v0.17.0

This release adds scikit-learn 1.9 support (fixing an import causalml.dataset failure present on the 0.16.0 wheel, #926), native Polars support across all meta-learners, a JAX/flax.nnx backend for DragonNet, and makes the meta-learners scikit-learn compliant (BaseEstimator, so clone() and get_params() work). It also adds the RATE evaluation metric and post-fit confidence intervals for BaseTLearner, plus numerous bug fixes.

New Features

• Add native Polars DataFrame, Series, and LazyFrame support for all meta-learners by @aman-coder03 in #901
• Polish Polars support and add bootstrap CI test coverage by @aman-coder03 in #921
• Add JAX/flax.nnx backend for DragonNet by @xrhd in #918
• docs: document JAX backend for DragonNet by @xrhd in #919
• Make meta-learners scikit-learn compliant via BaseEstimator by @aman-coder03 in #912
• Add Rank-weighted Average Treatment Effect (RATE) metric by @aman-coder03 in #887
• Add rate_score() with bootstrap confidence intervals and p-values by @aman-coder03 in #890
• Add AIPW docstring warning to get_toc() and rate_score() by @jeongyoonlee in #891
• Add post-fit confidence intervals to BaseTLearner via store_bootstraps and return_ci by @aman-coder03 in #886
• Support NaN values in UpliftTree and UpliftRandomForest by @aman-coder03 in #860

scikit-learn 1.9 Support

• Fix for compatibility with sklearn v1.9.0 by @jakevdp in #903
• Fix CausalRandomForestRegressor.fit() on scikit-learn 1.9 by @jeongyoonlee in #907
• Support CausalRandomForestRegressor.calculate_error() on scikit-learn >= 1.9 by @jeongyoonlee in #909

Bug Fixes

• Fix #904: Prevent deepcopy of fitted templates in bootstrap and correct predict validation ordering by @Saurav-Gupta-9741 in #910
• Optimised training, inference and memory for metalearners in multitreatment settings by @Ic3fr0g in #896
• Fix UpliftRandomForest predict shape mismatch with multiple treatments by @jeongyoonlee in #884
• Fix uplift tree p-value NaN from division by zero by @jeongyoonlee in #882
• Fix CausalRandomForestRegressor predicting inf from division by zero by @jeongyoonlee in #883
• Fix SensitivityPlaceboTreatment ignoring actual treatment groups by @jeongyoonlee in #880
• Fix seed parameter TypeError in BaseDRLearner bootstrap CI by @mohsinm-dev in #879
• Fix ValueError on read-only arrays in BaseSLearner.predict() by @mohsinm-dev in #878
• Add input validation to auuc_score for missing model columns by @jeongyoonlee in #881
• Make xgboost optional in synthetic dataset generation by @Si-ra-kri in #872
• Bug Fix: use iloc to index pd.Series by @bekojuniranjan in #877

Build / CI

• Make Cython line tracing opt-in to keep release wheels fast by @HSJung93 in #914
• Remove the PyPI token from GitHub Actions in favor of the Trusted Publishing by @jeongyoonlee in #871
• Upgrade GitHub Actions for Node 24 compatibility by @salmanmkc in #874
• Upgrade GitHub Actions to latest versions by @salmanmkc in #875
• ci: declare workflow-level contents: read on 4 workflows by @arpitjain099 in #900

⚠️ Breaking Changes

• Meta-learner __init__ signatures (#912): to become scikit-learn BaseEstimators, each learner now stores its constructor arguments verbatim and builds models in fit(). Most visibly, XGBRRegressor no longer accepts arbitrary **kwargs — pass XGBoost parameters via the explicit xgb_kwargs=<dict> argument.

Full Changelog: v0.16.0...v0.17.0

v0.16.0

This release primarily focuses on adding support for scipy>=1.16.0, numpy>=1.25.2, statsmodels>=0.14.5, and Python>=3.11, addressing Cython binary interface compatibility issues with updated dependencies.

