Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6,420 Commits 32 Branches 59 Tags
141a062e00d977e5d64c8be5574395df36658e6e
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Philip Hyunsu Cho 6676c28cbc [CI] Fix Windows wheel to be compatible with Poetry (#8991) 
* [CI] Fix Windows wheel to be compatible with Poetry

* Typo

* Eagerly scan globs to avoid patching same file twice
 2023-03-28 21:32:54 -07:00
Philip Hyunsu Cho 8dc1e4b3ea Improve doxygen (#8959) 
* Remove Sphinx build from GH Action

* Build Doxygen as part of RTD build

* Add jQuery
 2023-03-21 09:22:11 -07:00
Philip Hyunsu Cho 36263dd109 [jvm-packages] Use akka 2.6 (#8920) 2023-03-16 20:06:42 -07:00
Philip Hyunsu Cho a2cdba51ce Use hi-res SVG logo (#8923) 2023-03-15 10:02:38 -07:00
Philip Hyunsu Cho 6d8afb2218 [CI] Require C++17 + CMake 3.18; Use CUDA 11.8 in CI (#8853) 
* Update to C++17

* Turn off unity build

* Update CMake to 3.18

* Use MSVC 2022 + CUDA 11.8

* Re-create stack for worker images

* Allocate more disk space for Windows

* Tempiorarily disable clang-tidy

* RAPIDS now requires Python 3.10+

* Unpin cuda-python

* Use latest NCCL

* Use Ubuntu 20.04 in RMM image

* Mark failing mgpu test as xfail
 2023-03-01 09:22:24 -08:00
Philip Hyunsu Cho dd79ab846f [CI] Fix failing arm build (#8751) 
* Always install Conda env into /opt/python; use Mamba

* Change ownership of Conda env to buildkite-agent user

* Use unique name

* Fix
 2023-02-03 22:32:48 -08:00
Philip Hyunsu Cho d29e45371f [R-package] Alter xgb.train() to accept multiple eval metrics as a list (#8657) 2023-01-24 17:14:14 -08:00
Philip Hyunsu Cho 5bf9e79413 [CI] Disable gtest with RMM (#8620) 2022-12-21 01:41:34 -08:00
Philip Hyunsu Cho 35d8447282 [CI] Use conda-forge channel in conda (#8583) 2022-12-11 23:25:29 -08:00
Philip Hyunsu Cho 42c5ee5588 [jvm-packages] Bump version of akka packages (#8524) 2022-12-05 22:45:00 +08:00
Philip Hyunsu Cho 2546d139d6 [jvm-packages] Add missing commons-lang3 dependency to xgboost4j-gpu (#8508) 
* [jvm-packages] Add missing commons-lang3 dependency to xgboost4j-gpu

* Update commons-lang3
 2022-12-01 16:27:11 -08:00
Philip Hyunsu Cho 7c6f2346d3 [jvm-packages] Configure dependabot properly (#8507) 
* [jvm-packages] Configure dependabot properly

* Allow automatic updates for Scala and Spark within the same major version
 2022-12-01 16:26:47 -08:00
Philip Hyunsu Cho f550109641 Bump some old dependencies of JVM packages (#8456) 2022-11-30 23:04:08 -08:00
Philip Hyunsu Cho 9a98e79649 [jvm-packages] Set up dependabot (#8501) 2022-11-30 22:46:17 -08:00
Philip Hyunsu Cho 0db903b471 Fix formatting in NEWS.md [skip ci] 2022-10-31 15:42:31 -07:00
Philip Hyunsu Cho 8bb55949ef Fix building XGBoost with libomp 15 (#8384) 2022-10-25 12:01:11 -07:00
Philip Hyunsu Cho 29595102b9 [CI] Set up test analytics for CPU Python tests (#8333) 
* [CI] Set up test analytics for CPU Python tests

