v25.06.00

🚨 Breaking Changes

• Deprecate device selection (#6784) @jcrist
• Remove use of legacy FIL from Python cuML (#6728) @hcho3
• Use RBC from cuVS (#6644) @divyegala
• Map Barnes-Hut to FFT for T-SNE in cuml.accel. (#6619) @csadorf
• New Estimator Proxy architecture (#6613) @jcrist
• Disable building and uploading cuml-cpu in CI (#6529) @dantegd
• FIX Propagate random state to numpy rng in make_classification (#6518) @betatim

🐛 Bug Fixes

• UMAP batch nnd test hotfix (#6826) @jinsolp
• Update forest_inference_demo.ipynb for new FIL API changes (#6824) @dantegd
• Fix to compile FIL infer_kernel with CUDA arch 1210 (#6821) @viclafargue
• Reduce solver test threshold for ElasticNet test. (#6766) @csadorf
• Use _assert_allclose function for approximate equality check. (#6763) @csadorf
• UMAP spectral initialization fallback to random on error (#6750) @aamijar
• [Fix] include_self=False for duplicate points (tied zero distances) in kneighbors_graph (#6735) @aamijar
• FIX: Set scikit-learn KMeans n_init=10 to match cuML (#6727) @csadorf
• Fix output dtype of LinearSVC.predict (#6715) @jcrist
• CUDA 12.9 Correctly bound updated compression flags (#6713) @robertmaynard
• Fix type and support for PCA.noise_variance_ (#6693) @jcrist
• Update dask RF regressor partial_inference (#6691) @TomAugspurger
• Properly handle degenerate trees in FIL (#6673) @hcho3
• Fix strict marker in scikit-learn test xfail list. (#6661) @csadorf
• DOC Repair docstring formatting (#6659) @betatim
• Fix the field ids in the pr/issue status automation (#6656) @csadorf
• Adjust scikit-learn test failure handling (#6646) @csadorf
• Remove arbitrary score threshold for the test_qn test with softmax (#6636) @csadorf
• Do not set n_features_in_ attribute before fit (#6624) @betatim
• Map Barnes-Hut to FFT for T-SNE in cuml.accel. (#6619) @csadorf
• Mark sklearn test test_cross_val_predict[coo_array] as flaky. (#6610) @csadorf
• Do not run the scikit-learn test suite in parallel. (#6609) @csadorf
• Mark one kmeans and one t_sne sklearn test as flaky (#6598) @csadorf
• Relax threshold for failing ARIMA pytest (#6579) @divyegala
• Fix race condition in _block_gemv kernel (#6578) @divyegala
• Fix launch parameters for kernel shap (#6577) @divyegala
• Update SVC tests for CCCL update (#6569) @viclafargue
• Update FIL Model Loading Parameter in FIL demo notebook (#6562) @csadorf
• Localise output‑dtype; remove global set_api_output_dtype calls (#6561) @Ofek-Haim
• Lower logistic regression digits test threshold to 0.9. (#6552) @csadorf
• FIX Propagate random state to numpy rng in make_classification (#6518) @betatim
• Fix cudf.DataFrame construction with dict like inputs and ensure CumlArray.to_output('cudf') does not convert NaN to NA (#6517) @mroeschke
• Calling gc in SVC tests (#6514) @viclafargue
• Fix logger macros (#6511) @vyasr
• Simplicial set functions fixes and cleanups (#6493) @viclafargue
• Inspect KNN graph during smooth KNN generation and fail if necessary (#6491) @viclafargue
• More permanently disable BATCH_TREE_REORG (#6374) @wphicks

📖 Documentation

• Move FIL docs to user docs (#6762) @dantegd
• Document cuml.accel doesn't support set_output (#6717) @jcrist
• DOC Update zero-code-change documentation (#6697) @csadorf
• Document WSL2 limitation re UVM (#6663) @csadorf
• DOC Make MultinomialNB dask doctest more lax (#6482) @betatim
• Future-proof SVC documentation (#6481) @viclafargue
• DOC Add developer documentation about deprecation policy (#6476) @betatim

🚀 New Features

• Multi-gpu KNN build for UMAP using all-neighbors API (#6654) @jinsolp
• Add caching and object size metrics to conda CI builds (#6573) @divyegala
• Use Gram kernel from cuVS instead of Raft (#6538) @lowener

