Commit Graph
6831 Commits
Author SHA1 Message Date
Jiaming Yuan 0edd600f3d [doc] Brief introduction to base_score. (#9882) 2023-12-17 13:34:34 +08:00
david-cortes db7f952ed6 update docs for parameters (#9900) 2023-12-16 12:19:22 +08:00
Dmitry RazdoburdinandDmitry Razdoburdin <> 2a6ab2547d SYCL inference optimization (#9876)
---------

Co-authored-by: Dmitry Razdoburdin <>
2023-12-15 11:04:39 +08:00
Jiaming Yuan 1c6e031c75 [R] Fix clang warning. (#9874) 2023-12-15 01:30:43 +08:00
Jiaming Yuan 125bc812f8 [doc] Reference enable_categorical doc in sklearn. (#9884) 2023-12-14 23:29:19 +08:00
Jiaming Yuan 1aa8c8d9be Support more scipy types. (#9881) 2023-12-14 18:28:37 +08:00
david-cortes cd473c9da3 [R] enable multi-dimensional base_margin (#9885) 2023-12-14 09:16:53 +08:00
Philip Hyunsu Cho 936b22fdf3 [CI] Upload libxgboost4j.dylib (M1) to S3 bucket (#9886)
* [CI] Upload libxgboost4j.dylib (M1) to S3 bucket

* Fix typo
2023-12-13 15:25:51 -08:00
david-cortes 42173d7bc3 [doc] Clarify the effect of enable_categorical (#9877) 2023-12-13 08:39:41 +08:00
github-actions[bot] d530d37707 [CI] Update RAPIDS to latest stable (#9857) 2023-12-12 22:58:03 +09:00
Dmitry RazdoburdinandDmitry Razdoburdin <> 43897b8296 Sycl implementation for objective functions (#9846)
---------

Co-authored-by: Dmitry Razdoburdin <>
2023-12-12 14:41:50 +08:00
david-cortes ddab49a8be [doc][R] Update arguments for ellipsis in predict (#9868) 2023-12-12 12:13:47 +08:00
Jiaming Yuan faf0f2df10 Support dataframe data format in native XGBoost. (#9828)
- Implement a columnar adapter.
- Refactor Python pandas handling code to avoid converting into a single numpy array.
- Add support in R for transforming columns.
- Support R data.frame and factor type.
2023-12-12 09:56:31 +08:00
Jiaming Yuan b3700bbb3f Flexible find protobuf. (#9867) 2023-12-12 07:34:01 +08:00
david-cortes 562352101d [R] Move all DMatrix fields to function arguments (#9862) 2023-12-10 02:45:28 +08:00
Jiaming Yuan 1094d6015d [py] Use the first found native library. (#9860) 2023-12-08 17:23:16 +08:00
Jiaming Yuan 42de9206fc Support multi-target, fit intercept for hinge. (#9850) 2023-12-08 05:50:41 +08:00
Jiaming Yuan 39c637ee19 Use array interface in Python prediction return. (#9855) 2023-12-08 03:42:14 +08:00
david-cortes 2c0fc97306 Remove note about multi-quantile being python-only (#9854) 2023-12-07 05:17:15 +08:00
david-cortes 9e9d41b95c [R] Add note about serialization of DMatrix objects (#9853) 2023-12-07 03:11:15 +08:00
Jiaming Yuan 4bc1f3a388 [R] Bump requirement to 4.3.0. (#9847) 2023-12-07 00:12:45 +08:00
david-cortes 1de3f4135c [R] Enable vector-valued parameters (#9849) 2023-12-06 20:32:20 +08:00
david-cortes 0716c64ef7 [R] Error out on multidimensional arrays (#9852) 2023-12-06 17:43:51 +08:00
david-cortes 62571b79eb [R] Enable multi-output objectives (#9839) 2023-12-06 03:13:14 +08:00
david-cortes 9c56916fd7 [R] Very small performance tweaks (#9837) 2023-12-04 18:40:45 +08:00
381f1d3dc9 Add support inference on SYCL devices (#9800)
---------

Co-authored-by: Dmitry Razdoburdin <>
Co-authored-by: Nikolay Petrov <nikolay.a.petrov@intel.com>
Co-authored-by: Alexandra <alexandra.epanchinzeva@intel.com>
2023-12-04 16:15:57 +08:00
david-cortes 7196c9d95e [R] Fix memory safety issues (#9823) 2023-12-02 13:43:50 +08:00
Jiaming Yuan e78b46046e [CI] Update R version on Linux. (#9835) 2023-12-02 11:03:17 +08:00
Jiaming Yuan 2d8c67d6dc [jvm-packages] Bump dependencies. (#9827)
- #9811
- #9814
- #9826
- #9830
- #9833
- #9832
- #9831
- #9834
2023-12-02 07:34:56 +08:00
david-cortes 95af5c074b more usage of array interface, fix potential memory leaks of std::string (#9824) 2023-12-01 00:06:59 +08:00
david-cortesandJiaming Yuan 37da66f865 [R] Use array interface for dense DMatrix creation (#9816)
---------

Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
2023-11-30 04:35:05 +08:00
Yuan (Terry) Tang da3d55db5b Update affiliation (#9822) 2023-11-30 03:27:05 +08:00
Jiaming Yuan e2e089ce12 [jvm-packages] Bump rapids version. (#9820) 2023-11-29 15:51:07 +08:00
david-cortes c0ef2f8dce [R] Fix potential memory leaks in case of R allocation failures (#9817) 2023-11-29 13:14:17 +08:00
Jiaming Yuan 59684b2db6 [doc] Draft for language binding consistency. [skip ci] (#9755) 2023-11-29 13:13:40 +08:00
david-cortes bfa1252fca [R][doc] Update docs about fitting from CSR (#9818) 2023-11-29 05:42:41 +08:00
Jiaming Yuan 34a2616696 [jvm-packages] Update dependencies. (#9809)
- scalatest: 3.2.17
- maven-checkstyle-plugin: 3.3.1
- maven-surefire-plugin: 3.2.2
- maven-project-info-reports-plugin: 3.5.0
- maven-javadoc-plugin: 3.6.2
2023-11-27 20:09:25 +08:00
Jiaming Yuan e9f149481e [sklearn] Fix loading model attributes. (#9808) 2023-11-27 17:19:01 +08:00
Jiaming Yuan 3f4e22015a Mark NCCL python test optional. (#9804)
Skip the tests if XGBoost is not compiled with dlopen.
2023-11-25 11:25:47 +08:00
Jiaming Yuan 8fe1a2213c Cleanup code for distributed training. (#9805)
* Cleanup code for distributed training.

- Merge `GetNcclResult` into nccl stub.
- Split up utilities from the main dask module.
- Let Channel return `Result` to accommodate nccl channel.
- Remove old `use_label_encoder` parameter.
2023-11-25 09:10:56 +08:00
Jiaming Yuan e9260de3f3 [breaking] Remove dense libsvm parser plugin. (#9799) 2023-11-23 00:12:39 +08:00
Jiaming Yuan 1877cb8e83 Change default metric for gamma regression to deviance. (#9757)
* Change default metric for gamma regression to deviance.

- Cleanup the gamma implementation.
- Use deviance instead since the objective is derived from deviance.
2023-11-22 21:17:48 +08:00
Jiaming Yuan 0715ab3c10 Use dlopen to load NCCL. (#9796)
This PR adds optional support for loading nccl with `dlopen` as an alternative of compile time linking. This is to address the size bloat issue with the PyPI binary release.
- Add CMake option to load `nccl` at runtime.
- Add an NCCL stub.

After this, `nccl` will be fetched from PyPI when using pip to install XGBoost, either by a user or by `pyproject.toml`. Others who want to link the nccl at compile time can continue to do so without any change.

At the moment, this is Linux only since we only support MNMG on Linux.
2023-11-22 19:27:31 +08:00
Jiaming Yuan fedd9674c8 Implement column sampler in CUDA. (#9785)
- CUDA implementation.
- Extract the broadcasting logic, we will need the context parameter after revamping the collective implementation.
- Some changes to the event loop for fixing a deadlock in CI.
- Move argsort into algorithms.cuh, add support for cuda stream.
2023-11-17 04:29:08 +08:00
Bobby Wang 178cfe70a8 [pyspark][doc] Test and doc for stage-level scheduling. (#9786) 2023-11-16 18:15:59 +08:00
Jiaming Yuan ada377c57e [coll] Reduce the scope of lock in the event loop. (#9784) 2023-11-15 14:16:19 +08:00
Bobby Wang 36a552ac98 [jvm-packages] support stage-level scheduling (#9775) 2023-11-14 08:59:45 +08:00
Ken Geis 162da7b52b fix typo in Parameters doc (#9781) 2023-11-13 03:09:06 +08:00
Jiaming Yuan 6fd4a30667 [coll] Increase timeout for allgather test. (#9777) 2023-11-09 05:26:40 +08:00
Jiaming Yuan 44099f585d [coll] Add C API for the tracker. (#9773) 2023-11-08 18:17:14 +08:00