v0.6.1

What's Changed

• Convert .pylintrc file to pyproject.toml by @bpkroth in #797
• Pylint fixups by @bpkroth in #798
• Update azure credentials to be more flexible by @eujing in #787
• Fix mypy 1.11 warnings by @motus in #809
• Fix the coercion of scores to floats in the optimizer by @motus in #789
• Fixups and testing for cli config file parsing by @bpkroth in #722
• Use token-based authentication instead of the access key in the AzureFileShareService by @motus in #779
• Test that the config provided is serializable by @bpkroth in #790
• Validate the configuration produced by LlamaTune inverse_transform() by @bpkroth in #812
• make doc CI fixups by @bpkroth in #830
• Version identifier packaging tweaks by @bpkroth in #828
• Migrate to ConfigSpace 1.* by @motus in #802
• Restrict matplotlib version to < 3.9 for Windows conda env by @bpkroth in #827
• Proposed fixes for token-based auth in azure fileshare service by @eujing in #820
• Add CI build for Python 3.12 by @motus in #829
• Upgrade required dabl version and simplify related dependency rules by @bpkroth in #836
• Fixups to example notebook for new APIs by @bpkroth in #834
• Temporarily avoid new version of FLAML by @bpkroth in #840
• Fixup to release version tagged devcontainer publishing logic by @bpkroth in #841
• Use number of bins instead of quantization interval in mlos_bench tunables by @motus in #835
• Rename quantization -> quantization_bins by @motus in #844
• Bump version: 0.6.0 → 0.6.1 by @bpkroth in #845

Full Changelog: v0.6.0...v0.6.1

v0.6.0

What's Changed

• Move config_space to base Optimizer property by @bpkroth in #697
• Pin a devcontainer version to workaround library search path issue by @bpkroth in #699
• Typing tweaks by @bpkroth in #696
• Add cardinality and size related properties to Tunables by @bpkroth in #698
• Don't shell interpret commit messages by @bpkroth in #702
• Disable flaky logging line in CI workflow by @bpkroth in #703
• Pluggable Scheduler implementation by @motus in #691
• Conda Updates by @bpkroth in #705
• Add GridSearch Optimizer support by @bpkroth in #690
• Factor out the functionality of best config tracking into a separate abstract Optimizer class by @motus in #709
• Roll back forceful assignment of PATH when invoking a local process by @motus in #708
• A function for a popular None if value is None else func(value) code pattern by @motus in #707
• Tweak Storage schema related tests by @bpkroth in #712
• Pluggable Scheduler by @motus in #710
• Add more tests for Scheduler configurations (CLI and JSON) by @motus in #717
• UTC datetime handling improvements by @motus in #716
• Rename max_iterations to max_suggestions and track in Optimizer .suggest() instead of .register() by @motus in #713
• Add max_trials scheduler config parameter by @motus in #719
• bugfix: handle -Inf in try_parse_val() function by @motus in #724
• Pass multiple metrics into the optimizer by @motus in #723
• use Python 3.11 in Windows conda env by @motus in #728
• Allow multiple objectives in the optimizer config by @motus in #725
• Implement multi-objective optimization in mlos_bench base classes by @motus in #726
• CI fixups: changes in package naming via upstream libs (dash to underscore) by @bpkroth in #733
• Fix docstrings for mlos_core SMAC and FLAML optimizers by @motus in #736
• Issue a warning on non-null optimization context by @motus in #735
• make optimizer a fixture for 1-hot encoding/decoding tests by @motus in #737
• Pin a lower version of matplotlib while we wait for dabl to get fixed by @bpkroth in #743
• Update mlos_core API to support with multi-factor optimization by @motus in #730
• Cascading series of CI dependency problems by @bpkroth in #746
• Pass optional weights for optimization targets in mlos_core; implement multi-target optimization for FLAML. by @motus in #738
• Limit CI concurrency by @bpkroth in #753
• Added missing 'shell_env_params' in redis.jsonc by @DelphianCalamity in #754
• Temporarily avoid broken numpy dependency by @bpkroth in #757
• Require explicit arguments for mlos_core optimizers by @jsfreischuetz in #760
• Add some dev notes on contributing to the MLOS project by @bpkroth in #761
• Cleanup a warning issued by the SmacOptimizer cleanup code by @bpkroth in #767
• Staged Makefile Improvements by @bpkroth in #762
• Remove a few under used vscode extensions by @bpkroth in #765
• Add Metadata to optimizers by @jsfreischuetz in #770
• Better naming and values for MockEnv parameters + unit tests by @motus in #769
• Introduce pyproject.toml build changes by @bpkroth in #764
• Fix Linting Error in os_environ.py by @jsfreischuetz in #772
• Prepare isort black formatters and checks by @bpkroth in #766
• Make sure One Shot Optimizer always returns the same suggestion by @motus in #759
• CI fixups by @bpkroth in #783
• bugfix: is_defaults() is a method, not property by @motus in #782
• bugfix: MySQL can round microseconds into the future causing scheduler to skip trials by @motus in #775
• Temporarily restrict ConfigSpace version by @bpkroth in #791
• Enable black, isort, and doc formatters and checks by @bpkroth in #774
• Ignore reformatting revision by @bpkroth in #794
• Bump version: 0.5.1 → 0.6.0 by @bpkroth in #795

