v1.2.0

1.2.0 (2026-01-26)

• Compatibility with Python up to version 3.14, end-of-life Python 3.8 and 3.9
• Compatibility with newer lightning versions - @phoeenniixx in #625
• Support for model stacking - @taimo3810 in #520
• Support for Multi-GPU Training - @sorenmacbeth in #517

Enhancements

• [ENH] Protection for MDN Head misuse by @manujosephv in #448
• [ENH] Remove restriction for using missing and unknown category in SSL models by @sorenmacbeth in #470
• [ENH] Add multi target classification by @YonyBresler in #441
• [ENH] Feature/tuner multiple model by @ProgramadorArtificial in #461
• [ENH] Add dataloader_kwargs support in DataConfig by @snehilchatterjee in #492
• [ENH] Adding informative str and repr by @manujosephv in #507
• [ENH] Enable Support for Multi-GPU Training by @sorenmacbeth in #517
• [ENH] Add Built-in Support for Model Stacking by @taimo3810 in #520
• [ENH] Make tensor dtypes np.float32 for MPS devices by @sorenmacbeth in #540
• [ENH] Optimizer lr scheduler interval by @sorenmacbeth in #545
• [ENH] add conditional test skips to estimator specific tests by @fkiraly in #607
• [ENH] Ensure compatibility with lightning > 2.6 and above by @phoeenniixx in #625

Documentation

• [DOC] Create FUNDING.yml by @manujosephv in #376
• [DOC] Fix syntax error in experiment_tracking.md log_target by @furyhawk in #382
• [DOC] Fix documentation structure and change default variable by @ProgramadorArtificial in #394
• [DOC] docs(contributor): contributors readme action update by @github-actions[bot] in #390
• [DOC] Update README.md by @HernandoR in #410
• [DOC] lint: simplify used tools by @Borda in #431
• [DOC] Fixed typo in metrics_prob_input by @abhisharsinha in #455
• [DOC] README - collected important links at top, badges by @fkiraly in #624

Maintenance

• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #378
• [MNT] build(deps): update torchmetrics requirement from <1.3.0,>=0.10.0 to >=0.10.0,<1.4.0 by @dependabot[bot] in #391
• [MNT] build(deps): bump actions/cache from 3 to 4 by @dependabot[bot] in #392
• [MNT] build(deps): update pytorch-lightning requirement from <2.2.0,>=2.0.0 to >=2.0.0,<2.3.0 by @dependabot[bot] in #409
• [MNT] build(deps): bump pypa/gh-action-pypi-publish from 1.8.11 to 1.8.12 by @dependabot[bot] in #407
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #395
• [MNT] build(deps): update plotly requirement from <5.19.0,>=5.13.0 to >=5.13.0,<5.20.0 by @dependabot[bot] in #408
• [MNT] Update mkdocstrings[python] requirement from ==0.22.* to ==0.23.* by @dependabot[bot] in #282
• [MNT] Bump akhilmhdh/contributors-readme-action from 2.3.6 to 2.3.10 by @dependabot[bot] in #454
• [MNT] freeze numpy <2.0 & fix ci+docs by @Borda in #482
• [MNT] Update pytorch-lightning requirement from <2.3.0,>=2.0.0 to >=2.0.0,<2.5.0 by @dependabot[bot] in #476
• [MNT] Bump pypa/gh-action-pypi-publish from 1.8.12 to 1.10.1 by @dependabot[bot] in #484
• [MNT] Update torchmetrics requirement from <1.4.0,>=0.10.0 to >=0.10.0,<1.5.0 by @dependabot[bot] in #477
• [MNT] Update plotly requirement from <5.20.0,>=5.13.0 to >=5.13.0,<5.25.0 by @dependabot[bot] in #478
• [MNT] Update protobuf requirement from <4.26.0,>=3.20.0 to >=3.20.0,<5.29.0 by @dependabot[bot] in #479
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #417
• [MNT] docs(contributor): contributors readme action update by @github-actions[bot] in #468
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #485
• [MNT] docs(contributor): contributors readme action update by @github-actions[bot] in #488
• [MNT] docs(contributor): contributors readme action update by @github-actions[bot] in #505
• [MNT] Update wandb requirement from <0.17.0,>=0.15.0 to >=0.15.0,<0.19.0 by @dependabot[bot] in #489
• [MNT] Update torchmetrics requirement from <1.5.0,>=0.10.0 to >=0.10.0,<1.6.0 by @dependabot[bot] in #501
• [MNT] Update .pre-commit-config.yaml by @manujosephv in #506
• [MNT] Bump pypa/gh-action-pypi-publish from 1.10.1 to 1.11.0 by @dependabot[bot] in #500
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #486
• [MNT] Update broken link for Denoising AutoEncoder tutorial by @manujosephv in #511
• [MNT] Bump pypa/gh-action-pypi-publish from 1.11.0 to 1.12.2 by @dependabot[bot] in #513
• [MNT] Update protobuf requirement from <5.29.0,>=3.20.0 to >=3.20.0,<5.30.0 by @dependabot[bot] in #516
• [MNT] Update torchmetrics requirement from <1.6.0,>=0.10.0 to >=0.10.0,<1.7.0 by @dependabot[bot] in #514
• [pre-commit.ci] pre-commit suggestions by @pre-commit-ci[bot] in #509
• [MNT] Update mkdocstrings[python] requirement from ==0.26.* to ==0.27.* by @dependabot[bot] in #515
• [MNT] chore(ci): setting concurrency by @Borda in #524
• [MNT] docs(contributor): contributors readme action update by @github-actions[bot] in #523
• [MNT] Update base.txt by @manujosephv in #556
• [MNT] docs(contributor): contributors readme action update by @github-actions[bot] in #560
• [MNT] Update mkdocstrings[python] requirement from ==0.27.* to ==0.29.* by @dependabot[bot] in #559
• [MNT] Migrate packaging to pyproject.toml by @fkiraly in #594
• [MNT] fix CI jobs with problems by @fkiraly in #597
• [MNT] improved testing CI job - uv and installed dependencies display by @fkiraly in #599
• [MNT] Update dependency versions and complete migration to pyproject.toml by @phoeenniixx in #596
• [MNT] Bump actions/checkout from 4 to 6 by @dependabot[bot] in #591
• [MNT] CI test matrix with unix, windows, and all supported python versions by @fkiraly in #604
• [MNT] stop CI job that adds documentation build link to PR description by @fkiraly in #608
• [MNT] remove unused dependencies from pyproject.toml by @fkiraly in #602
• [MNT] update dependabot.yml for daily updates by @fkiraly in #609
• [MNT] isolate captum soft dependency in tests by @fkiraly in #613
• [MNT] Dependabot: Bump actions/cache from 4 to 5 by @dependabot[bot] in #614
• [MNT] add tests without and with all extra dependencies by @fkiraly in https://g...

