Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Release v4.0.0 #5952

Merged
merged 6 commits into from
Jul 13, 2023
Merged

Release v4.0.0 #5952

merged 6 commits into from
Jul 13, 2023

Conversation

jameslamb
Copy link
Collaborator

@jameslamb jameslamb commented Jun 29, 2023

Fixes #4788.
Fixes #5153.

Release checklist:

Copied from #5666, with a few new additions.

before merge

PRs:

after merge

What about the changelog?

As described in #4930 (comment), previous patch releases that didn't release latest master interfere with the automatic release notes generation tool used in this repo.

In each of those special releases, we copied in the list of PRs into the notes. I've listed all of those below. When I create the actual GitHub release, I'll combine all of them.

changelog from v3.3.1 to when v3.3.2 was created (click me)

NOTE: I wrapped this in a markdown text block so GitHub stops re-evaluating the links.

## Changes

## 💡 New Features

- [R-package] added argument eval_train_metric to lgb.cv() (fixes #4911) @mayer79 (#4918)
- Add support for Visual Studio 2022 @StrikerRUS (#4889)
- Add C API function that returns all parameter names with their aliases @StrikerRUS (#4829)
- [python][sklearn] respect parameters for predictions in `init()` and `set_params()` methods @StrikerRUS (#4822)
- Add customized parser support @chjinche (#4782)
- [R-package] Add `print()` and `summary()` methods for Booster @david-cortes (#4686)
- Add 'nrounds' as an alias for 'num_iterations' (fixes #4743) @mikemahoney218 (#4746)
- [python-package] early stopping min_delta (fixes #2526) @jmoralez (#4580)
- [python][sklearn] respect objective aliases @StrikerRUS (#4758)
- [python][sklearn] add `n_estimators_` and `n_iter_` post-fit attributes @StrikerRUS (#4753)

## 🔨 Breaking

- [python] remove `early_stopping_rounds` argument of `train()` and `cv()` functions @StrikerRUS (#4908)
- [python] remove `evals_result` argument of `train()` function @StrikerRUS (#4882)
- [python][sklearn] do not replace empty dict with `None` for `evals_result_` @StrikerRUS (#4884)
- [python] Drop Python 3.6 support @StrikerRUS (#4891)
- [python] remove `verbose_eval` argument of `train()` and `cv()` functions @StrikerRUS (#4878)
- [python] remove `verbose` argument of `model_from_string()` method of Booster class @StrikerRUS (#4877)
- [python][sklearn] Remove `early_stopping_rounds` argument of `fit()` method @StrikerRUS (#4846)
- [R-package] remove support for '...' in `slice()` @jameslamb (#4872)
- [R-package] remove support for '...' in `lgb.Dataset()` @jameslamb (#4874)
- [R-package] remove support for '...' in `dim.lgb.Dataset()` @jameslamb (#4873)
- [R-package] remove support for '...' in lgb.train() @jameslamb (#4863)
- [R-package] remove support for '...' in `create_valid()` @jameslamb (#4865)
- [R-package] remove support for 'info' in Dataset @jameslamb (#4866)
- [R-package] remove Dataset getinfo() @jameslamb (#4864)
- [R-package] remove support for '...' in lgb.cv() @jameslamb (#4860)
- [R-package] remove Dataset setinfo() @jameslamb (#4854)
- [R-package] remove support for '...' in predict() @jameslamb (#4857)
- [R-package] remove support for '...' in Booster reset_parameter() @jameslamb (#4856)
- [python][sklearn] unify values of `best_iteration` for sklearn and standard APIs @StrikerRUS (#4845)
- [ci] migrate CI from macOS 10.14 to 10.15 and drop support of Mojave @StrikerRUS (#4849)
- [R-package] enable saving Booster with saveRDS() and loading it with readRDS() (fixes #4296) @david-cortes (#4685)
- [python][sklearn] remove `verbose` argument from `fit()` method @StrikerRUS (#4832)
- [python] remove `learning_rates` argument of `train()` function @StrikerRUS (#4831)
- [python] remove "auto" value of `ylabel` argument of `plot_metric()` function @StrikerRUS (#4818)
- [python] Remove `print_evaluation()` function @StrikerRUS (#4819)
- [python] Remove `silent` argument @StrikerRUS (#4800)

## 🚀 Efficiency Improvement

- clear memory allocated for sampled data when constructing Dataset from text file @xuchuanyin (#4890)
- [python] Faster categorical column names selection @Neronuser (#4787)
- [R-package] parallelize compilation in CMake-based builds @jameslamb (#4525)

## 🐛 Bug Fixes

- [R-package] respect 'verbose' argument in lgb.cv() (fixes #4667) @jameslamb (#4903)
- [R-package] Apply patch for R4.2 on Windows @shiyu1994 (#4923)
- [R-package] respect aliases for objective and metric and lgb.train() and lgb.cv() @jameslamb (#4913)
- [python] raise an informative error instead of segfaulting when custom objective produces incorrect output @yaxxie (#4815)
- [R-package] fix handling of duplicate parameters (fixes #4521) @jameslamb (#4914)
- [R-package] update parameter 'verbosity' based on keyword arg 'verbose' @jameslamb (#4899)
- [R-package] fix CVBooster reset_parameter() method (fixes #4900) @jameslamb (#4901)
- [python] reset storage in record evaluation callback each time before starting training @StrikerRUS (#4885)
- [python] reset storages in early stopping callback after finishing training @StrikerRUS (#4868)
- [R-package] fix `--no-build-vignettes` option for `build-cran-package.sh` @jameslamb (#4848)
- [python][docs] fix type hints for custom functions and remove vague `array-like` wording @StrikerRUS (#4816)
- Always respect forced splits, even when feature_fraction < 1.0 (fixes #4601) @tongwu-msft (#4725)
- Reset OpenMP thread number if num_threads <= 0 @hzy46 (#4704)

## 📖 Documentation

- [R-package] [docs] fix calculation of R test coverage (fixes #4919) @jameslamb (#4922)
- [docs] Update link to FLAML code example @qingyun-wu (#4892)
- [R-package] [docs] add Michael Mayer to DESCRIPTION @jameslamb (#4867)
- [docs] document that `pred_early_stop` can be used only in normal and raw scores prediction @StrikerRUS (#4823)
- [R-package] [docs] add intro vignette (#3946) @jameslamb (#4775)
- [docs] fix broken SynapseML link @StrikerRUS (#4795)
- [docs] [dask] Add return information to Dask fit() docs (fixes #4402) @jameslamb (#4716)
- [docs] add specific estimator names to sklearn fit() docs @jameslamb (#4774)
- [docs] [R-package] update cran-comments for v3.3.1 release @jameslamb (#4738)
- [ci] only use conda-forge when installing R packages in docs builds @jameslamb (#4767)
- [docs] improve docs about `nthreads` parameter @StrikerRUS (#4756)
- [docs] update link to LightGBM Ruby @StrikerRUS (#4740)
- [docs] fix broken SynapseML link @jameslamb (#4712)

