Releases: ourownstory/neural_prophet
1.0.0rc10
Changes from Beta 0.9.0 are also included.
Major Changes
- [Major] Dataloader: Just-In-Time tabularization by @ourownstory in #1529
Other Changes
- [Devops] speed up future-reg-nn tests by @ourownstory in #1596
- [Minor] Make tests deterministic by @MaiBe-ctrl in #1600
- [Minor] Improve Season glocal reg invalid parameter handling by @ourownstory in #1601
Full Changelog: 0.9.0...1.0.0rc10
Beta 0.9.0
Major Changes
- [DevOps] Upgrade to lightning 2.0 by @ourownstory in #1514
New Features
- [Minor] add subdivisions to holidays by @MaiBe-ctrl in #1584
- [Minor] Support io.bytes output by @MaiBe-ctrl in #1583
Fixes
- [Fix] fix bug #1580 by @MaiBe-ctrl in #1582
- [Fix] fix future regressor by @MaiBe-ctrl in #1585
- [Fix| Fix neural nets regressor shape by @MaiBe-ctrl in https://github.com//pull/1589
- [Fix] Raise Error if ar_reg is set an dn_lags is null by @MaiBe-ctrl in #1588
- [Fix] Fix frequency inference if NaT by @MaiBe-ctrl in #1590
- [Docs] fix typo in get_latest_forecast example by @SimonWittner in #1551
Other Changes
- [DevOps] Cleaning CI workflow files and dependencies by @ourownstory in #1547
- [DevOps] update poetry by @ourownstory in #1587
- [DevOps] Run black by @ourownstory in #1593
- [DevOps] bump version to 0.9.0 by @ourownstory in #1594
New Contributors
- @MaiBe-ctrl made their first contribution in #1582
Full Changelog: 0.8.0...0.9.0
1.0.0rc9
What's Changed
- [Major] Add support for Python 3.12, drop support for Python 3.8 by @ourownstory in #1544
- [Docs] update website to 1.0.0rc8 by @ourownstory in #1538
- [Devops] Upgrade dependencies to more recent range of dependencies by @ourownstory in #1543
Full Changelog: 1.0.0rc8...1.0.0rc9
Beta 0.8.0
Note: identical to 1.0.0rc9
Major Changes
- [Major] Glocal Modelling V2: Local-Global mixture models by @alfonsogarciadecorral in #1008
- [Major] Add support for Python 3.12, drop support for Python 3.8 by @ourownstory in #1544
Other Changes
- [Fix] Accessing None type config_refgressors by @ourownstory in #1536
- [Docs] update website to 1.0.0rc8 by @ourownstory in #1538
- [Devops] Upgrade dependencies to more recent range of dependencies by @ourownstory in #1543
Full Changelog: 0.7.0...0.8.0
1.0.0rc8
What's Changed
- [Fix] Accessing None type config_refgressors (introduced by #1008) by @ourownstory in #1536
Full Changelog: 1.0.0rc7...1.0.0rc8
1.0.0rc7
What's Changed
- [Major] Glocal Modelling v2 by @alfonsogarciadecorral in #1008
Full Changelog: 0.7.0...1.0.0rc7
1.0.0rc6
Major changes
- [Major] Throw RuntimeError when trying to fit model second time (Fixes #1493) by @ourownstory in #1531
- [Major] reduce batch size for small datasets, reduce overall epochs for most datasets by @ourownstory in #1533
Fixes
- [Fix] loading model on machine with different accelerator by @McOffsky in #1509
- [Fix] Plot_parameter - Yearly Seasonality by @SimonWittner in #1501
Documentation
- [Docs] Fix MLFlow Sphinx test error by @ourownstory in #1508
- [Docs] MLflow tutorial by @SimonWittner in #1494
- [Docs] Adding experiment to MLflow_and_NeuralProphet.ipynb by @cargecla1 in #1499
Dependencies and tests
- [Devops] Bump jinja2 from 3.1.2 to 3.1.3 by @dependabot in #1506
- [Devops] Upgrade to Pandas 2.0.0 by @SimonWittner in #1498
- [Devops] Upgrade to torch >=2.0.0 by @ourownstory in #1512
- [Devops] Support python 3.11 by @ourownstory in #1513
- [Devops] Upgrade tests to Python 3.11 and remove windows by @ourownstory in #1515
- [Devops] Poetry Dev dependencies update by @ourownstory in #1530
New Contributors
- @McOffsky made their first contribution in #1509
- @cargecla1 made their first contribution in #1499
Full Changelog: 1.0.0rc5...1.0.0rc6
Beta 0.7.0
Note: Identical to 1.0.0rc6
About
This is the first non-pre-release of a large set of changes destined for 1.0.0. We decided to do a regular beta release, as there should be no breaking changes and the main branch has been without major issues for a few months. This release contains many fixes and improvements compared to 0.6.x.
