Flaml

Highlights

* Preserve the checkpoint when deleting `AutoML` objects.
* Create no eval set when setting `use_best_model` to False for catboost.

What's Changed
* add guideline collection by qingyun-wu in https://github.com/microsoft/FLAML/pull/687
* LightGBM notebook update by sonichi in https://github.com/microsoft/FLAML/pull/690
* Add preserve_checkpoint to preserve the checkpoint after del by liususan091219 in https://github.com/microsoft/FLAML/pull/692
* use_best_model for catboost by sonichi in https://github.com/microsoft/FLAML/pull/679


**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.10...v1.0.11

This release contains several new features to highlight:
* A major new feature is to support multiple time series in one dataset with a new task named "ts_forecast_panel" and a neural network estimator from pytorch-forecast.
* Allow disabling shuffle for custom splitter.
* Allow explicit specification of whether the choices of a hp have an inherent order.
* Allow skipping data transformation to avoid overhead.
* Support AzureML pipeline tuning.
* Allow log file name to be specified in `tune.run` and perform logging when ray is used.

There are other improvements for the transformer estimator and bug fixes for config constraints.

What's Changed
* Fixing the issue that FLAML trial number is significantly smaller than Transformers.hyperparameter_search by liususan091219 in https://github.com/microsoft/FLAML/pull/657
* make test result more stable by sonichi in https://github.com/microsoft/FLAML/pull/646
* Add pipeline tuner component and dependencies. by ruizhuanguw in https://github.com/microsoft/FLAML/pull/671
* Skip transform by jmrichardson in https://github.com/microsoft/FLAML/pull/665
* pull request template by sonichi in https://github.com/microsoft/FLAML/pull/668
* Update Research.md by liususan091219 in https://github.com/microsoft/FLAML/pull/672
* Documentation on search space and parallel/sequential tuning by qingyun-wu in https://github.com/microsoft/FLAML/pull/675
* time series forecasting with panel datasets by int-chaos in https://github.com/microsoft/FLAML/pull/541
* categorical choice can be ordered or unordered by sonichi in https://github.com/microsoft/FLAML/pull/677
* Disable shuffle for custom CV by jmrichardson in https://github.com/microsoft/FLAML/pull/659
* update time series forecast notebook by int-chaos in https://github.com/microsoft/FLAML/pull/682
* check config constraints for the initial config by sonichi in https://github.com/microsoft/FLAML/pull/685
* log_file_name in tune.run() by sonichi in https://github.com/microsoft/FLAML/pull/681
* updating nlp notebook by liususan091219 in https://github.com/microsoft/FLAML/pull/683
* VW version requirement and documentation on config_constraints vs metric_constraints by qingyun-wu in https://github.com/microsoft/FLAML/pull/686

New Contributors
* jmrichardson made their first contribution in https://github.com/microsoft/FLAML/pull/665

**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.9...v1.0.10

Highlight
* Add the feature names and importance in AutoML
* Update NLP search space and fix several bugs in NLP tasks
* Respect kwargs in AutoML.predict()

What's Changed
* Feature names and importances by sonichi in https://github.com/microsoft/FLAML/pull/621
* fix NER roberta bug by liususan091219 in https://github.com/microsoft/FLAML/pull/632
* updating search space by liususan091219 in https://github.com/microsoft/FLAML/pull/633
* Bump terser from 5.10.0 to 5.14.2 in /website by dependabot in https://github.com/microsoft/FLAML/pull/642
* This PR fixes the frequent NLP bugs in the other PRs by liususan091219 in https://github.com/microsoft/FLAML/pull/647
* added "**kwargs" to "predict" by zzheng93 in https://github.com/microsoft/FLAML/pull/641
* Fix alerts by skzhang1 in https://github.com/microsoft/FLAML/pull/644
* Update .NET documentation by luisquintanilla in https://github.com/microsoft/FLAML/pull/643
* Fix HPO evaluation bug by liususan091219 in https://github.com/microsoft/FLAML/pull/645

New Contributors
* dependabot made their first contribution in https://github.com/microsoft/FLAML/pull/642
* zzheng93 made their first contribution in https://github.com/microsoft/FLAML/pull/641
* luisquintanilla made their first contribution in https://github.com/microsoft/FLAML/pull/643

