Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6,252 Commits 32 Branches 59 Tags
c400fa1e8d8399ec867e0a18ffec7cd7df00de83
Commit Graph

1409 Commits

Author SHA1 Message Date
Jiaming Yuan
c400fa1e8d Predictor for vector leaf. (#8898) 2023-03-14 19:07:10 +08:00
Jiaming Yuan
8be6095ece Implement NDCG cache. (#8893) 2023-03-13 22:16:31 +08:00
Jiaming Yuan
9bade7203a Remove public access to tree model param. (#8902) 
* Make tree model param a private member.
* Number of features and targets are immutable after construction.

This is to reduce the number of places where we can run configuration.
 2023-03-13 20:55:10 +08:00
Jiaming Yuan
5ba3509dd3 Define multi expand entry. (#8895) 2023-03-13 19:31:05 +08:00
Jiaming Yuan
36a7396658 Replace dmlc any with std any. (#8892) 2023-03-11 06:11:04 +08:00
Jiaming Yuan
2aa838c75e Define multi-strategy parameter. (#8890) 2023-03-11 02:58:01 +08:00
Jiaming Yuan
6deaec8027 Pass obj info by reference instead of by value. (#8889) 
- Pass obj info into tree updater as const pointer.

This way we don't have to initialize the learner model param before configuring gbm, hence
breaking up the dependency of configurations.
 2023-03-11 01:38:28 +08:00
Jiaming Yuan
5feee8d4a9 Define core multi-target regression tree structure. (#8884) 
- Define a new tree struct embedded in the `RegTree`.
- Provide dispatching functions in `RegTree`.
- Fix some c++-17 warnings about the use of nodiscard (currently we disable the warning on
  the CI).
- Use uint32_t instead of size_t for `bst_target_t` as it has a defined size and can be used
  as part of dmlc parameter.
- Hide the `Segment` struct inside the categorical split matrix.
 2023-03-09 19:03:06 +08:00
Jiaming Yuan
46dfcc7d22 Define a new ranking parameter. (#8887) 2023-03-09 17:46:24 +08:00
Jiaming Yuan
f236640427 Support F order for the tensor type. (#8872) 
- Add F order support for tensor and view.
- Use parameter pack for automatic type cast. (avoid excessive static cast for shape).
 2023-03-08 03:27:49 +08:00
Jiaming Yuan
cad7401783 Disable gcc parallel extension if openmp is not available. (#8871) 
`<parallel/algorithm>` internally includes the <omp.h> header, which leads to an error
when openmp is not available.
 2023-03-06 22:51:06 +08:00
Jiaming Yuan
228a46e8ad Support learning rate for zero-hessian objectives. (#8866) 2023-03-06 20:33:28 +08:00
Jiaming Yuan
4d665b3fb0 Restore clang tidy test. (#8861) 2023-03-03 13:47:04 -08:00
Rong Ou
2dc22e7aad Take advantage of C++17 features (#8858) 
---------

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2023-03-04 00:24:13 +08:00
Rory Mitchell
69a50248b7 Fix scope of feature set pointers (#8850) 
---------

Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2023-03-02 12:37:14 +08:00
Rong Ou
7cbaee9916 Support column split in approx tree method (#8847) 2023-03-02 03:59:07 +08:00
Jiaming Yuan
d54ef56f6f Fix cache with gc (#8851) 
- Make DMatrixCache thread-safe.
- Remove the use of thread-local memory.
 2023-03-01 00:39:06 +08:00
Rong Ou
d9688f93c7 Support column-split in row partitioner (#8828) 2023-02-26 04:43:35 +08:00
Mauro Leggieri
90c0633a28 Fixes compilation errors on MSVC x86 targets (#8823) 2023-02-26 03:20:28 +08:00
Rong Ou
a65ad0bd9c Support column split in histogram builder (#8811) 2023-02-17 22:37:01 +08:00
Jiaming Yuan
c0afdb6786 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.
 2023-02-16 04:20:34 +08:00
Jiaming Yuan
cce4af4acf Initial support for quantile loss. (#8750) 
- Add support for Python.
- Add objective.
 2023-02-16 02:30:18 +08:00
Jiaming Yuan
282b1729da Specify the number of threads for parallel sort. (#8735) 
* Specify the number of threads for parallel sort.

