Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
5,837 Commits 32 Branches 59 Tags
2407381c3d3ebbf56b093701a9d9cb066216ebf8
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Jiaming YuanandHyunsu Cho e88d6e071d Fix compiler warning in JSON IO. (#8031) 
Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-06-30 01:13:22 +08:00
Jiaming Yuan dcaf580476 Fix Python package source install. (#8036) 
* Copy gputreeshap.
 2022-06-29 21:45:09 +08:00
Jiaming Yuan f0c1b842bf Implement sketching with adapter. (#8019) 2022-06-23 00:03:02 +08:00
Jiaming Yuan 142a208a90 Fix compiler warnings. (#8022) 
- Remove/fix unused parameters
- Remove deprecated code in rabit.
- Update dmlc-core.
 2022-06-22 21:29:10 +08:00
Jiaming Yuan 4a87ea49b8 Reduce regularization for CPU gblinear. (#8013) 2022-06-21 01:05:27 +08:00
Jiaming Yuan d285d6ba2a Reduce regularization in GPU gblinear test. (#8010) 2022-06-20 23:55:12 +08:00
Jiaming Yuan e58e417603 [CI] Fix lintr error. (#8011) 2022-06-20 22:17:14 +08:00
Jiaming Yuan 9b0eb66b78 Fix GPU driver test. (#8008) 
* Initialize the training parameter.
 2022-06-20 19:37:31 +08:00
Jiaming Yuan 637e42a0c0 Use 22.04 for RMM. (#8001) 
22.06 is not released yet.
 2022-06-17 04:07:31 +08:00
Jiaming Yuan bb47fd8c49 [jvm-packages] Change log level for tracker message. (#7968) 2022-06-09 18:15:08 +08:00
Jiaming Yuan 8f8bd8147a Fix LTR with weighted Quantile DMatrix. (#7975) 
* Fix LTR with weighted Quantile DMatrix.

* Better tests.
 2022-06-09 01:33:41 +08:00
Jiaming Yuan 1a33b50a0d Fix compiler warnings. (#7974) 
- Remove unused parameters. There are still many warnings that are not yet
addressed. Currently, the warnings in dmlc-core dominate the error log.
- Remove `distributed` parameter from metric.
- Fixes some warnings about signed comparison.
 2022-06-06 22:56:25 +08:00
Jiaming Yuanandjakirkham d48123d23b Fix rmm build (#7973) 
- Optionally switch to c++17
- Use rmm CMake target.
- Workaround compiler errors.
- Fix GPUMetric inheritance.
- Run death tests even if it's built with RMM support.

Co-authored-by: jakirkham <jakirkham@gmail.com>
 2022-06-06 20:18:32 +08:00
Jiaming Yuan b90c6d25e8 Implement max_cat_threshold for CPU. (#7957) 2022-06-04 11:02:46 +08:00
Jiaming Yuan 6b55150e80 Fix pylint errors. (#7967) 2022-06-02 18:04:46 +08:00
Jiaming Yuan 13b15e07e8 Handle formatted JSON input. (#7953) 2022-06-01 16:20:58 +08:00
Jiaming Yuan bde4f25794 Handle missing categorical value in CPU evaluator. (#7948) 2022-05-27 14:15:47 +08:00
Jiaming Yuan 18cbebaeb9 Unify the cat split storage for CPU. (#7937) 
* Unify the cat split storage for CPU.

* Cleanup.