What's Changed

• Add BaseDRClassifier for binary classification with probabilities by @jeongyoonlee in #844
• add ability to benchmark via synth validation by @IyarLin in #847
• Bug Fix: Update uplift.pyx with isinstance(v, Numbers.number)) by @00helloworld in #849
• Fix #848: Pass estimation_sample_size parameter to individual trees in UpliftRandomForestClassifier by @mohsinm-dev in #850
• #775: Support multiple treatments in CausalTreeRegressor and CausalRandomForestRegressor by @alexander-pv in #852
• fix: support scipy>=1.16.0 by removing sklearn internal dependency by @jeongyoonlee in #861
• Fix Ubuntu packaging failure - scipy manylinux compatibility by @jeongyoonlee in #865
• Upgrade cibuildwheel to v3.3.1 and remove deprecated macos-13 runner by @jeongyoonlee in #867
• Release v0.16.0: Upgrade to manylinux_2_28 and remove scipy version constraints by @jeongyoonlee in #869

New Contributors

• @IyarLin made their first contribution in #847
• @00helloworld made their first contribution in #849
• @mohsinm-dev made their first contribution in #850

Full Changelog: v0.15.5...v0.16.0

v0.15.5

What's Changed

• Fix runtime dependencies by @pavelzw in #831
• Fix build errors by @jeongyoonlee in #832
• Remove failing test-conda-forge-install.yml by @jeongyoonlee in #827
• Install via uv by @aaelony and @emilioMaddalena in #834 and #836
• Move calibrate() inside the PropensityModel class by @jeongyoonlee in #839
• v0.15.5 release by @jeongyoonlee in #842

New Contributors

• @pavelzw made their first contribution in #831
• @aaelony made their first contribution in #834
• @emilioMaddalena made their first contribution in #836

Full Changelog: v0.15.4...v0.15.5

v0.15.4

What's Changed

• add calibration example by @ras44 in #816
• resolves #811, update propensity.py by @ras44 in #817
• docs(validation): fixes some typos by @javiergarea in #822
• remove description of unused argument by @ras44 in #821
• #820: Update Jupyter Notebook with causal trees interpretation examples by @alexander-pv in #823
• Fix incorrect bootstrap mean calculation in meta-learners by @jeongyoonlee in #829

New Contributors

• @javiergarea made their first contribution in #822

Full Changelog: v0.15.3...v0.15.4

v0.15.3

• This release allows users to use the latest versions of Cython, numpy, scipy, and scikit-learn by updating dependencies.
• Pre-built wheels for Python 3.12, in addition to Python 3.9-3.11, are now available on PyPI for easy installation.

What's Changed

• Control to Treatment Matching by @spohngellert-o in #797
• Update uplift and Qini curve plots with a fixed random line by @jeongyoonlee in #799
• Precommit by @spohngellert-o in #800
• Remove Python 3.7 and 3.8 from and add Python 3.12 to the build tests by @jeongyoonlee in #807
• Pin scikit-learn<1.6 until xgboost fix gets released by @jeongyoonlee in #809
• Update CausalTree's base _tree module with sklearn 1.5.2 by @jeongyoonlee in #806
• resolves #796 remove dep on pygam, use IsotonicRegression by @ras44 in #803
• Remove the pygam dependency and allow the latest versions of Cython, numpy, scipy, and scikit-learn by @jeongyoonlee in #812
• Add the Python 3.12 test and wheel back by @jeongyoonlee in #814
• v0.15.3 release by @jeongyoonlee in #815

Full Changelog: v0.15.2...v0.15.3

v0.15.2

What's Changed

• fix typo in MAINTAINERS.md by @jeongyoonlee in #761
• Update python-publish GHA to build multi-platform wheels by @jeongyoonlee in #762
• Remove build for 32-bit systems that are failing by @jeongyoonlee in #763
• Remove musllinux build by @jeongyoonlee in #765
• Fix issue 764 to handle edge case in one hot encoding transformation by @paullo0106 in #766
• fix #771 and add a test for get/plot_tmlegain() by @jeongyoonlee in #773
• Add an explicit treatment input argument to causaltree/forest by @jeongyoonlee in #776
• Dev/install issues #767 #769 by @ras44 in #778
• Dev/ras44 remove envs 780 by @ras44 in #782
• Many to one propensity matching without replacement by @spohngellert-o in #786
• Update input arguments of XGBoost to be compatible with the latest APIs by @jeongyoonlee in #788
• Make torch optional. Update DragonNet w/ latest TF APIs by @jeongyoonlee in #790
• Update example references by @emmanuel-ferdman in #793
• Up the version to 0.15.2 by @jeongyoonlee in #792
• Update the cibuildwheel version to 2.21 by @jeongyoonlee in #794
• Pin the numpy version to be <2 by @jeongyoonlee in #795

