v1.2.4

This release contains:

• improved support for using a list of configurations (thanks to @BeibinLi ),
• using a filter to select from the responses out of a sequence of configurations (doc).
• a new experimental human-proxy agent (thanks to @qingyun-wu and @gagb).
• utility function to create config lists.
• method clear_cache added in oai.Completion.
• update of the default search space (thanks to @Kyoshiin and @LittleLittleCloud ).
• prepare for flaml v2 (thanks to @qingyun-wu for writing the blogpost).

Breaking change:

• cache_path is renamed into cache_path_root in set_cache.

Thanks to code reviewers @skzhang1 @jtongxin @pcdeadeasy @ZviBaratz @LittleLittleCloud @Borda , and to @liususan091219 @thinkall for fixing test error.

What's Changed

• Catch AuthenticationError trying different configs by @BeibinLi in #1023
• chat completion check by @sonichi in #1024
• update model of text summarization in test by @liususan091219 in #1030
• Human agent by @qingyun-wu in #1025
• fix of website link by @sonichi in #1042
• Blogpost by @qingyun-wu in #1026
• Update default search space by @Kyoshiin in #1044
• Fix PULL_REQUEST_TEMPLATE and improve test by removing unnecessary environment variable by @thinkall in #1043
• response filter by @sonichi in #1039

New Contributors

• @BeibinLi made their first contribution in #1023
• @Kyoshiin made their first contribution in #1044

Full Changelog: v1.2.3...v1.2.4

v1.2.3

This release contains a number of updates in autogen and automl.

• We added more utilities and documentation improvement to flaml.oai such as logging, templating and using multiple configs to make developing and experimentation with OpenAI models more convenient. (thanks to @afourney @victordibia @torronen @ekzhu )
• We added an experimental coding agent based on GPT-4. (thanks to @BeibinLi @qingyun-wu @skzhang1 )
• We added options to disable the default mlflow logging in automl. (thanks to @garar)
• We make better use of parallelism in auto-scale spark cluster for automl and tune. (thanks to @thinkall )

Thanks to @royninja @luckyklyist @Borda @qingyun-wu @thinkall @liususan091219 for other bug fixes, documentation improvement, and engineering improvement. Thanks to @victordibia @skzhang1 @kevin666aa @jtongxin @levscaut for code reviews.

What's Changed

• version update post release v1.2.2 by @sonichi in #1005
• fixing the typo #990 by @royninja in #994
• fixed sentence misplace #998 by @luckyklyist in #1010
• pyproject.toml & switch to Ruff by @Borda in #976
• update readme by @qingyun-wu in #1014
• raise content_filter error by @sonichi in #1018
• Fix catboost failure in mac-os python<3.9 by @thinkall in #1020
• coding agent; logging by @sonichi in #1011
• Add mlflow_logging param by @garar in #1015
• fix NLP zero division error by @liususan091219 in #1009
• update max_spark_parallelism to fit in auto-scale spark cluster by @thinkall in #1008
• update spark session in spark tests by @thinkall in #1006
• Mark experimental classes; doc; multi-config trial by @sonichi in #1021

New Contributors

• @luckyklyist made their first contribution in #1010
• @garar made their first contribution in #1015

Full Changelog: v1.2.2...v1.2.3

v1.2.2

What's Changed

• update nlp notebook by @liususan091219 in #940
• Blog post for LLM tuning by @sonichi in #986
• fix zerodivision by @liususan091219 in #1000
• extract code from text; solve_problem; request_timeout in config; improve code by @sonichi in #999

Full Changelog: v1.2.1...v1.2.2

v1.2.1

What's Changed

• handling nlp divide by zero by @liususan091219 in #926
• set black with 120 line length by @Borda in #975
• Post release update by @sonichi in #985
• fix tests failure caused by version incompatibility by @thinkall in #995
• Expose feature and label transformer in automl.py by @JaneIllario in #993
• make context optional; improve error handling and doc by @sonichi in #997

New Contributors

• @JaneIllario made their first contribution in #993

Full Changelog: v1.2.0...v1.2.1

v1.2.0

This is a major release with many new features.

Highlights

• autogen subpackage in preview: text generation optimization, including support for ChatGPT and GPT-4.
• Spark dataframe and SynapseML support in preview.
• Holt-Winters exponential smoothing for time series forecasting.
• Task class abstraction for easier task-specific development.

