Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
4,034 Commits 32 Branches 59 Tags
adc795929a52d90ecced21a5279fe175a2a129b3
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Philip Hyunsu Cho adc795929a Fix compilation error due to 64-bit integer narrowing to size_t (#5250) 2020-01-30 21:32:15 -08:00
Philip Hyunsu Cho 4240daed4e Make pip install xgboost*.tar.gz work by fixing build-python.sh (#5241) 
* Make pip install xgboost*.tar.gz work by fixing build-python.sh

* Simplify install doc

* Add test

* Install Miniconda for Linux target too

* Build XGBoost only once in sdist

* Try importing xgboost after installation

* Don't set PYTHONPATH env var for sdist test
 2020-01-28 23:18:23 -08:00
Philip Hyunsu Cho cb3ed404cf [R] Enable OpenMP with AppleClang in XGBoost R package (#5240) 
* [R] Enable OpenMP with AppleClang in XGBoost R package

* Dramatically simplify install doc
 2020-01-28 12:37:22 -08:00
Philip Hyunsu Cho f7105fa44f Update Rabit (#5237) 2020-01-28 02:05:01 -08:00
Philip Hyunsu Cho b513dcd352 Support FreeBSD (#5233) 
* Fix build on FreeBSD

* Use __linux__ macro
 2020-01-27 22:19:16 -08:00
Philip Hyunsu Cho 0c7455276d [R] Robust endian detection in CRAN xgboost build (#5232) 
* [R] Robust endian detection in CRAN xgboost build

* Check for external backtrace() lib

* Update Makevars.win
 2020-01-27 02:55:13 -08:00
Philip Hyunsu ChoandJiaming Yuan 44469a0ca9 Extensible binary serialization format for DMatrix::MetaInfo (#5187) 
* Turn xgboost::DataType into C++11 enum class

* New binary serialization format for DMatrix::MetaInfo

* Fix clang-tidy

* Fix c++ test

* Implement new format proposal

* Move helper functions to anonymous namespace; remove unneeded field

* Fix lint

* Add shape.

* Keep only roundtrip test.

* Fix test.

* various fixes

* Update data.cc

Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2020-01-23 11:33:17 -08:00
Philip Hyunsu Cho 2a071cebc5 Add CMake option to run Undefined Behavior Sanitizer (UBSan) (#5211) 
* Fix related errors.
 2020-01-20 16:57:44 +08:00
Philip Hyunsu Cho 0184f2e9f7 Explicitly use UTF-8 codepage when using MSVC (#5197) 
* Explicitly use UTF-8 codepage when using MSVC

* Fix build with CUDA enabled
 2020-01-14 13:30:34 -08:00
Philip Hyunsu Cho 9b0af6e882 Enable OpenMP with Apple Clang (Mac default compiler) (#5146) 
* Add OpenMP as CMake target

* Require CMake 3.12, to allow linking OpenMP target to objxgboost

* Specify OpenMP compiler flag for CUDA host compiler

* Require CMake 3.16+ if the OS is Mac OSX

* Use AppleClang in Mac tests.

* Update dmlc-core
 2019-12-26 16:53:12 +08:00
Philip Hyunsu Cho 74f545bde3 [CI] Repair download URL for Maven 3.6.1 (#5139) 2019-12-20 10:07:40 +08:00
Philip Hyunsu Cho 37fdfa03f8 [jvm-packages] Comply with scala style convention + fix broken unit test (#5134) 
* Fix scala style check

* fix messed unit test
 2019-12-18 17:26:58 -08:00
Philip Hyunsu Cho 64f4361b47 [CI] Locate vcomp140.dll from System32 directory (#5078) 2019-12-02 02:09:32 -08:00
Philip Hyunsu Cho 4d2779663e Require Python 3.5+ in setup.py (#5021) 2019-11-18 18:55:58 -08:00
Philip Hyunsu Cho f4e7b707c9 Revert #4529 (#5008) 
* Revert " Optimize ‘hist’ for multi-core CPU (#4529)"

This reverts commit 4d6590be3c.