🛠️ Improvements

• Migrate cuml.experimental.fil -> cuml.fil (#6799) @hcho3
• Mark test_uniform_grid as flaky in sklearn tests (#6791) @jcrist
• Deprecate a few missed cuml-only non-keyword-only kwargs (#6789) @jcrist
• Delay importing of dask (#6788) @jcrist
• Port DBSCAN to ProxyBase/InteropMixin (#6786) @jcrist
• Deprecate device selection (#6784) @jcrist
• use 'rapids-init-pip' in wheel CI, other CI changes (#6779) @jameslamb
• Support SVC and SVR in cuml.accel (#6778) @jcrist
• chore: reduce ceiling of cuml wheel size (#6771) @gforsyth
• Port KMeans to use ProxyBase/InteropMixin (#6770) @jcrist
• Port sklearn.manifold wrappers to use ProxyBase (#6761) @jcrist
• Fix dask warning in FIL demo notebook (#6758) @jcrist
• Remove deprecations for 25.06 (#6754) @jcrist
• Check that as_sklearn'ed models can be fitted (#6751) @betatim
• Implement sklearn_tags on mixin using vendored compat module. (#6746) @csadorf
• Test that proxied method signatures match their CPU versions (#6740) @jcrist
• Remove upper bound constraint on xgboost. (#6739) @csadorf
• Deprecate non-keyword-only cuml-only kwargs (#6738) @jcrist
• Fix DistributedDataHandler.calculate_parts_to_sizes (#6737) @TomAugspurger
• Finish CUDA 12.9 migration and use branch-25.06 workflows (#6733) @bdice
• Remove use of legacy FIL from Python cuML (#6728) @hcho3
• Update Dask RandomForest classifier (#6726) @TomAugspurger
• Update to clang 20 (#6724) @bdice
• Use matplotlib-base in docs/notebooks environments. (#6723) @bdice
• FIX Allow prereleases when evaluating xfail conditions (#6722) @betatim
• Quote head_rev in conda recipes (#6701) @bdice
• Don't setup managed memory if rmm isn't using default settings (#6700) @jcrist
• Expand sklearn matrix used in tests (#6698) @jcrist
• Port sklearn.decomposition to new proxy estimator structure (#6695) @jcrist
• fix(rattler): add librmm to host for libcuml to fix overlinking error (#6686) @gforsyth
• CUDA 12.9 use updated compression flags (#6684) @robertmaynard
• Increase scikit-learn compatibility to 1.5+ (#6666) @csadorf
• Revert "CI: Increase nightly CI check window to 14 days. (#6628)" (#6660) @csadorf
• Remove cuml-cpu and CPU <> GPU docs (#6651) @jcrist
• More dask==2025.4.1 compatibility (#6645) @TomAugspurger
• Use RBC from cuVS (#6644) @divyegala
• Exclude librmm.so from auditwheel (#6632) @bdice
• Enhance accelerator parity testing framework (#6629) @csadorf
• CI: Increase nightly CI check window to 14 days. (#6628) @csadorf
• Add support for Python 3.13 (#6616) @gforsyth
• Compatibility with Dask 2025.4.1 (#6614) @TomAugspurger
• New Estimator Proxy architecture (#6613) @jcrist
• Using cuVS NN Descent in UMAP (#6611) @jinsolp
• Make scikit-learn a required dependency (#6608) @betatim
• Speedup dask LogisticRegression tests (#6607) @jcrist
• Add GitHub Actions workflow for PR and issue project field automation (#6602) @csadorf
• Use nogil for libcuml calls (#6601) @jcrist
• Remove cuml.common.cuda (#6600) @jcrist
• Remove most of import_utils.py (#6599) @jcrist
• Remove guarded imports of scipy (#6596) @jcrist
• Remove remainder of code branching on cuml-cpu support (#6593) @jcrist
• Add timeout to scikit-learn test suite (#6591) @jcrist
• Remove safe_imports.py (#6588) @jcrist
• Remove usage of cudf.core.buffer (#6586) @mroeschke
• Split wheel tests into two jobs (#6585) @jcrist
• Only run one cudf.pandas enabled build per PR (#6584) @jcrist
• Don't run memleak tests in CI (#6583) @jcrist
• Documentation and Testing Infrastructure Updates (#6580) @csadorf
• Move Random Forest n_streams UserWarning out of execution (#6574) @ncclementi
• Remove conditional compilation blocks leftover from cuml-cpu (#6572) @jcrist
• Download build artifacts from Github for CI (#6566) @VenkateshJaya
• Don't exclude sklearn tests from acceleration (#6565) @jcrist
• Add .git-blame-ignore-revs (#6564) @csadorf
• Add cuml.accel.is_proxy (#6559) @jcrist
• Update KMeans test to improve stability by setting n_init to 2 (#6553) (#6555) @csadorf
• Support weights="distance" for KNeighbors* in cuml.accel (#6554) @jcrist
• Improve detection of scikit-learn parity regressions (#6553) @csadorf
• Revise cuml.accel module patching and import mechanism (#6550) @jcrist
• Add isort to pre-commit hooks (#6549) @csadorf
• Refactor dataset generation and testing utilities (#6548) @csadorf
• Move cuml.accel tests out into own directory (#6547) @jcrist
• refactor(rattler): enable strict channel priority for builds (#6546) @gforsyth
• Remove unnecessary CudfBuffer and vendor cuDF min_signed_type function (#6543) @mroeschke
• Remove _experimental_dispatching attribute (#6542) @jcrist
• Avoid private attributes on cuDF objects (#6537) @mroeschke
• Reduce CI test combinations for scikit-learn acceleration tests (#6536) @csadorf
• Vendor RAPIDS.cmake (#6531) @bdice
• Disable building and uploading cuml-cpu in CI (#6529) @dantegd
• Remove CUDA whole compilation ODR violations (#6525) @divyegala
• SLHC to Migrate to cuVS Agglomerative from RAFT (#6507) @tarang-jain
• Solve forward merger conflicts of 6435 (#6501) @csadorf
• Add minimal import check to CI (#6484) @jcrist
• Moving wheel builds to specified location and uploading build artifacts to Github (#6472) @VenkateshJaya

[NIGHTLY] v25.08.00

🔗 Links

• Development Branch
• Compare with main branch

🚨 Breaking Changes

• Deprecating data_on_host parameter for UMAP (#6953) @jinsolp
• Deprecate hdbscan prediction functions in cuml.cluster namespace (#6943) @jcrist
• Reflect output type in hdbscan prediction functions (#6939) @jcrist
• Deprecate connectivity parameter to HDBSCAN (#6936) @jcrist
• Deprecate penalty='none' in MBSGDClassifier/MBSGDRegressor/SGD (#6926) @jcrist
• Remove experimental.fil Python module (#6899) @hcho3
• Change KMeans random_state default to None (#6884) @jcrist
• Remove cuvs symbols in libcuml headers (#6883) @hcho3
• Migrate C++ files from src/experimental/fil to src/fil (#6882) @hcho3
• Add test for signatures of common methods (#6878) @jcrist
• Port cuml.neighbors to InteropMixin/ProxyBase (#6851) @jcrist
• Remove CUDA 11 from dependencies.yaml (#6847) @KyleFromNVIDIA
• Remove legacy FIL from libcuml (#6844) @hcho3
• stop uploading packages to downloads.rapids.ai (#6803) @jameslamb