What's Changed

• precommit: end-of-file-fixer by @Borda in #929
• adding precommit check by @Borda in #930
• Bump http-cache-semantics from 4.1.0 to 4.1.1 in /website by @dependabot in #913
• adding openai documentation by @liususan091219 in #938
• improve max_valid_n and doc by @sonichi in #933
• Update doc and notebook by @skzhang1 in #946
• ChatGPT support by @sonichi in #942
• Extract task class from automl by @markharley in #857
• fix delete in TransformersEstimator, so to remove the ckpt not found warning in PR 940 by @liususan091219 in #950
• Support spark dataframe as input dataset and spark models as estimators by @thinkall in #934
• gpt-4 support; openai workflow fix; model str; timeout; voting by @sonichi in #958
• more tolerant time limit for test_overtime by @levscaut in #960
• Adding a test function for OpenAI completion in flaml by @qingyun-wu in #951
• Noctus contribute by @AyeshGK in #961
• add timeout to openai create by @thinkall in #973
• ci: concurrency by @Borda in #931
• Bump eta from 1.12.3 to 2.0.0 in /website by @dependabot in #941
• Bump webpack from 5.74.0 to 5.76.1 in /website by @dependabot in #952
• Bump @sideway/formula from 3.0.0 to 3.0.1 in /website by @dependabot in #963
• Add Holt-Winters exponential smoothing by @coffepowered in #962
• autogen subpackage by @sonichi in #968

New Contributors

• @AyeshGK made their first contribution in #961

Full Changelog: v1.1.3...v1.2.0

v1.1.3

This version contains multiple updates, including support of Azure OpenAI APIs, user-defined meta features in zero-shot AutoML, and time monitoring for spark jobs. Thanks to all the contributors!

What's Changed

• workflow path->paths by @sonichi in #915
• azure oai by @sonichi in #920
• Added extras for synapse by @thinkall in #916
• Update hf version by @liususan091219 in #918
• bump version to 1.1.3; workflow trigger by @sonichi in #924
• added a documentation about small time budget by @Animaholic in #914
• Fix Dependabot alerts by @skzhang1 in #922
• pre-commit: update config by @Borda in #925
• Improve annotations in automl and ml modules by @coffepowered in #919
• Updated dict type args default value to None by @thinkall in #927
• fix #493 by @LittleLittleCloud in #932
• add PySparkOvertimeMonitor to avoid exceeding time budget by @levscaut in #923
• minor change to notebook by @sonichi in #928

New Contributors

• @Borda made their first contribution in #925
• @LittleLittleCloud made their first contribution in #932
• @levscaut made their first contribution in #923

Full Changelog: v1.1.2...v1.1.3

v1.1.2

News

• AAAI-23 tutorial: We will be giving a tutorial about FLAML in AAAI-23 on Feb 08, 2023. Please find the tutorial agenda in this page.
• Latest research to appear at ICLR-23: Our latest research on multiobjective HPO is accepted as one of the notable-top-5% research papers in ICLR 2023:
  • Please find more details about the method in our paper: Targeted Hyperparameter Optimization with Lexicographic Preferences Over Multiple Objectives. Shaokun Zhang, Feiran Jia, Chi Wang, Qingyun Wu. ICLR 2023 (notable-top-5%).
  • Please find detailed documentation about this new functionality in this page: https://microsoft.github.io/FLAML/docs/Use-Cases/Tune-User-Defined-Function#lexicographic-objectives.
• OpenAI support is added in this release. A notebook example can be found at: https://github.com/microsoft/FLAML/blob/main/notebook/integrate_openai.ipynb

What's Changed

• handle num_samples=-1 by @sonichi in #879
• Bump json5 from 2.2.1 to 2.2.3 in /website by @dependabot in #877
• fix the doc error of customized metrics in automl by @skzhang1 in #883
• Bump ua-parser-js from 0.7.32 to 0.7.33 in /website by @dependabot in #900
• display data head in notebook; exclude None by @sonichi in #885
• Support percentage tolerance for lexicographic optimization by @skzhang1 in #875
• Document how to use the group k-fold by @coffepowered in #894
• update doc for research papers by @qingyun-wu in #912
• stratified group kfold splitter by @sonichi in #899
• add cost budget; move loc of make_dir by @sonichi in #888
• Add bibtex entries for research publications by @Animaholic in #904
• Openai by @sonichi in #905

