GitLab

Clear split info buffer in cost efficient gradient boosting before every iteration (fix partially #3679) (#5164)

* clear split info buffer in cegb_ before every iteration

* check nullable of cegb_ in serial_tree_learner.cpp

* add a test case for checking the split buffer in CEGB

* swith to Threading::For instead of raw OpenMP

* apply review suggestions

* apply review comments

* remove device cpu

* Update test_basic.py

* Address review comment

* [python] add more type hints in basic.py

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

[R-package] upgrade docs to roxygen2==7.2.0

* dont copy dataframe on rename

* test with feature_name and 'auto'

Update GPU-Windows.rst

check nullable of bin mappers

* Update tree.cpp

* Update common.h

* Update common.h

* Update basic.py

* Update test_engine.py

* Add return type annotation

* check existence of inet_pton for win32 in CMakeLists.txt (fix #5019)

* remove extra spaces

* add check for inet_pton in configure.win

* Update CMakeLists.txt
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update src/network/socket_wrapper.hpp
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update src/network/socket_wrapper.hpp
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update CMakeLists.txt
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update R-package/configure.win
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

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

* fix comments in CMakeLists.txt

* include check for WIN64

* remove WIN64 flag checks

* fix #ifdef

* Update R-package/configure.win
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: James Lamb <jaylamb20@gmail.com>

[python-package] make a shallow copy when replacing categorical features with codes (fixes #4596) (#5225)

* [R-package] stop automatically calculating eval metrics on training data in lightgbm()

* update docs

* update docs

* roxygen cares about line breaks

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* allow using feature names when retrieving number of bins

* unname vector

* use default feature names when not defined

* lint

* apply suggestions

* remove extra comma

* add test with categorical feature

* make feature names sync more transparent

minor improvements on weight demo

* [R-package] remove lgb.unloader()

* update version reference

* Update dataset_loader.cpp

* Update gbdt.h

* Update regression_objective.hpp

* Update linker_topo.cpp

* Update xentropy_objective.hpp

* Update regression_objective.hpp

* investigate inf test failure

* avoid overflow in regression objective

* remove `test_inf_handle` test
Co-authored-by: Guolin Ke <guolin.ke@outlook.com>

* [R-package] ensure that callbacks respect verbosity from params

* added tests on lgb.train()

* add lgb.cv() tests

* add fix for lightgbm()

* fix early stopping tests

* start reduucing duplication in tests

* more simplification

* more simplification

* reduce more code duplication

* centrallize more details

* reuse more logic across tests

* more code re-use

* started condensing lgb.cv test code

* further condense lgb.cv() tests

* re-use more code lgb.cv() tests

* more condensing

* Update test_windows.ps1

* Update .appveyor.yml

* Update test_windows.ps1

* Update test_windows.ps1

* Update test_windows.ps1

* Update test_windows.ps1

* Update test_windows.ps1

* Update .appveyor.yml

* Update test_windows.ps1

* Update .appveyor.yml

* Update .appveyor.yml

* [ci] fix git checkout for comment-triggered CI jobs

* set locale prior to installing packages

* comment out cuda

* change strategy for setting locale

* comment out R jobs

* comment out more CI jobs

* update locales before installing other packages

* remove unnecessary packages

* add libc6 back

* restore libicu and libssl

* Revert "comment out more CI jobs"

This reverts commit 8fd92144ad1dafc33ae699d7c3e159d8846e41b2.

* uncomment CI jobs

* revert more changes

* more reverting

* remove r_package.yml from diff

* Update dataset_loader.cpp

* Update config.cpp

* Update application.cpp

* Update linkers_socket.cpp

[R-package] allow use of `categorical_features` in Dataset when raw data does not have column names (fixes #4374) (#5184)

* check for number of columns if data is matrixx for categorical indices check

* check for error when using a greater index than the number of columns

* apply suggestion
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* revert whitespace change

* check if is filename instead of matrix
Co-authored-by: James Lamb <jaylamb20@gmail.com>