v0.11: Bug fixes and better support for non-numeric classes

• Ensure same type for CF output and input features
• Return original class instead of encoded class in CFs
• Better error checking for input query instance
• Dynamic creation of datasets for testing
• Approximate equality checking for float values in dataframe
• Other bug fixes
  Thank you @gaugup, @bdavj, @DanieleMorotti for your contributions.

v0.10: Bug fixes and Python 3.11 support

• Support for python 3.11
• Update to requirements and tests
• Updated adult dataset download logic
• Misc. bug fixes

v0.9: Unified API for deep learning and sklearn models

• [Major refactor] Unified API for deep learning and sklearn models
• Support for generating CFs without training data (private data mode) for sklearn models
• Bug fixes and implementing iteration limits for search on CFs

New Contributors

• @lange-martin made their first contribution in #304
• @bgalvao made their first contribution in #313
• @daikikatsuragawa made their first contribution in #311

Full Changelog: v0.8...v0.9

v0.8: More configurable parameters for generating counterfactuals and better input checking

• Better error checking for user input parameters
• Improved error messages for feature importance functions
• Bug fixes for dimension mismatch with categorical variables
• Add capability to serialize and de-serialize dice-ml explainers
• Add more configurable parameters to generate_counterfactuals function
• Increased test coverage
• Replaced boston housing with california housing dataset
• Tests and support for Python 3.10, removed test support for 3.6

Contributors: @gaugup, @amit-sharma

Version bump 0.7.2 (Minor bug fixes)

The release contains the following:-

• Fix for logging
• Fix indentation in schema files
• Add documentation to save generated counterfactuals to disk

Minor version update: Better input checks

• Better input checks and minor bug fixes

Bug fixes and improvements to data interfaces and input validation

• Bug fixes for feature importance calculation and data interfaces
• Example notebooks: Use test data always to generate counterfactuals
• Updates to the genetic algorithm and KD-tree algorithm
• Better input validation and informative error messages for Dice explainers
• Improved test coverage

A big thank you to @gaugup and @soundarya98 for their contributions.

Minor version update: Fixed CF schema and better tests

• Added init so schema are packaged with distribution
• All flake8 lint errors fixed. Flake8 validation is now enabled on all new commits.
• Better automation for tests

Thank you @gaugup and @gregorybchris for your contributions!

Optimized methods and updated schema for counterfactuals

• Better exception handling for invalid user inputs (e.g., total_CFs=0)
• Updated schema for serializing counterfactual explanations object
• Better description for DiCE in getting started notebook
• Add progress bar for counterfactual computation
• Optimizations for genetic and random methods
• Better test coverage

Thank you @gaugup and @soundarya88 for your contributions!

Rolling out DiCE for sklearn and regression models

• [Major] DiCE now supports sklearn models. Added three model-agnostic methods: randomized, genetic algorithm, and kd-tree
• [Major] Support for regression and multi-class problems
• [Major] Added local and global feature importance scores based on counterfactuals
• [Major] Better support for customizing counterfactuals through features_to_vary and permitted_range parameters for both continuous and categorical features
• [Refactor] ML Model and DiCE Explainer can use different feature transformations. Model's transformation can be provided as an input to the dice_ml.Model constructor. DiCE accepts inputs in the original data frame and does its transformations internally
• Enhanced tests for the library
• Deep learning libraries (tensorflow and pytorch) marked as optional dependencies
• New notebooks showing applications of DiCE in docs/source/notebooks/

A big thanks to @raam93, @soundarya98 and @gaugup for this release!