Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6,038 Commits 32 Branches 59 Tags
Commit Graph

1150 Commits

Author SHA1 Message Date
Jiaming Yuan
c22f6db4bf
[backport] Fix CPU bin compression with categorical data. (#8809) (#8810) 
* [backport] Fix CPU bin compression with categorical data. (#8809)

* Fix CPU bin compression with categorical data.

* The bug causes the maximum category to be lesser than 256 or the maximum number of bins when
the input data is dense.

* Avoid test symbol.
 2023-02-16 06:39:25 +08:00
Jiaming Yuan
f15a6d2b19
[backport] Fix ranking with quantile dmatrix and group weight. (#8762) (#8800) 
* [backport] Fix ranking with quantile dmatrix and group weight. (#8762)

* backport test utilities.
 2023-02-15 02:45:09 +08:00
Jiaming Yuan
60303db2ee
[backport] Fix GPU L1 error. (#8749) (#8770) 
* [backport] Fix GPU L1 error. (#8749)

* Fix backport.
 2023-02-09 20:16:39 +08:00
Jiaming Yuan
df984f9c43
[backport] Fix different number of features in gpu_hist evaluator. (#8754) (#8769) 
Co-authored-by: Rory Mitchell <r.a.mitchell.nz@gmail.com>
 2023-02-09 18:31:49 +08:00
Jiaming Yuan
2f22f8d49b
[backport] Make sure input numpy array is aligned. (#8690) (#8696) (#8734) 
* [backport] Make sure input numpy array is aligned. (#8690)

- use `np.require` to specify that the alignment is required.
- scipy csr as well.
- validate input pointer in `ArrayInterface`.

* Workaround CUDA warning. (#8696)

* backport from half type support for alignment.

* fix import.
 2023-02-06 16:58:15 +08:00
Jiaming Yuan
e5bef4ffce
[backport] Fix threads in DMatrix slice. (#8667) (#8679) 2023-01-14 18:46:04 +08:00
Jiaming Yuan
10bb0a74ef
[backport] [CI] Skip pyspark sparse tests. (#8675) (#8678) 2023-01-14 06:40:17 +08:00
Jiaming Yuan
899e4c8988
[backport] Do not return internal value for get_params. (#8634) (#8642) 2023-01-06 02:28:39 +08:00
Jiaming Yuan
067b704e58
[backport] Fix inference with categorical feature. (#8591) (#8602) (#8638) 
* Fix inference with categorical feature. (#8591)

* Fix windows build on buildkite. (#8602)

* workaround.
 2023-01-06 01:17:49 +08:00
Jiaming Yuan
162b48a1a4
[backport] [CI] Disable gtest with RMM (#8620) (#8640) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2023-01-05 23:13:45 +08:00
Jiaming Yuan
83a078b7e5
[backport] Fix sklearn test that calls a removed field (#8579) (#8636) 
Co-authored-by: Rong Ou <rong.ou@gmail.com>
 2023-01-05 21:17:05 +08:00
Jiaming Yuan
575fba651b
[backport] [CI] Fix CI with updated dependencies. (#8631) (#8635) 2023-01-05 19:10:58 +08:00
Jiaming Yuan
60a8c8ebba
[pyspark] sort qid for SparkRanker (#8497) (#8555) 
* [pyspark] sort qid for SparkRandker

* resolve comments

Co-authored-by: Bobby Wang <wbo4958@gmail.com>
 2022-12-07 02:07:37 +08:00
Jiaming Yuan
58bc225657
[backport] [CI] Fix github action mismatched glibcxx. (#8551) (#8552) 
Split up the Linux test to use the toolchain from conda forge.
 2022-12-06 21:35:26 +08:00
Philip Hyunsu Cho
db14e3feb7
Support null value in CUDA array interface. (#8486) (#8499) 2022-11-30 11:44:54 -08:00
Philip Hyunsu Cho
5b76acccff
Add back xgboost.rabit for backwards compatibility (#8408) (#8411) 2022-11-02 07:56:55 -07:00
Dmitry Razdoburdin
5bd849f1b5
Unify the partitioner for hist and approx. 
Co-authored-by: dmitry.razdoburdin <drazdobu@jfldaal005.jf.intel.com>
Co-authored-by: jiamingy <jm.yuan@outlook.com>
 2022-10-20 02:49:20 +08:00
Jiaming Yuan
c884b9e888
Validate features for inplace predict. (#8359) 2022-10-19 23:05:36 +08:00
Bobby Wang
76f95a6667
[pyspark] Filter out the unsupported train parameters (#8355) 2022-10-18 23:26:02 +08:00
Jiaming Yuan
3901f5d9db
[pyspark] Cleanup data processing. (#8344) 
* Enable additional combinations of ctor parameters.
* Unify procedures for QuantileDMatrix and DMatrix.
 2022-10-18 14:56:23 +08:00
Rong Ou
8f3dee58be
Speed up tests with federated learning enabled (#8350) 
* Speed up tests with federated learning enabled