🐛 Bug Fixes

• Remove CuPy version constraint (#6979) @csadorf
• Fix test_typeerror_input test with xfail for known sklearn LARS broadcasting bug (#6978) @csadorf
• Add upper constraint for cupy 13.5 due to memory access issues (#6961) @csadorf
• Decrease nightly check window to 7 days (#6934) @csadorf
• Support empty categorical nodes in FIL (#6924) @hcho3
• Rewrite HDBSCAN python wrapper (#6913) @jcrist
• Mark test_warm_start_oob[RandomForestClassifier] as flaky (#6906) @csadorf
• Unxfail tests affected by numba compilation error on Python 3.13.4 (#6905) @csadorf
• Fix import in benchmark code in algorithms.py that was causing an error (#6902) @dantegd
• Add test for signatures of common methods (#6878) @jcrist
• Fix CI test failures: xfail tests for scikit-learn 1.7.0 and Python 3.13.4 compatibility (#6865) @csadorf
• Fix RandomForestRegressor default max_features (#6862) @jcrist
• Increase tolerance in Dask logistic regression test (#6848) @csadorf

📖 Documentation

• Remove async operation section from Python developer docs (#6980) @csadorf
• Revamp cuml.accel limitations docs page (#6965) @jcrist
• add docs on CI workflow inputs (#6952) @jameslamb
• Fixup rst formatting of 0cc limitations doc (#6945) @jcrist
• Documentation Updates for cuML Python Developer Guide (#6843) @csadorf

🛠️ Improvements

• Use pytest.ini for all upstream tests (#6994) @jcrist
• Support invoking cuml_accel_tests without extra pytest flags (#6993) @jcrist
• Fall back on unsupported metrics in TSNE (#6992) @jcrist
• Expand LogisticRegression limitations (#6991) @jcrist
• Expand sklearn.ensemble limitations (#6990) @jcrist
• Run umap-learn test suite in CI (#6989) @jcrist
• Move cuml.accel upstream tests to the cuml_accel_tests directory (#6986) @jcrist
• Compute components_ in DBSCAN (#6976) @jcrist
• Add the cython-lint commit to .git-blame-ignore-revs (#6972) @jcrist
• Several SVC cleanups (#6970) @jcrist
• Update cython-lint, fix long lines (#6969) @jcrist
• Enhance logging for cuml.accel proxy estimators (#6957) @csadorf
• Add a reason to all UnsupportedOnGPU/UnsupportedOnCPU errors (#6954) @jcrist
• Deprecating data_on_host parameter for UMAP (#6953) @jinsolp
• Implement metadata routing in cuml.accel (#6950) @jcrist
• xfail some tests due to new scipy 1.16 release (#6946) @jcrist
• Use CUDA 12.9 in Conda, Devcontainers, Spark, GHA, etc. (#6944) @jakirkham
• Deprecate hdbscan prediction functions in cuml.cluster namespace (#6943) @jcrist
• Support set_output and get_feature_names_out in cuml.accel (#6942) @jcrist
• Reflect output type in hdbscan prediction functions (#6939) @jcrist
• Remove Base.__getattr__/Base.__getstate__/Base.__setstate__ (#6937) @jcrist
• Deprecate connectivity parameter to HDBSCAN (#6936) @jcrist
• Remove nvidia and dask channels (#6935) @vyasr
• Fixup KernelRidge hypothesis test (#6931) @jcrist
• A few fixes for CI (#6928) @jcrist
• Deprecate penalty='none' in MBSGDClassifier/MBSGDRegressor/SGD (#6926) @jcrist
• Run FIL demo notebook in CI (#6923) @hcho3
• Restore old behavior for threshold in FIL (#6922) @hcho3
• De-Cython several modules (#6920) @jcrist
• Remove UniversalBase (#6919) @jcrist
• Add KernelRidge to cuml.accel (#6917) @jcrist
• Port HDBSCAN to InteropMixin/ProxyBase (#6916) @jcrist
• Expose n_iter_ for LogisticRegression models (#6911) @betatim
• Warn and fail on stale xfail list entries (#6907) @csadorf
• refactor(shellcheck): fix all shellcheck warnings/errors (#6901) @gforsyth
• Instantiate only specific RAFT linewise kernels (#6900) @aamijar
• Remove experimental.fil Python module (#6899) @hcho3
• Remove GlobalSettings.device_type (#6895) @jcrist
• Add xfail list auto-formatter (#6892) @csadorf
• Remove deprecated device selection functions (#6888) @jcrist
• Remove deprecations for 25.08 (#6887) @jcrist
• Change KMeans random_state default to None (#6884) @jcrist
• Remove cuvs symbols in libcuml headers (#6883) @hcho3
• Migrate C++ files from src/experimental/fil to src/fil (#6882) @hcho3
• Remove invalid xfail entry and mark correct test as flaky (#6877) @csadorf
• A few max_features fixups for cuml.ensemble (#6873) @jcrist
• Loosen constrain on numpy upper version (#6871) @csadorf
• Correct sys.argv when running cuml.accel (#6870) @jcrist
• Port cuml.ensemble to InteropMixin/ProxyBase (#6863) @jcrist
• Deprecate some cuml.accel CLI options (#6855) @jcrist
• Remove deprecated args in cuml.ensemble (#6854) @jcrist
• Move deprecated positional/keyword args to keyword-only (#6853) @jcrist
• Support -c and - options in cuml.accel CLI (#6852) @jcrist
• Port cuml.neighbors to InteropMixin/ProxyBase (#6851) @jcrist
• Remove CUDA 11 from dependencies.yaml (#6847) @KyleFromNVIDIA
• Add support for scikit-learn version 1.4 (#6845) @betatim
• Remove legacy FIL from libcuml (#6844) @hcho3
• Port UMAP to InteropMixin/ProxyBase (#6840) @jcrist
• Move FIL to subclass from Base (#6835) @jcrist
• Mark test_equivalence_components_pca_spca as flaky (#6834) @jcrist
• Fix several bugs in cuml.accel CLI (#6832) @jcrist
• stop uploading packages to downloads.rapids.ai (pt 2) (#6828) @jameslamb
• Remove CUDA 11 devcontainers and update CI scripts (#6818) @bdice
• refactor(rattler): remove cuda11 options and general cleanup (#6817) @gforsyth
• Add GitHub Actions workflow to assign PRs to their authors (#6816) @csadorf
• stop uploading packages to downloads.rapids.ai (#6803) @jameslamb
• Instantiate only specific RAFT reduction kernels (#6780) @divyegala
• Forward-merge branch-25.06 into branch-25.08 (#6647) @gforsyth
• [FEA] HDBSCAN to Migrate to cuVS cluster prims from raft::cluster (#6560) @tarang-jain

v25.04.00

🚨 Breaking Changes

• Promote experimental FIL to stable (#6464) @wphicks
• Support non-trivial classes_ in LogisticRegression (#6346) @jcrist
• Use new rapids-logger library (#6289) @vyasr

