries to CSV and database files by @sokorn in #645
- Added support for a more comprehensive list of Scikit-Learn models by @davidusb-geek in #644
Full Changelog: v0.14.1...v0.15.0