* Install test collector
 2022-10-12 23:15:50 -07:00
Philip Hyunsu Cho 2faa744aba [CI] Test federated learning plugin in the CI (#8325) 2022-10-12 13:57:39 -07:00
Philip Hyunsu Cho a71421e825 [CI] Update GitHub Actions to use macos-11 (#8321) 2022-10-08 00:40:43 -07:00
Philip Hyunsu Cho d70e59fefc Fix Intel's link [skip ci] 2022-10-06 16:55:42 -07:00
Philip Hyunsu Cho 50ff8a2623 More CI improvements (#8313) 
* Reduce clutter in log of Python test

* Set up BuildKite test analytics

* Add separate step for building containers

* Enable incremental update of CI stack; custom agent IAM policy
 2022-10-06 06:33:46 -08:00
Philip Hyunsu Cho bc7a6ec603 Fix clang tidy (#8314) 
* Fix clang-tidy

* Exempt clang-tidy from budget check

* Move clang-tidy
 2022-10-06 05:16:06 -08:00
Philip Hyunsu Cho 66fd9f5207 Update sponsors list [skip ci] (#8309) 2022-10-05 16:40:46 -08:00
Philip Hyunsu Cho b2bbf49015 Additional improvements to CI (#8303) 
* Wait until budget check is complete

* Ensure that multi-GPU tests run for the master branch

* Fix
 2022-10-04 03:03:38 -08:00
Philip Hyunsu Cho ca0547bb65 [CI] Use RAPIDS 22.10 (#8298) 
* [CI] Use RAPIDS 22.10

* Store CUDA and RAPIDS versions in one place

* Fix

* Add missing #include

* Update gputreeshap submodule

* Fix

* Remove outdated distributed tests
 2022-10-03 23:18:07 -08:00
Philip Hyunsu Cho 37886a5dff [CI] Document the use of Docker wrapper script (#8297) 
* [CI] Document the use of Docker wrapper script

* Grammer fixes

* Document buildkite pipeline defs

* tests/buildkite/*.sh isn't meant to run locally
 2022-10-02 12:45:00 -07:00
Philip Hyunsu Cho 9af99760d4 Various CI savings (#8291) 2022-09-30 05:42:56 -07:00
Philip Hyunsu Cho b14c44ee5e [CI] Put Multi-GPU test suites in separate pipeline (#8286) 
* [CI] Put Multi-GPU test suites in separate pipeline

* Avoid unset var error in Bash
 2022-09-29 00:41:48 -08:00
Philip Hyunsu Cho 70df36c99c [CI] Retire Jenkins server (#8243) 2022-09-14 08:46:23 -07:00
Philip Hyunsu Cho 23faf656ad [CI] Don't require manual approval for master branch (#8235) 2022-09-08 09:26:22 -08:00
Philip Hyunsu Cho e888eb2fa9 [CI] Migrate CI pipelines from Jenkins to BuildKite (#8142) 
* [CI] Migrate CI pipelines from Jenkins to BuildKite

* Require manual approval

* Less verbose output when pulling Docker

* Remove us-east-2 from metadata.py

* Add documentation

* Add missing underscore

* Add missing punctuation

* More specific instruction

* Better paragraph structure
 2022-09-07 16:29:25 -08:00
Philip Hyunsu Cho b397d64c96 Drop use of deleted virtual function to support older MacOS (#8226) 
* Support older MacOS

* Update json.h
 2022-09-07 11:25:59 -08:00
Philip Hyunsu Cho 56395d120b Work around MSVC behavior wrt constexpr capture (#8211) 
* Work around MSVC behavior wrt constexpr capture

* Fix lint
 2022-08-31 11:42:08 -08:00
Philip Hyunsu Cho 35ef8abc27 [CI] Prune unused archs from libnccl (#8179) 
* [CI] Prune unused archs from libnccl