* Fix build
 2019-11-12 09:35:03 -08:00
Philip Hyunsu Cho a37691428f Document minimum version required for gtest [skip ci] (#5001) 2019-10-31 15:47:50 -07:00
Philip Hyunsu Cho 96cd7ec2bb [CI] Upload master branch artifacts to S3 root [skip ci] (#4979) 2019-10-23 22:39:04 -07:00
Philip Hyunsu Cho da6e74f7bb [CI] Upload nightly builds to S3 (#4976) 
* Do not store built artifacts in the Jenkins master

* Add wheel renaming script

* Upload wheels to S3 bucket

* Use env.GIT_COMMIT

* Capture git hash correctly

* Add missing import in Jenkinsfile

* Address reviewer's comments

* Put artifacts for pull requests in separate directory

* No wildcard expansion in Windows CMD
 2019-10-23 21:16:05 -07:00
Philip Hyunsu Cho 95295ce026 [CI] Use latest dask (#4973) 
* Remove version spec, to use latest dask always
 2019-10-22 07:00:13 -04:00
Philip Hyunsu Cho 741fbf47c4 [CI] Update lint configuration to support latest pylint convention (#4971) 
* Update lint configuration

* Use gcc 8 consistently in build instruction
 2019-10-21 16:40:57 -07:00
Philip Hyunsu Cho f7487e4c2a [CI] Run cuDF tests in Jenkins CI server (#4927) 2019-10-13 00:04:54 -04:00
Philip Hyunsu Cho 5b4f28cc46 [CI] Raise timeout threshold in Jenkins (#4938) 2019-10-12 23:47:35 -04:00
Philip Hyunsu Cho 0e0955a6d8 Add link to Ruby XGBoost gem (#4856) 2019-09-17 10:40:44 -07:00
Philip Hyunsu Cho 166def9f75 [CI] Fix broken installation of Pandas (#4722) 
* [CI] Fix broken installation of Pandas

* Update Dockerfile.gpu
 2019-07-30 22:03:11 -07:00
Philip Hyunsu Cho 2758c5acea [CI] Fix broken installation of Pandas (#4704) 2019-07-24 19:03:35 -07:00
Philip Hyunsu Cho d4e0a30582 Upgrade dmlc-core submodule (#4688) 2019-07-20 11:30:05 -07:00
Philip Hyunsu Cho 4ef6d216b9 Upgrade dmlc-core submodule (#4674) 2019-07-17 18:54:49 -07:00
Philip Hyunsu Cho 9975c533c7 Re-organize contributor's guide (#4659) 
* Reorganize contributor's doc

* Address comments from @trivialfis

* Address @sriramch's comment: include ABI compatibility guarantee

* Address @rongou's comment

* Postpone ABI compatibility guarantee for now
 2019-07-15 20:56:05 -07:00
Philip Hyunsu Cho d333918f5e [jvm-packages] Expose setMissing method in XGBoostClassificationModel / XGBoostRegressionModel (#4643) 2019-07-07 16:02:44 -07:00
Philip Hyunsu Cho 1aaf4a679d Fix early stopping in the Python package (#4638) 
* Fix #4630, #4421: Preserve correct ordering between metrics, and always use last metric for early stopping

* Clarify semantics of early stopping in presence of multiple valid sets and metrics

* Add a test

* Fix lint
 2019-07-07 01:01:03 -07:00
Philip Hyunsu Cho b7a1f22d24 Empty evaluation list in early stopping should produce meaningful error message (#4633) 
* Empty evaluation list should not break early stopping

