v1.0.0

What's Changed

1. Codebase Migration: Consolidated tpot2 into tpot; removed deprecated/experimental features.
2. Performance Enhancements: Optimized pipeline evaluation processes and genetic programming operators, leading to faster convergence and reduced computational overhead. ​
3. Graph-Based Pipelines: Introduced a flexible graph-based representation of machine learning pipelines, enhancing the exploration of complex model architectures. ​
4. Dependency Updates: Updated dependencies to ensure compatibility with the latest versions of scikit-learn and other essential libraries.​
5. Stability Improvements: Resolved various bugs and improved error handling to enhance overall stability and user experience.​
6. Genetic Feature Selection: Implemented genetic feature selection mechanisms, enabling automatic identification of relevant features during pipeline optimization. ​
7. Expanded Search Spaces: Enhanced the flexibility in defining search spaces, allowing for more comprehensive exploration of potential pipeline configurations. ​
8. Modular Framework: Refactored the codebase into a more modular structure, simplifying customization and extension of the evolutionary algorithm components. ​
9. Documentation Overhaul: Revised and expanded documentation, including updated examples and comprehensive guides to reflect the new features and API changes.​

Key Contributors

• Pedro Henrique Ribeiro (Lead developer - https://github.com/perib, https://www.linkedin.com/in/pedro-ribeiro/)
• Anil Saini ([email protected])
• Jose Hernandez ([email protected])
• Jay Moran ([email protected])
• Nicholas Matsumoto ([email protected])
• Gabriel Ketron ([email protected])
• Hyunjun Choi ([email protected])
• Miguel E. Hernandez ([email protected])
• Jason Moore ([email protected])

Full Changelog: https://github.com/EpistasisLab/tpot/commits/v1.0.0

v0.12.2

What's Changed

• estimator type by @perib in #1319
• Update requirements by @gatl in #1335
• Bump torch from 1.3.1 to 1.13.1 by @dependabot in #1336
• update sklearn version to 1.1.3 by @perib in #1337
• remove deprecated imp, fix docstring warning by @perib in #1331
• update compatibility with scikitlearn 1.4 by @perib in #1343
• Improve error message by @gatl in #1338
• Mate operator fix by @perib in #1268

New Contributors

• @gatl made their first contribution in #1335
• @dependabot made their first contribution in #1336

Full Changelog: v0.12.1...v0.12.2

v0.12.1

Fixes issue with runs terminating too early

What's Changed

• Early crash fix by @perib in #1315
  Full Changelog: v0.12.0...v0.12.1

v0.12.0 release

• Fix numpy compatibility
• Dask optimizations
• Minor bug fixes

v0.11.7 minor release

• Fix compatibility issue with scikit-learn 0.24 and xgboost 1.3.0
• Fix a bug causing that TPOT does not work when classifying more than 50 classes
• Add initial support Resampler from imblearn
• Fix minor bugs

0.11.6.post3

• A patch to fix compatibility issues with the latest version of xgboost (v1.3.0)

v0.11.6.post2

• make XGBoost as a required dependency

v0.11.6.post1

• Refine the logic of checking the type of an operator.

Version 0.11.6

• Fix a bug causing point mutation function does not work properly with using template option
• Add a new built configuration called "TPOT cuML" which TPOT will search over a restricted configuration using the GPU-accelerated estimators in RAPIDS cuML and DMLC XGBoost. This configuration requires an NVIDIA Pascal architecture or better GPU with compute capability 6.0+, and that the library cuML is installed.
• Add string path support for log/log_file parameter
• Fix a bug in version 0.11.5 causing no update in stdout after each generation
• Fix minor bugs

Covariate adjustments branch

• Development branch based on TPOT 0.11.1 for adjusting covariate without data leakage.