* Put pruning logic in CI directory

* Don't use --color in grep
 2022-08-21 00:46:16 -08:00
Philip Hyunsu Cho 24c2373080 [Doc] Indicate lack of py-xgboost-gpu on Windows (#8127) 2022-07-28 12:57:16 -07:00
Philip Hyunsu Cho 4325178822 [CI] Clear workspace after budget check (#8092) 
* [CI] Clear workspace after budget check

* Windows too
 2022-07-18 19:17:33 -07:00
Philip Hyunsu Cho 36cf979b82 [CI] Fix S3 uploads (#8069) 
* [CI] Fix S3 upload issues

* Don't launch Docker containers when uploading to S3
 2022-07-13 16:23:00 -07:00
Philip Hyunsu Cho 1ced638165 Document how to reproduce Docker environment from Jenkins (#7971) 2022-06-04 20:56:53 +09:00
Philip Hyunsu Cho 47224dd6d3 Use private mirror to host llvm-openmp tarballs (#7950) 2022-05-27 14:56:59 -07:00
Philip Hyunsu Cho 2070afea02 [CI] Rotate package repository keys (#7943) 2022-05-26 17:06:46 -07:00
Philip Hyunsu Cho 6f424d8d6c [Doc] Warn against loading JSON from external source (#7918) 2022-05-18 17:02:36 -07:00
Philip Hyunsu Cho 4cd14aee5a Rename misspelled config parameter for pseudo-Huber (#7904) 2022-05-15 06:38:33 -07:00
Philip Hyunsu Cho 65e6d73b95 [CI] Automate artifact fetch step in JVM release process (#7882) 2022-05-11 00:35:22 -07:00
Philip Hyunsu Cho d2bc0f0f08 Allow loading old models from RDS (#7864) 2022-05-06 22:49:38 -07:00
Philip Hyunsu Cho e5ab8f3ebe [CI] Speed up CPU test pipeline (#7772) 2022-04-01 02:39:04 +08:00
Philip Hyunsu Cho e8eff3581b [CI] Enable faulthandler to show details when 0xC0000005 error occurs (#7771) (#7775) 2022-03-31 17:40:06 +08:00
Philip Hyunsu Cho 66cb4afc6c Update install doc (#7747) 2022-03-23 17:20:01 +08:00
Philip Hyunsu Cho 1b25dd59f9 Use CUDA 11 in clang-tidy (#7701) 
* Show command args when clang-tidy fails

* Add option to specify CUDA args

* Use clang-tidy 11

* [CI] Use CUDA 11
 2022-02-24 15:15:07 -08:00
Philip Hyunsu Cho 0149f81a5a [CI] Fix S3 upload (#7662) 2022-02-16 01:35:27 -08:00
Philip Hyunsu Cho 34a238ca98 [CI] Clean up Python wheel build pipeline (#7626) 
* [CI] Always upload artifacts to [branch_name]/

* [CI] Move detailed setup inside build_python_wheels.sh

* Fix typo
 2022-02-03 00:55:44 -08:00
Philip Hyunsu Cho f6e6d0b2c0 [CI] Build Python wheels for MacOS (x86_64 and arm64) (#7621) 
* Build Python wheels for OSX (x86_64 and arm64)

* Use Conda's libomp when running Python tests

* fix

* Add comment to explain CIBW_TARGET_OSX_ARM64

* Update release script

* Add comments in build_python_wheels.sh

* Document wheel pipeline
 2022-02-02 17:35:48 -08:00
Philip Hyunsu Cho 271a7c5d43 [Doc] fix typo in install doc (#7623) 2022-01-31 13:35:56 -08:00
Philip Hyunsu Choandfis c621775f34 Replace all uses of deprecated function sklearn.datasets.load_boston (#7373) 
* Replace all uses of deprecated function sklearn.datasets.load_boston

* More renaming

* Fix bad name

* Update assertion

* Fix n boosted rounds.

* Avoid over regularization.

* Rebase.

* Avoid over regularization.