🐛 Bug Fixes

• Unblock CI for 25.04 (#6519) @csadorf
• Skip test_rf_classification_seed for cudf.pandas tests. (#6500) @csadorf
• Remove dask to sparse workarounds (#6489) @TomAugspurger
• Fix accidental required sklearn import (#6483) @jcrist
• FIX Pass instances to the constructor of the meta-estimator in SVC (#6471) @betatim
• Fix compiler dependency on ARM (#6456) @bdice
• Revert "Temporarily increase max_days_without_success (#6390)" (#6455) @divyegala
• Enhance argument handling in UniversalBase to accept NoneType (#6453) @csadorf
• Fix UMAP transform (#6449) @viclafargue
• Fallback to CPU for sparse inputs for KMeans (#6448) @csadorf
• Limited support for array-like inputs (#6442) @csadorf
• Fix test_accuracy_score test on cudf.pandas build (#6439) @jcrist
• Support positional args in cuml.accel estimators (#6423) @jcrist
• Fix HDBSCAN python docs related to metric (#6422) @divyegala
• declare runtime dependency on 'packaging', update scikit-learn and hdbscan for cuml-cpu to match cuml (#6420) @jameslamb
• Fix forwarding of initial_alpha and learning_rate in UMAP (#6417) @jcrist
• Implement Ridge .solver_ estimated attribute (#6415) @csadorf
• Fix multi-target prediction in linear models (#6414) @csadorf
• Ensure output_type="pandas" returns user-facing pandas (#6407) @jcrist
• Ensure LinearSVC supports all input types (#6404) @jcrist
• Make UMAP callback pickleable (#6402) @jcrist
• Correctly align trees in experimental FIL (#6397) @wphicks
• A few log level handling cleanups (#6393) @jcrist
• Fix cupy error in KernelDensity with epanechnikov (#6388) @jcrist
• Check if both ground truth and result are NaN in MSEObjectiveTest for RF (#6387) @wphicks
• Support non-native endian inputs to LabelEncoder (#6384) @jcrist
• Fix hypothesis test in test_kernel_ridge.py (#6382) @jcrist
• Remove debug logging from nearest neighbors test (#6376) @csadorf
• Ensure FIL CPU can be run without an available GPU (#6373) @wphicks
• Correctly pass through unhashable objects during hyperparam lookup (#6369) @wphicks
• Correct translation of RandomForest criterion hyperparameter (#6363) @wphicks
• UMAP fallback improvements (#6358) @viclafargue
• A few GPU<->CPU interop fixes (#6355) @jcrist
• Fix output type of KernelRidge.predict (#6354) @jcrist
• Skip test_extract_partitions_shape test. (#6338) @csadorf
• Require sphinx<8.2.0 (#6336) @csadorf
• Fix call to NearestNeighbors.kneighbors() without args. (#6333) @csadorf
• Commenting out debug print statements in UMAP (#6332) @viclafargue
• Correct UMAP dispatch trigger condition (#6330) @viclafargue
• Fix issue caused by None arg provided by sklearn pipeline (#6326) @viclafargue
• Improvements for interoperability of Random Forest (#6320) @dantegd
• Improvements for PCA CPU interoperability (#6319) @dantegd
• Plumb metric and metric_kwds through to UMAP with nn_descent (#6304) @jcrist
• Use L2Expanded/L2SqrtExpanded instead of the *Unexpanded variants (#6303) @jcrist
• Fix UMAP issues with large inputs (#6245) @viclafargue
• Fix LogisticRegression.decision_function output shape (#6235) @jcrist
• Compute noise_variance_ in PCA implementation (#6234) @jcrist
• Multiple CPU interop fixes for serialization and cloning (#6223) @dantegd

📖 Documentation

• Clarify support in docs (#6452) @bdice
• Provide docs for cuml.accel command line feature (#6322) @wphicks

🚀 New Features

• Promote experimental FIL to stable (#6464) @wphicks
• UMAP 32bits dispatch mechanism (#6314) @viclafargue
• Add aarch64 conda environments (#6226) @jcrist
• SVM cpu/gpu interop (#6089) @viclafargue