- Pass context object into argsort.
- Replace macros with inline functions.
 2023-02-16 00:20:19 +08:00
Jiaming Yuan
c7c485d052 Extract fit intercept. (#8793) 2023-02-15 22:41:31 +08:00
Jiaming Yuan
594371e35b Fix CPP lint. (#8807) 2023-02-15 20:16:35 +08:00
Rong Ou
74572b5d45 Add convenience method for allgather (#8804) 2023-02-15 11:37:11 +08:00
Jiaming Yuan
52d0230b58 Fix merge conflict. (#8791) 2023-02-13 23:43:42 +08:00
Jiaming Yuan
81b2ee1153 Pass DMatrix into metric for caching. (#8790) 2023-02-13 22:15:05 +08:00
Jiaming Yuan
31d3ec07af Extract device algorithms. (#8789) 2023-02-13 20:53:53 +08:00
Jiaming Yuan
457f704e3d Add quantile metric. (#8761) 2023-02-13 19:07:40 +08:00
Jiaming Yuan
d11a0044cf Generalize prediction cache. (#8783) 
* Extract most of the functionality into `DMatrixCache`.
* Move API entry to independent file to reduce dependency on `predictor.h` file.
* Add test.

---------

Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2023-02-13 12:36:43 +08:00
Jiaming Yuan
17b709acb9 Rename ranking utils to threading utils. (#8785) 2023-02-12 05:41:18 +08:00
Jiaming Yuan
70c9b885ef Extract floating point rounding routines. (#8771) 2023-02-12 04:26:41 +08:00
Jiaming Yuan
8a16944664 Fix ranking with quantile dmatrix and group weight. (#8762) 2023-02-10 20:32:35 +08:00
Jiaming Yuan
199c421d60 Send default configuration from metric to objective. (#8760) 2023-02-09 20:18:07 +08:00
Jiaming Yuan
5f76edd296 Extract make metric name from ranking metric. (#8768) 
- Extract the metric parsing routine from ranking.
- Add a test.
- Accept null for string view.
 2023-02-09 18:30:21 +08:00
Rong Ou
cbf98cb9c6 Add Allgather to collective communicator (#8765) 
* Add Allgather to collective communicator
 2023-02-09 11:31:22 +08:00
Jiaming Yuan
48cefa012e Support multiple alphas for segmented quantile. (#8758) 2023-02-07 17:17:59 +08:00
Jiaming Yuan
28bb01aa22 Extract optional weight. (#8747) 
- Extract optional weight from coommon.h to reduce dependency on this header.
- Add test.
 2023-02-07 03:11:53 +08:00
Jiaming Yuan
a2e433a089 Fix empty DMatrix with categorical features. (#8739) 2023-02-07 00:40:11 +08:00
Rory Mitchell
7214a45e83 Fix different number of features in gpu_hist evaluator. (#8754) 2023-02-06 23:15:16 +08:00
Rong Ou
66191e9926 Support cpu quantile sketch with column-wise data split (#8742) 2023-02-05 14:26:24 +08:00
Jiaming Yuan
c1786849e3 Use array interface for CSC matrix. (#8672) 
* Use array interface for CSC matrix.

Use array interface for CSC matrix and align the interface with CSR and dense.

- Fix nthread issue in the R package DMatrix.
- Unify the behavior of handling `missing` with other inputs.
- Unify the behavior of handling `missing` around R, Python, Java, and Scala DMatrix.
- Expose `num_non_missing` to the JVM interface.
- Deprecate old CSR and CSC constructors.
 2023-02-05 01:59:46 +08:00
Jiaming Yuan
0e61ba57d6 Fix GPU L1 error. (#8749) 2023-02-04 03:02:00 +08:00
Jiaming Yuan
3760cede0f Consistent use of context to specify number of threads. (#8733) 
- Use context in all tests.
- Use context in R.
- Use context in C API DMatrix initialization. (0 threads is used as dft).
 2023-01-30 15:25:31 +08:00
Jiaming Yuan
21a28f2cc5 Small refactor for hist builder. (#8698) 
- Use span instead of vector as parameter. No perf change as the builder work on pointer.
- Use const pointer for reg tree.
 2023-01-30 14:06:41 +08:00
Jiaming Yuan
34eee56256 Fix compiler warnings. (#8703) 
Fix warnings about signed/unsigned comparisons.
 2023-01-21 15:16:23 +08:00
Jiaming Yuan
e49e0998c0 Extract CPU sampling routines. (#8697) 2023-01-19 23:28:18 +08:00
Jiaming Yuan
7a068af1a3 Workaround CUDA warning. (#8696) 2023-01-19 09:16:08 +08:00
Jiaming Yuan
4416452f94 Return single thread from context when called inside omp region. (#8693) 2023-01-18 09:23:37 +08:00