* Whac-a-mole

Co-authored-by: fis <jm.yuan@outlook.com>
 2022-01-30 04:27:57 -08:00
Philip Hyunsu Cho b4340abf56 Add special handling for multi:softmax in sklearn predict (#7607) 
* Add special handling for multi:softmax in sklearn predict

* Add test coverage
 2022-01-29 15:54:49 -08:00
Philip Hyunsu Cho f21301c749 [Doc] Add instruction to install XGBoost for Apple Silicon using Conda (#7612) 2022-01-28 01:06:39 -08:00
Philip Hyunsu Cho 028bdc1740 [R] Fix typo in docstring (#7606) 2022-01-26 23:33:25 +08:00
Philip Hyunsu Cho 9fd510faa5 [CI] Clarify steps for publishing artifacts to Maven Central (#7582) 2022-01-20 14:23:07 -08:00
Philip Hyunsu Cho 20c0d60ac7 Restore functionality of max_depth=0 in hist (#7551) 
* Restore functionality of max_depth=0 in hist

* Add test case
 2022-01-11 01:37:44 +08:00
Philip Hyunsu Cho 619c450a49 [CI] Add missing step extract_branch (#7479) 2021-11-24 17:35:59 +08:00
Philip Hyunsu Cho 96a9848c9e [CI] Fix continuous delivery pipeline for MacOS (#7472) 2021-11-23 22:22:08 +08:00
Philip Hyunsu Cho 0c67685e43 [CI] Add a helper script to aid Maven release (#7470) 
* [CI] Add a helper script to aid Maven release

* Move script to dev/ [skip ci]

* Update command [skip ci]
 2021-11-23 00:11:07 -08:00
Philip Hyunsu Cho 2adf222fb2 [CI] CI cost saving (#7407) 
* [CI] Drop CUDA 10.1; Require 11.0

* Change NCCL version

* Use CUDA 10.1 for clang-tidy, for now

* Remove JDK 11 and 12

* Fix NCCL version

* Don't require 11.0 just yet, until clang-tidy is fixed

* Skip MultiClassesSerializationTest.GpuHist
 2021-11-17 21:02:20 -08:00
Philip Hyunsu Cho b8e8f0fcd9 [doc] Use latest Sphinx RTD theme (#7347) 2021-10-20 00:04:43 -07:00
Philip Hyunsu Cho 2a0368b7ca Add CMake option to use /MD runtime (#7277) 2021-09-30 13:13:57 +08:00
Philip Hyunsu Cho d27a427dc5 [CI] Rotate access keys for uploading MacOS artifacts from Travis CI (#7253) 2021-09-24 10:44:00 +08:00
Philip Hyunsu Cho b70e07da1f [CI] Clean up in beginning of each task in Win CI (#7189) 2021-08-25 04:15:22 -07:00
Philip Hyunsu Cho 3060f0b562 [CI] Automatically build GPU-enabled R package for Windows (#7185) 
* [CI] Automatically build GPU-enabled R package for Windows

* Update Jenkinsfile-win64

* Build R package for the release branch only

* Update install doc
 2021-08-25 02:11:01 -07:00
Philip Hyunsu Cho d04312b9c0 [CI] Fix hanging Python setup in Windows CI (#7186) 2021-08-24 22:03:51 -07:00
Philip Hyunsu Cho 336af4f974 Work around a segfault observed in SparsePage::Push() (#7161) 
* Work around a segfault observed in SparsePage::Push()

* Revert "Work around a segfault observed in SparsePage::Push()"

This reverts commit 30934844d00908750a5442082eb4769b1489f6a9.

