What's Changed
- Lightning Migration by @karl-richter in #837
- GPU support by @karl-richter in #961
- Support plotting of only specific panels in plot_parameters and plot_components by @LeonieFreisinger in #884
- Uncertainty: Conformal Prediction V1 with .conformal_predict() by @Kevin-Chen0 in #802 and #1044
- Support for holidays of multiple countries by @leoniewgnr in #1001
- Deep model parameter interpretation by @karl-richter in #883
- Add type annotations for main NeuralProphet class by @noxan in #981
Migrations and breaking changes
- Remove support for df dict as an input by @judussoari in #929
- Change plotting_backend deprecation warning for implicit and explicit matplotlib use by @LeonieFreisinger in #1006
- Refactored fit() interface by @karl-richter in #1039
List of all changes
- [lightning] Remove rich progress bar by @karl-richter in #952
- Improvement/type annotations by @noxan in #954
- Commented out test_progress_display in test_integration.py. by @Kevin-Chen0 in #956
- Added plot param of the NeuralProphet fit() and uncomment test_progress_display in test_integration.py. by @Kevin-Chen0 in #958
- [lightning] refactored Trainer args by @karl-richter in #959
- refactored test failure by @karl-richter in #960
- Add alias function plot_last_forecast() for plot_latest_forecast() and give deprecation warning by @judussoari in #919
- Mask seasonality on Sat/Sun when data frequency is Business days by @judussoari in #808
- [bug] Forecast output dataframe contains all float dtypes instead of having some objects by @Kevin-Chen0 in #978
- Fix labels of yhatx and origin-x in plots by @judussoari in #942
- Improve github action speed for docs test by @noxan in #936
- Remove residuals in forecast dataframe returned by m.predict() by @judussoari in #927
- Resolved discontinuous trend when changepoints are not sorted chronologically by @leoniewgnr in #968
- Address small tech debt to use a conditional instead of try/except for Python 3.7 support by @kapoor1992 in #934
- [dev-ops] Collapsible metrics reports by @noxan in #926
- Cleanup document files by @noxan in #975
- Rename plot_forecast.py and plot_model_parameters.py to xxx_matplotli… by @SaumyaBhushan in #971
- [naming] Renamed tutorial notebooks by @karl-richter in #921
- Labels in CONTRIBUTING.md by @karl-richter in #946
- [enhancement] add pytests for matplotlib by @LeonieFreisinger in #914
- [devops] Improve codecov threshold by @noxan in #980
- [devops] Upgrade GitHub action versions by @noxan in #979
- change deprecation message to switch plotting backend to plotly by @LeonieFreisinger in #984
- [feature] Plot components of averaged df (and quants) if multiple IDs, but no df_name specified in plot_components() by @judussoari in #888
- [devops] Code coverage threshold fix by @noxan in #990
- Time in model training benchmark by @karl-richter in #995
- Unified the naming of variables - num and nb by @SaumyaBhushan in #994
- [tests] Add tests for configure by @noxan in #983
- [cleanup] Remove unused from_kwargs config function by @noxan in #988
- [devops] Add shell script to setup basic development environment by @noxan in #986
- [lightning] Speed-up by @karl-richter in #973
- Update progress bar only every epoch by @karl-richter in #1003
- [typing] Resolve type annotation issues in configure file by @noxan in #987
- [typing] Add ConfigEvents type annotations by @noxan in #989
- [refactor] Extract plot utils into dedicated file by @LeonieFreisinger in #1000
- [website] Fix typo in preview of website description by @noxan in #1012
- Move tensors to correct device by @karl-richter in #1010
- [docs] New tutorials section for docs by @noxan in #976
- Meta name bool by @alfonsogarciadecorral in #997
- [feature] Add a main script to print neuralprophet version by @noxan in #974
- [dev-ops] Type annotation validation GitHub action by @noxan in #902
- Add pull request template for Github by @karl-richter in #915
- [typing] fix typings and constructor super call for pinnball loss by @noxan in #1019
- Follow-up tensors on wrong device by @karl-richter in #1023
- Fix string Interpolation by @poweroftrue in #1026
- Fix label links in contributing guidelines by @noxan in #1024
- [bugfix] enable events while predicting into known future by @judussoari in #1029
- [fix] Replace all deprecated plot_last_forecast with plot_latest_forecast by @noxan in #1022
- [refactor] Rename config_seasonality and add typings by @noxan in #1021
- [refactor] Remove duplicate season reg lambda by @noxan in #1020
- [fix] Renamed remaining self.config_season variables to self.config_seasonality in time_net.py by @Kevin-Chen0 in #1035
- Removed commented tests in test_uncertainty.py. by @Kevin-Chen0 in #1032
- [refactor] Rename uncertainty_estimation.ipynb to uncertainty_quantile_regression.ipynb by @Kevin-Chen0 in #1034
- Normalize training speed metric by @karl-richter in #1041
- Improved system performance tracker by @karl-richter in #1042
- Update metrics.yml by @karl-richter in #1043
- [devops] Cleanup and limit CodeQL Github action to main branch by @noxan in #1031
- [docs] Tuturials use prepared dataset by @noxan in #1037
- Refactored dataloader by @karl-richter in #1040
- Normalize training time metric by @karl-richter in #1045
- Removed unnecessary if statement in predict_season_from_dates by @judussoari in #1049
New Contributors
- @leoniewgnr made their first contribution in #968
- @SaumyaBhushan made their first contribution in #971
- @poweroftrue made their first contribution in #1026
Full Changelog: 0.4.2...0.5.0