## 🧰 Maintenance

- [R-package] [tests] remove uses of testthat::context() @jameslamb (#4915)
- [R-package][tests] remove uses of `testthat::expect_is()` @jameslamb (#4916)
- [R-package] remove unnecessary comments and fix typos in comments @jameslamb (#4902)
- [R-package] reduce verbosity in tests using lgb.train() @jameslamb (#4896)
- [tests][python-package] change boston dataset to synthetic dataset in tests that don't check score @jmoralez (#4895)
- [R-package] reduce verbosity in some unit tests @jameslamb (#4879)
- [R-package] remove unused callback cb.reset.parameter @jameslamb (#4871)
- [python-package][dask] handle failures parsing worker host names @jameslamb (#4852)
- [python-package] [dask] fix mypy error about worker_addresses @jameslamb (#4851)
- [python] remove workaround for `UnboundLocalError` in early stopping callback for Python 2 @StrikerRUS (#4855)
- [python-package] remove unused imports @jameslamb (#4850)
- [python-package] fix mypy errors in engine.py @jameslamb (#4839)
- [python-package] [dask] fix mypy errors from dask.py docstrings @jameslamb (#4844)
- [python-package] fix mypy errors in sklearn.py @jameslamb (#4837)
- [tests][dask] fix argument names in custom eval function in Dask test @StrikerRUS (#4833)
- [python-package] fix mypy errors in plotting.py @jameslamb (#4838)
- [python-package] fix mypy error about missing type hint in dask.py @jameslamb (#4840)
- [python-package] fix mypy errors in compat.py and setup.py @jameslamb (#4836)
- [docs][python] simplify mocking in docs @StrikerRUS (#4830)
- [python] add type hints to `_compare_params_for_warning()` and make it reusable @StrikerRUS (#4824)
- [R-package] Move R6 to Imports @david-cortes (#4812)
- [ci] Fix CI job for R artifact @StrikerRUS (#4811)
- [ci] fix getting status of optional workflows in PRs with a lot of comments @StrikerRUS (#4806)
- [python] add type hints for custom objective and metric functions in scikit-learn interface @jameslamb (#4547)
- [c_api] Improve ANSI compatibility by avoiding <stdbool.h> @drewmiller (#4697)
- [ci] ignore CMakeLint errors related to package names @StrikerRUS (#4801)
- [ci] fix CMakeLint errors related to function naming case @StrikerRUS (#4794)
- [ci] simplify processing of `CMAKE_CUDA_FLAGS` variable in `CMakeLists.txt` @StrikerRUS (#4799)
- [ci] fix CMakeLint `linelength` errors @StrikerRUS (#4796)
- [ci] use pure `endif()` and `endfunction()` without expression inside in `CMakeLists.txt` @StrikerRUS (#4798)
- cmake: use object library to avoid duplicate compilation. @cyfdecyf (#4489)
- Suppress categorical warning (fixes #3379) @hzy46 (#4768)
- [ci][tests][python] remove assertion for `filename` that is no longer true with new version of graphviz @StrikerRUS (#4778)
- [ci] Revert temp workaround for `freetype` version @StrikerRUS (#4776)
- [ci] increase timeout for valgrind job to 240 minutes @jameslamb (#4773)
- [python] improve warning message about aliases in `cv()` function @StrikerRUS (#4766)
- [ci] Temporary pin dask version at CI @StrikerRUS (#4770)
- [ci] use wch1/r-debug image in Solaris tests @jameslamb (#4765)
- [ci] upgrade actions/checkout to v2.4.0 @jameslamb (#4763)
- [docs][python] add `datatable` to the mocked modules during docs building process and sort them alphabetically @StrikerRUS (#4750)
- [R-package] allow use of custom R executable when building CRAN package @jameslamb (#4754)
- [R-package] remove temporary files created in configure.win @jameslamb (#4752)
- [ci] indicate support of Monterey @StrikerRUS (#4732)
- [python] Make dummy classes constructible with any arguments @StrikerRUS (#4749)
- [ci] Temporary pin freetype version at Windows CI @StrikerRUS (#4747)
- Remove checks for label when loading dataset from binary file because label is ignored in that case @StrikerRUS (#4737)
- [tests][python] add test for non-serializable callback @StrikerRUS (#4741)
- Improve warning wordings @StrikerRUS (#4731)
- [ci] Bump version for development @StrikerRUS (#4730)
- Add some warnings when loading dataset from binary file @StrikerRUS (#4724)
changelog from when v3.3.2 was created to date of v3.3.3 (click me)

NOTE: I wrapped this in a markdown text block so GitHub stops re-evaluating the links.

## Changes

## 💡 New Features

- [CUDA] Add multiclass objective for cuda_exp @shiyu1994 (#5473)
- [CUDA] Add feature interaction constraint for cuda_exp (fix #4785) @shiyu1994 (#5474)
- [CUDA] Add rank_xendcg objective for cuda_exp @shiyu1994 (#5472)
- [CUDA] Add fair regression objective for cuda_exp @shiyu1994 (#5469)
- [CUDA] Add lambdarank objective for cuda_exp @shiyu1994 (#5453)
- [CUDA] Add Huber regression objective for cuda_exp @shiyu1994 (#5462)
- [CUDA] Add L1 regression objective for cuda_exp @shiyu1994 (#5457)
- [CUDA] L2 regression objective for cuda_exp @shiyu1994 (#5452)
- [CUDA] Add binary objective for cuda_exp @shiyu1994 (#5425)
- [R-package] Add remainder of prediction funtions @david-cortes (#5312)
- [python-package] support saving and loading CVBooster (fixes #3556) @nyanp (#5160)
- feature: Add true streaming APIs to reduce client-side memory usage @svotaw (#5299)
- [python-package] highlight the path a sample takes through a tree in `plot_tree` and `create_tree_digraph` (fixes #4784) @jmoralez (#5119)
- reproducible parameter alias resolution for wrappers (fixes #5304) @jmoralez (#5338)
- [CUDA] Initial work for boosting and evaluation with CUDA @shiyu1994 (#5279)
- [python-package] add validate_features argument to refit @jmoralez (#5331)
- [python-package] check feature names in predict with dataframe (fixes #812) @jmoralez (#4909)
- [R-package] Add sparse feature contribution predictions @david-cortes (#5108)
- [python-package][R-package] allow using feature names when retrieving number of bins @jmoralez (#5116)
- [R-package] Keep row names in output from `predict` @david-cortes (#4977)
- [python] make `reset_parameter` callback pickleable @StrikerRUS (#5109)
- [python] make `record_evaluation` callback pickleable @StrikerRUS (#5107)
- [python] make `log_evaluation` callback pickleable @StrikerRUS (#5101)
- [python] allow to register any custom logger (fixes #4783) @RustingSword (#4880)
- Load initial scores with binary data files in CLI version @shiyu1994 (#4807)
- [R-package] Rename `weight` -> `weights` @david-cortes (#4975)
- [CUDA] New CUDA version Part 1 @shiyu1994 (#4630)
- [python] make `early_stopping` callback pickleable @Yard1 (#5012)
- [c-api][python-package][R-package] expose feature num bin @jmoralez (#5048)
- [python-package] [R-package] propagate the best iteration of cvbooster into the individual boosters @jmoralez (#5066)
- [python-package] add support for pandas nullable types @jmoralez (#4927)
- [R-package] Promote objective and init_score to top-level arguments in `lightgbm()` @david-cortes (#4976)
- [python] Start supporting Python 3.10 @StrikerRUS (#4893)
- [python-package] support customizing Dataset creation in Booster.refit() (fixes #3038) @TremaMiguel (#4894)
- [dask] add support for custom objective functions (fixes #3934) @jameslamb (#4920)

