GitLab

* [R-package] fix segfaults caused by missing Booster and Dataset handles (fixes #4208)

* fix test errors

* fixes for cpplint

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

* fix tests

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

* move asserts inside try-catch
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [R-package] introduce Dataset set_field() and get_field()

* fix incorrect fields

* update pkgdown

* fix example

* fix another example

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

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

* [R-package] avoid unnecessary computation in Dataset set_reference() method

* re-arrange conditions

* do more validation upfront and add tests

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

* Update R-package/tests/testthat/test_dataset.R
Co-authored-by: Nikita Titov <nekit94-12@hotmail.com>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [R-package] fix inaccurate log message in Dataset get_colnames() method

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

* [R-package] preserve uses of '...' in Dataset slice() method

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

* [R-package] introduce keyword arguments for Dataset properties

* adding examples

* trying to fix loading issues

* add check on use of info and fix examples

* eliminate one '...'

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

* Update R-package/R/lgb.Dataset.R

* Dataset get_params() is public
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [R-package] deprecate uses of '...' in Dataset slice() method

* Update R-package/R/lgb.Dataset.R

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

* [R-package] fix warnings in examples

* fix silently-ignored parameter

* [R-package] add deprecation warnings on uses of '...' in predict()

* add importFrom(utils, modifyList) in NAMESPACE

[docs] Clarify the fact that predict() on a file does not support saved Datasets (fixes #4034) (#4545)

* documentation changes

* add list of supported formats to error message

* add unit tests

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

* update per review comments

* make references consistent
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

Co-authored-by: Nikita Titov <nekit94-12@hotmail.com>

Co-authored-by: Nikita Titov <nekit94-12@hotmail.com>

* [R-package] remove unnecessary comments

* Update R-package/R/lgb.Booster.R

* [R-package] consolidate duplicate lists of Dataset info keys

* remove extra parentheses

* [R-package] move more finalizer logic intoo C++ side

* add C finalizers

* use gc()

* put skip() back

* Update R-package/tests/testthat/test_lgb.Booster.R
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [R-package] avoid unnecessary computation of std deviations in lgb.cv()

* use expect_equal()

* [R-package] factor out lgb.check.r6.class()

* Predictor

* started converting handles

* more changes

* sort of working for Dataset

* yay all the tests are passing for Dataset handle changes

* working for other handle types

* remove debugging logging

* remove unnecessary spaces

* fix null logic

* more NULL

* updates

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

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

* [R-package] move creation of character vectors in some methods to C++ side

* convert LGBM_BoosterGetEvalNames_R

* convert LGBM_BoosterDumpModel_R and LGBM_BoosterSaveModelToString_R

* remove debugging code

* update docs

* remove comment

* add handling for larger model strings

* handle large strings in feature and eval names

* got long feature names working

* more fixes

* linting

* resize

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

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

* converted LGBM_BoosterSaveModel_R

* switch all other non-params cases

* handle params

* [R-package] move some functions over to SEXP objects

* clarify comment

* remove paste0()

* fix registration table

* fix cmake builds

* [R-package] prevent symbol lookup conflicts

* add unit tests

* only run test on Windows

* move to .Call() calls

* fix references

* testing registration for CMake builds

* revert NAMESPACE changes

* revert testing changes

* [R-package] raise errors from C++ side

* working but a lot of warnings

* more changes

* simplify

* cleanup

* linting

* fix valgrind issues

* revert unnecessary change

* [R-package] clarify parameter documentation

* fixes to braces

* linting

* use same default arguments as python

* Update R-package/R/lgb.cv.R

Changed.
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* revert change in docs
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* update the explanation of num_threads in lightgbm.R

* adjusted sentences for linting.

* remove white space for lintr pass

* add note about num_thread param in files

* Return feature importances as visible

Feature importances are returned as invisible. I don't get why would anyone want them invisible. This PR fixes it.

* Update R-package/R/lgb.importance.R

Thanks, didn't realize that, changed now.
Co-authored-by: James Lamb <jaylamb20@gmail.com>
Co-authored-by: James Lamb <jaylamb20@gmail.com>