Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
5,045 Commits 32 Branches 59 Tags
81c37c28d5f3fd6637449fde6186f1c68171409a
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Jiaming Yuan 81c37c28d5 Time the CPU tests on Jenkins. (#6257) 
* Time the CPU tests on Jenkins.
* Reduce thread contention.
* Add doc.
* Skip heavy tests on ARM.
 2020-10-20 17:19:07 -07:00
Jiaming Yuan ddf37cca30 Unify thread configuration. (#6186) 2020-10-19 16:05:42 +08:00
Jiaming Yuan 5037abeb86 Fix linear gpu input (#6255) 2020-10-19 12:02:36 +08:00
Jiaming Yuan 6d293020fb Disable JSON full serialization for now. (#6248) 
* Disable JSON serialization for now.

* Multi-class classification is checkpointing for each iteration.
This brings significant overhead.

Revert: 90355b4f00

* Set R tests to use binary.
 2020-10-16 17:59:54 +08:00
Jiaming Yuan 52452bebb9 Fix cls typo. (#6247) 2020-10-16 16:40:44 +08:00
Jiaming Yuan 3da5a69dc9 Fix typo in dask interface. (#6240) 2020-10-15 15:26:29 +08:00
Jiaming Yuan bed7ae4083 Loop over thrust::reduce. (#6229) 
* Check input chunk size of dqdm.
* Add doc for current limitation.
 2020-10-14 10:40:56 +13:00
Jiaming Yuan b05073bda5 [dask] Test for data initializaton. (#6226) 2020-10-13 11:08:35 +08:00
Jiaming Yuan 2443275891 Cleanup Python code. (#6223) 
* Remove pathlike as XGBoost 1.2 requires Python 3.6.
* Move conditional import of dask/distributed into dask module.
 2020-10-12 15:44:41 +08:00
Jiaming Yuan 70c2039748 Catch all standard exceptions in C API. (#6220) 
* `std::bad_alloc` is not guaranteed to be caught.
 2020-10-12 14:01:46 +08:00
Jiaming Yuan 2241563f23 Handle duplicated values in sketching. (#6178) 
* Accumulate weights in duplicated values.
* Fix device id in iterative dmatrix.
 2020-10-10 19:32:44 +08:00
Jiaming Yuan ab5b35134f Rework Python callback functions. (#6199) 
* Define a new callback interface for Python.
* Deprecate the old callbacks.
* Enable early stopping on dask.
 2020-10-10 17:52:36 +08:00
Jiaming Yuan b5b24354b8 More categorical tests and disable shap sparse test. (#6219) 
* Fix tree load with 32 category.
 2020-10-10 16:12:37 +08:00
Jiaming Yuan 70ce5216b5 Add high level tests for categorical data. (#6179) 
* Fix unique.
 2020-10-09 09:27:23 +08:00
Jiaming Yuan 4cfdcaaf7b Move non-OpenMP gtest to GitHub Actions (#6210) 2020-10-08 00:58:21 -07:00
Jiaming Yuan ddc4f20e54 Add JSON schema for categorical splits. (#6194) 2020-10-07 17:33:31 +08:00
Jiaming Yuan fd58005edf Ignore cachedir by joblib. [skip ci] (#6193) 2020-10-04 14:54:32 +08:00
Jiaming Yuan f0c63902ff Use default allocator in sketching. (#6182) 2020-09-30 14:55:59 +08:00
Jiaming Yuan 444131a2e6 Add categorical data support to GPU Hist. (#6164) 2020-09-29 11:27:25 +08:00
Jiaming Yuan 798af22ff4 Add categorical data support to GPU predictor. (#6165) 2020-09-29 11:25:34 +08:00
Jiaming Yuan 7622b8cdb8 Enable categorical data support on Python DMatrix. (#6166) 
* Only pandas is recognized.
 2020-09-29 11:22:56 +08:00
Jiaming Yuan 52c0b3f100 Fix error message. (#6176) 2020-09-29 11:18:25 +08:00
Jiaming Yuan 434a3f35a3 Add TAGS to gitignore. [skip ci] (#6175) 2020-09-27 21:27:40 +08:00
Jiaming Yuan 07355599c2 Option for generating device debug info. (#6168) 
* Supply `-G;-src-in-ptx` when `USE_DEVICE_DEBUG` is set and debug mode is selected.
* Refactor CMake script to gather all CUDA configuration.
* Use CMAKE_CUDA_ARCHITECTURES.  Close #6029.
* Add compute 80.  Close #5999
 2020-09-27 03:26:56 +08:00
Jiaming Yuan 14afdb4d92 Support categorical data in ellpack. (#6140) 2020-09-24 19:28:57 +08:00
Jiaming Yuan 78d72ef936 Add DaskDeviceQuantileDMatrix demo. (#6156) 2020-09-24 14:08:28 +08:00
Jiaming Yuan e033caa3ba Remove linking RMM library. (#6146) 
* Remove linking RMM library.