🛠️ Improvements

• Don't use private APIs in FIL notebook (#6488) @jcrist
• Only use released versions of dask (#6485) @jcrist
• Move SVM working set log to debug (#6477) @jcrist
• C++ SG_RF_TEST OOMs in 11.4 nightly (#6474) @divyegala
• Revert "ci: pre-filter 11.4 jobs before they are enabled in shared workflows (#6447)" (#6470) @divyegala
• Support __len__ in ensemble estimators (#6468) @jcrist
• Add deprecation warning to cuml-cpu builds (#6466) @jcrist
• Fix template argument of DeviceSegmentedReduce kernels (#6459) @miscco
• Add scikit-learn acceleration tests to nightly CI tests (#6457) @csadorf
• Fix benchmark builds in devcontainers (#6450) @bdice
• ci: pre-filter 11.4 jobs before they are enabled in shared workflows (#6447) @gforsyth
• Reenable more UMAP tests on ARM (#6446) @jcrist
• Clean up after removing KMeans deprecation warning (#6445) @betatim
• Port all conda recipes to rattler-build (#6440) @gforsyth
• Remove the deprecation warning and switch to new default (#6433) @betatim
• Remove old deprecations (#6427) @jcrist
• Fix oob_score=True handling with accelerator (#6425) @betatim
• remove unused imports (#6424) @jameslamb
• Disable codecov comments (#6419) @jcrist
• Require hypothesis tests have explicit examples (#6409) @aamijar
• Improve accuracy_score compatibility with sklearn. (#6406) @jcrist
• Drop cub{Min, Max} (#6405) @miscco
• Move accel implementation fully to cuml.accel (#6401) @jcrist
• cuml-cpu: fix import issues, enable conda import tests (#6400) @jameslamb
• Drop numba_utils (#6396) @jakirkham
• Simplify cpu_to_gpu and gpu_to_cpu (#6395) @jcrist
• Adds get_level() (#6392) @Intron7
• Temporarily increase max_days_without_success (#6390) @jcrist
• Drop deprecated CCCL functors (#6389) @miscco
• Solve forward merger conflicts of #6313 (#6385) @dantegd
• CI: Run extended accel mode tests (#6377) @csadorf
• Use conda-build instead of conda-mambabuild (#6368) @bdice
• Don't log coverage output to terminal in CI (#6367) @jcrist
• Remove KernelRidge from cuml.accel (#6361) @jcrist
• Enable UVM in cuml.accel (#6359) @wphicks
• Drop deprecated CUB iterators in favor of Thrust iterators (#6356) @miscco
• Adjust accel verbosity (#6353) @csadorf
• Avoid importing umap with import cuml (#6352) @jcrist
• Make proxy module patching skip patching if patched library is not installed (#6348) @dantegd
• Add new cuml.accel module (#6347) @jcrist
• Support non-trivial classes_ in LogisticRegression (#6346) @jcrist
• Make r2_score compatible with Scikit-Learn (#6337) @jcrist
• Remove left-over code related to strict mode. (#6335) @csadorf
• Make sure specifying default value or nothing behaves the same in KMeans (#6331) @betatim
• Make KernelRidge inherit from UniversalBase (#6327) @betatim
• Reduce peak memory in UMAP.fit/UMAP.fit_transform (#6323) @jcrist
• Consolidate more Conda solves in CI (#6321) @KyleFromNVIDIA
• Require CMake 3.30.4 (#6318) @robertmaynard
• Copy attributes from the original class to the proxy (#6306) @betatim
• Use shared-workflows branch-25.04 (#6298) @bdice
• Add build_type input field for test.yaml (#6296) @gforsyth
• Do not use LEGACY_PTX_ARCH argument for BlockReduce (#6290) @miscco
• Use new rapids-logger library (#6289) @vyasr
• disallow fallback to Make in wheel builds (#6286) @jameslamb
• Forward-merge branch-25.02 to branch-25.04 (#6283) @bdice
• Migrate to NVKS for amd64 CI runners (#6280) @bdice
• Add verify-codeowners hook (#6275) @KyleFromNVIDIA
• Speed up FIL and TSNE pytests in CI (#6270) @dantegd
• Bracnh 25.04 merge branch 25.02 (#6268) @vyasr
• Add shellcheck to pre-commit and fix warnings (#6246) @gforsyth
• Prepare for n_init=auto in KMeans (#6142) @betatim

v25.02.01

🚨 Breaking Changes

• update pip devcontainers to UCX 1.18 (#6249) @jameslamb

🐛 Bug Fixes

• Remove straggling click option (#6381) @dantegd
• Fix dask logistic regression segfault/hang due to wrongly typed variable (#6281) @dantegd
• Log UMAP arrays at trace verbosity level. (#6274) @csadorf
• Ensure all method signatures are sklearn compatible (#6260) @jcrist
• Fix UMAP transform illegal memory access error when data_on_host=True (#6259) @csadorf
• Ignore cudf's dataframe deprecation. (#6229) @bdice
• Fix for porter stemmer for cuDF change and ARIMA pytest adjustments (#6227) @dantegd
• Avoid duplicate log entries (#6222) @jcrist
• Further fixes for Scipy 1.15 update for PR and nightly CI (#6213) @dantegd
• Update setulb call from Scipy for updated signature of 1.15 (#6207) @dantegd
• Adjust test_kmeans to avoid false positive failures (#6193) @dantegd
• Adjust margin of logistic regression log_proba pytest to avoid false positive failures (#6188) @dantegd
• Skip flaky test of kernel_density in nightly job of CUDA 12.0.1 (#6184) @dantegd
• Try to reduce network usage in cuML tests. (#6174) @bdice
• cuML dask fixes to unblock CI (#6170) @dantegd
• Remap BATCH_TREE_REORG to TREE_REORG in FIL (#6161) @wphicks

📖 Documentation

• Fixes github link from pyx files (#6202) @thomasjpfan

🚀 New Features

• Allow CUDA ODR violations in 25.02 (#6264) @robertmaynard
• Define block size for sm_120 (#6250) @robertmaynard
• Support alpha=0 in Ridge (#6236) @jcrist
• Add as_sklearn and from_sklearn APIs to serialize to CPU sklearn-estimators for supported models (#6102) @dantegd

🛠️ Improvements

• Backport release 25.04 PRs for patch release version 25.02.01 (#6329) @dantegd
• Use rapids-pip-retry in CI jobs that might need retries (#6293) @gforsyth
• Avoid large device allocation in UMAP with nndescent (#6292) @jcrist
• Revert CUDA 12.8 shared workflow branch changes (#6282) @vyasr
• Build and test with CUDA 12.8.0 (#6266) @bdice
• update pip devcontainers to UCX 1.18 (#6249) @jameslamb
• Drop deprecated thrust features and replace with libcu++ ones (#6248) @miscco
• Add upper bound to prevent usage of numba 0.61.0 (#6244) @galipremsagar
• Normalize whitespace (#6238) @bdice
• Rename cpp/test to cpp/tests. (#6237) @bdice
• Use cuda.bindings layout. (#6233) @bdice
• Skip dispatching to GPU for unimplemented metrics in UMAP (#6224) @betatim
• Use GCC 13 in CUDA 12 conda builds. (#6221) @bdice
• declare cuda-python dependency for wheels, other small packaging changes (#6217) @jameslamb
• Bump Treelite to 4.4.1 (#6212) @hcho3
• Support raft's logger targets (#6208) @vyasr
• Use rapids-cmake for the logger (#6205) @vyasr
• consolidate pytest config in pyproject.toml (#6201) @jameslamb
• introduce libcuml wheels (#6199) @jameslamb
• Check if nightlies have succeeded recently enough (#6196) @vyasr
• Remove sphinx pinning (#6195) @vyasr
• simplify wheel CI scripts, other small packaging changes (#6190) @jameslamb
• Update for raft logger changes (#6187) @vyasr
• update telemetry actions to fluent-bit friendly style (#6186) @msarahan
• HDBSCAN and KMeans API improvements for improving CPU interoperability (#6181) @dantegd
• CPU/GPU interop with RandomForest (#6175) @hcho3
• Update version references in workflow (#6172) @AyodeAwe
• gate telemetry dispatch calls on TELEMETRY_ENABLED env var (#6171) @msarahan
• Use rapids-logger to generate the cuml logger (#6162) @vyasr
• Use estimator tags to improve sparse error handling (#6151) @dantegd
• FEA Add support for accepting a Numpy RandomState (#6150) @betatim
• prefer system install of UCX in pip devcontainers, update outdated RAPIDS references (#6149) @jameslamb
• Improve infrastructure for experimental dispatching of non existing methods in cuML (#6148) @dantegd
• Adapt to rmm logger changes (#6147) @vyasr
• Require approval to run CI on draft PRs (#6145) @bdice
• Add breaking change workflow trigger (#6130) @AyodeAwe
• Switch to native traceback in cuml (#6078) @galipremsagar

