Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
5,376 Commits 32 Branches 59 Tags
dd4db347f34c6f114b9b8db74b2706cb7f8f73fa
Commit Graph
5376 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Jiaming Yuan f5ff90cd87 Support _estimator_type. (#6582) 
* Use `_estimator_type`.

For more info, see: https://scikit-learn.org/stable/developers/develop.html#estimator-types

* Model trained from dask can be loaded by single node skl interface.
 2021-01-08 10:01:16 +08:00
Jiaming YuanandHyunsu Cho 8747885a8b Support Solaris. (#6578) 
* Add system header.

* Remove use of TR1 on Solaris

Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2021-01-07 09:05:05 +08:00
TP BoudreauandHyunsu Cho b2246ae7ef Update dmlc-core submodule and conform to new API (#6431) 
* Update dmlc-core submodule and conform to new API

* Remove unsupported parameter from method signature

* Update dmlc-core submodule and conform to new API

* Update dmlc-core

Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2021-01-05 16:12:22 -08:00
Jiaming Yuan 60cfd14349 [dask, sklearn] Fix predict proba. (#6566) 
* For sklearn:
  - Handles user defined objective function.
  - Handles `softmax`.

* For dask:
  - Use the implementation from sklearn, the previous implementation doesn't perform any extra handling.
 2021-01-05 08:29:06 +08:00
Jiaming Yuan 516a93d25c Fix best_ntree_limit. (#6569) 2021-01-03 05:58:54 +08:00
James Lamb 195a41cef1 [python-package] remove unnecessary files to reduce sdist size (fixes #6560) (#6565) 2021-01-02 15:56:39 +08:00
Jiaming Yuan 2b049b32e9 Document various tree methods. (#6564) 2021-01-02 15:40:46 +08:00
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
Jiaming Yuan 5e9e525223 Remove warnings in tests. (#6554) 2020-12-31 13:41:18 +08:00
James Lamb 8ad22bf4e7 Add credentials to .gitignore (#6559) 2020-12-30 15:58:14 -08:00
Jiaming YuanandHyunsu Cho de8fd852a5 [dask] Add type hints. (#6519) 
* Add validate_features.
* Show type hints in doc.

Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2020-12-29 19:41:02 +08:00
Jiaming Yuan 610ee632cc [Breaking] Rename data to X in predict_proba. (#6555) 
New Scikit-Learn version uses keyword argument, and `X` is the predefined
keyword.

* Use pip to install latest Python graphviz on Windows CI.
 2020-12-28 21:36:03 +08:00
Jiaming Yuan cb207a355d Add script for generating release tarball. (#6544) 2020-12-23 16:08:10 +08:00
Gorkem Ozkaya 2231940d1d Clip small positive values in gamma-nloglik (#6537) 
For the `gamma-nloglik` eval metric, small positive values in the labels are causing `NaN`'s in the outputs, as reported here: https://github.com/dmlc/xgboost/issues/5349. This will add clipping on them, similar to what is done in other metrics like `poisson-nloglik` and `logloss`.
 2020-12-22 03:11:40 +08:00
MBSMachineLearning 95cbfad990 "featue_map" typo changed to "feature_map" (#6540) 2020-12-21 22:11:11 +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
Jiaming YuanandPhilip Hyunsu Cho ca3da55de4 Support early stopping with training continuation, correct num boosted rounds. (#6506) 
* Implement early stopping with training continuation.

* Add new C API for obtaining boosted rounds.

* Fix off by 1 in `save_best`.

Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2020-12-17 19:59:19 +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
Jiaming Yuan d8d684538c [CI] Split up main.yml, add mypy. (#6515) 2020-12-17 00:15:44 +08:00
Jiaming Yuan c5876277a8 Drop saving binary format for memory snapshot. (#6513) 2020-12-17 00:14:57 +08:00
Jiaming Yuan 0e97d97d50 Fix merge conflict. (#6512) 2020-12-16 18:02:25 +08:00
hzy001andHao Ziyu 749364f25d Update the C API comments (#6457) 
Signed-off-by: Hao Ziyu <haoziyu@qiyi.com>

Co-authored-by: Hao Ziyu <haoziyu@qiyi.com>
 2020-12-16 14:56:13 +08:00
Jiaming Yuan 347f593169 Accept numpy array for DMatrix slice index. (#6368) 2020-12-16 14:42:52 +08:00
Jiaming YuanandPhilip Hyunsu Cho ef4a0e0aac Fix DMatrix feature names/types IO. (#6507) 
* Fix feature names/types IO

Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2020-12-16 14:24:27 +08:00
Jiaming Yuan 886486a519 Support categorical data in GPU weighted sketching. (#6508) 2020-12-16 14:23:28 +08:00
Igor Rukhovich 5c8ccf4455 Improved InitSampling function speed by 2.12 times (#6410) 
* Improved InitSampling function speed by 2.12 times

* Added explicit conversion
 2020-12-15 20:59:24 -08:00
Jiaming Yuan 3c3f026ec1 Move metric configuration into booster. (#6504) 2020-12-16 05:35:04 +08:00
Jiaming Yuan d45c0d843b Show partition status in dask error. (#6366) 2020-12-16 02:58:21 +08:00
James Lambandfis 1e2c3ade9e [doc] [dask] Add example on early stopping with Dask (#6501) 
Co-authored-by: fis <jm.yuan@outlook.com>
 2020-12-15 22:23:23 +08:00
ShvetsKSandKirill Shvets 8139849ab6 Fix handling of print period in EvaluationMonitor (#6499) 
Co-authored-by: Kirill Shvets <kirill.shvets@intel.com>
 2020-12-15 19:20:19 +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
James Lamb afc4567268 [doc] [dask] fix partitioning in Dask example (#6389) 2020-12-14 18:37:49 +08:00
Jiaming Yuan a30461cf87 [dask] Support all parameters in regressor and classifier. (#6471) 
* Add eval_metric.
* Add callback.
* Add feature weights.
* Add custom objective.
 2020-12-14 07:35:56 +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
Philip Hyunsu Cho 0d483cb7c1 Bump version to 1.4.0 snapshot in master (#6486) 2020-12-10 07:38:08 -08:00
Philip Hyunsu Cho b8044e6136 [CI] Use manylinux2010_x86_64 container to vendor libgomp (#6485) 2020-12-10 07:37:15 -08:00
Jiaming Yuan 0ffaf0f5be Fix dask ip resolution. (#6475) 
This adopts the solution used in dask/dask-xgboost#40 which employs the get_host_ip from dmlc-core tracker.
 2020-12-07 16:36:23 -08:00
Jiaming Yuan 47b86180f6 Don't validate feature when number of rows is 0. (#6472) 2020-12-07 18:08:51 +08:00
Philip Hyunsu Cho 55bdf084cb [Doc] Document that AUC and AUCPR are for binary classification/ranking [skip ci] (#5899) 2020-12-06 22:17:20 -08:00
Jiaming Yuan 703c2d06aa Fix global config default value. (#6470) 2020-12-06 06:15:33 +08:00
Jiaming YuanandHyunsu Cho d6386e45e8 Fix filtering callable objects in skl xgb param. (#6466) 
Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2020-12-05 17:20:36 +08:00
Philip Hyunsu Cho 05e5563c2c [CI] Fix CentOS 6 Docker images (#6467) 2020-12-04 21:33:11 -08:00
Philip Hyunsu Cho 84b726ef53 Vendor libgomp in the manylinux Python wheel (#6461) 
* Vendor libgomp in the manylinux2014_aarch64 wheel

* Use vault repo, since CentOS 6 has reached End-of-Life on Nov 30

* Vendor libgomp in the manylinux2010_x86_64 wheel

* Run verification step inside the container
 2020-12-03 19:55:32 -08:00
Philip Hyunsu Cho c103ec51d8 Enforce row-major order in cuPy array (#6459) 2020-12-03 18:29:10 -08:00
Philip Hyunsu Cho 4f70e14031 Fix docstring of config.py to use correct versionadded (#6458) 2020-12-03 10:41:53 -08:00