* RMM is now header only.

* Remove remaining reference.
 2020-09-22 16:59:33 -07:00
Jiaming Yuan 452ac8ea62 Time GPU tests on CI. (#6141) 2020-09-22 14:25:10 +08:00
Jiaming Yuan 33d80ffad0 [dask] Support more meta data on functional interface. (#6132) 
* Add base_margin, label_(lower|upper)_bound.
* Test survival training with dask.
 2020-09-21 16:56:37 +08:00
Jiaming Yuan 7065779afa Improve JSON format for categorical features. (#6128) 
* Gather categories for all nodes.
 2020-09-21 15:35:05 +08:00
Jiaming Yuan 210c131ce7 Support categorical data in GPU sketching. (#6137) 2020-09-21 13:53:06 +08:00
Jiaming Yuan a069a21e03 Implement intrusive ptr (#6129) 
* Use intrusive ptr for JSON.
 2020-09-20 20:07:16 +08:00
Jiaming Yuan e319b63f9e Merge extract cuts into QuantileContainer. (#6125) 
* Use pruning for initial summary construction.
 2020-09-18 16:36:39 +08:00
Jiaming Yuan cc82ca167a [dask] Refactor meta data handling. (#6130) 2020-09-18 13:26:40 +08:00
Jiaming Yuan 5384ed85c8 Use caching allocator from RMM, when RMM is enabled (#6131) 2020-09-17 21:51:49 -07:00
Jiaming Yuan b5f52f0b1b Validate weights are positive values. (#6115) 2020-09-15 09:03:55 +08:00
Jiaming YuanandRory Mitchell c6f2b8c841 Upgrade gputreeshap. (#6099) 
* Upgrade gputreeshap.

Co-authored-by: Rory Mitchell <r.a.mitchell.nz@gmail.com>
 2020-09-15 12:57:22 +12:00
Jiaming Yuan 07945290a2 Remove unused RABIT targets. (#6110) 
* Remove rabit mock.
* Remove rabit base.
 2020-09-11 14:09:44 +08:00
Jiaming Yuan c92d751ad1 Enable building rabit on Windows (#6105) 2020-09-11 11:54:46 +08:00
Jiaming Yuan 08bdb2efc8 Fix dask doc. [skip ci] (#6108) 2020-09-11 10:56:12 +08:00
Jiaming Yuan 3dcd85fab5 Refactor rabit tests (#6096) 
* Merge rabit tests into XGBoost.
* Run them On CI.
* Simplification for CMake scripts.
 2020-09-09 12:30:29 +08:00
Jiaming Yuan 318bffaa10 Fix custom obj link. [skip ci] (#6100) 2020-09-09 10:55:38 +08:00
Jiaming Yuan b0001a6e29 Correct style warnings from clang-tidy for rabit. (#6095) 2020-09-08 12:13:58 +08:00
Jiaming Yuan 93e9af43bb Unify set index data. (#6062) 2020-09-08 11:38:41 +08:00
Jiaming Yuan e5d40b39cd [Breaking] Don't save leaf child count in JSON. (#6094) 
The field is deprecated and not used anywhere in XGBoost.
 2020-09-08 11:11:13 +08:00
Jiaming Yuan 5994f3b14c Don't link imported target. (#6093) 2020-09-07 02:51:09 -07:00
Jiaming Yuan 2fcc4f2886 Unify evaluation functions. (#6037) 2020-08-26 14:23:27 +08:00
Jiaming Yuan 80c8547147 Make binary bin search reusable. (#6058) 
* Move binary search row to hist util.
* Remove dead code.
 2020-08-26 05:05:11 +08:00
Jiaming Yuan 81d8dd79ca Bump header version. (#6056) 2020-08-26 00:29:00 +08:00
Jiaming YuanandPhilip Hyunsu Cho 20c95be625 Expand categorical node. (#6028) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2020-08-25 18:53:57 +08:00
Jiaming Yuan a144daf034 Limit tree depth for GPU hist. (#6045) 2020-08-22 19:34:52 +08:00
Jiaming Yuan b9ebbffc57 Fix plotting test. (#6040) 
Previously the test loads a model generated by `test_basic.py`, now we generate
the model explicitly.

