Skip to content

CHANGELOG.md

Latest commit

 

History

History
167 lines (96 loc) · 8.39 KB

CHANGELOG.md

File metadata and controls

167 lines (96 loc) · 8.39 KB

Changelog

Unreleased

Full Changelog

Implemented features and enhancements:

  • Implement ModelSearchCV with grid search and halving grid search #33

Fixed bugs:

  • Calling Authenticity metric raises an error on the \_\_repr\_\_ method #67
  • Error when importing set\_matplotlib\_style #60

Documentation updates:

  • Add mlresearch.show_versions() to documentation page #65
  • Review and add examples to documentation #4

v0.5.1 (2024-07-11)

Full Changelog

Deprecated:

  • Format code with new version of black and fix errors coming from new sklearn version #66

v0.5.0 (2024-04-04)

Full Changelog

Implemented features and enhancements:

  • Implement metrics for evaluating synthetic data #63
  • Set utils.parallel\_loop to track job completions instead of job starts #51

Fixed bugs:

  • Error when running make\_mean\_sem\_table (AttributeError: 'DataFrame' object has no attribute 'map') #61
  • Fix applymap FutureWarning #42

Documentation updates:

  • Add mlresearch.show\_versions\(\) functions for bug reports #56
  • Add issues template and fix broken link on PR template #49
  • Add Changelog generator #45
  • Add CONTRIBUTING file #44

Closed issues:

  • Add type hints to code and fix make code-analysis warnings #57
  • Add Python 3.12 support #53
  • Make package rich an optional dependency #47
  • Rename data\_augmentation submodule to synthetic\_data #32

v0.4.2 (2023-09-28)

Full Changelog

Fixed bugs:

  • mlresearch.preprocessing.PipelineEncoder is not preserving feature order #43
  • Fix bugs and deprecation warnings arising from dependency updates #41
  • census dataset URL is broken #40
  • Replace df.append functions to comply with pandas 2.0 #39

v0.4.1 (2023-06-02)

Full Changelog

Fixed bugs:

  • adult dataset contains trailing white spaces #36
  • The number of non-metric + metric features in mlresearch.datasets.Datasets.summarize\_datasets\(\) is not matching the total features #35

Closed issues:

  • Add check\_random\_states to \_\_init\_\_.py in utils submodule #37

v0.4.0 (2023-01-24)

Full Changelog

Implemented features and enhancements:

  • Create function dataframe\_to\_image #28
  • Include a pipeline-compatible version for one-hot encoding #19
  • Add more dataset information in summarize\_datasets #18
  • sklearn.metrics.SCORERS is deprecated #16
  • Modify parameter names in make\_bold function to ensure parameter name consistency across utils function #15
  • Make pytorch models sklearn compatible #9
  • Move secondary dependencies to optional dependencies #8

Documentation updates:

  • Move LaTeX-related functions to their own submodule #29
  • Fix some of the errors/warnings in the documentation #22
  • Add missing docstring to GeometricSMOTE's \_encode\_categorical #17

Closed issues:

  • Fix deprecation warnings #21
  • Consider modifying default BYOL hyper-parameters for smaller batch sizes #11
  • Add Semi-supervised learning implementation #7

v0.4a2 (2022-10-10)

Full Changelog

Closed issues:

  • Add ML-Research to conda #5

v0.4a1 (2022-04-14)

Full Changelog

Implemented features and enhancements:

  • Add function to describe and imbalance datasets #10
  • Get to 80% code coverage #1

Documentation updates:

  • Update make commands #12

Closed issues:

  • Finish BYOL implementation and tests #6
  • Move all CI/CD to GitHub Actions #3

v0.3.4 (2021-12-28)

Full Changelog

v0.3.3 (2021-10-07)

Full Changelog

v0.3.2 (2021-09-03)

Full Changelog

v0.3.1 (2021-07-26)

Full Changelog

v0.3.0 (2021-07-22)

Full Changelog

v0.2.1 (2021-07-09)

Full Changelog

v0.2.0 (2021-07-07)

Full Changelog

0.1.0 (2021-07-05)

Full Changelog

* This Changelog was automatically generated by github_changelog_generator