* Workaround.
 2022-05-26 04:14:40 -07:00
Jiaming Yuan 606be9e663 Handle missing values in one hot splits. (#7934) 2022-05-24 20:48:41 +08:00
Jiaming Yuan 18a38f7ca0 Refactor for GHistIndex. (#7923) 
* Pass sparse page as adapter, which prepares for quantile dmatrix.
* Remove old external memory code like `rbegin` and extra `Init` function.
* Simplify type dispatch.
 2022-05-23 23:04:53 +08:00
Jiaming Yuan d314680a15 Verify shared object version at load. (#7928) 2022-05-23 20:53:30 +08:00
Jiaming Yuan 474366c020 Add convergence test for sparse datasets. (#7922) 2022-05-23 18:07:26 +08:00
Jiaming Yuan f93a727869 Address remaining mypy errors in python package. (#7914) 2022-05-18 22:46:15 +08:00
Jiaming Yuan edf9a9608e Fix type conversion warning. (#7916) 2022-05-18 20:14:14 +08:00
Jiaming Yuan 765097d514 Simplify inplace-predict. (#7910) 
Pass the `X` as part of Proxy DMatrix instead of an independent `dmlc::any`.
 2022-05-18 17:52:00 +08:00
Jiaming Yuan 19775ffe15 Use adapter to initialize column matrix. (#7912) 2022-05-18 16:15:12 +08:00
Jiaming Yuan 4fcfd9c96e Fix and cleanup for column matrix. (#7901) 
* Fix missed type dispatching for dense columns with missing values.
* Code cleanup to reduce special cases.
* Reduce memory usage.
 2022-05-16 21:11:50 +08:00
Jiaming Yuan 1baad8650c Small cleanup to Column. (#7898) 
* Define forward iterator to hide the internal state.
 2022-05-15 12:39:10 +08:00
Jiaming Yuan ee382c4153 Update news for 1.6.1 (#7877) 2022-05-14 15:38:18 -07:00
Jiaming YuanandPhilip Hyunsu Cho 1b6538b4e5 [breaking] Drop single precision histogram (#7892) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-05-13 19:54:55 +08:00
Jiaming Yuan c8f9d4b6e6 Show libxgboost.so path in build info. (#7893) 2022-05-13 18:08:56 +08:00
Jiaming Yuan 11d65fcb21 Extract partial sum into an independent function. (#7889) 2022-05-13 14:30:35 +08:00
Jiaming Yuan db80671d6b Fix monotone constraint with tuple input. (#7891) 2022-05-13 04:00:03 +08:00
Jiaming Yuan 94ca52b7b7 Fix overflow in prediction size. (#7885) 2022-05-12 02:44:03 +08:00
Jiaming Yuan 8ba4722d04 Remove pyarrow workaround. (#7884) 2022-05-11 20:54:48 +08:00
Jiaming Yuan 16ba74d008 Update CUDA version requirement in CMake script. (#7876) 2022-05-09 04:16:22 +08:00
Jiaming Yuan 46e0bce212 Use maximum category in sketch. (#7853) 2022-05-05 19:56:49 +08:00
Jiaming Yuan 8ab5e13b5d Fix typo [skip ci] (#7861) 2022-05-04 18:34:45 +08:00
Jiaming Yuan 317d7be6ee Always use partition based categorical splits. (#7857) 2022-05-03 22:30:32 +08:00
Jiaming Yuan 50d854e02e [CI] Test with latest RAPIDS. (#7816) 2022-04-30 11:55:10 -07:00
Jiaming Yuan 288c52596c Define bin type. (#7850) 2022-04-29 19:41:39 +08:00
Jiaming Yuan fdf533f2b9 [POC] Experimental support for l1 error. (#7812) 
Support adaptive tree, a feature supported by both sklearn and lightgbm.  The tree leaf is recomputed based on residue of labels and predictions after construction.

For l1 error, the optimal value is the median (50 percentile).

This is marked as experimental support for the following reasons:
- The value is not well defined for distributed training, where we might have empty leaves for local workers. Right now I just use the original leaf value for computing the average with other workers, which might cause significant errors.
- Some follow-ups are required, for exact, pruner, and optimization for quantile function. Also, we need to calculate the initial estimation.
 2022-04-26 21:41:55 +08:00
Jiaming Yuan ad06172c6b Refactor pandas dataframe handling. (#7843) 2022-04-26 18:53:43 +08:00
Jiaming Yuan 332380479b Avoid warning in np primitive type tests. (#7833) 2022-04-23 02:07:01 +08:00
Jiaming Yuan f0f76259c9 Remove STRING_TYPES. (#7827) 2022-04-22 19:07:51 +08:00
Jiaming Yuan c70fa502a5 Expose feature_types to sklearn interface. (#7821) 2022-04-21 20:23:35 +08:00
Jiaming Yuan 401d451569 Clear configuration cache. (#7826) 2022-04-21 19:09:54 +08:00
Jiaming Yuan 52d4eda786 Deprecate use_label_encoder in XGBClassifier. (#7822) 
* Deprecate `use_label_encoder` in XGBClassifier.

* We have removed the encoder, now prepare to remove the indicator.
 2022-04-21 13:14:02 +08:00
Jiaming Yuan 5815df4c46 Remove warning in 1.4. (#7815) 2022-04-20 01:19:09 +08:00
Jiaming YuanandPhilip Hyunsu Cho d0de954af2 v1.6.0 release note. [skip ci] (#7746) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-04-16 16:27:54 +08:00
Jiaming Yuan 5dea21273a Fix training continuation with categorical model. (#7810) 
* Make sure the task is initialized before construction of tree updater.

