f8ec57b8eb
[python-package] Correctly recognize LGBMClassifier(num_class=2, objective="multiclass") as multiclass classification ( #6524 )
2024-07-16 16:05:00 -05:00
2bc3ab86b7
[python-package] respect 'verbose' setting when using custom objective function ( fixes #6014 ) ( #6428 )
2024-07-12 11:21:39 -07:00
cd4459a1db
[test][python] fix pandas DeprecationWarning about `is_sparse` function ( #6532 )
2024-07-10 09:20:27 -07:00
6662d557d5
[tests][python] fix nan in pandas bool column ( #6530 )
2024-07-09 21:54:59 -07:00
3e6156a120
[ci] prevent `lgb.model` and `lgb.pkl` files being left behind after testing ( #6518 )
2024-07-04 22:06:27 -05:00
f811c82708
[python-package] Add `feature_names_in_` attribute for scikit-learn estimators ( fixes #6279 ) ( #6310 )
2024-07-03 18:13:47 -05:00
1e7ebc5194
[python-package] Add support for NumPy 2.0, test against nightly versions of dependencies ( fixes #6454 ) ( #6467 )
2024-06-12 23:22:22 -05:00
8579d5e34f
[python-package] clarify max_depth warning and limit when it is emitted ( #6402 )
2024-06-04 19:46:07 -05:00
e0cda880fc
[python-package] remove uses of deprecated NumPy random number generation APIs, require 'numpy>=1.17.0' ( #6468 )
2024-06-03 20:17:40 -05:00
a70e832782
[R-package] [python-package] deprecate Dataset arguments to cv() and train() ( #6446 )
2024-05-10 19:26:39 -05:00
9f5fbb6479
[python-package] Allow to pass early stopping min delta in params ( #6274 )
...
* [python-package] Allow to pass early stopping min delta in params
* Fix test
* Add separate test
* Fix
* Add to cpp config
* Adjust test
* Adjust test
* Debug
* Revert
* Apply suggestions from code review
---------
Co-authored-by: James Lamb <jaylamb20@gmail.com>
2024-05-01 19:57:55 +02:00
92a8741b9c
[ci] Skip Arrow tests on AppVeyor, use Intel macOS runners, upgrade to XCode 14.3 on macOS jobs, disable MacOS MPI jobs ( #6425 )
2024-04-29 10:36:53 -05:00
1871350515
remove unnecessary omp single that cause deadlock ( fixes #6273 ) ( #6394 )
2024-04-22 21:52:02 -05:00
1ce2357193
[python-package] allow use of early_stopping_round<=0 to turn off early stopping ( fixes #6401 ) ( #6406 )
2024-04-19 23:31:30 -05:00
4332b9d515
[docs] [ci] encourage use of `cmake --build` ( #6368 )
2024-03-20 12:21:46 -05:00
631e0a2a7b
[ci] prevent trailing whitespace, ensure files end with newline ( #6373 )
2024-03-18 23:24:14 -05:00
faba817787
[python-package] Allow to pass Arrow table with boolean columns to dataset ( #6353 )
2024-03-18 21:03:26 -05:00
0a3e1a550c
Fix single row prediction performance in a multi-threaded environment ( #6024 )
2024-03-18 16:46:48 -05:00
cb4972eeef
[ci] Fix R 3.6 tests, dask tests, compatibility with dask>=2024.3.1 ( #6357 )
2024-03-17 21:26:26 -05:00
6f19edde52
[ci] [python-package] enforce 'pylint' checks ( fixes #4308 ) ( #6334 )
2024-02-23 23:07:03 -06:00
776c5c3c49
[c++][fix] Support Quantized Training with Categorical Features on CPU ( #6301 )
...
* support quantized training with categorical features on cpu
* remove white spaces
* add tests for quantized training with categorical features
* skip tests for cuda version
* fix cases when only 1 data block in row-wise quantized histogram construction with 8 inner bits
* remove useless capture
* fix compilation warnings
revert useless changes
* revert useless change
* separate functions in feature histogram into cpp file
* add feature_histogram.o in Makevars
2024-02-23 13:41:21 +08:00
1b792e7166
[ci] [python-package] enable ruff-format on tests and examples ( #6317 )
2024-02-21 12:15:38 -06:00
2dfb9a4047
[python-package] Allow to pass Arrow table for prediction ( #6168 )
2023-12-13 21:35:46 -06:00
522f0f07b0
[python-package] Add tests for passing Arrow arrays with empty chunks ( #6210 )
2023-12-10 11:29:04 -06:00
1548b42bac
[R-package] [c++] add tighter multithreading control, avoid global OpenMP side effects ( fixes #4705 , fixes #5102 ) ( #6226 )
2023-12-07 17:03:16 -06:00
e797985227
[python-package] take shallow copy of dataframe in predict ( fixes #6195 ) ( #6218 )
2023-12-07 08:54:18 -06:00
d84582b746
Fix null handling for Arrow data ( #6227 )
2023-12-06 10:18:28 -06:00
f5b6bd60d9
[python-package] Allow to pass Arrow table and array as init scores ( #6167 )
2023-12-04 13:26:55 -06:00
516bde9501
[python-package] Allow to pass Arrow array as groups ( #6166 )
2023-11-22 15:50:31 -06:00
deb70773b8
[python-package] Allow to pass Arrow array as weights ( #6164 )
2023-11-13 12:03:58 -06:00
501e6e62fe
[python-package] Accept numpy generators as `random_state` ( #6174 )
2023-11-08 22:56:08 -06:00
aeafccfbfb
[python-package] fix access to Dataset metadata in scikit-learn custom metrics and objectives ( #6108 )
2023-11-07 15:01:52 -06:00
b7f6311f27
[python-package] Allow to pass Arrow array as labels ( #6163 )
2023-11-07 12:14:09 -06:00
3405ee82b9
[python-package] Allow to pass Arrow table as training data ( #6034 )
2023-11-01 08:21:48 -05:00
fcf76bceb9
[ci] resolve warning in tests ( #6154 )
2023-10-26 12:58:21 -05:00
b793cd821c
ignore unknown parameters when loading from model file ( #6126 )
2023-10-06 22:43:40 -05:00
8f577de01f
fix compiler warnings for CPP tests ( #6124 )
2023-10-06 12:11:53 -05:00
d45dca702e
[python-package] reorganize early stopping callback ( #6114 )
2023-10-05 12:45:51 -05:00
ab1eaa832d
[python-package] [R-package] include more params in model text representation ( fixes #6010 ) ( #6077 )
2023-09-13 17:35:38 -05:00
0b3d9da2eb
[python-package] mark EarlyStopException as part of public API ( #6095 )
2023-09-13 13:47:31 -05:00
a92bf3742b
[fix] fix quantized training ( fixes #5982 ) ( fixes #5994 ) ( #6092 )
...
