Fix Changelog Merge Conflicts for branch-21.12

CHANGELOG.md

@@ -2,9 +2,75 @@
 
 Please see https://github.com/rapidsai/cuml/releases/tag/v21.12.00a for the latest changes to this development branch.
 
-# cuML 21.10.00 (Date TBD)
+# cuML 21.10.00 (7 Oct 2021)
 
-Please see https://github.com/rapidsai/cuml/releases/tag/v21.10.00a for the latest changes to this development branch.
+## 🚨 Breaking Changes
+
+- RF: python api behaviour refactor ([#4207](https://github.com/rapidsai/cuml/pull/4207)) [@venkywonka](https://github.com/venkywonka)
+- Implement vector leaf for random forest ([#4191](https://github.com/rapidsai/cuml/pull/4191)) [@RAMitchell](https://github.com/RAMitchell)
+- Random forest refactoring ([#4166](https://github.com/rapidsai/cuml/pull/4166)) [@RAMitchell](https://github.com/RAMitchell)
+- RF: Add Poisson deviance impurity criterion ([#4156](https://github.com/rapidsai/cuml/pull/4156)) [@venkywonka](https://github.com/venkywonka)
+- avoid paramsSolver::{n_rows,n_cols} shadowing their base class counterparts ([#4130](https://github.com/rapidsai/cuml/pull/4130)) [@yitao-li](https://github.com/yitao-li)
+- Apply modifications to account for RAFT changes ([#4077](https://github.com/rapidsai/cuml/pull/4077)) [@viclafargue](https://github.com/viclafargue)
+
+## 🐛 Bug Fixes
+
+- Update scikit-learn version in conda dev envs to 0.24 ([#4241](https://github.com/rapidsai/cuml/pull/4241)) [@dantegd](https://github.com/dantegd)
+- Using pinned host memory for Random Forest and DBSCAN ([#4215](https://github.com/rapidsai/cuml/pull/4215)) [@divyegala](https://github.com/divyegala)
+- Make sure we keep the rapids-cmake and cuml cal version in sync ([#4213](https://github.com/rapidsai/cuml/pull/4213)) [@robertmaynard](https://github.com/robertmaynard)
+- Add thrust_create_target to install export in CMakeLists ([#4209](https://github.com/rapidsai/cuml/pull/4209)) [@dantegd](https://github.com/dantegd)
+- Change the error type to match sklearn. ([#4198](https://github.com/rapidsai/cuml/pull/4198)) [@achirkin](https://github.com/achirkin)
+- Fixing remaining hdbscan bug ([#4179](https://github.com/rapidsai/cuml/pull/4179)) [@cjnolet](https://github.com/cjnolet)
+- Fix for cuDF changes to cudf.core ([#4168](https://github.com/rapidsai/cuml/pull/4168)) [@dantegd](https://github.com/dantegd)
+- Fixing UMAP reproducibility pytest failures in 11.4 by using random init for now ([#4152](https://github.com/rapidsai/cuml/pull/4152)) [@cjnolet](https://github.com/cjnolet)
+- avoid paramsSolver::{n_rows,n_cols} shadowing their base class counterparts ([#4130](https://github.com/rapidsai/cuml/pull/4130)) [@yitao-li](https://github.com/yitao-li)
+- Use the new RAPIDS.cmake to fetch rapids-cmake ([#4102](https://github.com/rapidsai/cuml/pull/4102)) [@robertmaynard](https://github.com/robertmaynard)
+
+## 📖 Documentation
+
+- Expose train_test_split in API doc ([#4234](https://github.com/rapidsai/cuml/pull/4234)) [@hcho3](https://github.com/hcho3)
+- Adding docs for `.get_feature_names()` inside `TfidfVectorizer` ([#4226](https://github.com/rapidsai/cuml/pull/4226)) [@mayankanand007](https://github.com/mayankanand007)
+- Removing experimental flag from hdbscan description in docs ([#4211](https://github.com/rapidsai/cuml/pull/4211)) [@cjnolet](https://github.com/cjnolet)
+- updated build instructions ([#4200](https://github.com/rapidsai/cuml/pull/4200)) [@shaneding](https://github.com/shaneding)
+- Forward-merge branch-21.08 to branch-21.10 ([#4171](https://github.com/rapidsai/cuml/pull/4171)) [@jakirkham](https://github.com/jakirkham)
+
+## 🚀 New Features
+
+- Experimental option to build libcuml++ only with FIL ([#4225](https://github.com/rapidsai/cuml/pull/4225)) [@dantegd](https://github.com/dantegd)
+- FIL to import categorical models from treelite ([#4173](https://github.com/rapidsai/cuml/pull/4173)) [@levsnv](https://github.com/levsnv)
+- Add hamming, jensen-shannon, kl-divergence, correlation and russellrao distance metrics ([#4155](https://github.com/rapidsai/cuml/pull/4155)) [@mdoijade](https://github.com/mdoijade)