## 🔨 Breaking

- fix: Adjust LGBM_DatasetCreateFromSampledColumn to handle distributed data @svotaw (#5344)
- [python-package] allow custom weighing in fobj for scikit-learn API (closes #5027) @jmoralez (#5211)
- [R-package] Use `type` argument to control prediction types @david-cortes (#5133)
- [python-package] Use scikit-learn interpretation of negative `n_jobs` and change default to number of cores @david-cortes (#5105)
- [python-package] remove Booster.set_attr() and Booster.attr() @jameslamb (#5272)
- remove support for Solaris (fixes #5216) @jameslamb (#5226)
- [R-package] stop automatically calculating eval metrics on training data in lightgbm() @jameslamb (#5209)
- [R-package] remove lgb.unloader() @jameslamb (#5204)
- [python-package] remove 'fobj' in favor of passing custom objective function in params @TremaMiguel (#5052)
- [python-package] remove is_reshape argument in Booster.predict (fixes #5115) @jmoralez (#5117)
- [R-package] Remove `reshape` argument in `predict` @david-cortes (#4971)
- [R-package] Promote number of threads to top-level argument in `lightgbm()` and change default to number of cores @david-cortes (#4972)
- [R-package] Rename `data` -> `newdata` in `predict` @david-cortes (#4973)
- Build Windows artifacts in `windows-2019` image instead of `vs2017-win2016` @StrikerRUS (#5059)
- [python-package] use 2d collections for predictions, grads and hess in multiclass custom objective @jmoralez (#4925)
- [R-package] prefer params to keyword argument in lgb.train() @jameslamb (#5007)
- [R-package] remove behavior where lightgbm() saves model to disk @david-cortes (#4974)
- [python-package] make record_evaluation compatible with cv (fixes #4943) @jmoralez (#4947)

## 🚀 Efficiency Improvement

- [python-package] simplify Dataset processing of label @jameslamb (#5456)
- [python-package] make a shallow copy on dataframe rename (fixes #4596) @jmoralez (#5254)
- [python-package] make a shallow copy when replacing categorical features with codes (fixes #4596) @jmoralez (#5225)
- [R-package] reduce cost of repeated parameter alias checks @jameslamb (#5141)
- reduce duplicate computation in poisson, gamma, and tweedie objectives @lorentzenchr (#4950)

## 🐛 Bug Fixes

- include parameters from reference dataset on subset (fixes #5402) @jmoralez (#5416)
- [python-package] ignore training set on early stopping callback (fixes #5354) @jmoralez (#5412)
- [fix] change the destructor of ScoreUpdater to virtual (fixes #5400) @shiyu1994 (#5403)
- Add default definition for GetColWiseData and GetColWiseData @shiyu1994 (#5413)
- Fix potential overflow in linear trees @StrikerRUS (#5395)
- Use double precision in threaded calculation of linear tree coefficients (fixes #5226) @btrotta (#5368)
- [R-package] raise an informative error when custom objective produces incorrect output (fixes #5323) @jmoralez (#5329)
- Clear split info buffer in cost efficient gradient boosting before every iteration (fix partially #3679) @shiyu1994 (#5164)
- [c++][fix] check nullable of bin mappers in dataset_loader.cpp (fix #5221) @shiyu1994 (#5258)
- [python] Fix training on subset constructed without params @StrikerRUS (#5213)
- Check existence of inet_pton for win32 in CMakeLists.txt (fixes #5019) @shiyu1994 (#5159)
- Fix potential overflow "Multiplication result converted to larger type" @StrikerRUS (#5189)
- [R-package] ensure that callbacks respect verbosity from params @jameslamb (#5199)
- fix precision lost in tree's ToIfElse @Grass-CLP (#5187)
- fix some wrong format specifiers @StrikerRUS (#5190)
- [R-package] allow use of categorical_features in Dataset when raw data does not have column names (fixes #4374) @jmoralez (#5184)
- [c-api] check number of features when retrieving number of bins @jmoralez (#5183)
- [CUDA] Fix integer overflow in cuda row-wise data @shiyu1994 (#5167)
- [R-package] ensure values in params override keyword arguments to predict() (fixes #4670) @jameslamb (#5122)
- check nullable of bin_mappers in DatasetLoader::CheckCategoricalFeatureNumBin (fix #5145) @shiyu1994 (#5146)
- [CUDA] Fix row-wise histogram construction with dense data matrix @shiyu1994 (#5103)
- [fix] fix duplicate added initial scores for single-leaf trees @shiyu1994 (#5050)
- [python] fixes for supporting 2d numpy arrays for predictions, grads and hess in multiclass custom objective and eval @StrikerRUS (#5030)
- CUDATreeLearner: free GPU memory in destructor if any allocated @denmoroz (#4963)
- Use `delete[]` where appropriate instead of `delete` @david-cortes (#4984)
- Pass train dataset parser config to valid dataset loading parser @chjinche (#4985)
- [R-package] Fix custom objective detection in `print.lgb.Booster()` @StrikerRUS (#4941)
- gpu allocate memory overflow (fixes #4926) @jiapengwen (#4928)

## 📖 Documentation

- [R-package] [docs] use CRAN canonical form for package links @jameslamb (#5504)
- [docs] Fix link to Optuna's LightGBMTuner example @knshnb (#5519)
- [docs]fix a typo in docs/Features.rst @JiantingFeng (#5463)
- Use ROC-AUC metric for classification model in examples @Green-16 (#5440)
- [docs] Fix links @StrikerRUS (#5451)
- [R-package] Fix docstrings for predict functions @StrikerRUS (#5444)
- [R-package] [docs] clarify shape of predictions @jameslamb (#5384)
- [python-package] Improve readme on experimental cuda usage @thomasaarholt (#5386)
- [docs] [R-package] document how to regenerate roxygen docs @jameslamb (#5382)
- [R-package] [docs] fix typo in documentation on init_model @sebffischer (#5379)
- [docs] [R-package] upgrade to roxygen2 7.2.1 @jameslamb (#5381)
- [ci] make check-docs job compatible with rstcheck 6.x @jameslamb (#5388)
- [docs][python] Fix return types in docstrings @StrikerRUS (#5326)
- [docs] set language = 'en' in Sphinx config @jameslamb (#5322)
- [R-package] [docs] document difference between lightgbm() and lgb.train() (fixes #5203) @jameslamb (#5273)
- [docs] Use https links in docs @StrikerRUS (#5284)
- [R-package] [docs] upgrade docs to roxygen2==7.2.0 @jameslamb (#5251)
- [docs] Update broken links in docs @StrikerRUS (#5265)
- [docs] add Microsoft security policy (fixes #5235) @jameslamb (#5264)
- [R-package] update roles in DESCRIPTION @StrikerRUS (#5192)
- [docs] Fix formula in path smoothing docs (fixes #5139) @samFarrellDay (#5154)
- [docs] Document behaviour of the first linear estimator @Pablo-Davila (#5132)
- [docs] use 'docker run --rm' in valgrind instructions @jameslamb (#5127)
- [docs] use 'docker run --rm' in autoconf instructions @jameslamb (#5125)
- [docs] Improve rendering of class properties in docs @StrikerRUS (#5078)
- [python-package][docs] Booster eval methods accept list of callables @TremaMiguel (#5054)
- [docs] update categorical feature description in Advanced Topics @StrikerRUS (#5044)
- Correct documentation for sparse predictions @david-cortes (#4979)
- [docs][R-package] Use direct link for the list of supported objectives @StrikerRUS (#5029)
- [Docs] Weights non-negative for train data @TremaMiguel (#5013)
- [docs] improve docs for sklearn wrapper @StrikerRUS (#5026)
- [docs] clarify that categorical features will be converted to integers internally @jmoralez (#4959)
- [docs] clarify that custom eval functions are not only used on training data @jameslamb (#5011)
- [docs] document rounding behavior of floating point numbers in categorical features @StrikerRUS (#5009)
- Change docs for feval @akshitadixit (#5002)
- [docs] document `conda-forge` channel preference over `default` one and describe possible workaround for OpenMP conflicts in FAQ @StrikerRUS (#4994)
- Add Nyoka to README @StrikerRUS (#4992)
- [docs] minor docs improvements @StrikerRUS (#4938)