This is a quick fix meant to be backported to 1.6, for a full fix we should pass the model
param into tree updater by reference instead.
 2022-04-15 18:21:02 +08:00
Jiaming Yuan bcce17e688 Remove text loading in basic walk through demo. (#7753) 2022-04-01 00:59:42 +08:00
Jiaming YuanandPhilip Hyunsu Cho fd78af404b Drop support for deprecated CUDA architectures. (#7774) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-03-31 21:42:23 +08:00
Jiaming Yuan 02dd7b6913 Remove use of distutils. (#7770) 
distutils is deprecated and replaced by other stdlib constructs.
 2022-03-31 19:03:10 +08:00
Jiaming Yuan 6fa1afdffc Avoid compiler warning about comparison. (#7768) 2022-03-31 08:52:14 +08:00
Jiaming Yuan 522636cb52 Bump version. (#7769) 2022-03-31 06:33:22 +08:00
Jiaming Yuan 9150fdbd4d Support pandas nullable types. (#7760) 2022-03-30 08:51:52 +08:00
Jiaming Yuan d4796482b5 Fix failures on R hub and Win builder. (#7763) 
* Update date.
* Workaround amalgamation build with clang. (SimpleDMatrix instantiation)
* Workaround compiler error with driver push.
* Revert autoconf requirement.
* Fix model IO on 32-bit environment. (i386)
* Clarify the function name.
 2022-03-30 07:14:33 +08:00
Jiaming Yuan a50b84244e Cleanup configuration for constraints. (#7758) 2022-03-29 04:22:46 +08:00
Jiaming Yuan 3c9b04460a Move num_parallel_tree to model parameter. (#7751) 
The size of forest should be a property of model itself instead of a training
hyper-parameter.
 2022-03-29 02:32:42 +08:00
Jiaming Yuan 8b3ecfca25 Mitigate flaky tests. (#7749) 
* Skip non-increasing test with external memory when subsample is used.
* Increase bin numbers for boost from prediction test. This mitigates the effect of
  non-deterministic partitioning.
 2022-03-28 21:20:50 +08:00
Jiaming Yuan b3ba0e8708 Check cupy lazily. (#7752) 2022-03-26 06:09:58 +08:00
Jiaming Yuan af0cf88921 Workaround compiler error. (#7745) 2022-03-25 17:05:14 +08:00
Jiaming Yuan 64575591d8 Use context in SetInfo. (#7687) 
* Use the name `Context`.
* Pass a context object into `SetInfo`.
* Add context to proxy matrix.
* Add context to iterative DMatrix.

This is to remove the use of the default number of threads during `SetInfo` as a follow-up on
removing the global omp variable while preparing for CUDA stream semantic.  Currently, XGBoost
uses the legacy CUDA stream, we will gradually remove them in the future in favor of non-blocking streams.
 2022-03-24 22:16:26 +08:00
Jiaming Yuan 4d81c741e9 External memory support for hist (#7531) 
* Generate column matrix from gHistIndex.
* Avoid synchronization with the sparse page once the cache is written.
* Cleanups: Remove member variables/functions, change the update routine to look like approx and gpu_hist.
* Remove pruner.
 2022-03-22 00:13:20 +08:00
Jiaming Yuan cd55823112 Demo for using custom objective with multi-target regression. (#7736) 2022-03-20 17:44:25 +08:00
Jiaming Yuan 996cc705af Small cleanup to hist tree method. (#7735) 
* Remove special optimization using number of bins.
* Remove 1-based index for column sampling.
* Remove data layout.
* Unify update prediction cache.
 2022-03-20 03:44:55 +08:00
Jiaming Yuan 718472dbe2 [CI] Upgrade GitHub action Windows workers. (#7739) 2022-03-20 01:44:33 +08:00
Jiaming Yuan 9a400731d9 Replace device sync with stream sync. (#7737) 2022-03-19 23:22:23 +08:00
Jiaming Yuan da351621a1 [R] Fix parsing decision stump. (#7689) 2022-03-17 01:08:22 +08:00
Jiaming Yuan e78a38b837 Sort sparse page index when constructing DMatrix. (#7731) 2022-03-16 18:01:05 +08:00
Jiaming Yuan 98d6faefd6 Implement slope for Pseduo-Huber. (#7727) 
* Add objective and metric.
* Some refactoring for CPU/GPU dispatching using linalg module.
 2022-03-14 21:42:38 +08:00
Jiaming Yuan a62a3d991d [dask] prediction with categorical data. (#7708) 2022-03-10 00:21:48 +08:00
Jiaming Yuan 1d468e20a4 Optimize GPU evaluation function for categorical data. (#7705) 
* Use transform and cache.
 2022-02-28 17:46:29 +08:00
Jiaming Yuan 18a4af63aa Update documents and tests. (#7659) 
* Revise documents after recent refactoring and cat support.
* Add tests for behavior of max_depth and max_leaves.
 2022-02-26 03:57:47 +08:00
Jiaming Yuan 5eed2990ad Fix file descriptor leak. (#7704) 2022-02-25 17:49:33 +08:00
Jiaming Yuan 83a66b4994 Support categorical data for hist. (#7695) 
* Extract partitioner from hist.
* Implement categorical data support by passing the gradient index directly into the partitioner.
* Organize/update document.
* Remove code for negative hessian.
 2022-02-25 03:47:14 +08:00
Jiaming Yuan f60d95b0ba [R] Construct booster object in load.raw. (#7686) 2022-02-24 10:06:18 +08:00
Jiaming Yuan 6762c45494 Small cleanup to gradient index and hist. (#7668) 
* Code comments.
* Const accessor to index.
* Remove some weird variables in the `Index` class.
* Simplify the `MemStackAllocator`.
 2022-02-23 11:37:21 +08:00
Jiaming Yuan 49c74a5369 Update R package description. (#7691) 
* Change role.
* Remove cmake file when building the package.
 2022-02-23 08:36:37 +08:00
Jiaming Yuan c859764d29 [doc] Clarify that states in callbacks are mutated. (#7685) 
* Fix copy for cv.  This prevents inserting default callbacks into the input list.
* Clarify the behavior of callbacks in training/cv.
* Fix typos in doc.
 2022-02-22 11:45:00 +08:00
Jiaming Yuan 584bae1fc6 Fix document build with scikit-learn (#7684) 
* Require sphinx >= 4.4 for RTD.

