Beta 0.5.0

New / major improvements

• 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

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
• Remove residuals in forecast dataframe returned by m.predict() by @judussoari in #927

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
• 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