v1.1.1

Release Notes

---

New Features

• Support for Multi-Model Tuning: Added support for tuning multiple models using the tuner functionality, enhancing flexibility in hyperparameter optimization.
  [Commit: 560dec6]

• Multi-Target Classification: Introduced multi-target classification capabilities, expanding the library's support for more complex use cases.
  [Commit: 25691f5]

• dataloader_kwargs in DataConfig: Added support for customizing dataloader_kwargs in the DataConfig module for improved data-loading flexibility.
  [Commit: caa3ea1]

---

Enhancements

• Improved Informative str and repr: Added more informative str and repr methods to enhance debugging and readability of objects.
  [Commit: 495803c]

• Bug Fixes for Categorical Dtype: Fixed issues with Categorical data type handling to ensure smoother model training and predictions.
  [Commit: cf1454a]

• Removed Restrictions on Missing and Unknown Values: Enhanced the framework to handle missing and unknown values more robustly.
  [Commit: fc6060e]

• Protection Against Misuse of MDN Head: Added safeguards to prevent improper usage of the MDN Head in models.
  [Commit: cc3504a]

---

Bug Fixes

• Fixed an SSL finetuning bug to ensure secure operations during model fine-tuning.
  [Commit: 3d978f9]

• Fixed errors in saving and loading custom loss functions to enhance reproducibility and reliability.
  [Commit: 3f0a15c]

• Addressed a bug in cross-validation, ensuring accurate evaluation metrics.
  [Commit: 0d088fc]

• Fixed a KeyError issue with nn.activation in Tab Transformer and FT Transformer models.
  [Commit: 11adefa]

---

Other Improvements

• Multiple pre-commit configuration updates and enhancements for code linting and formatting.
  [Commits: f354b9c, 75b21c4, a890dda]

• Various CI improvements, including dependency bump for gh-action-pypi-publish and caching updates.
  [Commits: cb78a6e, e49a999, da20ed3]

• Fixed typos and minor issues in documentation and code for improved clarity and maintainability.
  [Commits: 7285787, 6586705]

For more details, you can refer to the respective commits on the library's GitHub repository.