* fix leaf splits update after split in quantized training
* fix preparation ordered gradients for quantized training
* remove force_row_wise in distributed test for quantized training
* Update src/treelearner/leaf_splits.hpp
---------
Co-authored-by: James Lamb <jaylamb20@gmail.com>
2023-09-13 01:06:20 +08:00
5e592fe6ff
[python-package] Fix misdetected objective after multiple calls to `LGBMClassifier.fit` ( #6002 )
2023-09-11 21:53:36 -05:00
ee51120118
[python-package] simplify processing of pandas data ( #6066 )
2023-09-06 08:14:20 -05:00
7e34d23c05
Treat position bias via GAM in LambdaMART ( #5929 )
...
* Update dataset.h
* Update metadata.cpp
* Update rank_objective.hpp
* Update metadata.cpp
* Update rank_objective.hpp
* Update metadata.cpp
* Update dataset.h
* Update rank_objective.hpp
* Update metadata.cpp
* Update test_engine.py
* Update test_engine.py
* Add files via upload
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update _rank.train.position
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update _rank.train.position
* Update _rank.train.position
* Update test_engine.py
* Update _rank.train.position
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update the position of import statement
* Update rank_objective.hpp
* Update config.h
* Update config_auto.cpp
* Update rank_objective.hpp
* Update rank_objective.hpp
* update documentation
* remove extra blank line
* Update src/io/metadata.cpp
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update src/io/metadata.cpp
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* remove _rank.train.position
* add position in python API
* fix set_positions in basic.py
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* Update Advanced-Topics.rst
* remove List from _LGBM_PositionType
* move new position parameter to the last in Dataset constructor
* add position_filename as a parameter
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update Advanced-Topics.rst
* Update src/objective/rank_objective.hpp
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update src/io/metadata.cpp
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update metadata.cpp
* Update python-package/lightgbm/basic.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update python-package/lightgbm/basic.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update python-package/lightgbm/basic.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update python-package/lightgbm/basic.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update src/io/metadata.cpp
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* more infomrative fatal message
address more comments
* update documentation for more flexible position specification
* fix SetPosition
add tests for get_position and set_position
* remove position_filename
* remove useless changes
* Update python-package/lightgbm/basic.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* remove useless files
* move position file when position set in Dataset
* warn when positions are overwritten
* skip ranking with position test in cuda
* split test case
* remove useless import
* Update test_engine.py
* Update test_engine.py
* Update test_engine.py
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* Update Parameters.rst
* Update rank_objective.hpp
* Update config.h
* update config_auto.cppp
* Update docs/Advanced-Topics.rst
Co-authored-by: James Lamb <jaylamb20@gmail.com>
* fix randomness in test case for gpu
---------
Co-authored-by: shiyu1994 <shiyu_k1994@qq.com>
Co-authored-by: James Lamb <jaylamb20@gmail.com>
2023-09-04 17:05:46 +08:00
c60bc739b0
[ci] [python-package] add more linting checks ( #6049 )
2023-08-19 07:01:28 -05:00
37c3d3fb80
fix percentile computation for regression objectives ( #5848 )
...
* Update regression_objective.hpp
* Update regression_objective.hpp
Maybe still need a (1.0 - alpha)
* fix position in percentile calculation
* fix regression metric threshold for l1
---------
Co-authored-by: shiyu1994 <shiyu_k1994@qq.com>
2023-08-15 10:11:39 +08:00
b8cc873882
[python-package] make Booster and Dataset 'handle' attributes private ( fixes #5313 ) ( #5947 )
2023-06-29 08:02:58 -05:00
ac57d5a45a
[dask] hold ports until training ( #5890 )
2023-06-20 10:08:46 -06:00
15e3aecce5
[python-package] adding max_category_values parameter to create_tree_digraph method ( fixes #5687 ) ( #5818 )
2023-06-09 21:14:13 -05:00
d47006f4d0
[ci] [python-package] use ruff, enforce flake8-bugbear and flake8-comprehensions checks ( #5871 )
2023-05-15 22:49:31 -05:00
RektPunk
James Lamb
Nikita Titov
Nick Miller
Oliver Borchert
Yohei Kishimoto
Darcie Delzell
Thomas B
shiyu1994
José Morales
david-cortes
Pavel Metrikov
Zachary Huang
Mohamed Ziada