GitLab

[python] raise an informative error instead of segfaulting when custom objective produces incorrect output (#4815)

* fix for bad grads causing segfault

* adjust checking criteria to properly reflect reality of multi-class classifiers

* fix styling

* Line break before operator

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

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

* add a note to the C-API docs

* rearrange text s;ightly

* add some tests to python package

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

* PR comments

* match argument is a regex and our expression has brackets ..

* rework tests

* isorting imports

* updating test to relfect that the python APi does not take pres/labels as a fobj function
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update test_sklearn.py

* Update python_package.yml

* Update python_package.yml

* Update callback.py

* Update callback.py

* Update sklearn.py

* Update sklearn.py

* Update test_sklearn.py

* remove `verbose_eval` argument

* update example Notebook

* [python-package][dask] handle failures parsing work host names

* add tests

* revert local testing changes

* unify values of `best_iteration` for sklearn and standard APIs

* update Dask test

* add C API function that returns all param names with aliases

* add C API function that returns all param names with aliases

* add R code

* test R code

* remove debug CI

* fix R lint

* refactor

* run CI

* fix R

* fix

* revert CI checks

* revert changes in docs

* Try to make function `const`
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* add `const` in cpp file

* address review comments and sync with `master`
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* [python-package] fix mypy errors in engine.py

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

* allow for stdv

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

* [python-package] fix mypy errors in sklearn.py

* use ignore comments

* in predict(), respect params set via `set_params()` after fit()

* continue

* add test

* fix return name

* hotfix

* simplify

* [python-package] fix mypy errors in plotting.py

* empty commit

* [python-package] fix mypy error about missing type hint in dask.py

* list of lists

* use different variable

* use append()

* Update __init__.py

* Update callback.py

* Update sklearn.py

* Update engine.py

* Update sklearn.py

* Update engine.py

* Update basic.py

* Update engine.py

* Update test_plotting.py

* Update dask.py

* Update sklearn.py

* Update test_sklearn.py

* Update basic.py

* Update engine.py

* Update test_engine.py

* Update basic.py

* Update basic.py

* Update engine.py

* [python] add type hints for custom objective and metric functions in scikit-learn interface

* update type hints

* remote unnecessary input

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

* remove type hint on objective being callable
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [c_api] Improve ANSI compatibility by avoiding <stdbool.h>

* fixes in response to CI linting

* inline NOLINT instead of separate test

* moving length declaration to non-ANSI C conditional

* [c_api] Align expected return type in `basic.py` with new c_api type.

* Faster categorical column names selection (#1)

* Faster categorical column names selection

Change slow and redundant dataframe query by select_dtypes into a dataframe.dtypes list comprehension

* Update compat with CategoricalDtype

* sort imports

* import CategoricalDtype from pandas.api.types

* add categorical import try/except

* Add 'nrounds' as an alias for 'num_iterations'

* Improve tests

* Compare against nrounds directly

* Fix whitespace lints

* initial changes

* initial version

* better handling of cases

* warn only with positive threshold

* remove early_stopping_threshold from high-level functions

* remove remaining early_stopping_threshold

* update test to use callback

* better handling of cases

* rename threshold to min_delta

enhance parameter description

update tests

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

* reduce num_boost_round in tests

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

* trigger ci
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-12@hotmail.com>

* respect objective aliases

* Update test_sklearn.py

* revert removal of blank lines

* add argument name which is being overwritten in warning message

* [docs] [dask] Add return information to Dask fit() docs (fixes #4402)

* put return before note

* add n_estimators_ and n_iter_ post-fit attributes

* address review comments