New Contributors

• @coffepowered made their first contribution in #894
• @Animaholic made their first contribution in #904

Full Changelog: v1.1.1...v1.1.2

v1.1.1

What's Changed

• create dir for log file name by @sonichi in #867
• Do not persist entire AutoMLState in Searcher by @Yard1 in #870
• fix #871: call check_spark only when necessary by @thinkall in #872
• notebook test; spark warning message; reproducibility bug; sequential tuning stop condition by @sonichi in #869

Full Changelog: v1.1.0...v1.1.1

v1.1.0

Highlights

• Spark is now supported as a new parallel tuning backend.
• New tuning capability: targeted tuning with multiple lexicographic objectives. Check out documentation and an example for this new tuning capability.
• New metrics: roc_auc_weighted, roc_auc_ovr_weighted, roc_auc_ovo_weighted.
• New reproducible learner selection method when time_budget is not specified.
• AutoML-related functionaility is moved into a new automl subpackage.

Thanks to all contributors who contributed to this release!

What's Changed

• Bump actions/checkout from 2 to 3 by @dependabot in #699
• fix dependably alert by @skzhang1 in #818
• fix typo by @skzhang1 in #823
• install editable package in codespace by @sonichi in #826
• skip test_hf_data in py 3.6 by @sonichi in #832
• fix typo of output directory by @thinkall in #828
• catch TFT logger bugs by @int-chaos in #833
• roc_auc_weighted metric addition by @shreyas36 in #827
• make performance test reproducible by @sonichi in #837
• Refactor into automl subpackage by @markharley in #809
• Edit the announcement of AAAI-23 tutorial and the KDD tutorial announcement. by @HangHouCheong in #820
• Use get to avoid KeyError by @sonichi in #824
• Update doc by @skzhang1 in #843
• fix bug related to choice by @sonichi in #848
• FAQ about OOM by @sonichi in #849
• Update .NET documentation links by @luisquintanilla in #847
• Added an info reminding user that if no time_budget and no max_iter is specified, then effectively zero-shot AutoML is used by @jingdong00 in #850
• Fix example tune-pytorch where the checkpoint path may be named differently by @jingdong00 in #853
• Format errors on the web. by @skzhang1 in #855
• Add supporting using Spark as the backend of parallel training by @thinkall in #846
• Info and naming by @sonichi in #864

New Contributors

• @thinkall made their first contribution in #828
• @markharley made their first contribution in #809
• @HangHouCheong made their first contribution in #820

Full Changelog: v1.0.14...v1.1.0

v1.0.14

Highlights

• Preparing alpha release of multi-objective hyperparameter tuning with lexicographic preference.
• Fixed issues related to zero-shot automl.
• Multiple improvements to documentation.

What's Changed

• Discord Badge Added by @royninja in #760
• fix bug in current nlp documentation by @liususan091219 in #763
• Multiple objectives hyperparameter tuning with lexicographic preference by @Anonymous-submission-repo in #752
• Indentation corrected by @Kirito-Excalibur in #778
• Included hint to escape brackets for pip setup by @evensure in #786
• Docs by @velezbeltran in #765
• Bump actions/setup-python from 2 to 4 by @dependabot in #700
• Bump codecov/codecov-action from 1 to 3 by @dependabot in #697
• Removed extra | in documentation by @satya-vinay in #790
• fix_alert by @skzhang1 in #793
• Fixed typo by @ElinaAndreeva in #797
• fix_alerts by @skzhang1 in #799
• Documentation about classification/regression task #753 by @royninja in #802
• Added a link to documentation webpage in notebook time_series_forcast by @jingdong00 in #791
• Fix issues related to zero-shot automl by @sonichi in #783
• added the models used for forecasting in documentation by @shreyas36 in #811
• Add performance test for LexiFlow by @Anonymous-submission-repo in #812

New Contributors

• @royninja made their first contribution in #760
• @Anonymous-submission-repo made their first contribution in #752
• @Kirito-Excalibur made their first contribution in #778
• @evensure made their first contribution in #786
• @velezbeltran made their first contribution in #765
• @satya-vinay made their first contribution in #790
• @ElinaAndreeva made their first contribution in #797
• @jingdong00 made their first contribution in #791
• @shreyas36 made their first contribution in #811

Full Changelog: v1.0.13...v1.0.14