[NIGHTLY] v22.12.00

🔗 Links

• Development Branch
• Compare with main branch

🚨 Breaking Changes

• Change docs theme to pydata-sphinx theme (#4985) @galipremsagar
• Remove "Open In Colab" link from Estimator Intro notebook. (#4980) @bdice
• Remove CumlArray.copy() (#4958) @madsbk

🐛 Bug Fixes

• Backport "Don't initialize CUDA context if RAPIDS_NO_INITIALIZE env variable is set" (#5069) @dantegd
• Remove cupy.cusparse custom serialization (#5024) @dantegd
• Restore LinearRegression documentation (#5020) @viclafargue
• Don't use CMake 3.25.0 as it has a FindCUDAToolkit show stopping bug (#5007) @robertmaynard
• verifying cusparse wrapper revert passes CI (#4990) @cjnolet
• Use rapdsi_cpm_find(COMPONENTS ) for proper component tracking (#4989) @robertmaynard
• Fix integer overflow in AutoARIMA due to bool-to-int cub scan (#4971) @Nyrio
• Add missing includes (#4947) @vyasr
• Fix the CMake option for disabling deprecation warnings. (#4946) @vyasr
• Make doctest resilient to changes in cupy reprs (#4945) @vyasr
• Assign python/ sub-directory to python-codeowners (#4940) @csadorf
• Fix for non-contiguous strides (#4736) @viclafargue

📖 Documentation

• Change docs theme to pydata-sphinx theme (#4985) @galipremsagar
• Remove "Open In Colab" link from Estimator Intro notebook. (#4980) @bdice
• Updating build instructions (#4979) @cjnolet

🚀 New Features

• Reenable copy_prs. (#5010) @vyasr
• Add wheel builds (#5009) @vyasr
• LinearRegression: add support for multiple targets (#4988) @ahendriksen
• CPU/GPU interoperability POC (#4874) @viclafargue

🛠️ Improvements

• Upgrade Treelite to 3.0.1 (#5018) @hcho3
• fix addition of nan_euclidean_distances to public api (#5015) @mattf
• Fixing raft pin to 22.12 (#5000) @cjnolet
• Pin dask and distributed for release (#4999) @galipremsagar
• Update dask nightly install command in CI (#4978) @galipremsagar
• Improve error message for array_equal asserts. (#4973) @csadorf
• Use new rapids-cmake functionality for rpath handling. (#4966) @vyasr
• Impl. CumlArray.deserialize() (#4965) @madsbk
• Update cuda-python dependency to 11.7.1 (#4961) @galipremsagar
• Add check for nsys utility version in the nvtx_benchmarks.py script (#4959) @viclafargue
• Remove CumlArray.copy() (#4958) @madsbk
• Implement hypothesis-based tests for linear models (#4952) @csadorf
• Switch to using rapids-cmake for gbench. (#4950) @vyasr
• Remove stale labeler (#4949) @raydouglass
• Fix url in python/setup.py setuptools metadata. (#4937) @csadorf
• Updates to fix cuml build (#4928) @cjnolet
• Documenting hdbscan module to add prediction functions (#4925) @cjnolet
• Unpin dask and distributed for development (#4912) @galipremsagar
• Use KMeans from Raft (#4713) @lowener
• Update cuml raft header extensions (#4599) @cjnolet
• Reconciling primitives moved to RAFT (#4583) @cjnolet