* Re-enable timeouts

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
 2022-10-17 15:17:04 -07:00
Jiaming Yuan
031d66ec27
Configuration for init estimation. (#8343) 
* Configuration for init estimation.

* Check whether the model needs configuration based on const attribute `ModelFitted`
instead of a mutable state.
* Add parameter `boost_from_average` to tell whether the user has specified base score.
* Add tests.
 2022-10-18 01:52:24 +08:00
Jiaming Yuan
2176e511fc
Disable pytest-timeout for now. (#8348) 2022-10-17 23:06:10 +08:00
Jiaming Yuan
748d516c50
[pyspark] Enable running GPU tests on variable number of GPUs. (#8335) 2022-10-13 21:03:45 +08:00
Jiaming Yuan
3ef1703553
Allow using string view to find JSON value. (#8332) 
- Allow comparison between string and string view.
- Fix compiler warnings.
 2022-10-13 17:10:13 +08:00
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
Jiaming Yuan
97a5b088a5
[pyspark] Use quantile dmatrix. (#8284) 2022-10-12 20:38:53 +08:00
Rory Mitchell
ce0382dcb0
[CI] Refactor tests to reduce CI time. (#8312) 2022-10-12 11:32:06 +02:00
Rong Ou
39afdac3be
Better error message when world size and rank are set as strings (#8316) 
Co-authored-by: jiamingy <jm.yuan@outlook.com>
 2022-10-12 15:53:25 +08:00
Rory Mitchell
210915c985
Use integer gradients in gpu_hist split evaluation (#8274) 2022-10-11 12:16:27 +02:00
Jiaming Yuan
5545c49cfc
Require keyword args for data iterator. (#8327) 2022-10-10 17:47:13 +08:00
Jiaming Yuan
e1f9f80df2
Use gpu predictor for get csr test. (#8323) 2022-10-10 16:12:37 +08:00
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
Rory Mitchell
909e49e214
Reduce docker image size. (#8306) 2022-10-05 15:55:51 -08:00
Rong Ou
668b8a0ea4
[Breaking] Switch from rabit to the collective communicator (#8257) 
* Switch from rabit to the collective communicator

* fix size_t specialization

* really fix size_t

* try again

* add include

* more include

* fix lint errors

* remove rabit includes

* fix pylint error

* return dict from communicator context

* fix communicator shutdown

* fix dask test

* reset communicator mocklist

* fix distributed tests

* do not save device communicator

* fix jvm gpu tests

* add python test for federated communicator

* Update gputreeshap submodule

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
 2022-10-05 14:39:01 -08:00
Jiaming Yuan
e47b3a3da3
Upgrade mypy. (#8302) 
Some breaking changes were made in mypy.
 2022-10-05 14:31:59 +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
Rory Mitchell
d686bf52a6
Reduce time for some multi-gpu tests (#8288) 
* Faster dask tests

* Reuse AllReducer objects in tests.

* Faster boost from prediction tests.

* Use rmm dask fixture.

* Speed up dask demo.

* mypy

* Format with black.

* mypy

* Clang-tidy

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
 2022-10-04 02:49:33 -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
Jiaming Yuan
299e5000a4
Fix buildkite label. (#8287) 2022-09-29 17:33:19 -07:00
Jiaming Yuan
55cf24cc32
Obtain CSR matrix from DMatrix. (#8269) 2022-09-29 20:41:43 +08:00
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
Jiaming Yuan
6925b222e0
Fix mixed types with cuDF. (#8280) 2022-09-29 00:57:52 +08:00
Jiaming Yuan
6d1452074a
Remove MGPU cpp tests. (#8276) 
Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
 2022-09-27 21:18:23 +08:00
Jiaming Yuan
fcab51aa82
Support more pandas nullable types (#8262) 
- Float32/64
- Category.
 2022-09-27 01:59:50 +08:00
Rory Mitchell
8f77677193
Use quantised gradients in gpu_hist histograms (#8246) 2022-09-26 17:35:35 +02:00