Major changes
Introduced in 1.0.0rc1:
- [major] speed up nested additive model by @LeonieFreisinger in #1311
- [major] Drop Python 3.7 support and complete poetry migration by @noxan in #1271
Introduced in 1.0.0rc3:
- [major] Speed up split_df by vectorizing for loops by @leoniewgnr in #1356
Introduced in 1.0.0rc4:
Introduced in 1.0.0rc5:
- [major] Training: increase default batch size by @ourownstory in #1462
- [major] Training: update 1cycle lr to 3phase by @ourownstory in #1450
- [major] Training: SmoothL1Loss update beta from 1.0 to 0.3 by @ourownstory in #1461
Introduced in 1.0.0rc6:
- [Major] Throw RuntimeError when trying to fit model second time (Fixes #1493) by @ourownstory in #1531
- [Major] reduce batch size for small datasets, reduce overall epochs for most datasets by @ourownstory in #1533
Everything that has changed since 0.6.0
- [major] Drop Python 3.7 support and complete poetry migration by @noxan in #1271
- [docs] Revamp of feature guides by @leoniewgnr in #1301
- [website] Fix the deployment of the website with poetry by @noxan in #1307
- Uncertainty quantification tutorial by @christymctse in #1225
- [minor] Add static plotly backend by @hxyue1 in #1286
- [docs] Update documentation website by @leoniewgnr in #1309
- [minor] Update dependencies to latest versions by @noxan in #1305
- [docs] added reproducibility to tutorial 10 by @leoniewgnr in #1310
- [minor] Unify version number handling, remove setup.py and migrate content by @noxan in #1316
- [refactor] Uncertainty streamline naming by @leoniewgnr in #1262
- [major] speed up nested additive model by @LeonieFreisinger in #1311
- bump version number to 1.0.0rc1 by @leoniewgnr in #1319
- [fix] Downgrade kaleido version to 0.2.1 to fix installs by @noxan in #1320
- bump version number to 1.0.0rc2 by @leoniewgnr in #1321
- Bump flask from 2.3.1 to 2.3.2 by @dependabot in #1318
- [docs] fix showing of plots in tutorials and feature guides by @leoniewgnr in #1322
- update selective forecasting tests by @ourownstory in #1312
- [devops] Remove macos and windows from Github actions by @noxan in #1325
- [docs] prettify application examples by @leoniewgnr in #1328
- [devops] Add ruff github action by @noxan in #1324
- [refactor] Ruff fixes by @noxan in #1327
- [docs] Fix prophet tutorial by @leoniewgnr in #1332
- [docs] update prophet tutorials by @leoniewgnr in #1334
- Remove type conversion for _stride_lagged_features and _stride_future… by @hxyue1 in #1331
- Bump requests from 2.30.0 to 2.31.0 by @dependabot in #1337
- [docs] Clear plotting tutorial by @leoniewgnr in #1343
- (Fix) pytorch_lightning<2.0.0 callbacks named ProgressBarBase instead of ProgressBar by @JSarsfield in #1344
- [minor] fixed prediction_frequency when quantiles are set by @leoniewgnr in #1347
- [minor] fix sum of components by @leoniewgnr in #1349
- fix tutorials link in readme by @ShreyaKhurana in #1353
- [major] Speed up split_df by vectorizing for loops by @leoniewgnr in #1356
- [docs] ar layers fixed in uncertainty tutorial by @leoniewgnr in #1364
- Fixes failing tests by @leoniewgnr in #1381
- [docs] Small fixes in documentation by @leoniewgnr in #1380
- [docs] New feature guide with industry application and tool by @leoniewgnr in #1369
- [bug] Fix pyright and flake8 warnings by @leoniewgnr in #1392
- Update poetry dependencies by @noxan in #1390
- [bug] Trend plotting when using numpy 1.24 by @leoniewgnr in #1395
- bump version number to 1.0.0rc3 by @leoniewgnr in #1396
- [bug] Fix correct definition of torchmetrics inside pytorch lightning module by @leoniewgnr in #1365
- [minor] Memory optimized stride function by @leoniewgnr in #1400
- [minor] Speed up dataloader by @leoniewgnr in #1360
- [minor] Memory efficient float32 types instead of float64 types by @leoniewgnr in #1402
- [bug] fixed ruff warnings by @leoniewgnr in #1408
- [minor] Add a verbose option to NeuralProphet.test by @c3-ziqin in #1407
- [fix] Remove wrong plotting warning by @Maisa-bs in #1411
- [major] Upgrade to PyTorch 2 by @noxan in #1404
- Bump tornado from 6.3.2 to 6.3.3 by @dependabot in #1412
- Add the trainer back to the forecaster after saving model to file by @c3-ziqin in #1416
- [fix] Tutorials on Colab Crash by @SimonWittner in #1418
- bump version number to 1.0.0rc4 by @leoniewgnr in #1424
- [docs] Removed incorrect comments. by @thekoc in #1421
- [fix] Reduce saved model size by @leoniewgnr in #1425
- [fix] Correct holiday test and update version by @leoniewgnr in #1431
- [docs] typo in README.md by @eltociear in #1436
- [fix] Pytest for fix ValueError when plotting events components by @SimonWittner in #1434
- [minor] Helperfunction to support equidistant data without datestamps by @SimonWittner in #1432
- [devops] bump version-number to 1.0.0rc5 by @ourownstory in #1440
- [devops] Update poetry lock to most recent dependencies by @ourownstory in #1439
- [devops] Dev dependencies groups restructure by @ourownstory in #1442
- [docs] Update CONTRIBUTING.md Poetry dev install by @ourownstory in #1357
- [devops] bump-pandas-stubs-2 by @ourownstory in #1441
- [Devops] group dev dependencies and update lock by @ourownstory in #1448
- [Devops] Skip windows python 3.8 tests in ci.yml by @ourownstory in #1449
- [fix] Pytests Plots by @SimonWittner in #1445
- [tests] Model tests expansion (partially undone in #1464) by @ourownstory in #1455
- [Devops] Bump urllib3 from 2.0.6 to 2.0.7 by @dependabot in #1452
- [Devops] update metrics ci with new tests by @ourownstory in #1457
- [Devops] Relax dependency ranges to next major number by @ourownstory in https...