* Fix lint

* Update callback.py
 2019-07-04 13:27:18 -07:00
Philip Hyunsu Cho 4df246191f Add warning when save_model() is called from scikit-learn interface (#4632) 2019-07-03 23:37:53 -07:00
Philip Hyunsu Cho 96bf91725b Support ndcg- and map- (#4635) 2019-07-03 22:51:48 -07:00
Philip Hyunsu Cho 4e9fad74eb [R] Use built-in label when xgb.DMatrix is given to xgb.cv() (#4631) 
* Use built-in label when xgb.DMatrix is given to xgb.cv()

* Add a test

* Fix test

* Bump version number
 2019-07-03 01:32:40 -07:00
Philip Hyunsu Cho a30176907f Support Dask 2.0 (#4617) 2019-06-27 20:42:35 -07:00
Philip Hyunsu Cho dd01f7c4f5 Use Sphinx 2.1+ to compile documentation [skip ci] (#4609) 2019-06-26 16:26:22 -07:00
Philip Hyunsu Cho cd3a3f99da Fix doc for customized objective/metric [skip ci] (#4608) 2019-06-26 13:40:34 -07:00
Philip Hyunsu Cho 1f98f18cb8 Add instruction to run formatting checks locally [skip ci] (#4591) 2019-06-24 00:09:09 -07:00
Philip Hyunsu Cho 30e1cb4e9e Fix docstring for XGBModel.predict() [skip ci] (#4592) 2019-06-21 12:44:42 +08:00
Philip Hyunsu Cho 0c50f8417a [CI] Specify account ID when logging into ECR Docker registry (#4584) 
* [CI] Specify account ID when logging into ECR Docker registry

* Do not display awscli login command
 2019-06-19 15:08:42 -07:00
Philip Hyunsu Cho 570374effe [CI] Remove CUDA 8.0 from CI pipeline (#4580) 2019-06-18 23:38:03 -07:00
Philip Hyunsu ChoandJiaming Yuan 3f2fe25a32 Fix C++11 config parser (#4521) 
* Fix C++11 config parser
* Use raw strings to improve readability of regex
* Fix compilation for GCC 5.x

Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2019-06-03 22:18:16 +08:00
Philip Hyunsu Cho c2a3902ba3 Fix #4497: Enable feature importance property for DART booster (#4525) 2019-05-31 15:11:57 -07:00
Philip Hyunsu Cho ea44417754 Enforce exclusion between pred_interactions=True and pred_interactions=True (#4522) 2019-05-31 12:29:23 -07:00
Philip Hyunsu Cho 515f5f5c47 [RFC] Version 0.90 release candidate (#4475) 
* Release 0.90

* Add script to automatically generate acknowledgment

* Update NEWS.md
 2019-05-20 01:02:44 -07:00
Philip Hyunsu Cho cf2400036e [CI] Add Python and C++ tests for Windows GPU target (#4469) 
* Add CMake option to use bundled gtest from dmlc-core, so that it is easy to build XGBoost with gtest on Windows

* Consistently apply OpenMP flag to all targets. Force enable OpenMP when USE_CUDA is turned on.

* Insert vcomp140.dll into Windows wheels

* Add C++ and Python tests for CPU and GPU targets (CUDA 9.0, 10.0, 10.1)

* Prevent spurious msbuild failure

* Add GPU tests

* Upgrade dmlc-core
 2019-05-16 01:06:46 +00:00
Philip Hyunsu Cho c6f2a7e186 [CI] Add Windows GPU to Jenkins CI pipeline (#4463) 
* Fix #4462: Use /MT flag consistently for MSVC target

* First attempt at Windows CI

* Distinguish stages in Linux and Windows pipelines

* Try running CMake in Windows pipeline

* Add build step
 2019-05-14 04:45:06 +00:00
Philip Hyunsu Cho e7d17ec4f4 [CI] Build XGBoost wheels with CUDA 9.0 (#4459) 
* [CI] Build XGBoost wheels with CUDA 9.0