## 🧰 Maintenance

- [ci] run Appveyor checks on PRs targeting release/ branches @jameslamb (#5528)
- [ci] run CI on pull requests targeting release/ branches @jameslamb (#5527)
- [python-package] prefix is_numpy_1d_array with _ @Madnex (#5520)
- replace `pandas.Series.iteritems` with `pandas.Series.items` @jmoralez (#5506)
- [ci] prefer CPython in Windows test environment and use safer approach for cleaning up network (fixes #5509) @jameslamb (#5510)
- [ci] test against R 4.2 for macOS and Linux CI jobs @jameslamb (#5484)
- [R-package] improve safety of index selection in plotting @jameslamb (#5485)
- [ci] [R-package] ensure that MSVC jobs fail when tests fail (fixes #5439) @jameslamb (#5448)
- [python-package] add more hints in sklearn.py @jameslamb (#5460)
- [R-package] [ci] restore R 3.6 Windows cran CI job (fixes #5036) @jameslamb (#5479)
- Remove redundant whitespaces @ch3rn0v (#5480)
- fix references to 'object function' @jameslamb (#5468)
- [python] Fix typehints in Sequence API @StrikerRUS (#5465)
- fix: Revert leftover debugging in streaming test @svotaw (#5467)
- Fix CUDA `#ifndef` guards @StrikerRUS (#5466)
- [python-package] add type hints on Dataset constructors @jameslamb (#5458)
- Add streaming concurrency tests @svotaw (#5437)
- Rename Metadata num_classes to be more clear @svotaw (#5461)
- [R-package] fix function references in error messages @jameslamb (#5455)
- [python-package] fix type hints on ctypes array converters @jameslamb (#5446)
- [python-package] fix mypy errors about callbacks @jameslamb (#5450)
- [python-package] add type hints on empty initializations @jameslamb (#5445)
- [python-package] fix mypy error about type change in Dataset.feature_num_bin() @jameslamb (#5447)
- [ci][fix] Fix cuda_exp ci @shiyu1994 (#5438)
- [python] remove unused import @StrikerRUS (#5443)
- [ci] bump CUDA version from `11.7.0` to `11.7.1` at CI @StrikerRUS (#5442)
- [python-package] add type hints on Booster eval methods @jameslamb (#5433)
- update tree to if-else @jmoralez (#5422)
- [ci] bump CUDA version from `11.6.2` to `11.7.0` at CI  @StrikerRUS (#5287)
- [python-package] add more type hints on Dataset @jameslamb (#5431)
- [python-package] add more type hints on Booster @jameslamb (#5432)
- [ci] increase valgrind bytes lost limit to 352 @jameslamb (#5429)
- [python] Prefix `basic.is_numeric()` with _ @ak04p (#5421)
- [ci] increase valgrind timeout to 5 hours @jameslamb (#5414)
- [python-package] add more type hints in basic.py @jameslamb (#5407)
- [python-package] prefix NUMERIC_TYPES with _ to make it a internal object @thomasjpfan (#5409)
- [python-package] add type hints on Booster.save_model() @jameslamb (#5406)
- Minor CUDA cleanup @StrikerRUS (#5394)
- [ci] temporarily pin scipy version and fix Python linting error @StrikerRUS (#5398)
- [python-package] add type hints on some functions in basic.py @makquel (#5362)
- [python-package] add more type hints on Booster @jameslamb (#5360)
- [python-package] make library-loading stricter @jameslamb (#5357)
- [python-package] add type hints on Booster.update() @jameslamb (#5359)
- [python-package] add more type hints on basic.py @jameslamb (#5356)
- [python-package] add type hints on Dataset @jameslamb (#5353)
- [python-package] remove inner function _construct_dataset() in LGBMModel.fiit() @jameslamb (#5333)
- [python-package] add type hints on predict() methods @jameslamb (#5334)
- [python-package] add more type hints in basic.py @jameslamb (#5330)
- [ci][R-package] Minor refactoring for `lintr` code @StrikerRUS (#5327)
- [python-package] add more type hints in engine.py @jameslamb (#5301)
- [ci] [R-package] Add string_boundary_linter @CuriousCorrelation (#5324)
- [python-package] add more type hints on Booster @jameslamb (#5309)
- [ci] [R-package] Add paste_linter @CuriousCorrelation (#5320)
- [ci] [R-package] Add duplicate_argument_linter @SaumyaBhushan (#5310)
- [ci] [R-package] Add a few more linters @CuriousCorrelation (#5311)
- [ci] [R-package] Add class_equals linter @CuriousCorrelation (#5307)
- [python][sklearn] Simplify params handling in `predict()` with `_choose_param_value()` @StrikerRUS (#5308)
- [ci] [R-package] Add any_is_na_linter @CuriousCorrelation (#5306)
- [python-package] add some type hints on Booster @jameslamb (#5302)
- [python] preserve None in _choose_param_value() @jameslamb (#5289)
- [ci] Install PoCL in Docker for old Ubuntu 14.04 @StrikerRUS (#5286)
- [R-package] Specify concrete linting rule that is being ignored via `nolint` mark @StrikerRUS (#5300)
- [python-package] add type hints on cv() @jameslamb (#5271)
- [ci] [R-package] upgrade to lintr v3.0 (fixes #5228) @jameslamb (#5294)
- [ci] Pin lintr to <3.0 @jgiannuzzi (#5290)
- [ci] Run Linux OpenCL tests against POCL instead of the AMD App SDK @jgiannuzzi (#5282)
- [tests][python] Make test that checks original pandas data isn't modified more strict @StrikerRUS (#5267)
- [python] add more type hints in basic.py @jameslamb (#5255)
- [python] add more type hints on LGBMModel methods @jameslamb (#5239)
- [R-package] address linter warnings about portable paths @jameslamb (#5249)
- [R-package] [ci] silence more logs in tests (fixes #4862) @jameslamb (#5250)
- Remove leftovers after the drop of Solaris support @StrikerRUS (#5248)
- [python] add type hints on sklearn metric and eval wrappers @jameslamb (#5238)
- [R-package] silence logs in print(), show(), summary() tests @jameslamb (#5237)
- [R-package] standardize style for placement of braces @jameslamb (#5240)
- [R-package] remove semicolon in R code @jameslamb (#5232)
- [R-package] remove Solaris reference in test message @jameslamb (#5231)
- [R-package] silence more logs in unit tests @jameslamb (#5227)
- [ci] switch CRAN mirror to RStudio @jameslamb (#5230)
- Cleanup codeowners @StrikerRUS (#5215)
- [R-package] minor improvements on weight demo @jmoralez (#5212)
- [ci] fix R Hub token @jameslamb (#5210)
- [R-package] silence more logs in tests @jameslamb (#5208)
- [ci] Prevent Python downgrading to pypy on Windows (2) @StrikerRUS (#5198)
- [ci] Prevent Python downgrading to pypy on Windows @StrikerRUS (#5197)
- [ci] fix git checkout for comment-triggered CI jobs @jameslamb (#5169)
- [python] remove unused import from `basic.py` @StrikerRUS (#5188)
- [tests] replace `fobj` with `custom objective` in test comments and make tests stricter @StrikerRUS (#5173)
- [ci] Update version of Azure REST API @StrikerRUS (#5172)
- [ci] skip auto-injected tasks at Azure Pipelines @StrikerRUS (#5175)
- fix typo in CEGB method name @jameslamb (#5168)
- [ci] temporarily pin Azure Devops jobs to ubuntu 20.04 @jameslamb (#5174)
- [ci] fix git checkout for CI jobs (fixes #5151) @jameslamb (#5152)
- [ci] bump CUDA version from `11.5.1` to `11.6.2` at CI @StrikerRUS (#5149)
- [python] Use predefined constant in feature importance type comparison instead of raw int literal @StrikerRUS (#5148)
- [ci] update to R 4.1.3 and use macOS-latest for R jobs (fixes #4990) @jameslamb (#5137)
- [ci] have no-response bot post as github-actions user @jameslamb (#5136)
- [ci] fail R macOS CI jobs earlier when installations fail @jameslamb (#5129)
- [ci] use lee-dohm/no-response to close stale issues (fixes #5060) @jameslamb (#5120)
- [ci] Remove temp workaround for graphviz installation @StrikerRUS (#5126)
- [R-package] rename internal callback functions @mayer79 (#5123)
- [R-package] ensure boosting happens in tests on small datasets @jameslamb (#5121)
- [R-package] Add missed packages into dependencies list @StrikerRUS (#5118)
- [R-package] add assertions to test on lightgbm() weights @jameslamb (#5110)
- Log warnings for number of bins of categorical features @shiyu1994 (#4448)
- clarify no-meaningful-features warning in Dataset construction (fixes #5081) @jameslamb (#5083)
- [ci] Add text file with GitHub commit hash to nightly builds @StrikerRUS (#5082)
- [ci] remove Visual Studio 2017 CI job for R-package @StrikerRUS (#5079)
- [python-package] Better column dtype logging when column has "bad dtype" @hsorsky (#5065)
- [ci] update CODEOWNERS @jameslamb (#5063)
- [ci] fix current `master` fails with graphviz-related error @StrikerRUS (#5068)
- [ci] temporarily remove R3.6 CRAN CI build @jameslamb (#5049)
- [tests][python] move tests that use `train()` function defined in `engine.py` from `test_basic.py` to `test_engine.py` @StrikerRUS (#5034)
- [R-package] remove internal function lgb.check.obj() @jameslamb (#5021)
- [ci] use Python 3.8 for `bdist` CI job in old Ubuntu container @StrikerRUS (#5022)
- [R-package] raise informative errors directly when Booster creation fails @jameslamb (#5014)
- [ci] [docs] use miniforge for readthedocs builds (fixes #4954) @jameslamb (#4957)
- [ci] Enclose Python versions in quotes in config yml files @StrikerRUS (#5015)
- [python] enforce a floor of Python 3.6 (fixes #5004) @PyVCEchecker (#5006)
- Update versions of LightGBM dependencies @StrikerRUS (#4935)
- [tests][python] remove compatibility code for old versions in tests @StrikerRUS (#4978)
- [ci] use conda-forge in Windows CI jobs and Docker images @StrikerRUS (#4993)
- [ci] use `latest` tag again for Ubuntu 14 CI Docker @StrikerRUS (#5001)
- [ci] use conda-forge in Linux and macOS CI jobs @jameslamb (#4953)
- [ci] upgrade to R 4.1.2 in CI, change approach to macOS-latest R CI builds (fixes #4988) @jameslamb (#4989)
- [ci] bump CUDA version from `11.5.0` to `11.5.1` at CI @StrikerRUS (#4967)
- [ci] use `-1` for exit code in case of failure @StrikerRUS (#4939)
- [tests][R-package] use one `#` symbol for comments @StrikerRUS (#4940)
- [ci] bump CUDA version from 11.4.2 to 11.5.0 at CI @StrikerRUS (#4937)
- [ci] Bump version for development @StrikerRUS (#4933)
changelog from when v3.3.3 was created to v3.3.4 (click me)