v24.12.00

🚨 Breaking Changes

• Forward merge Branch 24.10 into 24.12 (#6106) @divyegala

🐛 Bug Fixes

• Fix SSL error. (#6177) @bdice
• Fix scikit-learn version specifier (#6158) @trxcllnt
• Correctly handle missing categorical data in experimental FIL (#6132) @wphicks
• Put a ceiling on cuda-python (#6131) @bdice
• Don't presume pointers are mutually exclusive for device/host. (#6128) @robertmaynard
• cuml SINGLEGPU now tells cuvs to not build with nccl/mg support (#6127) @robertmaynard
• Remove type from pickle header for CumlArray (#6120) @wphicks
• Forward merge Branch 24.10 into 24.12 (#6106) @divyegala
• Fix Dask estimators serialization prior to training (#6065) @viclafargue

🚀 New Features

• Enable HDBSCAN gpu training and cpu inference (#6108) @divyegala

🛠️ Improvements

• Update FIL tests to use XGBoost UBJSON instead of binary (#6153) @hcho3
• Use sparse knn / distances from cuvs (#6143) @benfred
• Ensure MG to have the same number of allreduce calls in mean_stddev for sparse matrix to avoid hanging (#6141) @lijinf2
• Stop excluding cutlass from symbol exclusion check (#6140) @vyasr
• Optimize MG variance calculation for dataset standardization for logistic regression (#6138) @lijinf2
• enforce wheel size limits, README formatting in CI (#6136) @jameslamb
• Experimental command line interface UX (#6135) @dantegd
• add telemetry (#6126) @msarahan
• Make cuVS optional if CUML_ALGORITHMS is set (#6125) @hcho3
• devcontainer: replace VAULT_HOST with AWS_ROLE_ARN (#6118) @jjacobelli
• print sccache stats in builds (#6111) @jameslamb
• fix version in Doxygen docs (#6104) @jameslamb
• make conda installs in CI stricter (#6103) @jameslamb
• Make get_param_names a class method on single GPU estimators to match Scikit-learn closer (#6101) @dantegd
• Prune workflows based on changed files (#6094) @KyleFromNVIDIA
• Update all rmm imports to use pylibrmm/librmm (#6084) @Matt711
• Merge branch 24.10 into branch 24.12 (#6083) @jameslamb

v24.10.00

🚨 Breaking Changes

• Remove old dask-glm based logistic regression (#6028) @dantegd

🐛 Bug Fixes

• Fix train_test_split for string columns (#6088) @dantegd
• Stop shadowing free function (#6076) @vyasr
• Set default values for conftest options. (#6067) @bdice
• Add license file to conda packages (#6061) @raydouglass
• Fix np.NAN to np.nan. (#6056) @bdice
• Reenable pytest cuml-dask for CUDA 12.5 wheel CI tests (#6051) @divyegala
• Fix for simplicial_set_embedding (#6043) @viclafargue
• MAINT: Allow for error message to contain np.float32(1.0) (#6030) @seberg
• Stop exporting fill_k kernel as that causes ODR violations (#6021) @robertmaynard
• Avoid cudf column APIs after cudf.Series disallows column inputs (#6019) @mroeschke
• Use HDBSCAN package pin to 0.8.38 (#5906) @divyegala

📖 Documentation

• Update UMAP doc (#6064) @viclafargue
• Update README in experimental FIL (#6052) @hcho3
• add docs for simplicial_set (#6042) @Intron7

🚀 New Features

• TSNE CPU/GPU Interop (#6063) @divyegala
• Enable GPU fit and CPU transform in UMAP (#6032) @divyegala

🛠️ Improvements

• Migrate to use cuVS for vector search (#6085) @benfred
• Support all-zeroes feature vectors for MG sparse logistic regression (#6082) @lijinf2
• Update update-version.sh to use packaging lib (#6081) @AyodeAwe
• Use CI workflow branch 'branch-24.10' again (#6072) @jameslamb
• Update fmt (to 11.0.2) and spdlog (to 1.14.1), add those libraries to libcuml conda host dependencies (#6071) @jameslamb
• Update flake8 to 7.1.1. (#6070) @bdice
• Add support for Python 3.12, update to umap-learn==0.5.6 (#6060) @jameslamb
• Fix compiler warning about signed vs unsigned ints (#6053) @hcho3
• Update rapidsai/pre-commit-hooks (#6048) @KyleFromNVIDIA
• Drop Python 3.9 support (#6040) @jameslamb
• Add use_cuda_wheels matrix entry (#6038) @KyleFromNVIDIA
• Switch debug build to RelWithDebInfo (#6033) @rongou
• Remove NumPy <2 pin (#6031) @seberg
• Remove old dask-glm based logistic regression (#6028) @dantegd
• [FEA] UMAP API for building with batched NN Descent (#6022) @jinsolp
• Enabling CPU/GPU interop for SVM, DBSCAN and KMeans (#6020) @viclafargue
• Update pre-commit hooks (#6016) @KyleFromNVIDIA
• Improve update-version.sh (#6014) @bdice
• Use tool.scikit-build.cmake.version, set scikit-build-core minimum-version (#6012) @jameslamb
• Merge branch-24.08 into branch-24.10 (#5981) @jameslamb
• Use CUDA math wheels (#5966) @KyleFromNVIDIA

