GitLab

* initial work for boosting and evaluation with CUDA

* fix compatibility with CPU code

* fix creating objective without USE_CUDA_EXP

* fix static analysis errors

* fix static analysis errors

* Improve readme on experimental cuda usage

* Revert header style change

* [python-package] add more type hints on Booster

* add return to Callable

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* docs: typo in documentation of parameter init_model

* docs: actually fix typo and build docs
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Adjust LGBM_DatasetCreateFromSampledColumn to handle distributed data better

* linting fix

* switch to 1 API with breaking change

* Fix pything native call

* more python test fixes

[R-package] raise an informative error when custom objective produces incorrect output (fixes #5323) (#5329)

* [python-package] add type hints on Booster.update()

* add __boost()

add validate_features to refit

* [python-package] remove inner function _construct_dataset() in LGBMModel.fit()

* switch order

* [python-package] add type hints on predict() methods

* formatting

* allow custom weighing in sklearn api

* add suggestions from review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* switch to single prediction type argument

* linter

* missing piece of code

* comments

* linter

* fix test

* revert incorrect 'fix'

* fix failing test

* fix test again

* modify recently introduced tests after changes here

* rename prediction types

* rebase

* restore tests for prediction type in params

* Update R-package/tests/testthat/test_Predictor.R
Co-authored-by: James Lamb <jaylamb20@gmail.com>
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* check feature names and order in predict with dataframe

* slice df in predict to remove the target

* scramble features

* handle int column names

* only change column order when needed

* include validate_features param in booster and sklearn estimators

* document validate_features argument

* use all_close in preds checks and check for assertion error to compare different arrays

* perform remapping and checks in cpp

* remove extra logs

* fixes

* revert cpp

* proposal

* remove extra arg

* lint

* restore _data_from_pandas arguments

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* move data conversion to Predictor.predict

* use Vector2Ptr
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

Update basic.py

* Update utils.R

* Update lint_r_code.R

* Update lint_r_code.R

* Update lint_r_code.R

* Update lint_r_code.R

* [python-package] add more type hints in engine.py

* Update python-package/lightgbm/engine.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

[python-package] add more hints on Booster

[ci] [R-package] Add paste_linter (#5303)

* Add duplicate_argument_linter

* Add duplicate_argument_linter

* added acomment for duplicate argument

* Update R-package/tests/testthat/test_utils.R
Co-authored-by: José Morales <jmoralz92@gmail.com>
Co-authored-by: José Morales <jmoralz92@gmail.com>

* [ci] Fix any_is_na linter key

* [ci] Add several linter checks (#5303)

Includes following linters

- any_duplicated_linter()
- inner_combine_linter()
- literal_coercion_linter()
- missing_argument_linter()
- numeric_leading_zero_linter()
- outer_negation_linter()
- package_hooks_linter()
- sprintf_linter()
- regex_subset_linter()
- unreachable_code_linter()
- vector_logic_linter()

These required no changes to the code base

Simplify params handling in `predict()` with `_choose_param_value()`

* [python] preserve None in _choose_param_value()

* Update python-package/lightgbm/basic.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Install PoCL in Docker for old Ubuntu 14.04

* Update .vsts-ci.yml

* Try with Clang

* switch back to gcc

* Switch from dev container to stable one

* Update make-r-def.R

* Update make-r-def.R

* Update build_r.R

* Update utils.R

* Update install.libs.R

* [python-package] add type hints on cv()

* remove inadvertent changes

* Apply suggestions from code review
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

[python-package] Use scikit-learn interpretation of negative `n_jobs` and change default to number of cores (#5105)

* use joblib formula for negative n_jobs

* correction for n_jobs calculation

* use more robust cpu_count from joblib

* change default n_jobs to number of cores

* fix detection of num_threads under parameters

* better handling of n_jobs at prediction time

* fix incorrect usage of list.pop

* correct pop/remove yet again

* Update python-package/lightgbm/sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update tests/python_package_test/test_sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update tests/python_package_test/test_sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* add comments clarifying negative n_jobs

* fix CI (code taken from PR comment)

* change default to n_jobs=None in dask interface

* corrections for handling of n_jobs

* linter

* corrections for predict-time n_jobs

* linter

* add more comments about n_jobs values

* linter

* more corrections

* linter

* linter

* linter

* Update python-package/lightgbm/compat.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update python-package/lightgbm/sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update python-package/lightgbm/sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update python-package/lightgbm/sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update python-package/lightgbm/sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* workaround for passing test about outputs with multiple threads

* Update tests/python_package_test/test_sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update tests/python_package_test/test_sklearn.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* add predcontrib for sparse inputs

* register newly-added function

* comments

* correct wrong types in test

* forcibly take transpose function from Matrix

* keep row names, test comparison to dense inputs

* workaround for passing test while PR for row names is not merged

* Update R-package/R/lgb.Predictor.R
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update R-package/R/lgb.Predictor.R
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update R-package/R/lgb.Predictor.R
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* proper handling of integer overflow

* add test for CSR contrib row names

* add more tests for predict(<sparse>, predcontrib=TRUE)

* make linter happy

* linter

* linter

* check error messages for bad input shapes

* fix regex

* hard-coded number of columns in regex for tests
Co-authored-by: James Lamb <jaylamb20@gmail.com>