+- Add Categorical Naive Bayes ([#4150](https://github.com/rapidsai/cuml/pull/4150)) [@lowener](https://github.com/lowener)
+- FIL to infer categorical forests and generate them in C++ tests ([#4092](https://github.com/rapidsai/cuml/pull/4092)) [@levsnv](https://github.com/levsnv)
+- Add Gaussian Naive Bayes ([#4079](https://github.com/rapidsai/cuml/pull/4079)) [@lowener](https://github.com/lowener)
+- ARIMA - Add support for missing observations and padding ([#4058](https://github.com/rapidsai/cuml/pull/4058)) [@Nyrio](https://github.com/Nyrio)
+
+## 🛠️ Improvements
+
+- Pin max `dask` and `distributed` versions to 2021.09.1 ([#4229](https://github.com/rapidsai/cuml/pull/4229)) [@galipremsagar](https://github.com/galipremsagar)
+- Fea/umap refine ([#4228](https://github.com/rapidsai/cuml/pull/4228)) [@AjayThorve](https://github.com/AjayThorve)
+- Upgrade Treelite to 2.1.0 ([#4220](https://github.com/rapidsai/cuml/pull/4220)) [@hcho3](https://github.com/hcho3)
+- Add option to clone RAFT even if it is in the environment ([#4217](https://github.com/rapidsai/cuml/pull/4217)) [@dantegd](https://github.com/dantegd)
+- RF: python api behaviour refactor ([#4207](https://github.com/rapidsai/cuml/pull/4207)) [@venkywonka](https://github.com/venkywonka)
+- Pytest updates for Scikit-learn 0.24 ([#4205](https://github.com/rapidsai/cuml/pull/4205)) [@dantegd](https://github.com/dantegd)
+- Faster glm ols-via-eigendecomposition algorithm ([#4201](https://github.com/rapidsai/cuml/pull/4201)) [@achirkin](https://github.com/achirkin)
+- Implement vector leaf for random forest ([#4191](https://github.com/rapidsai/cuml/pull/4191)) [@RAMitchell](https://github.com/RAMitchell)
+- Refactor kmeans sampling code ([#4190](https://github.com/rapidsai/cuml/pull/4190)) [@Nanthini10](https://github.com/Nanthini10)
+- Gracefully accept 'n_jobs', a common sklearn parameter, in NearestNeighbors Estimator ([#4178](https://github.com/rapidsai/cuml/pull/4178)) [@NV-jpt](https://github.com/NV-jpt)
+- Update with rapids cmake new features ([#4175](https://github.com/rapidsai/cuml/pull/4175)) [@robertmaynard](https://github.com/robertmaynard)
+- Update to UCX-Py 0.22 ([#4174](https://github.com/rapidsai/cuml/pull/4174)) [@pentschev](https://github.com/pentschev)
+- Random forest refactoring ([#4166](https://github.com/rapidsai/cuml/pull/4166)) [@RAMitchell](https://github.com/RAMitchell)
+- Fix log level for dask tree_reduce ([#4163](https://github.com/rapidsai/cuml/pull/4163)) [@lowener](https://github.com/lowener)
+- Add CUDA 11.4 development environment ([#4160](https://github.com/rapidsai/cuml/pull/4160)) [@dantegd](https://github.com/dantegd)
+- RF: Add Poisson deviance impurity criterion ([#4156](https://github.com/rapidsai/cuml/pull/4156)) [@venkywonka](https://github.com/venkywonka)
+- Split FIL infer_k into phases to speed up compilation (when a patch is applied) ([#4148](https://github.com/rapidsai/cuml/pull/4148)) [@levsnv](https://github.com/levsnv)
+- RF node queue rewrite ([#4125](https://github.com/rapidsai/cuml/pull/4125)) [@RAMitchell](https://github.com/RAMitchell)
+- Remove max version pin for `dask` & `distributed` on development branch ([#4118](https://github.com/rapidsai/cuml/pull/4118)) [@galipremsagar](https://github.com/galipremsagar)
+- Correct name of a cmake function in get_spdlog.cmake ([#4106](https://github.com/rapidsai/cuml/pull/4106)) [@robertmaynard](https://github.com/robertmaynard)
+- Apply modifications to account for RAFT changes ([#4077](https://github.com/rapidsai/cuml/pull/4077)) [@viclafargue](https://github.com/viclafargue)
+- Warnings are errors ([#4075](https://github.com/rapidsai/cuml/pull/4075)) [@harrism](https://github.com/harrism)
+- ENH Replace gpuci_conda_retry with gpuci_mamba_retry ([#4065](https://github.com/rapidsai/cuml/pull/4065)) [@dillon-cullinan](https://github.com/dillon-cullinan)
+- Changes to NearestNeighbors to call 2d random ball cover ([#4003](https://github.com/rapidsai/cuml/pull/4003)) [@cjnolet](https://github.com/cjnolet)
+- support space in workspace ([#3752](https://github.com/rapidsai/cuml/pull/3752)) [@jolorunyomi](https://github.com/jolorunyomi)
 
 # cuML 21.08.00 (4 Aug 2021)