* Do not call archiveArtifacts for 8.0 wheel
 2019-05-11 16:35:02 -07:00
Philip Hyunsu Cho b5f7cbfadf [CI] Cache two R build Docker containers (#4458) 2019-05-11 10:54:00 -07:00
Philip Hyunsu Cho 6ff994126a [BLOCKING][CI] Upgrade to Spark 2.4.3 (#4414) 
* [CI] Upgrade to Spark 2.4.2

* Pass Spark version to build script

* Allow multiple --build-arg in ci_build.sh

* Fix syntax

* Fix container name

* Update pom.xml

* Fix container name

* Update Jenkinsfile

* Update pom.xml

* Update Dockerfile.jvm_cross
 2019-05-09 21:36:59 -07:00
Philip Hyunsu Cho ade3f30237 Fix list formatting in missing value tutorial in XGBoost4J-Spark 2019-05-06 14:24:02 -07:00
Philip Hyunsu Cho b511638ca1 Fix list formatting in missing value tutorial in XGBoost4J-Spark 2019-05-06 14:21:49 -07:00
Philip Hyunsu Cho 9252b686ae Make AUCPR work with multiple query groups (#4436) 
* Make AUCPR work with multiple query groups

* Check AUCPR <= 1.0 in distributed setting
 2019-05-03 10:34:44 -07:00
Philip Hyunsu Cho c1e4a0f2c6 Upgrade dmlc-core (#4430) 2019-05-02 18:38:22 +08:00
Philip Hyunsu Cho bfddc2c42c Make CMakeLists.txt compatible with CMake 3.3 (#4420) 
* Make CMakeLists.txt compatible with CMake 3.3; require CMake 3.11 for MSVC

* Use CMake 3.12 when sanitizer is enabled

* Disable funroll-loops for MSVC

* Use cmake version in container name

* Add missing arg

* Fix egrep use in ci_build.sh

* Display CMake version

* Do not set OpenMP_CXX_LIBRARIES for MSVC

* Use cmake_minimum_required()
 2019-05-02 11:49:32 +08:00
Philip Hyunsu Cho 17df5fd296 Simplify bound checking in feature interaction constraints (#4428) 2019-05-01 16:59:53 -07:00
Philip Hyunsu Cho ba98e0cdf2 Add additional Python tests to test training under constraints (#4426) 2019-04-30 18:23:39 -07:00
Philip Hyunsu Cho ea850ecd20 [CI] Refactor Jenkins CI pipeline + migrate all Linux tests to Jenkins (#4401) 
* All Linux tests are now in Jenkins CI
* Tests are now de-coupled from builds. We can now build XGBoost with one version of CUDA/JDK and test it with another version of CUDA/JDK
* Builds (compilation) are significantly faster because 1) They use C5 instances with faster CPU cores; and 2) build environment setup is cached using Docker containers
 2019-04-26 18:39:12 -07:00
Philip Hyunsu Cho 503cc42f48 [CI] Fix Windows tests (#4403) 
* Install binary igraph

* Include Graphviz in PATH
 2019-04-25 20:25:43 -07:00
Philip Hyunsu Cho bbe0dbd7ec Migrate pylint check to Python 3 (#4381) 
* Migrate lint to Python 3

* Fix lint errors

* Use Miniconda3 to use Python 3.7

* Use latest pylint and astroid
 2019-04-21 01:01:54 -07:00
Philip Hyunsu Cho 70be1e38c2 [CI] Optimize external Docker build cache (#4334) 
* When building pull requests, use Docker cache for master branch

Docker build caches are per-branch, so new pull requests will initially
have no build cache, causing the Docker containers to be built from
scratch. New pull requests should use the cache associated with the
master branch. This makes sense, since most pull requests do not modify
the Dockerfile.