1.0.0rc5
What's Changed
Breaking
(None)
Major
- [major] Training: increase default batch size by @ourownstory in #1462
- [major] Training: update 1cycle lr to 3phase by @ourownstory in #1450
- [major] Training: SmoothL1Loss update beta from 1.0 to 0.3 by @ourownstory in #1461
Minor
- [minor] Helperfunction to support equidistant data without datestamps by @SimonWittner in #1432
- [minor] Torch Load by @SimonWittner in #1473
Bugs
- [fix] Reduce saved model size by @leoniewgnr in #1425
- [fix] Correct holiday test and update version by @leoniewgnr in #1431
- [fix] Pytest for fix ValueError when plotting events components by @SimonWittner in #1434
- [fix] Pytests Plots by @SimonWittner in #1445
Documentation
- [docs] Removed incorrect comments. by @thekoc in #1421
- [docs] typo in README.md by @eltociear in #1436
- [docs] Update CONTRIBUTING.md Poetry dev install by @ourownstory in #1357
- [Docs] fix-tutorial-data-descript daily not hourly by @ourownstory in #1453
- [docs] SmoothL1Loss correctly mentioned instead of Huber by @ourownstory in #1458
- [docs] README.md typos by @VinayKokate22 in #1472
- [docs] fix typos by @Mayureshd-18 in #1470
- [docs] Update README.md by @harshhere905 in #1467
- [docs] update lagged regressor tutorial by @ourownstory in #1463
- [docs] added full_stops by @bhargavshirin in #1465
- [docs] Added contributors section to README.md by @Kalyanimhala in #1474
- [docs] fix typo in df_utils.py by @eltociear in #1478
Tests
- [tests] Model tests expansion (partially undone in #1464) by @ourownstory in #1455
- [tests] add new test and remove 3 duplicate tests by @ourownstory in #1464
Dev Ops
- [devops] bump version-number to 1.0.0rc5 by @ourownstory in #1440
- [devops] Update poetry lock to most recent dependencies by @ourownstory in #1439
- [devops] Dev dependencies groups restructure by @ourownstory in #1442
- [devops] bump-pandas-stubs-2 by @ourownstory in #1441
- [Devops] group dev dependencies and update lock by @ourownstory in #1448
- [Devops] Skip windows python 3.8 tests in ci.yml by @ourownstory in #1449
- [Devops] Bump urllib3 from 2.0.6 to 2.0.7 by @dependabot in #1452
- [Devops] update metrics ci with new tests by @ourownstory in #1457
- [Devops] Relax dependency ranges to next major number by @ourownstory in #1459
- [devops] Bump werkzeug from 3.0.0 to 3.0.1 by @dependabot in #1468
- [devops] Bump aiohttp from 3.8.6 to 3.9.0 by @dependabot in #1484
New Contributors
- @thekoc made their first contribution in #1421
- @eltociear made their first contribution in #1436
- @VinayKokate22 made their first contribution in #1472
- @Mayureshd-18 made their first contribution in #1470
- @harshhere905 made their first contribution in #1467
- @bhargavshirin made their first contribution in #1465
- @Kalyanimhala made their first contribution in #1474
Full Changelog: 1.0.0rc4...1.0.0rc5
1.0.0rc4
What's Changed
Major changes
Minor changes
- [minor] Memory optimized stride function by @leoniewgnr in #1400
- [minor] Speed up dataloader by @leoniewgnr in #1360
- [minor] Memory efficient float32 types instead of float64 types by @leoniewgnr in #1402
- [minor] Add a verbose option to NeuralProphet.test by @c3-ziqin in #1407
Bugs fixed
- [bug] Fix correct definition of torchmetrics inside pytorch lightning module by @leoniewgnr in #1365
- [bug] fixed ruff warnings by @leoniewgnr in #1408
- [fix] Remove wrong plotting warning by @Maisa-bs in #1411
- Add the trainer back to the forecaster after saving model to file by @c3-ziqin in #1416
- [fix] Tutorials on Colab Crash by @SimonWittner in #1418
New Contributors
- @c3-ziqin made their first contribution in #1407
- @Maisa-bs made their first contribution in #1411
- @SimonWittner made their first contribution in #1418
Full Changelog: 1.0.0rc3...1.0.0rc4