|
| 1 | +# cuml 24.12.00 (11 Dec 2024) |
| 2 | + |
| 3 | +## 🚨 Breaking Changes |
| 4 | + |
| 5 | +- Forward merge Branch 24.10 into 24.12 ([#6106](https://github.com/rapidsai/cuml/pull/6106)) [@divyegala](https://github.com/divyegala) |
| 6 | + |
| 7 | +## 🐛 Bug Fixes |
| 8 | + |
| 9 | +- Fix `scikit-learn` version specifier ([#6158](https://github.com/rapidsai/cuml/pull/6158)) [@trxcllnt](https://github.com/trxcllnt) |
| 10 | +- Correctly handle missing categorical data in experimental FIL ([#6132](https://github.com/rapidsai/cuml/pull/6132)) [@wphicks](https://github.com/wphicks) |
| 11 | +- Put a ceiling on cuda-python ([#6131](https://github.com/rapidsai/cuml/pull/6131)) [@bdice](https://github.com/bdice) |
| 12 | +- Don't presume pointers are mutually exclusive for device/host. ([#6128](https://github.com/rapidsai/cuml/pull/6128)) [@robertmaynard](https://github.com/robertmaynard) |
| 13 | +- cuml SINGLEGPU now tells cuvs to not build with nccl/mg support ([#6127](https://github.com/rapidsai/cuml/pull/6127)) [@robertmaynard](https://github.com/robertmaynard) |
| 14 | +- Remove type from pickle header for CumlArray ([#6120](https://github.com/rapidsai/cuml/pull/6120)) [@wphicks](https://github.com/wphicks) |
| 15 | +- Forward merge Branch 24.10 into 24.12 ([#6106](https://github.com/rapidsai/cuml/pull/6106)) [@divyegala](https://github.com/divyegala) |
| 16 | +- Fix Dask estimators serialization prior to training ([#6065](https://github.com/rapidsai/cuml/pull/6065)) [@viclafargue](https://github.com/viclafargue) |
| 17 | + |
| 18 | +## 🚀 New Features |
| 19 | + |
| 20 | +- Enable HDBSCAN `gpu` training and `cpu` inference ([#6108](https://github.com/rapidsai/cuml/pull/6108)) [@divyegala](https://github.com/divyegala) |
| 21 | + |
| 22 | +## 🛠️ Improvements |
| 23 | + |
| 24 | +- Update FIL tests to use XGBoost UBJSON instead of binary ([#6153](https://github.com/rapidsai/cuml/pull/6153)) [@hcho3](https://github.com/hcho3) |
| 25 | +- Use sparse knn / distances from cuvs ([#6143](https://github.com/rapidsai/cuml/pull/6143)) [@benfred](https://github.com/benfred) |
| 26 | +- Ensure MG to have the same number of allreduce calls in mean_stddev for sparse matrix to avoid hanging ([#6141](https://github.com/rapidsai/cuml/pull/6141)) [@lijinf2](https://github.com/lijinf2) |
| 27 | +- Stop excluding cutlass from symbol exclusion check ([#6140](https://github.com/rapidsai/cuml/pull/6140)) [@vyasr](https://github.com/vyasr) |
| 28 | +- Optimize MG variance calculation for dataset standardization for logistic regression ([#6138](https://github.com/rapidsai/cuml/pull/6138)) [@lijinf2](https://github.com/lijinf2) |
| 29 | +- enforce wheel size limits, README formatting in CI ([#6136](https://github.com/rapidsai/cuml/pull/6136)) [@jameslamb](https://github.com/jameslamb) |
| 30 | +- Experimental command line interface UX ([#6135](https://github.com/rapidsai/cuml/pull/6135)) [@dantegd](https://github.com/dantegd) |
| 31 | +- add telemetry ([#6126](https://github.com/rapidsai/cuml/pull/6126)) [@msarahan](https://github.com/msarahan) |
| 32 | +- Make cuVS optional if CUML_ALGORITHMS is set ([#6125](https://github.com/rapidsai/cuml/pull/6125)) [@hcho3](https://github.com/hcho3) |
| 33 | +- devcontainer: replace `VAULT_HOST` with `AWS_ROLE_ARN` ([#6118](https://github.com/rapidsai/cuml/pull/6118)) [@jjacobelli](https://github.com/jjacobelli) |
| 34 | +- print sccache stats in builds ([#6111](https://github.com/rapidsai/cuml/pull/6111)) [@jameslamb](https://github.com/jameslamb) |
| 35 | +- fix version in Doxygen docs ([#6104](https://github.com/rapidsai/cuml/pull/6104)) [@jameslamb](https://github.com/jameslamb) |
| 36 | +- make conda installs in CI stricter ([#6103](https://github.com/rapidsai/cuml/pull/6103)) [@jameslamb](https://github.com/jameslamb) |
| 37 | +- Make `get_param_names` a class method on single GPU estimators to match Scikit-learn closer ([#6101](https://github.com/rapidsai/cuml/pull/6101)) [@dantegd](https://github.com/dantegd) |
| 38 | +- Prune workflows based on changed files ([#6094](https://github.com/rapidsai/cuml/pull/6094)) [@KyleFromNVIDIA](https://github.com/KyleFromNVIDIA) |
| 39 | +- Update all rmm imports to use pylibrmm/librmm ([#6084](https://github.com/rapidsai/cuml/pull/6084)) [@Matt711](https://github.com/Matt711) |
| 40 | +- Merge branch 24.10 into branch 24.12 ([#6083](https://github.com/rapidsai/cuml/pull/6083)) [@jameslamb](https://github.com/jameslamb) |
| 41 | + |
1 | 42 | # cuml 24.10.00 (9 Oct 2024)
|
2 | 43 |
|
3 | 44 | ## 🚨 Breaking Changes
|
|
0 commit comments