**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.8...v1.0.9

* Support latest xgboost version
* Reproducibility improvement for blendsearch
* Allow custom GroupKFold object as split_type
* Bug fix in token classification tasks such as NER
* Allow FLAML_sample_size in starting_points

What's Changed
* log msg about ensemble by sonichi in https://github.com/microsoft/FLAML/pull/597
* support latest xgboost version by sonichi in https://github.com/microsoft/FLAML/pull/599
* Fix automl settings in scikit-learn pipeline integration example by ZviBaratz in https://github.com/microsoft/FLAML/pull/602
* update got version by sonichi in https://github.com/microsoft/FLAML/pull/607
* min eci depends on cost_attr; cost_attr in ls by sonichi in https://github.com/microsoft/FLAML/pull/612
* Replaced !pip calls with %pip magic command by ZviBaratz in https://github.com/microsoft/FLAML/pull/604
* cath URLError by sonichi in https://github.com/microsoft/FLAML/pull/613
* Updated pre-commit hooks by ZviBaratz in https://github.com/microsoft/FLAML/pull/609
* Py36 by sonichi in https://github.com/microsoft/FLAML/pull/614
* Allow custom GroupKFold object as split_type by sonichi in https://github.com/microsoft/FLAML/pull/616
* Typo fix by ZviBaratz in https://github.com/microsoft/FLAML/pull/618
* use relative url in doc by sonichi in https://github.com/microsoft/FLAML/pull/620
* This PR will solve issue, code example format in the doc 622 by 31Sanskrati in https://github.com/microsoft/FLAML/pull/623
* fix ner bug; refactor post processing of TransformersEstimator prediction by liususan091219 in https://github.com/microsoft/FLAML/pull/615
* isinstance(x, int) -> isinstance(x, (int, np.integer)) by liususan091219 in https://github.com/microsoft/FLAML/pull/627
* Allow FLAML_sample_size in starting_points by qingyun-wu in https://github.com/microsoft/FLAML/pull/619
* disable max_len for ner by liususan091219 in https://github.com/microsoft/FLAML/pull/629
* fix 630 by adi611 in https://github.com/microsoft/FLAML/pull/631

New Contributors
* ZviBaratz made their first contribution in https://github.com/microsoft/FLAML/pull/602
* 31Sanskrati made their first contribution in https://github.com/microsoft/FLAML/pull/623
* adi611 made their first contribution in https://github.com/microsoft/FLAML/pull/631

**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.7...v1.0.8

* Add support of Python 3.10.
* Enable ensemble when using ray.
* Enable nested tuning runs.
* Made BlendSearch reproducible when constructed outside tune.run().
* Fix resource limit issue in some macos version.
* Bug fix in nlp.
* Make set_search_properties() compatible with ray tune.

What's Changed
* enable ensemble when using ray by sonichi in https://github.com/microsoft/FLAML/pull/583
* update time from start when using ray by sonichi in https://github.com/microsoft/FLAML/pull/586
* Class variables, cost_attr, and reproducibility by qingyun-wu in https://github.com/microsoft/FLAML/pull/587
* backup & recover global vars for nested tune.run by sonichi in https://github.com/microsoft/FLAML/pull/584
* fixing a bug in nlp/utils.py by liususan091219 in https://github.com/microsoft/FLAML/pull/590
* fix resource limit issue by sonichi in https://github.com/microsoft/FLAML/pull/589
* Modified setup instructions by daniel-555 in https://github.com/microsoft/FLAML/pull/593
* Add python 3.10 in the CI by sonichi in https://github.com/microsoft/FLAML/pull/591
* trying to fix the indexerror for ner by liususan091219 in https://github.com/microsoft/FLAML/pull/596
* Update documentation for NLP by liususan091219 in https://github.com/microsoft/FLAML/pull/594
* set_search_properties by sonichi in https://github.com/microsoft/FLAML/pull/595

New Contributors
* daniel-555 made their first contribution in https://github.com/microsoft/FLAML/pull/593

**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.6...v1.0.7

What's Changed
* init value type match by sonichi in https://github.com/microsoft/FLAML/pull/575


**Full Changelog**: https://github.com/microsoft/FLAML/compare/v1.0.5...v1.0.6