GitLab

* find all needed ports in each worker at once

* lint

* better naming

* use _HostWorkers in test

* Update c_api LGBM_SampleIndices() comment.

rand.Sample() now returns exactly given number of samples, thus the
comment should be fixed.

* Update include/LightGBM/c_api.h
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [python] support Dataset.get_data for Sequence input.

* Tweaks according to review comments.

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

* Add test cases.

* fix import order in test_basic.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* remove R docs

* bump deps

* Update parameter_generator.py

* Update parameter_generator.py

* Update parameter_generator.py

* Update parameter_generator.py

* Update parameter_generator.py

* Update parameter_generator.py

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* Update helpers/parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* additional hinting fixes

* Update parameter_generator.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>

This removes code duplication for creating sample indices.

* Fix undefined behavior with NaN input in CategoricalDecision()

* Always associate the right child with NaN inputs

* remove preinstalled possibly conflicting software from PATH in CI jobs

* preserve pandoc

* [ci] add CI job running rchk

* try commenting out more stuff

* ignore R internal error

* pipes

* remove PROTECT()

* try removing testthat

* revert temporary testing changes

* use absolute() on paths first

* migrate to pathlib in setup.py

* call predict on one row of data to determine output shape

* make DaskLGBMRanker predict method equal to the others

* remove extra drop_axis

* allow to pass some params as pathlib.Path objects

* fix lint

* improve indentation

[dask] Make output of feature contribution predictions for sparse matrices match those from sklearn estimators (fixes #3881) (#4378)

* test_classifier working

* adding tests

* docs

* tests

* revert unnecessary changes in tests

* test output type

* linting

* linting

* use from_delayed() instead

* docstring pycodestyle is happy with

* isort

* put pytest skips back

* respect sparse return type

* fix doc

* remove unnecessary dask_array_concatenate()

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

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

* update predict_proba() docstring

* remove unnecessary np.array()

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

* fix assertion

* fix test use of len()

* restore np.array() in tests

* use np.asarray() instead

* use toarray()

* remove empty functions in compat
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update test_sklearn.py

* Update test_basic.py

* Update dask.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update callback.py

* refactor file loading in C API test

* continue

* [python-package] convert string concatenation to f-strings in test_engine.py (fixes #4136)

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

* revert get_workflow_status changes
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* include distributed tests

* remove github action file

* try CI

* build shared library and fix linting error

* ignore files created for testing. add type hints and check with mypy. include docstrings

* lint

* use pre_partition and write separate model files. remove mypy

* update docs

* remove ci. lower rtol. pass num_machines in config

* write predict.conf in the predict method. more robust port setup. use subprocess.run and check returncode

* add paths to tests and binary. remove lgb dependency. update .igtignore.

* lint

* allow to pass executable dir as argument to pytest

* pass execfile to pytest instead of execdir

* add suggestions

* use os.path and add type hint to predict_config

* Update tests/distributed/_test_distributed.py
Co-authored-by: James Lamb <jaylamb20@gmail.com>