New Contributors

• @spohngellert-o made their first contribution in #786
• @emmanuel-ferdman made their first contribution in #793

Full Changelog: v0.15.1...v0.15.2

v0.15.1

• This release fixes the build failure on macOS and a few bugs in UpliftTreeClassifier.
• We have two new contributors, @lee-junseok and @IanDelbridge. Thanks for your contributions!

What's Changed

• Relax pandas version requirement by @jeongyoonlee in #743
• Remove undefined variables in match.__main__() by @jeongyoonlee in #749
• Fix distr_plot_single_sim() by @jeongyoonlee in #750
• Add with_std, with_counts to create_table_one by @lee-junseok in #748
• fix stratified sampling call by @IanDelbridge in #756
• 20240207 honest leaf size by @IanDelbridge in #753
• 757: add return_ci=True in sensitivity by @lee-junseok in #758
• Update sensitivity tests with more meta-learners by @jeongyoonlee in #759
• manually specify multiprocessing use fork in setup.py by @IanDelbridge in #754
• v0.15.1 release by @jeongyoonlee in #760

New Contributors

• @lee-junseok made their first contribution in #748
• @IanDelbridge made their first contribution in #756

Full Changelog: v0.15.0...v0.15.1

v0.15.0

• In this release, we revamped documentation, cleaned up dependencies, and improved installation - in addition to the long list of bug fixes.
• We have four new contributors, @peterloleungyau, @SuperBo, @ZiJiaW, and @erikcs who submitted their first PRs to CausalML. Thanks for your contributions!

Updates

• Update python-publish.yml by @jeongyoonlee in #673
• Add build.[os, tools.python] to .readthedocs.yml by @jeongyoonlee in #676
• Update notebook example with causal trees interpretation by @alexander-pv in #683
• Remove the numpy and pandas version restriction in pyproject.toml by @jeongyoonlee in #681
• Add governance documents by @jeongyoonlee in #688
• Update GOVERNANCE.md by @ras44 in #691
• Dev/governance docs to snake-case by @ras44 in #693
• Reduce sklearn dependency in causalml by @alexander-pv in #686
• Update MAINTAINERS.md by @jeongyoonlee in #696
• Modified to speed up UpliftTreeClassifier.growDecisionTreeFrom. by @peterloleungyau in #695
• Update README.md by @ras44 in #698
• Add notebook examples to docs by @jeongyoonlee in #697
• resolves change requests in #166 by @ras44 in #701
• Fix the readthedocs build error by @jeongyoonlee in #702
• Replace Stack and PriorityHeap with cpp stack/heap methods in trees by @SuperBo in #700
• Hotfix for #701 by @jeongyoonlee in #705
• Dev/699 win build fix by @ras44 in #710
• expose n_jobs for rlearner by @ZiJiaW in #714
• minimal fix to resolve #707 by @ras44 in #720
• Add Python 3.10, 3.11, 3.12 to the testing by @cclauss in #454
• Remove Python 3.12 from the build tests in python-test.yaml by @jeongyoonlee in #726
• fix plot_std_diffs, add bal_tol, condense to one plot by @ras44 in #723
• Dev/677 documentation by @ras44 in #725
• documentation updates by @ras44 in #728
• resolves #730, docs clean conda install by @ras44 in #731
• minimal wrapper of MAQ #662 by @ras44 in #729
• Temporary fix for causal trees missing values support #733 by @alexander-pv in #734
• resolves #639, credit due to Dong Liu by @ras44 in #722
• v0.15.0 release by @jeongyoonlee in #737
• Remove maq git+pip dependency by @erikcs in #739

New Contributors

• @peterloleungyau made their first contribution in #695
• @SuperBo made their first contribution in #700
• @ZiJiaW made their first contribution in #714
• @erikcs made their first contribution in #739

Full Changelog: v0.14.1...v0.15.0

v0.14.1

• This release mainly addressed installation issues and updated documentation accordingly.
• We have 4 new contributors. @bsaunders27, @xhulianoThe1, @zpppy, and @bsaunders23. Thanks for your contributions!