New Contributors

---

• @furyhawk made their first contribution in #382
• @HernandoR made their first contribution in #410
• @charitarthchugh made their first contribution in #420
• @abhisharsinha made their first contribution in #455
• @YonyBresler made their first contribution in #441
• @snehilchatterjee made their first contribution in #492

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

v1.1.0

New Features and Enhancements

• Added DANet Model: Added a new model, DANet, for tabular data.
• Explainability: Integrated Captum for explainability
• Hyperparameter Tuner: Added Grid and Random Search functionality to search through hyperparameters and return best model.
• Model Sweep: Added an easy "Model Sweep" method with which we can sweep a list of models with given data and quickly assess performance.
• Documentation Enhancements: Improved documentation to make it more user-friendly and informative
• Dependency Updates: Updated various dependencies for improved compatibility and security
• Graceful Out-of-Memory Handling: Added graceful out-of-memory handling for tabular models
• GhostBatchNorm: Added GhostBatchNorm to the library

Deprecations

• Deprecations: Handled deprecations and updated the library accordingly
• Entmax Dependency Removed: Removed dependency on entmax

Infrastructure and CI/CD

• Continuous Integration: Improved CI with new actions and labels
• Dependency Management: Updated dependencies and restructured requirements

API Changes

• [BREAKING CHANGE] SSL API Change: Addressed SSL API change, along with documentation and tutorial updates.
• Model Changes: Added is_fitted and other markers to the tabular model.
• Custom Optimizer: Allow custom optimizer in the model config.

Contributors

• Thanks to all the contributors who helped shape this release! (List of Contributors)

Upgrading

• Ensure to check the updated documentation for any breaking changes or new features.
• If you are using SSL, please check the updated API and documentation.

v1.0.2

New Features:

• Added Feature Importance: The library now includes a new method in TabularModel and BaseModel for enabling feature importance. Feature Importance has been enabled for FTTransformer and GATE models. [Commit: dc2a49e]

Enhancements:

• Enabled two more parameters in the GATE model. [Commit: 3680413]
• Included metric_prob_input parameter in the library configuration. This update allows for better control over metrics in the models. [Commit: 0612db5]
• Slight improvements to the GATE model, including changes to defaults for better performance. [Commit: c30a6c3]
• Minor bug fixes and improvements, including accelerator options in the configuration and progress bar enhancements. [Commit: f932230, bdd9adb, f932230]

Dependency Updates:

• Updated dependencies, including docformatter, pyupgrade, and ruff-pre-commit. [Commits: 4aae9a8, b3df4ce, bdd9adb, 55e800c, c6c4679, c01154b, 107cd2f]

Documentation Updates:

• Updated the library's README.md file. [Commits: db8f3b2, cab6bf1, 669faec, 1e6c400, 3097799, 7fabf6b]

Other Improvements:

• Various code optimizations, bug fixes, and CI enhancements. [Commits: 5637020, e5171bf, 812b40f]

For more details, you can refer to the respective commits on the library's GitHub repository.

v1.0.1

• Added a new task - Self Supervised Learning (SSL) and a separate training API for it.
• Added new SOTA model - Gated Additive Tree Ensembles (GATE).
• Added one SSL model - Denoising AutoEncoder.
• Added lots of new tutorials and updated entire documentation.
• Improved code documentation and type hints.
• Separated a Model into separate Embedding, Backbone, and Head.
• Refactored all models to separate Backbone as native PyTorch Model(nn.Module).
• Refactored commonly used modules (layers, activations etc. to a common module).
• Changed MixedDensityNetworks completely (breaking change). Now MDN is a head you can use with any model.
• Enabled a low level api for training model.
• Enabled saving and loading of datamodule.
• Added trainer_kwargs to pass any trainer argument PyTorch Lightning supports.
• Added Early Stopping and Model Checkpoint kwargs to use all the arguments in PyTorch Lightining.
• Enabled prediction using GPUs in predict method.
• Added reset_model to reset model weights to random.
• Added many save and load functions including ONNX(experimental).
• Added random seed as a parameter.
• Switched over completely to Rich progressbars from tqdm.
• Fixed class-balancing / mu propagation and set default to 1.0.
• Added PyTorch Profiler for debugging performance issues.
• Fixed bugs with FTTransformer and TabTransformer.
• Updated MixedDensityNetworks fixing a bug with lambda_pi.
• Many CI/CD improvements including complete integration with GitHub Actions.
• Upgraded all dependencies, including PyTorch Lightning, pandas, to latest versions and added dependabot to manage it going forward.
• Added pre-commit to ensure code integrity and standardization.

v0.7.0-alpha

• Added a few more SOTA models - TabTransformer, FTTransformer
• Made improvements in the model save and load capability
• Made installation less restrictive by unfreezing some dependencies.

v0.5.0-alpha

First Alpha Release