NOTE: I wrapped this in a markdown text block so GitHub stops re-evaluating the links.

## Changes

## 💡 New Features

- [CUDA] Add binary logloss metric for new CUDA version @shiyu1994 (#5635)
- Decouple Boosting Types (fixes #3128) @lyf-00 (#4827)
- [CUDA] Add L2 metric for new CUDA version @shiyu1994 (#5633)
- [CUDA] Add rmse metric for new CUDA version @shiyu1994 (#5611)
- Build integrated OpenCL Linux wheels @jgiannuzzi (#5252)
- [CUDA] Add Poisson regression objective for cuda_exp and refactor objective functions for cuda_exp @shiyu1994 (#5486)
- [CUDA] Add multiclass_ova objective for cuda_exp @shiyu1994 (#5491)
- [python-package] add install option to enable printing of time costs @Remy-Luciani (#5497)
- [python-package][R-package] load parameters from model file (fixes #2613) @jmoralez (#5424)

## 🔨 Breaking

- [ci] migrate CI from macOS 10.15 to 11 (fixes #5391) @StrikerRUS (#5396)
- [ci] switch to manylinux_2_28 for Linux artifacts (fixes #5514, fixes #5589) @jameslamb (#5580)

## 🚀 Efficiency Improvement

- [python-package] replace .values usage with .to_numpy() @superlaut (#5612)

## 🐛 Bug Fixes

- Check feature indexes in forced split file (fixes #5517) @btrotta (#5653)
- fix feature index in Dataset::AddFeaturesFrom (fixes #5410) @jameslamb (#5650)
- [ci] [python-package] fix missing import, test that lightgbm can be imported with only required dependencies (fixes #5631) @jameslamb (#5632)
- Fix OpenMP thread allocation in Linux @svotaw (#5551)
- [R-package] correctly quote paths on Windows for CMake-based builds @jameslamb (#5607)
- [ci] [python-package] correct tag on x86_64 wheels @jameslamb (#5598)
- [tests][dask] fix workers without data test (fixes #5537) @jmoralez (#5544)
- prefer 'vsnprintf' to 'vsprintf' @jameslamb (#5561)
- [ci] fix R-package CI jobs and compatibility with OpenMP 15+ (fixes #5549, #5562) @jameslamb (#5563)

## 📖 Documentation

- [docs] update CPU dockerfiles (fixes #5550) @jameslamb (#5601)
- [docs] update cran-comments.md for v3.3.4 release @jameslamb (#5646)
- [docs] add postgresml to list of external repos @jameslamb (#5565)
- [docs] Improve docs: fix consistency of dots in C API and add notes about new ``time-costs`` Python-package build option @StrikerRUS (#5554)