* Cleanup saved files for basic tests.
 2020-08-22 13:18:48 +08:00
Jiaming Yuan 7a46515d3d Remove win2016 jvm github action test. (#6042) 2020-08-20 19:39:46 -07:00
Jiaming Yuan 7be2e04bd4 Fix scikit learn cls doc. (#6041) 2020-08-20 19:23:06 -07:00
Jiaming Yuan 29b7fea572 Optimize cpu sketch allreduce for sparse data. (#6009) 
* Bypass RABIT serialization reducer and use custom allgather based merging.
 2020-08-19 10:03:45 +08:00
Jiaming Yuan 90355b4f00 Make JSON the default full serialization format. (#6027) 2020-08-19 09:57:43 +08:00
Jiaming Yuan 4d99c58a5f Feature weights (#5962) 2020-08-18 19:55:41 +08:00
Jiaming Yuan a418278064 Merge pull request #6023 from trivialfis/merge-rabit 
Merge rabit
 2020-08-18 09:01:56 +08:00
fis 111968ca58 Merge rabit 2020-08-18 03:52:33 +08:00
fis 1c5904df3f Remove rabit. 2020-08-18 03:48:36 +08:00
Jiaming Yuan d240463b38 Revert "Remove warning about memset. (#6003)" (#6020) 
This reverts commit 12e3fb6a6c.
 2020-08-17 20:10:15 +08:00
Jiaming Yuan 674c409e9d Remove rabit dependency on public headers. (#6005) 2020-08-13 08:26:20 +08:00
Jiaming Yuan 12e3fb6a6c Remove warning about memset. (#6003) 2020-08-13 08:25:46 +08:00
Jiaming Yuan c3ea3b7e37 Fix nightly build doc. [skip ci] (#6004) 
* Fix nightly build doc. [skip ci]

* Fix title too short. [skip ci]
 2020-08-12 15:00:40 +08:00
Jiaming Yuan ee70a2380b Unify CPU hist sketching (#5880) 2020-08-12 01:33:06 +08:00
Jiaming Yuan 6f7112a848 Move warning about empty dataset. (#5998) 2020-08-11 14:10:51 +08:00
Jiaming Yuan f93f1c03fc Rabit update. (#5978) 
* Remove parameter on JVM Packages.
 2020-08-11 09:17:32 +08:00
Jiaming Yuan 0b2a26fa74 Remove skmaker. (#5971) 2020-08-09 15:23:31 +08:00
Jiaming Yuan 801e6b6800 Fix dask predict shape infer. (#5989) 2020-08-08 14:29:22 +08:00
Jiaming Yuan 4acdd7c6f6 Remove stop process. (#143) 2020-08-05 10:12:00 -07:00
Jiaming Yuan 9c6e791e64 Enforce tree order in JSON. (#5974) 
* Make JSON model IO more future proof by using tree id in model loading.
 2020-08-05 16:44:52 +08:00
Jiaming Yuan dde9c5aaff Fix missing data warning. (#5969) 
* Fix data warning.