* Don't call vector::resize() inside OpenMP block

* Set GITHUB_PAT env var to fix R tests

* Use built-in GITHUB_TOKEN
 2021-08-08 02:12:30 -07:00
Philip Hyunsu Cho f1a4a1ac95 [CI] Upgrade build image to CentOS 7 + GCC 8; require CUDA 10.1 and later (#7141) 2021-07-29 10:54:33 -07:00
Philip Hyunsu Cho 2801d69fb7 [CI] Pin libomp to 11.1.0 (#7107) 2021-07-15 11:16:51 +08:00
Philip Hyunsu Cho dd4db347f3 Fix early stopping behavior with MAPE metric (#7061) 2021-06-26 03:02:33 +08:00
Philip Hyunsu Cho b2d300e727 [CI] Upgrade to CMake 3.14 (#7060) 
* [CI] Upgrade to CMake 3.14

* Add FATAL_ERROR directive, for users with CMake 2.x
 2021-06-24 18:07:24 -07:00
Philip Hyunsu Cho 655e6992f6 [Dask] Add example of using custom callback in Dask (#6995) 2021-06-03 07:05:55 +08:00
Philip Hyunsu Cho 05db6a6c29 [CI] Upgrade cuDF and RMM to 21.06 nightly (#7012) 
* [CI] Upgrade cuDF and RMM to 21.06 nightly

* Trim outdated test cases

* Pin Dask version to 2021.05.0 for now
 2021-06-02 11:59:30 -07:00
Philip Hyunsu Cho c6d87e5e18 [CI] Remove stray build artifact to avoid error in artifact packaging (#6994) 2021-05-25 19:48:27 +08:00
Philip Hyunsu Cho 90cd724be1 [CI] Fix CI/CD pipeline broken by latest auditwheel (4.0.0) (#6951) 2021-05-10 22:43:15 -07:00
Philip Hyunsu Cho ec6ce08cd0 [jvm-packages] Make it easier to release GPU/CPU code artifacts to Maven Central (#6940) 2021-05-04 14:00:03 -07:00
Philip Hyunsu Cho 5472ef626c [R] Re-generate Roxygen2 doc (#6915) 2021-04-29 11:55:07 -07:00
Philip Hyunsu Cho 4224c08cac Add demo for using AFT survival with Dask (#6853) 2021-04-13 16:18:33 -07:00
Philip Hyunsu Cho 878b990fcd [CI] Upload Doxygen to correct destination (#6854) 2021-04-13 16:18:13 -07:00
Philip Hyunsu Cho ea7a6a0321 [CI] Pack R package tarball with pre-built xgboost.so (with GPU support) (#6827) 
* Add scripts for packaging R package with GPU-enabled libxgboost.so

* [CI] Automatically build R package tarball

* Add comments

* Don't build tarball for pull requests

* Update the installation doc
 2021-04-07 21:15:34 -07:00
Philip Hyunsu Cho 744c46995c [CI] Upload xgboost4j.dll to S3 (#6781) 2021-03-25 11:34:34 -07:00
Philip Hyunsu Cho 4230dcb614 Re-introduce double buffer in UpdatePosition, to fix perf regression in gpu_hist (#6757) 
* Revert "gpu_hist performance tweaks (#5707)"

This reverts commit f779980f7e.

* Address reviewer's comment

* Fix build error
 2021-03-18 13:56:10 -07:00
Philip Hyunsu Cho 366f3cb9d8 Add use_rmm flag to global configuration (#6656) 
* Ensure RMM is 0.18 or later

* Add use_rmm flag to global configuration

* Modify XGBCachingDeviceAllocatorImpl to skip CUB when use_rmm=True

* Update the demo

* [CI] Pin NumPy to 1.19.4, since NumPy 1.19.5 doesn't work with latest Shap
 2021-03-09 14:53:05 -08:00
Philip Hyunsu Cho e4894111ba Update dmlc-core submodule (#6745) 2021-03-07 00:30:26 -08:00
Philip Hyunsu Cho 5ae7f9944b [CI] Clear R package cache (#6746) 2021-03-06 08:37:16 -08:00
Philip Hyunsu Cho 0f2ed21a9d [Breaking] Change default evaluation metric for binary:logitraw objective to logloss (#6647) 2021-01-29 00:12:12 +09:00
Philip Hyunsu Cho 0ad6e18a2a [CI] Do not mix up stashed executable built for ARM and x86_64 platforms (#6646) 2021-01-27 23:57:26 +09:00
Philip Hyunsu Cho 55ee2bd77f [CI] Add ARM64 test to Jenkins pipeline (#6643) 
* Add ARM64 test to Jenkins pipeline