v24.08.00

🐛 Bug Fixes

• Fixes for encoders/transformers for cudf.pandas (#5990) @dantegd
• BUG: remove sample parameter from pca call to mean (#5980) @mfoerste4
• Fix segfault and other errors in ForestInference.load_from_sklearn (#5973) @hcho3
• Rename .devcontainers for CUDA 12.5 (#5967) @jakirkham
• [MNT] Small NumPy 2 related fixes (#5954) @seberg
• CI Fix: use ld_preload to avoid libgomp issue on ARM jobs (#5949) @dantegd
• Fix for benchmark runner to handle parameter sweeps of multiple data types (#5938) @dantegd
• Avoid extra memory copy when using cp.concatenate in cuml.dask kmeans (#5937) @dantegd
• Assign correct labels_ in cuml.dask.kmeans (#5931) @dantegd
• Fix nightly jobs by updating hypothesis strategies to account for sklearn change (#5925) @dantegd
• Fix for SVC fit_proba not using class weights (#5912) @pablotanner
• Fix cudf.pandas failure on test_convert_input_dtype (#5885) @dantegd
• Fix cudf.pandas failure on test_convert_matrix_order_cuml_array (#5882) @dantegd
• Simplify cuml array (#5166) @wence-

🚀 New Features

• [FEA] Enable UMAP to build knn graph using NN Descent (#5910) @jinsolp
• Allow estimators to accept any dtype (#5888) @dantegd

🛠️ Improvements

• Add support for XGBoost UBJSON in FIL (#6009) @hcho3
• split up CUDA-suffixed dependencies in dependencies.yaml (#5974) @jameslamb
• Use workflow branch 24.08 again (#5970) @KyleFromNVIDIA
• Bump Treelite to 4.3.0 (#5968) @hcho3
• reduce memory_footprint for sparse PCA transform (#5964) @Intron7
• Build and test with CUDA 12.5.1 (#5963) @KyleFromNVIDIA
• Support int64 index type in MG sparse LogisticRegression (#5962) @lijinf2
• Add CUDA_STATIC_MATH_LIBRARIES (#5959) @KyleFromNVIDIA
• skip CMake 3.30.0 (#5956) @jameslamb
• Make ci/run_cuml_dask_pytests.sh environment-agnostic again (#5950) @trxcllnt
• Use verify-alpha-spec hook (#5948) @KyleFromNVIDIA
• nest cuml one level deeper in python (#5944) @msarahan
• resolve dependency-file-generator warning, other rapids-build-backend followup (#5928) @jameslamb
• Adopt CI/packaging codeowners (#5923) @bdice
• Remove text builds of documentation (#5921) @vyasr
• Fix conflict of forward-merge #5905 of branch-24.06 into branch-24.08 (#5911) @dantegd
• Bump Treelite to 4.2.1 (#5908) @hcho3
• remove unnecessary 'setuptools' dependency (#5901) @jameslamb
• [FEA] PCA Initialization for TSNE (#5897) @aamijar
• Use rapids-build-backend (#5804) @KyleFromNVIDIA

v24.06.01

🐛 Bug Fixes

• [HOTFIX] Fix import of sklearn by using cpu_only_import (#5914) @dantegd
• Fix label binarize for binary class (#5900) @jinsolp
• Fix RandomForestClassifier return type (#5896) @jinsolp
• Fix nightly CI: remove deprecated creation of columns by using explicit dtype (#5880) @dantegd
• Fix DBSCAN allocates rbc index even if deactivated (#5859) @mfoerste4
• Remove gtest from dependencies.yaml (#5854) @robertmaynard
• Support expression-based Dask Dataframe API (#5835) @rjzamora
• Mark all kernels with internal linkage (#5764) @robertmaynard
• Fix build.sh clean command (#5730) @csadorf

📖 Documentation

• Update the developer's guide with new copyright hook (#5848) @KyleFromNVIDIA

🚀 New Features

• Always use a static gtest and gbench (#5847) @robertmaynard

🛠️ Improvements

• [HOTFIX] Add compatibility of imports with multiple Scikit-learn versions (#5922) @dantegd
• Support double precision in MNMG Logistic Regression (#5898) @lijinf2
• Reduce and rename cudf.pandas integrations jobs (#5890) @dantegd
• Fix building cuml with CCCL main (#5886) @trxcllnt
• Add optional CI job for integration tests with cudf.pandas (#5881) @dantegd
• Enable pytest failures on FutureWarnings/DeprecationWarnings (#5877) @mroeschke
• Remove return in test_lbfgs (#5875) @mroeschke
• Avoid dask_cudf.core imports (#5874) @bdice
• Support CPU object for train_test_split (#5873) @isVoid
• Only use functions in the limited API (#5871) @vyasr
• Replace deprecated disutils.version with packaging.version (#5868) @mroeschke
• Adjust deprecated cupy.sparse usage (#5867) @mroeschke
• Fix numpy 2.0 deprecations (#5866) @mroeschke
• Fix deprecated positional arg usage (#5865) @mroeschke
• Use int instead of float in random.randint (#5864) @mroeschke
• Migrate to {{ stdlib("c") }} (#5863) @hcho3
• Avoid deprecated API in notebook (#5862) @rjzamora
• Add dedicated handling for cudf.pandas wrapped Numpy arrays (#5861) @betatim
• Prepend devcontainer name with the username (#5860) @trxcllnt
• add --rm and --name to devcontainer run args (#5857) @trxcllnt
• Update pip devcontainers to UCX v1.15.0 (#5856) @trxcllnt
• Replace rmm::mr::device_memory_resource* with rmm::device_async_resource_ref (#5853) @harrism
• Update scikit-learn to 1.4 (#5851) @betatim
• Prevent undefined behavior when passing handle from Treelite to cuML FIL (#5849) @hcho3
• Adds missing files to update-version.sh (#5830) @AyodeAwe
• Enable all tests for arm arch (#5824) @galipremsagar
• Address PytestReturnNotNoneWarning in cuml tests (#5819) @mroeschke
• Handle binary classifier with all-0 labels (#5810) @hcho3
• Use pytest_cases.fixture to fix warnings. (#5798) @bdice
• Enable Dask tests with UCX-Py/UCXX in CI (#5697) @pentschev