## 🧰 Maintenance

- [ci][dask][gpu] Run Dask tests with LightGBM GPU version @jgiannuzzi (#5292)
- [python-package] prefix internal objects with '_' @jameslamb (#5654)
- [ci] speed up Windows jobs (fixes #5647) @jameslamb (#5648)
- [ci] automatically cancel GitHub Actions runs for outdated commits @jameslamb (#5651)
- [ci] use LightGBM CI image for building aarch64 wheels (fixes #5595) @jameslamb (#5622)
- [ci] allow ubuntu-latest to float for some GitHub Actions jobs @jameslamb (#5644)
- [ci] fix locale-setting in jobs running in ubuntu container @jameslamb (#5643)
- [ci] make GitHub Actions branch protection stricter (fixes #5501) @jameslamb (#5645)
- [ci] Use Ubuntu 22.04 as `ubuntu-latest` at CI (fixes #5186) @StrikerRUS (#5288)
- [tests][python-package] remove remaining tests using load_boston (fixes #4793) @jmoralez (#5581)
- [python-package] prefix c_int_array and c_float_array with _ @OmarManzoor (#5614)
- [ci] test against R 4.2.2 @jameslamb (#5621)
- [python-package] prefix basic.convert_from_sliced_object with _ @OmarManzoor (#5613)
- [ci] detect non-default dynamic symbols in check_dynamic_dependencies.py @jameslamb (#5610)
- [ci] add docs and specific error messages in check_dynamic_dependencies.py @jameslamb (#5582)
- [ci] update versions of GitHub Actions @jameslamb (#5590)
- [swig] switch to SWIG_ADD_LIBRARY() in CMakeLists.txt (fixes #5586) @shogohida (#5603)
- [ci] switch from MiKTeX to tinytex on Windows R jobs (fixes #5600) @jameslamb (#5602)
- [python-package] prefix several internal functions with _ @Madnex (#5545)
- [python-package] remove unused imports @jameslamb (#5559)
- [ci] prefer CPython in Linux and macOS test environment @StrikerRUS (#5555)
- [docs] bump development version to 3.3.3.99 @jameslamb (#5533)
- [ci] upgrade macOS_regular job to Python 3.9 (fixes #5569) @jameslamb (#5570)
- [tests] Fix cpp streaming data tests @StrikerRUS (#5481)
- [python-package] prefix is_numpy_column_array with _ @Madnex (#5531)
- renamed cur_cat => cur_cat_idx and added some comments @zyxue (#5522)
- [python-package] prefix cast_numpy_array_to_dtype with _ @Madnex (#5532)
- suppress alias warnings with verbosity<0 (fixes #4518) @jmoralez (#5253)
- renamed tmp_num_sample_values to non_na_cnt @zyxue (#5521)
changelog from when v3.3.4 was created to v3.3.5 (click me)

NOTE: I wrapped this in a markdown text block so GitHub stops re-evaluating the links.

## Changes

## 💡 New Features

- [CUDA] Add binary logloss metric for new CUDA version @shiyu1994 (#5635)

## 🚀 Efficiency Improvement

- [python-package] replace .values usage with .to_numpy() @superlaut (#5612)

## 🐛 Bug Fixes

- [ci] [R-package] fix clang 15 warning about unqualified calls (fixes #5661) @jameslamb (#5662)
- Check feature indexes in forced split file (fixes #5517) @btrotta (#5653)
- fix feature index in Dataset::AddFeaturesFrom (fixes #5410) @jameslamb (#5650)

## 🧰 Maintenance

- [ci][dask][gpu] Run Dask tests with LightGBM GPU version @jgiannuzzi (#5292)
- [python-package] prefix internal objects with '_' @jameslamb (#5654)
- [ci] speed up Windows jobs (fixes #5647) @jameslamb (#5648)
- [ci] automatically cancel GitHub Actions runs for outdated commits @jameslamb (#5651)
changelog from when v3.3.5 was created to today (click me)

NOTE: I wrapped this in a markdown text block so GitHub stops re-evaluating the links.

## Changes

- [c++] virtual destructor for gradient discretizer @shiyu1994 (#5965)

## 💡 New Features

- [python-package] add 'pandas' extra @jameslamb (#5937)
- [CUDA] Add more CUDA Regression Metrics @Xuweijia-buaa (#5924)
- [python-package] adding max_category_values parameter to create_tree_digraph method (fixes #5687) @moziada (#5818)
- [c++] support building with Ninja on Linux @jameslamb (#5910)
- add CMakeLists options to disable building CLI, installing headers @jameslamb (#5880)
- Add quantized training (CPU part) @shiyu1994 (#5800)
- [CUDA] Add quantile regression objective for new CUDA version @shiyu1994 (#5605)
- [CUDA] Add quantile metric for new CUDA version (contribute to #5163) @shiyu1994 (#5665)
- [python-package] add Booster.set_leaf_output method @jmoralez (#5712)
- feature: Add serialization of reference dataset @svotaw (#5427)
- [R-package] Accept factor labels and use their levels @david-cortes (#5341)

## 🔨 Breaking

- [python-package] make Booster and Dataset 'handle' attributes private (fixes #5313) @jameslamb (#5947)
- [python-package] remove hard dependency on 'scikit-learn', fix minimal runtime dependencies @jameslamb (#5942)
- [python-package] [ci] switch to PEP 517 / 518 builds (remove `setup.py`) (fixes #5061) @jameslamb (#5759)
- [ci] [python-package] replace 'python setup.py' with a shell script @jameslamb (#5837)
- [R-package] use C++17 in the CRAN package @jameslamb (#5690)
- [python-package] make some Booster and Dataset attributes private @jameslamb (#5723)
- [CUDA] consolidate CUDA versions @jameslamb (#5677)
- [python-package] make public API members explicit with module-level __all__ variables @jameslamb (#5655)

## 🚀 Efficiency Improvement

- Add quantized training (CPU part) @shiyu1994 (#5800)

## 🐛 Bug Fixes

- move LightGBM-vendored json11 into a LightGBM-specific namespace (fixes #5944) @maskedcoder1337 (#5946)
- [dask] hold ports until training @jmoralez (#5890)
- update MSBuild solution to Windows SDK v10.0, add inet_pton define (fixes #5856) @jameslamb (#5884)
- Fix DEBUG-mode GPU builds @GinkoBalboa (#5778)
- cast data_index as size_t in cuda_row_data to avoid integer overflow @SiNZeRo (#5706)

## 📖 Documentation

- [docs] fix broken link to contributor GitHub account @jameslamb (#5959)
- [docs] add versionadded notes for v4.0.0 features @jameslamb (#5948)
- [docs] Add scikit-learn for intersphinx (fixes #5954) @thomasjpfan (#5956)
- [docs] more fixes for broken links @jameslamb (#5941)
- [docs] remove outdated information in Python install docs @jameslamb (#5938)
- [docs] fix broken links in docs @jameslamb (#5939)
- [docs] add quantized training paper to docs @jameslamb (#5923)
- [docs] Update Quick Start @jaceklaskowski (#5900)
- [docs] recommend a more efficient source installation of the python package @jameslamb (#5881)
- adjusted bagging_freq parameter description @moziada (#5698)
- [docs] remove unnecessary interactivity in 'docker run' instructions @jameslamb (#5692)
- [docs] [R-package] mention {bonsai} in the docs @jameslamb (#5652)
- [docs] encourage use of releases for GUI-only installations @jameslamb (#5649)