* Add comments
 2019-04-04 15:59:07 -07:00
Philip Hyunsu Cho 37c75aac41 [CI] Add external Docker build cache (#4331) 2019-04-04 13:36:39 -07:00
Philip Hyunsu Cho 7aed8f3d48 [CI] Upgrade to GCC 5.3.1, CMake 3.6.0 (#4306) 
* Upgrade to GCC 5.3.1, CMake 3.6.0

* <regex> is now okay
 2019-03-28 00:21:21 -07:00
Philip Hyunsu Cho 263e2038e9 Bump Python version number (#4285) 2019-03-21 14:40:44 -07:00
Philip Hyunsu Cho 99a714be64 Simplify README page (#4254) 2019-03-12 11:58:08 -07:00
Philip Hyunsu Cho f83e62dca5 Address #4042: Prevent out-of-range access in column matrix (#4231) 2019-03-08 17:11:42 -08:00
Philip Hyunsu Cho 331cd3e4f7 Document limitation of one-split-at-a-time Greedy tree learning heuristic (#4233) 2019-03-08 10:05:39 -08:00
Philip Hyunsu Cho 20845e8ccf Broken link for NCCL: cannot use CUDA 10.1 (#4232) 2019-03-08 09:10:29 -08:00
Philip Hyunsu Cho 28bd6cde22 Add sponsors (#4222) 2019-03-06 13:11:06 -08:00
Philip Hyunsu Cho 67c38805a1 Update build doc: PyPI wheel now support multi-GPU (#4219) 2019-03-05 13:25:31 -08:00
Philip Hyunsu Cho 3f83dcd502 Release 0.82 (#4201) 2019-03-04 18:14:36 -08:00
Philip Hyunsu Cho 2aaae2e7bb Fix #4163: always copy sliced data (#4165) 
* Revert "Accept numpy array view. (#4147)"

This reverts commit a985a99cf0.

* Fix #4163: always copy sliced data

* Remove print() from the test; check shape equality

* Check if 'base' attribute exists

* Fix lint

* Address reviewer comment

* Fix lint
 2019-02-20 14:46:34 -08:00
Philip Hyunsu Cho 549c8d6ae9 Prevent empty quantiles in fast hist (#4155) 
* Prevent empty quantiles

* Revise and improve unit tests for quantile hist

* Remove unnecessary comment

* Add #2943 as a test case

* Skip test if no sklearn

* Revise misleading comments
 2019-02-17 16:01:07 -08:00
Philip Hyunsu Cho 37ddfd7d6e Fix broken R test: Install Homebrew GCC (#4142) 
* Fix broken R test: Install Homebrew GCC

Missing GCC Fortran causes installation failure of a dependency package
(igraph)

* Register gfortran system-wide

* Use correct keg

* Set env vars to change compiler choice

* Do not break other Mac builds

* Nuclear option: symlink gfortran

* Use /usr/local/bin instead of /usr/bin

* Symlink library path too

* Update run_test.sh
 2019-02-15 07:23:05 -08:00
Philip Hyunsu Cho 99a290489c Update Python docstring for ranking functions (#4121) 
* Update Python docstring for ranking functions

* Fix formatting
 2019-02-10 12:22:02 -08:00
Philip Hyunsu Cho 7a652a8c64 Speed up Jenkins by not compiling CMake (#4099) 2019-02-03 00:08:14 -08:00
Philip Hyunsu Cho a60e224484 Add Jenkins status badge (#4090) 2019-01-30 14:03:18 -08:00
Philip Hyunsu Cho a1c35cadf0 Fix failing Travis CI on Mac (#4086) 
* Fix failing Travis CI on Mac