v24.06.00

🐛 Bug Fixes

• [HOTFIX] Fix import of sklearn by using cpu_only_import (#5914) @dantegd
• Fix label binarize for binary class (#5900) @jinsolp
• Fix RandomForestClassifier return type (#5896) @jinsolp
• Fix nightly CI: remove deprecated creation of columns by using explicit dtype (#5880) @dantegd
• Fix DBSCAN allocates rbc index even if deactivated (#5859) @mfoerste4
• Remove gtest from dependencies.yaml (#5854) @robertmaynard
• Support expression-based Dask Dataframe API (#5835) @rjzamora
• Mark all kernels with internal linkage (#5764) @robertmaynard
• Fix build.sh clean command (#5730) @csadorf

📖 Documentation

• Update the developer's guide with new copyright hook (#5848) @KyleFromNVIDIA

🚀 New Features

• Always use a static gtest and gbench (#5847) @robertmaynard

🛠️ Improvements

• Support double precision in MNMG Logistic Regression (#5898) @lijinf2
• Reduce and rename cudf.pandas integrations jobs (#5890) @dantegd
• Fix building cuml with CCCL main (#5886) @trxcllnt
• Add optional CI job for integration tests with cudf.pandas (#5881) @dantegd
• Enable pytest failures on FutureWarnings/DeprecationWarnings (#5877) @mroeschke
• Remove return in test_lbfgs (#5875) @mroeschke
• Avoid dask_cudf.core imports (#5874) @bdice
• Support CPU object for train_test_split (#5873) @isVoid
• Only use functions in the limited API (#5871) @vyasr
• Replace deprecated disutils.version with packaging.version (#5868) @mroeschke
• Adjust deprecated cupy.sparse usage (#5867) @mroeschke
• Fix numpy 2.0 deprecations (#5866) @mroeschke
• Fix deprecated positional arg usage (#5865) @mroeschke
• Use int instead of float in random.randint (#5864) @mroeschke
• Migrate to {{ stdlib("c") }} (#5863) @hcho3
• Avoid deprecated API in notebook (#5862) @rjzamora
• Add dedicated handling for cudf.pandas wrapped Numpy arrays (#5861) @betatim
• Prepend devcontainer name with the username (#5860) @trxcllnt
• add --rm and --name to devcontainer run args (#5857) @trxcllnt
• Update pip devcontainers to UCX v1.15.0 (#5856) @trxcllnt
• Replace rmm::mr::device_memory_resource* with rmm::device_async_resource_ref (#5853) @harrism
• Update scikit-learn to 1.4 (#5851) @betatim
• Prevent undefined behavior when passing handle from Treelite to cuML FIL (#5849) @hcho3
• Adds missing files to update-version.sh (#5830) @AyodeAwe
• Enable all tests for arm arch (#5824) @galipremsagar
• Address PytestReturnNotNoneWarning in cuml tests (#5819) @mroeschke
• Handle binary classifier with all-0 labels (#5810) @hcho3
• Use pytest_cases.fixture to fix warnings. (#5798) @bdice
• Enable Dask tests with UCX-Py/UCXX in CI (#5697) @pentschev

v24.04.00

🐛 Bug Fixes

• Update pre-commit-hooks to v0.0.3 (#5816) @KyleFromNVIDIA
• Correct and adjust tolerances of mnmg logreg pytests (#5812) @dantegd
• Remove use of cudf.core.column.full. (#5794) @bdice
• Suppress all HealthChecks on test_split_datasets. (#5791) @bdice
• Suppress a hypothesis HealthCheck on test_split_datasets that fails in nightly CI. (#5790) @bdice
• [BUG] Fix MAX_THREADS_PER_SM on sm 89. (#5785) @trivialfis
• fix device to host copy not sync stream in logistic regression mg (#5766) @lijinf2
• Use cudf.Index instead of cudf.GenericIndex. (#5738) @bdice
• update RAPIDS dependencies to 24.4, refactor dependencies.yaml (#5726) @jameslamb

🚀 New Features

• Support CUDA 12.2 (#5711) @jameslamb

🛠️ Improvements

• Use conda env create --yes instead of --force (#5822) @bdice
• Bump Treelite to 4.1.2 (#5814) @hcho3
• Support standardization for sparse vectors in logistic regression MG (#5806) @lijinf2
• Update script input name (#5802) @AyodeAwe
• Add upper bound to prevent usage of NumPy 2 (#5797) @bdice
• Enable pytest failures on warnings from cudf (#5796) @mroeschke
• Use public cudf APIs where possible (#5795) @mroeschke
• Remove hard-coding of RAPIDS version where possible (#5793) @KyleFromNVIDIA
• Switch pytest-xdist algorithm to worksteal (#5792) @bdice
• Automate C++ include file grouping and ordering using clang-format (#5787) @harrism
• Add support for Python 3.11, require NumPy 1.23+ (#5786) @jameslamb
• [ENH] Let cuDF handle input types for label encoder. (#5783) @trivialfis
• Install test dependencies at the same time as cuml packages. (#5781) @bdice
• Update devcontainers to CUDA Toolkit 12.2 (#5778) @trxcllnt
• target branch-24.04 for GitHub Actions workflows (#5776) @jameslamb
• Add environment-agnostic scripts for running ctests and pytests (#5761) @trxcllnt
• Pandas 2.x support (#5758) @dantegd
• Update ops-bot.yaml (#5752) @AyodeAwe
• Forward-merge branch-24.02 to branch-24.04 (#5735) @bdice
• Replace local copyright check with pre-commit-hooks verify-copyright (#5732) @KyleFromNVIDIA
• DBSCAN utilize rbc eps_neighbors (#5728) @mfoerste4