New Contributors

• @DelphianCalamity made their first contribution in #754
• @jsfreischuetz made their first contribution in #760

Full Changelog: v0.5.1...v0.6.0

v0.5.1

What's Changed

• Add tests for mixed numerical type conversions of tunable config data from storage by @bpkroth in #670
• Fixtures context fixup by @bpkroth in #673
• Address upcoming changes to default GITHUB_TOKEN permissions by @bpkroth in #674
• Return status and actual timestamp from all environments by @motus in #671
• Handle some mypy errors and increase test coverage by @bpkroth in #680
• New pytest version dependency tweaks by @bpkroth in #681
• Address pandas warnings about future pyarrow requirements by @bpkroth in #682
• Address pandas timestamp warnings by @bpkroth in #683
• Scheduling queue: Allow submitting trials with future timestamps and use watermarks to retrieve configs and results by @motus in #676
• use named constants to define columns containing IDs and generic values by @motus in #678
• Allow use of Azure VM Service for only remote exec. by @bpkroth in #675
• Tweaks for anchored stdout parsing by @bpkroth in #685
• Allow no tunable configs for benchmarking by @bpkroth in #686
• Add tests for config data retrieval for experiments with no tunables by @bpkroth in #689
• Initial work to separate the scheduler and the optimizer (no async yet) by @motus in #684
• Bump version: 0.5.0 → 0.5.1 to mark a stable version before the big changes (async Environment etc.) by @motus in #695

Full Changelog: v0.5.0...v0.5.1

v0.5.0

What's Changed

• Reorg doc deploy step for new GITHUB_TOKEN permissions requirements by @bpkroth in #659
• Omit tests from docs by @bpkroth in #660
• Add weights for tunables with categorical or special numerical values by @motus in #646
• Add log scale and quantization to the tunables by @motus in #662
• Fix mixed numeric datatypes for optimizers by @ephoris in #667
• Add support for ConfigSpace distributions by @motus in #663
• Bump version: 0.4.2 → 0.5.0 by @motus in #672

New Contributors

• @ephoris made their first contribution in #667

Full Changelog: v0.4.2...v0.5.0

v0.4.2

What's Changed

• README documentation updates by @bpkroth in #647
• Remove TunableGroups from Storage classes and add TrialData tests by @bpkroth in #644
• Refactor mlos_bench.storage and add TunableConfigTrialGroup property for TrialData and ExperimentData by @bpkroth in #648
• Add default_tunable_config_id property to ExperimentData by @bpkroth in #650
• Remove mistakenly added columns by @bpkroth in #652
• Fixes to max_iterations calculation when trial_config_repeat_count is given. by @bpkroth in #651
• Address some pandas FuturesWarnings by @bpkroth in #653
• Improve mlos-viz for multiple repeats of a config and add tests by @bpkroth in #633
• Workaround pytest version bump incompatibility with pytest-lazy-fixtures plugin by @bpkroth in #656
• Add demo video link to README by @bpkroth in #657
• Bump version: 0.4.1 → 0.4.2 by @bpkroth in #658

Full Changelog: v0.3.2...v0.4.2

v0.3.2

What's Changed

• Rename and improve tunable vs config API and documentation by @bpkroth in #629
• CI tweak: use "release" event instead of "tag push" for pypi publishing by @bpkroth in #631
• Add support for storing the optimization targets and direction of an experiment by @bpkroth in #628
• Introduce basic experiment visualization module mlos_viz via dabl by @bpkroth in #624
• Github Action syntax fixup by @bpkroth in #632
• Add labels to the optimizer trends plot by @bpkroth in #634
• Add more info on const_args vs. tunable_params; start the DEVNOTES by @motus in #630
• Bump actions/cache from 3 to 4 in /.github/workflows by @dependabot in #637
• Support for special tunable values outside of the range by @motus in #617
• Tell pylint to ignore FIXME comments in our code. by @bpkroth in #639
• Add unit tests for MlosCoreoptimizer._to_df() functionality; use special values in tunables by @motus in #638
• Fixup a vscode config file. by @bpkroth in #643
• Disallow ! characters in tunable parameter names. by @bpkroth in #641
• Add basic support for repeating a trial config. by @bpkroth in #642
• PyPi publish makefile rules fixup by @bpkroth in #640

Full Changelog: v0.2.5...v0.3.2