Use Homebrew Addon + latest Mac image

* Use long command for pytest

* Downgrade OSX image to xcode9.3, to use Java 8

* Install pytest in Python 2 environment

* Remove clang-tidy from Travis
 2019-01-30 09:43:57 -08:00
Philip Hyunsu Cho 9af6b689d6 Use int instead of char in CLI config parser (#3976) 2018-12-07 01:00:21 -08:00
Philip Hyunsu Cho 4f26053b09 Fix typo in Feature Interaction Constraints tutorial (#3975) 2018-12-06 19:38:40 -08:00
Philip Hyunsu Cho c5f92df475 Disable retries in Jenkins CI, since we're now using On-Demand instances instead of Spot (#3948) 2018-11-28 14:57:09 -08:00
Philip Hyunsu Cho c5130e487a Fix #3894: Allow loading pickles without self.booster attributes (redux) (#3944) 2018-11-28 09:31:46 -08:00
Philip Hyunsu Cho f9302a56fb Fix #3894: Allow loading pickles without self.booster attributes (#3938) 
The addition of self.booster attribute broke backward compatibility.
 2018-11-23 12:15:50 -08:00
Philip Hyunsu Cho 7d3149a21f Add AUC-PR to list of metrics to maximize for early stopping (#3936) 2018-11-23 12:15:34 -08:00
Philip Hyunsu Cho 86aac98e54 [jvm-packages] Fix #3898: use correct group ID for maven-site-plugin (#3937) 2018-11-23 09:46:27 -08:00
Philip Hyunsu Cho e9ab4a1c6c Address #3933: document limitation of DMLC CSV parser + recommend Pandas (#3934) 2018-11-23 04:13:36 -08:00
Philip Hyunsu Cho 973fc8b1ff Use consistent type for sharding GPU data in GPU coordinate updater (#3917) 
* Use consistent type for sharding GPU data in GPU coordinate updater

* Use fast integer ceiling trick
 2018-11-18 00:20:00 -08:00
Philip Hyunsu Cho 3a150742c7 Update dmlc-core submodule (#3907) 2018-11-15 18:50:49 -08:00
Philip Hyunsu Cho c76d993681 Enforce naming style in Python lint (#3896) 2018-11-14 10:35:25 -08:00
Philip Hyunsu Cho a2a8954659 Update dmlc-core submodule (#3891) 2018-11-14 01:51:27 -08:00
Philip Hyunsu Cho be0bb7dd90 Remove unnecessary warning when 'gblinear' is selected (#3888) 2018-11-09 12:30:38 -08:00
Philip Hyunsu Cho e38d5a6831 Document current limitation in number of features (#3886) 2018-11-09 00:32:43 -08:00
Philip Hyunsu Cho 828d75714d Fix #3857: take down AWS YARN tutorial, as it is outdated (#3885) 2018-11-08 23:08:32 -08:00
Philip Hyunsu Cho ad6e0d55f1 Fix coef_ and intercept_ signature to be compatible with sklearn.RFECV (#3873) 
* Fix coef_ and intercept_ signature to be compatible with sklearn.RFECV

* Fix lint

* Fix lint
 2018-11-08 19:41:35 -08:00
Philip Hyunsu Cho 2b045aa805 Make C++ unit tests run and pass on Windows (#3869) 
* Make C++ unit tests run and pass on Windows

* Fix logic for external memory. The letter ':' is part of drive letter,
so remove the drive letter before splitting on ':'.
* Cosmetic syntax changes to keep MSVC happy.

* Fix lint

* Add Windows guard
 2018-11-06 17:17:24 -08:00
Philip Hyunsu Cho 20d5abf919 Disallow std::regex since it's not supported by GCC 4.8.x (#3870) 2018-11-05 22:57:04 -08:00
Philip Hyunsu Cho 91cc14ea70 Add another contributor for rabit update 2018-11-04 10:29:21 -08:00
Philip Hyunsu Cho 78ec77fa97 Release 0.81 version (#3864) 
* Release 0.81 version

* Update NEWS.md
 2018-11-04 05:49:11 -08:00
Philip Hyunsu Cho c22e90d5d2 Correct typo 2018-11-04 05:22:53 -08:00
Philip Hyunsu Cho 6da462234e Move MinGW-w64 + Python section to the end, since it's 'advanced' (#3863) 2018-11-04 05:12:27 -08:00