Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6,355 Commits 32 Branches 59 Tags
Commit Graph

799 Commits

Author SHA1 Message Date
Jiaming Yuan
7eba285a1e
Support sklearn cross validation for ranker. (#8859) 
* Support sklearn cross validation for ranker.

- Add a convention for X to include a special `qid` column.

sklearn utilities consider only `X`, `y` and `sample_weight` for supervised learning
algorithms, but we need an additional qid array for ranking.

It's important to be able to support the cross validation function in sklearn since all
other tuning functions like grid search are based on cross validation.
 2023-03-07 00:22:08 +08:00
Jiaming Yuan
6a892ce281
Specify src path for isort. (#8867) 2023-03-06 17:30:27 +08:00
mzzhang95
6cef9a08e9
[pyspark] Update eval_metric validation to support list of strings (#8826) 2023-03-02 08:24:12 +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
WeichenXu
f27a7258c6
Fix feature types param (#8772) 
Signed-off-by: Weichen Xu <weichen.xu@databricks.com>
 2023-02-14 02:16:42 +08:00
Jiaming Yuan
457f704e3d
Add quantile metric. (#8761) 2023-02-13 19:07:40 +08:00
Jiaming Yuan
225b3158f6
Support custom metric in sklearn ranker. (#8786) 2023-02-12 13:14:07 +08:00
Jiaming Yuan
8a16944664
Fix ranking with quantile dmatrix and group weight. (#8762) 2023-02-10 20:32:35 +08:00
Jiaming Yuan
c4802bfcd0
Cleanup booster param types. (#8756) 2023-02-07 15:52:19 +08:00
Jiaming Yuan
0f37a01dd9
Require black formatter for the python package. (#8748) 2023-02-07 01:53:33 +08:00
Jiaming Yuan
a2e433a089
Fix empty DMatrix with categorical features. (#8739) 2023-02-07 00:40:11 +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
BenEfrati
213b5602d9
Add sample_weight to eval_metric (#8706) 2023-02-05 00:06:38 +08:00
Jiaming Yuan
0e61ba57d6
Fix GPU L1 error. (#8749) 2023-02-04 03:02:00 +08:00
Jiaming Yuan
1325ba9251
Support primitive types of pyarrow-backed pandas dataframe. (#8653) 
Categorical data (dictionary) is not supported at the moment.
 2023-01-30 17:53:29 +08:00
Jiaming Yuan
9fb12b20a4
Cleanup the callback module. (#8702) 
- Cleanup pylint markers.
- run formatter.
- Update examples of using callback.
 2023-01-22 00:13:49 +08:00
James Lamb
6933240837
[python-package] remove unused functions in xgboost.data (#8695) 2023-01-19 08:02:54 +08:00
Jiaming Yuan
31b9cbab3d
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`.
 2023-01-18 08:12:13 +08:00
Jiaming Yuan
175986b739
[doc] Add missing document for pyspark ranker. [skip ci] (#8692) 2023-01-18 07:52:18 +08:00
Jiaming Yuan
247946a875
Cache transformed in QuantileDMatrix for efficiency. (#8666) 2023-01-17 06:02:40 +08:00
Jiaming Yuan
d6018eb4b9
Remove all use of DeviceQuantileDMatrix. (#8665) 2023-01-17 00:04:10 +08:00
Bobby Wang
72ec0c5484
[pyspark] support pred_contribs (#8633) 2023-01-11 16:51:12 +08:00
Jiaming Yuan
cfa994d57f
Multi-target support for L1 error. (#8652) 
- Add matrix support to the median function.
- Iterate through each target for quantile computation.
 2023-01-11 05:51:14 +08:00
Jiaming Yuan
badeff1d74
Init estimation for regression. (#8272) 2023-01-11 02:04:56 +08:00
Jiaming Yuan
1b58d81315
[doc] Document Python inputs. (#8643) 2023-01-10 15:39:32 +08:00
Jiaming Yuan
e68a152d9e
Do not return internal value for get_params. (#8634) 2023-01-05 17:48:26 +08:00
Bobby Wang
d3ad0524e7
[pyspark] Re-work _fit function (#8630) 2023-01-04 18:21:57 +08:00
Rong Ou
3ceeb8c61c
Add data split mode to DMatrix MetaInfo (#8568) 2022-12-25 20:37:37 +08:00
Rong Ou
77b069c25d
Support bitwise allreduce operations in the communicator (#8623) 2022-12-25 06:40:05 +08:00
Jiaming Yuan
c430ae52f3
Fix mypy errors with the latest numpy. (#8617) 2022-12-21 01:42:05 -08:00
Jiaming Yuan
f6effa1734
Support Series and Python primitives in inplace_predict and QDM (#8547) 2022-12-17 00:15:15 +08:00
Jiaming Yuan
001e663d42
Set enable_categorical to True in predict. (#8592) 2022-12-15 05:27:06 +08:00
James Lamb
06ea6c7e79
[python] remove unnecessary conversions between data structures (#8546) 
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
 2022-12-14 18:32:02 +08:00
Jiaming Yuan
40343c8ee1
Test dask demos. (#8557) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-12-13 18:37:31 +08:00
Jiaming Yuan
deb3edf562
Support list and tuple for QDM. (#8542) 2022-12-10 01:14:44 +08:00
Bobby Wang
40a1a2ffa8
[pyspark] check use_qdm across all the workers (#8496) 2022-12-08 18:09:17 +08:00
Gianfrancesco Angelini
5540019373
feat(py, plot_importance): + values_format as arg (#8540) 2022-12-08 00:47:28 +08:00
Matthew Rocklin
b7ffdcdbb9
Properly await async method client.wait_for_workers (#8558) 
* Properly await async method client.wait_for_workers