What's Changed

• Update the python-publish workflow file to fix the package publish Gi… by @jeongyoonlee in #633
• Update Cython dependency by @alexander-pv in #640
• Fix for builds on Mac M1 infrastructure by @bsaunders27 in #641
• code cleanups by @xhulianoThe1 in #634
• support valid error early stopping by @zpppy in #614
• Revert "support valid error early stopping" by @jeongyoonlee in #645
• "support valid error early stopping" by @zpppy in #648
• fix: update to envs/ conda build for precompiled M1 installs by @bsaunders27 in #646
• Installation updates to README and .github/workflows by @ras44 in #637
• fix: simulate_randomized_trial by @bsaunders23 in #656
• issue 252 by @vincewu51 in #660
• ras44/651 graph viz, resolves #651 by @ras44 in #661
• linted with black by @ras44 in #663
• Fix issue 650 by @vincewu51 in #659
• Install graphviz in the workflow builds by @jeongyoonlee in #668
• Update docs/installation.rst by @jeongyoonlee in #667
• Schedule monthly PyPI install tests by @jeongyoonlee in #670
• v0.14.1 release by @jeongyoonlee in #672

New Contributors

• @bsaunders27 made their first contribution in #641
• @xhulianoThe1 made their first contribution in #634
• @zpppy made their first contribution in #614
• @bsaunders23 made their first contribution in #656

Full Changelog: v0.14.0...v0.14.1

v0.14.0

• CausalML surpassed 2MM downloads on PyPI and 4,100 stars on GitHub. Thanks for choosing CausalML and supporting us on GitHub.
• We have 7 new contributors: @darthtrevino, @ras44, @AbhishekVermaDH, @joel-mcmurry, @AlxClt, @kklein, and @volico. Welcome to the CausalML development team, and thanks for your contributions!

What's Changed

• Fix the readthedocs build failure by @jeongyoonlee in #545
• Add pyproject.toml with basic build dependencies for PEP518 compliance by @darthtrevino in #553
• bump numpy==1.23.2 in environment-py38.yml #338 by @ras44 in #550
• CausalTree split criterions fix and fit optimization by @alexander-pv in #557
• fixing math notations for proper rendering by @AbhishekVermaDH in #558
• Update methodology.rst by @joel-mcmurry in #568
• Causal trees bootstrapping and max_leaf_nodes fixes with minor update by @alexander-pv in #583
• Fix #596 by @AlxClt in #597
• Add **kwargs to Explainer.plot_shap_values() by @jeongyoonlee in #603
• Make the Adam optimization optional and learning rate/epochs configurable in DragonNet by @jeongyoonlee in #604
• Fix bug in variance calculation in drivlearner. by @huigangchen in #606
• Bug Fix in Dragonnet: Adam parameter name lr depreciation by @huigangchen in #617
• Fix AttributeError in builds with numpy>=1.24 and pandas>=2.0 by @jeongyoonlee in #631
• Pass on **kwargs in plot_shap_values of base meta leaner by @kklein in #627
• Bump scipy from 1.4.1 to 1.10.0 by @dependabot in #629
• Feature/ttest criterion by @volico in #570
• Added Interaction Tree (IT), Causal Inference Tree (CIT), and Invariant DDP (IDDP) by @jroessler in #562
• Causal trees option to return counterfactual outcomes by @alexander-pv in #623
• Up the version to 0.14 by @jeongyoonlee in #632

New Contributors

• @darthtrevino made their first contribution in #553
• @ras44 made their first contribution in #550
• @AbhishekVermaDH made their first contribution in #558
• @joel-mcmurry made their first contribution in #568
• @AlxClt made their first contribution in #597
• @kklein made their first contribution in #627
• @volico made their first contribution in #570

Full Changelog: v0.13.0...v0.14.0