## 🧰 Maintenance

- [python-package] avoid data_has_header check in predict() @jameslamb (#5970)
- [ci] [python-package] Add CI job testing compatibility with oldest possible versions @jameslamb (#5936)
- [ci] add Python 3.11 test coverage @jameslamb (#5922)
- [ci] increase valgrind timeout to 6 hours @jameslamb (#5953)
- [python-package] ignore pip cache dir in build-python.sh @jameslamb (#5945)
- [ci] remove brew update-reset again (fixes #5670) @jameslamb (#5925)
- [python-package] ensure 'build-python.sh' always reinstalls (fixes #5906) @jameslamb (#5907)
- [python-package] remove unnecessary Boost files from sdists @jameslamb (#5912)
- [ci] stop relying on cran.microsoft.com in CI jobs @jameslamb (#5909)
- [ci] use an MRAN snapshot for R 3.6 Linux jobs (fixes #5898) @jameslamb (#5903)
- [ci] [python-package] use ruff, enforce flake8-bugbear and flake8-comprehensions checks @jameslamb (#5871)
- [ci] [python-package] fix mypy errors in sklearn.py @jameslamb (#5886)
- Add SWIG byte pointers @svotaw (#5832)
- [ci] [python-package] fix mypy error in Dataset.set_categorical_feature() @jameslamb (#5883)
- [ci] fix wheel renaming in test.sh @jameslamb (#5877)
- [ci] [python-package] tell mypy 'auto' has a special meaning for feature_name and categorical_feature @jameslamb (#5874)
- [ci] [python-package] move tool configuration to pyproject.toml @jameslamb (#5872)
- [ci] use Python 3.11 in linting CI job @jameslamb (#5873)
- [ci] use lowercase library names in linker flags @characat0 (#5870)
- [ci] minor fixes to build-python.sh @jameslamb (#5869)
- [ci] [python-package] resolve remaining mypy errors in dask.py @jameslamb (#5858)
- [ci] [python-package] check distributions with pydistcheck @jameslamb (#5838)
- [ci] remove R 3.6 macOS CI jobs @jameslamb (#5859)
- [ci] [python-package] fix mypy errors about Dataset._set_init_score_by_predictor() @jameslamb (#5850)
- [python-package] [ci] fix mypy errors in Booster.__inner_predict() @jameslamb (#5852)
- [ci] [python-package] fix mypy errors in Booster.refit() @jameslamb (#5853)
- [ci] add tests on cv() with an init_model @jameslamb (#5851)
- [ci] [python-package] fix mypy error about dask._predict_part() @jameslamb (#5849)
- [ci] [python-package] fix mypy error about return_cvbooster in cv() @jameslamb (#5845)
- [python-package] move validation up earlier in cv() and train() @jameslamb (#5836)
- [python-package] remove default arguments in internal functions @jameslamb (#5834)
- [python-package] fix mypy error about used_indices in Dataset.subset() @jameslamb (#5833)
- [ci] put a temporary ceiling on pip @jameslamb (#5835)
- [ci] [dask] fix mypy errors about predict() signature @jameslamb (#5827)
- [python-package] fix mypy error in Dataset.__init_from_list_np2d() @jameslamb (#5826)
- [ci] [python-package] fix mypy errors in cv() @jameslamb (#5828)
- [python-package] [docs] complete type annotations for scikit-learn fit() methods @jameslamb (#5816)
- [ci] remove pin on tinytex in R CI jobs @jameslamb (#5821)
- [ci] moving linting into small scripts (fixes #5142) @jameslamb (#5809)
- [python-package] add more type annotations in basic.py @jameslamb (#5812)
- [python-package] add type annotations on some array methods in basic.py @jameslamb (#5813)
- [python-package] add type annotations on Booster.trees_to_dataframe() inner functions @jameslamb (#5811)
- [python-package] fix type annotations for eval result tracking @jameslamb (#5793)
- [python-package] fix mypy error in Booster.get_split_value_histogram() @jameslamb (#5794)
- [python-package] fix mypy errors about early stopping rounds @jameslamb (#5795)
- [python-package] fix mypy errors about custom eval and metric functions @jameslamb (#5790)
- [python-packages] [docs] add type hints and define 'array-like' for X, y, group in scikit-learn interface @jameslamb (#5757)
- remove unused variables from bin.cpp @jmoralez (#5801)
- [ci] fix LaTeX tools installation in R jobs (fixes #5802) @jameslamb (#5807)
- [python-package] fix mypy errors about scikit-learn properties @jameslamb (#5788)
- [python-package] add hint on dtypes in Dataset.set_field() @jameslamb (#5787)
- [python-package] replace uses of scipy.sparse.issparse() with isinstance() @jameslamb (#5784)
- [python-package] fix mypy error about cpu_count() methods @jameslamb (#5786)
- [python-package] fix mypy errors about Dataset.params @jameslamb (#5780)
- [python-package] ignore some mypy errors in _DaskLGBMModel @jameslamb (#5774)
- [python-package] fix mypy errors about ctypes pointers @jameslamb (#5779)
- [python-package] fix mypy error about dask._HostWorkers.__eq__() @jameslamb (#5782)
- [python-package] fix mypy errors about validation sets in sklearn.py @jameslamb (#5724)
- [python-package] fix mypy errors about missing annotations and incompatible types @IdoKendo (#5672)
- [python-package] add type hints on sklearn properties @jameslamb (#5771)
- [python-package] add type hints for functions accepting dtypes @jameslamb (#5773)
- [ci] isolate c_api_test library-loading from Python source tree @jameslamb (#5761)
- [python-package] [dask] add type annotations on dask._HostWorkers @jameslamb (#5766)
- [python-package] [dask] fix mypy errors about Dask fit() return types @jameslamb (#5756)
- [python-package] use keyword arguments in predict() calls @jameslamb (#5755)
- [python-package] add type hints on label @jameslamb (#5754)
- [python-package] add type hints on group, init_score, and weight @jameslamb (#5753)
- [python-package] add type hints on raw data passed to Dataset and Booster @jameslamb (#5752)
- [python-package] add type hints on feature_name and categorical_feature in sklearn interface @jameslamb (#5747)
- [python-package] add type hints on fit() return and callbacks in sklearn.py @jameslamb (#5748)
- [python-package] add type hints on Booster predict() methods @jameslamb (#5749)
- [python-package] add more type hints in basic.py @jameslamb (#5729)
- [python-package] fix mypy error about ctypes array creation @jameslamb (#5726)
- [python-package] add type hints on Dataset feature processing @jameslamb (#5745)
- [ci] speed up conda setup for macOS and Linux jobs @jameslamb (#5743)
- [python-package] remove default values in internal functions @jameslamb (#5730)
- [python-package] use _DatasetHandle type hint @jameslamb (#5740)
- [ci] skip Dask tests on GPU builds @jameslamb (#5741)
- [python-package] fix _LGBM_CustomMetricFunction typing @Deimos357 (#5736)
- [python-package] fix mypy errors about Dataset handle @jameslamb (#5722)
- [python-package] fix mypy errors about cv() internals @jameslamb (#5725)
- [python-package] change types in sklearn.py initializations @jameslamb (#5718)
- [python-package] [dask] fix mypy errors regarding predict_proba @IdoKendo (#5728)
- [python-package] fix mypy errors about custom eval functions in sklearn.py @jameslamb (#5721)
- [python-package] fix mypy error about log callback @jameslamb (#5720)
- [python-package] fix mypy errors around eval metrics in sklearn.py @jameslamb (#5719)
- [python-package] fix mypy errors about _InnerPredictor @jameslamb (#5714)
- [dask] fix mypy errors about padded eval_results @jameslamb (#5716)
- [python-package] fix mypy errors in compat.py and libpath.py @jameslamb (#5711)
- [python-package] add more type hints in sklearn.py @jameslamb (#5710)
- [python-package] add type hints on plotting code @jameslamb (#5708)
- [ci] [python-package] enforce 'twine check' and 'check-wheel-contents' on Python distributions @jameslamb (#5707)
- [python-package] declare the use of inline type hints @jameslamb (#5709)
- [python-package] remove some inner function definitions @jameslamb (#5704)
- [python-package] add more type hints @jameslamb (#5694)
- [ci] restore flake8 checks @jameslamb (#5703)
- [python-package] Fix mypy errors for fit() incompatible signature @IdoKendo (#5679)
- [R-package] discourage use of regex for fixed string comparisons @jameslamb (#5685)
- [ci] get clang-15 from bookworm repository (fixes #5688) @jameslamb (#5689)
- [ci] speed up conda setup for some jobs @jameslamb (#5668)
- bump development version to v3.3.5.99 @jameslamb (#5681)
- [python-package] Fix mypy errors for predict() method @IdoKendo (#5678)
- [python-package] fix mypy errors for eval_at and basic @IdoKendo (#5674)
- [ci] [python-package] enforce flake8 checks (fixes #5566) @jameslamb (#5659)
- [python-package] remove `Dataset.feature_penalty` @jameslamb (#5656)
- [ci] fix brew errors in R macOS jobs (fixes #5670) @jameslamb (#5671)
- [ci] run r-package Linux jobs in containers @jameslamb (#5638)