* ignore mypy error.

Co-authored-by: jiamingy <jm.yuan@outlook.com>
 2022-12-07 21:49:30 +08:00
Bobby Wang
8e41ad24f5
[pyspark] sort qid for SparkRanker (#8497) 
* [pyspark] sort qid for SparkRandker

* resolve comments
 2022-12-01 16:40:35 -08:00
Jiaming Yuan
d666ba775e
Support all pandas nullable integer types. (#8480) 
- Enumerate all pandas integer types.
- Tests for `None`, `nan`, and `pd.NA`
 2022-11-28 22:38:16 +08:00
Jiaming Yuan
f2209c1fe4
Don't shuffle columns in categorical tests. (#8446) 2022-11-28 20:28:06 +08:00
WeichenXu
67ea1c3435
[pyspark] Make QDM optional based on cuDF check (#8471) 2022-11-27 14:58:54 +08:00
Jiaming Yuan
8f97c92541
Support half type for pandas. (#8481) 2022-11-24 12:47:40 +08:00
Jiaming Yuan
e07245f110
Take datatable as row major input. (#8472) 
* Take datatable as row major input.

Try to avoid a transform with dense table.
 2022-11-24 09:20:13 +08:00
Jiaming Yuan
9dd8d70f0e
Fix mypy errors. (#8444) 2022-11-09 13:19:11 +08:00
Jiaming Yuan
0d3da9869c
Require isort on all Python files. (#8420) 2022-11-08 12:59:06 +08:00
James Lamb
bf8de227a9
[CI] remove unused import in python tests (#8409) 2022-11-03 22:27:25 +08:00
Rong Ou
99fa8dad2d
Add back xgboost.rabit for backwards compatibility (#8408) 
* Add back xgboost.rabit for backwards compatibility

* fix my errors

* Fix lint

* Use FutureWarning

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
 2022-11-01 21:47:41 -07:00
Jiaming Yuan
a408c34558
Update JSON parser demo with categorical feature. (#8401) 
- Parse categorical features in the Python example.
- Add tests.
- Update document.
 2022-10-28 20:57:43 +08:00
Jiaming Yuan
cfd2a9f872
Extract dask and spark test into distributed test. (#8395) 
- Move test files.
- Run spark and dask separately to prevent conflicts.
- Gather common code into the testing module.
 2022-10-28 16:24:32 +08:00