* Add numpy/scipy test.
 2020-08-05 16:19:12 +08:00
Jiaming Yuan 8599f87597 Update JSON schema. (#5982) 
* Update JSON schema for pseudo huber.
* Update JSON model schema.
 2020-08-05 15:21:11 +08:00
Jiaming Yuan 9c93531709 Update Python custom objective demo. (#5981) 2020-08-05 12:27:19 +08:00
Jiaming Yuan 1149a7a292 Fix sklearn doc. (#5980) 2020-08-05 12:26:19 +08:00
Jiaming Yuan b069431c28 Export DaskDeviceQuantileDMatrix in doc. [skip ci] (#5975) 2020-08-05 00:48:10 +08:00
Jiaming Yuan 70903c872f Force colored output for ninja build. (#5959) 2020-07-30 20:48:03 +08:00
Jiaming Yuan fa3715f584 [Dask] Asyncio support. (#5862) 2020-07-30 06:23:58 +08:00
Jiaming Yuan e4a273e1da Fix evaluate root split. (#5948) 2020-07-29 19:33:29 +08:00
Jiaming Yuan f5fdcbe194 Disable feature validation on sklearn predict prob. (#5953) 
* Fix issue when scikit learn interface receives transformed inputs.
 2020-07-29 19:26:44 +08:00
Jiaming Yuan 18349a7ccf [Breaking] Fix custom metric for multi output. (#5954) 
* Set output margin to true for custom metric.  This fixes only R and Python.
 2020-07-29 19:25:27 +08:00
Jiaming Yuan 75b8c22b0b Fix prediction heuristic (#5955) 
* Relax check for prediction.
* Relax test in spark test.
* Add tests in C++.
 2020-07-29 19:24:07 +08:00
Jiaming Yuan 40361043ae [BLOCKING] Remove to_string. (#5934) 2020-07-26 10:21:26 +08:00
Jiaming Yuan a4de2f68e4 Use cudaOccupancyMaxPotentialBlockSize to calculate the block size. (#5926) 2020-07-23 14:24:42 +08:00
Jiaming Yuan fbfbd525d8 Cache dependencies on Github Action. (#5928) 2020-07-23 14:00:19 +08:00
Jiaming Yuan bc1d3ee230 Fix r early stop with custom objective. (#5923) 
* Specify `ntreelimit`.
 2020-07-23 03:28:17 +08:00
Jiaming Yuan 30363d9c35 Remove R and JVM from appveyor. (#5922) 2020-07-23 03:26:48 +08:00
Jiaming Yuan 66cc1e02aa Setup github action. (#5917) 2020-07-22 15:05:25 +08:00
Jiaming Yuan 9b688aca3b Fix mingw build with R. (#5918) 2020-07-22 02:56:49 +08:00
Jiaming Yuan 03fb98fbde Fix typo in CI. [skip ci] (#5919) 2020-07-21 14:25:27 +08:00
Jiaming Yuan 8b1afce316 Add Github Action for R. (#5911) 
* Fix lintr errors.
 2020-07-20 19:23:36 +08:00
Jiaming Yuan 6c0c87216f Fix Windows 2016 build. (#5902) 2020-07-18 05:50:17 +08:00
Jiaming Yuan 7c2686146e Dask device dmatrix (#5901) 
* Fix softprob with empty dmatrix.
 2020-07-17 13:17:43 +08:00
Jiaming Yuan e471056ec4 Fix sketch size calculation. (#5898) 2020-07-17 08:33:16 +08:00
Jiaming Yuan 029a8b533f Simplify the data backends. (#5893) 2020-07-16 15:17:31 +08:00
Jiaming Yuan 3cae287dea Fix NDK Build. (#5886) 
* Explicit cast for slice.
 2020-07-14 18:34:19 +08:00
Jiaming Yuan dd445af56e Cleanup on device sketch. (#5874) 
* Remove old functions.

* Merge weighted and un-weighted into a common interface.
 2020-07-14 10:15:54 +08:00
Jiaming Yuan d0a29c3135 Remove print. (#5867) 2020-07-08 04:12:14 +08:00
Jiaming Yuan a3ec964346 Accept iterator in device dmatrix. (#5783) 
* Remove Device DMatrix.
 2020-07-07 21:44:48 +08:00
Jiaming Yuan 048d969be4 Implement GK sketching on GPU. (#5846) 
* Implement GK sketching on GPU.
* Strong tests on quantile building.
* Handle sparse dataset by binary searching the column index.
* Hypothesis test on dask.
 2020-07-07 12:16:21 +08:00