* Check for bundled libgomp

* Use a separate test suite for ARM64

* Ensure that x86 jobs don't run on ARM workers
 2021-01-27 21:51:17 +09:00
Philip Hyunsu Cho fa13992264 Calling XGBModel.fit() should clear the Booster by default (#6562) 
* Calling XGBModel.fit() should clear the Booster by default

* Document the behavior of fit()

* Allow sklearn object to be passed in directly via xgb_model argument

* Fix lint
 2020-12-31 11:02:08 -08:00
Philip Hyunsu Cho fbb980d9d3 Expand ~ into the home directory on Linux and MacOS (#6531) 2020-12-19 23:35:13 -08:00
Philip Hyunsu Cho cd0821500c Add Saturn Cloud Dask XGBoost tutorial to Awesome XGBoost [skip ci] (#6532) 2020-12-19 15:57:05 -08:00
Philip Hyunsu Cho 380f6f4ab8 Remove cupy.array_equal, since it's not compatible with cuPy 7.8 (#6528) 2020-12-18 09:16:52 -08:00
Philip Hyunsu ChoandPSEUDOTENSOR / Jonathan McKinney 125b3c0f2d Lazy import cuDF and Dask (#6522) 
* Lazy import cuDF

* Lazy import Dask

Co-authored-by: PSEUDOTENSOR / Jonathan McKinney <pseudotensor@gmail.com>

* Fix lint

Co-authored-by: PSEUDOTENSOR / Jonathan McKinney <pseudotensor@gmail.com>
 2020-12-17 01:51:35 -08:00
Philip Hyunsu ChoandJiaming Yuan ad1a527709 Enable loading model from <1.0.0 trained with objective='binary:logitraw' (#6517) 
* Enable loading model from <1.0.0 trained with objective='binary:logitraw'

* Add binary:logitraw in model compatibility testing suite

* Feedback from @trivialfis: Override ProbToMargin() for LogisticRaw

Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2020-12-16 16:53:46 -08:00
Philip Hyunsu ChoandMark Harris bf6cfe3b99 [Breaking] Upgrade cuDF and RMM to 0.18 nightlies; require RMM 0.18+ for RMM plugin (#6510) 
* [CI] Upgrade cuDF and RMM to 0.18 nightlies

* Modify RMM plugin to be compatible with RMM 0.18

* Update src/common/device_helpers.cuh

Co-authored-by: Mark Harris <mharris@nvidia.com>

Co-authored-by: Mark Harris <mharris@nvidia.com>
 2020-12-16 10:07:52 -08:00
Philip Hyunsu Cho 9a194273cd Add conda-forge badge (#6502) 2020-12-14 18:58:03 -08:00
Philip Hyunsu ChoandJohn Zedlewski aac4eba2ef Add release note for 1.3.0 in NEWS.md (#6495) 
* Add release note for 1.3.0

* Address reviewer's comment

* Fix silly mistake

* Apply suggestions from code review

Co-authored-by: John Zedlewski <904524+JohnZed@users.noreply.github.com>

Co-authored-by: John Zedlewski <904524+JohnZed@users.noreply.github.com>
 2020-12-14 14:42:30 -08:00
Philip Hyunsu Cho c31e3efa7c Pass correct split_type to GPU predictor (#6491) 
* Pass correct split_type to GPU predictor

* Add a test
 2020-12-11 19:30:00 -08:00