* Install sklearn.
 2022-02-22 08:58:54 +08:00
Jiaming Yuan e56d1779e1 Require Python 3.7. (#7682) 
* Update setup.py.
 2022-02-21 05:46:48 +08:00
Jiaming Yuan 549f3bd781 Honor CPU counts from CFS. (#7654) 2022-02-21 03:13:26 +08:00
Jiaming Yuan 671b3c8d8e Fix typo. (#7680) 2022-02-20 03:42:47 +08:00
Jiaming Yuan b2341eab0c [R] Fix broken links. (#7670) 2022-02-20 00:55:48 +08:00
Jiaming Yuan f08c5dcb06 Cleanup some pylint errors. (#7667) 
* Cleanup some pylint errors.

* Cleanup pylint errors in rabit modules.
* Make data iter an abstract class and cleanup private access.
* Cleanup no-self-use for booster.
 2022-02-19 18:53:12 +08:00
Jiaming Yuan b76c5d54bf Define export symbols in callback module. (#7665) 2022-02-19 18:52:41 +08:00
Jiaming Yuan 7366d3b20c Ensure models with categorical splits don't use old binary format. (#7666) 2022-02-19 08:05:28 +08:00
Jiaming Yuan 14d61b0141 [doc] Update document for building from source. (#7664) 
- Mention standard install command for R package.
- Remove repeated "get source" step.
- Remove troubleshooting on Windows.  It's outdated considering VS 2022 is already out.
 2022-02-19 04:57:03 +08:00
Jiaming Yuan d625dc2047 Work around nvcc error. (#7673) 2022-02-19 01:41:46 +08:00
Jiaming Yuan 3877043d41 Avoid print for R package. (#7672) 2022-02-18 08:06:24 +08:00
Jiaming Yuan 711f7f3851 Avoid std::terminate for R package. (#7661) 
This is part of CRAN policies.
 2022-02-17 01:27:20 +08:00
Jiaming Yuan 12949c6b31 [R] Implement feature weights. (#7660) 2022-02-16 22:20:52 +08:00
Jiaming Yuan 93eebe8664 [doc] Fix broken link. [skip ci] (#7655) 2022-02-15 14:07:34 +08:00
Jiaming Yuan 0da7d872ef [doc] Update for prediction. (#7648) 2022-02-15 05:01:55 +08:00
Jiaming Yuan 0d0abe1845 Support optimal partitioning for GPU hist. (#7652) 
* Implement `MaxCategory` in quantile.
* Implement partition-based split for GPU evaluation.  Currently, it's based on the existing evaluation function.
* Extract an evaluator from GPU Hist to store the needed states.
* Added some CUDA stream/event utilities.
* Update document with references.
* Fixed a bug in approx evaluator where the number of data points is less than the number of categories.
 2022-02-15 03:03:12 +08:00
Jiaming Yuan 2369d55e9a Add tests for prediction cache. (#7650) 
* Extract the test from approx for other tree methods.
* Add note on how it works.
 2022-02-15 00:28:00 +08:00
Jiaming Yuan 5cd1f71b51 [dask] Improve configuration for port. (#7645) 
- Try port 0 to let the OS return the available port.
- Add port configuration.
 2022-02-14 21:34:34 +08:00
Jiaming Yuan b52c4e13b0 [dask] Fix empty partition with pandas input. (#7644) 
Empty partition is different from empty dataset.  For the former case, each worker has
non-empty dask collections, but each collection might contain empty partition.
 2022-02-14 19:35:51 +08:00