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jun 29, 2023

/gha run r-valgrind

Workflow R valgrind tests has been triggered! 🚀
https://github.com/microsoft/LightGBM/actions/runs/5416400117

Status: failure ❌.

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jul 4, 2023

/gha run r-valgrind

Workflow R valgrind tests has been triggered! 🚀
https://github.com/microsoft/LightGBM/actions/runs/5456202408

Status: success ✔️.

@jameslamb jameslamb changed the title WIP: Release v4.0.0 Release v4.0.0 Jul 6, 2023
@jameslamb jameslamb marked this pull request as ready for review July 6, 2023 03:46
@jameslamb
Copy link
Collaborator Author

I would like to get #5948 in if possible, to make the changes in this version clearer in the docs.

But otherwise, I think this is ready!

@guolinke @shiyu1994 @jmoralez could you please look at this and #5153, and:

  • approve this PR if you believe the v4.0.0 release is ready
  • comment on [RFC] 4.0.0 Release #5153 if you feel strongly that some other change(s) need to be merged before we release

If we can get v4.0.0 out in the next week, I'll announce it at SciPy 2023 (link) next week if I'm able to get a lightning talk slot there 😊

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jul 6, 2023

/gha run r-valgrind

Workflow R valgrind tests has been triggered! 🚀
https://github.com/microsoft/LightGBM/actions/runs/5471362730

Status: success ✔️.

Copy link
Collaborator

@jmoralez jmoralez left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🚀

@jameslamb
Copy link
Collaborator Author

Thanks!

@shiyu1994 what do you think? Can I move forward with releasing v4.0.0?

Also note that CRAN will not be accepting submissions from July 21-August 7... I hope we can get this submitted before then.

per https://cran.r-project.org/

This year's CRAN submission summer break will be from Jul 21, 2023 to Aug 7, 2023.

@shiyu1994
Copy link
Collaborator

I'm OK with that. Let's release a 4.0.0 version first. I still have several on going PRs. But let's left them in subsequent releases (hopefully very soon).

@jameslamb
Copy link
Collaborator Author

Ok great, thank you so much!

Yeah nothing prevents us from doing a 4.0.1 or 4.1.0 shortly after this... I would actually love to see the project get into the habit of more frequent, smaller releases 😊

I will work on the remaining release tasks over the next few days.

@jameslamb
Copy link
Collaborator Author

Merged in master to get the fix from #5965. Because that's about destruction of objects, I'm going to run the valgrind checks again.

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jul 12, 2023

/gha run r-valgrind

Workflow R valgrind tests has been triggered! 🚀
https://github.com/microsoft/LightGBM/actions/runs/5533840576

Status: success ✔️.

@jameslamb
Copy link
Collaborator Author

Add release branch to RTD versions, trigger a new build, check docs

Most recent build (20 minutes ago) passed on this branch: https://readthedocs.org/projects/lightgbm/builds/21280161/

And the docs look to me:

@jameslamb
Copy link
Collaborator Author

manually test on M1/M2 mac

I'm traveling right now and don't have access to my M2 Mac laptop, but I just submitted the R package to the mac-builder testing service. The results will show up here: https://mac.r-project.org/macbuilder/results/1689263124-d89c2bc709b122ea/

If that passes, I'll merge this PR and officially create the v4.0.0 release 😁

@jameslamb
Copy link
Collaborator Author

Wow that was VERY fast... running R CMD check --as-cran on an M1 Mac on mac-builder already completed, and succeeded 🎉

...
* checking for portable use of $(BLAS_LIBS) and $(LAPACK_LIBS) ... OK
* checking use of PKG_*FLAGS in Makefiles ... OK
* checking compiled code ... OK
* checking installed files from ‘inst/doc’ ... OK
* checking files in ‘vignettes’ ... OK
* checking examples ... [1s/1s] OK
* checking for unstated dependencies in ‘tests’ ... OK
* checking tests ... [9s/10s] OK
  Running ‘testthat.R’ [9s/10s]
* checking for unstated dependencies in vignettes ... OK
* checking package vignettes in ‘inst/doc’ ... OK
* checking running R code from vignettes ... [0s/0s] NONE
  ‘basic_walkthrough.Rmd’ using ‘UTF-8’... [0s/0s] OK
* checking re-building of vignette outputs ... [4s/5s] OK
* checking PDF version of manual ... [3s/3s] OK
* DONE
Status: 1 NOTE
* using check arguments '--no-clean-on-error '

* elapsed time (check, wall clock): 3:39

I'll begin with the release shortly.

@jameslamb jameslamb merged commit d73c6b5 into master Jul 13, 2023
@jameslamb jameslamb deleted the release/v4.0.0 branch July 13, 2023 16:37
@jameslamb
Copy link
Collaborator Author

I just published the v4.0.0 release on GitHub: https://github.com/microsoft/LightGBM/releases/tag/v4.0.0

That triggered this Azure DevOps build to build all the artifacts: https://dev.azure.com/lightgbm-ci/lightgbm-ci/_build/results?buildId=14917&view=results

Once that's done, I'll publish wheels to PyPI and submit the R package to CRAN.

@jameslamb
Copy link
Collaborator Author

Made changes at https://readthedocs.org/projects/lightgbm/versions/.

Remove the release branch from RTD versions

Done.

Add new tag to RTD versions and trigger a new build

Done.

successful build: https://readthedocs.org/projects/lightgbm/builds/21293502/

docs: https://lightgbm.readthedocs.io/en/v4.0.0/

@jameslamb
Copy link
Collaborator Author

All artifacts have been built successfully, and attached to the release: https://github.com/microsoft/LightGBM/releases

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jul 13, 2023

Updated stable tag to point to the same commit as v4.0.0.

git fetch upstream --tags
git tag -d stable
git push upstream :refs/tags/stable
git tag stable v4.0.0
git push upstream stable

https://github.com/microsoft/LightGBM/tags

Screen Shot 2023-07-13 at 3 40 17 PM

@jameslamb
Copy link
Collaborator Author

The CRAN release is currently blocked. Getting the following error when trying to upload it.

Unpacking failed. Please make sure the tar.gz was created with R CMD build.

https://twitter.com/henrikbengtsson/status/1679036506382430209

From https://cran.r-project.org/

CRAN submissions are currently partly not possible due to some infrastructure issues. Please so NOT conact us if you see "Unpacking failed. Please make sure the tar.gz was created with R CMD build....". CRAN sysadmins will work on it once they are available.

🙃 I'll check back later tonight, and keep checking until CRAN is available again.

@jameslamb
Copy link
Collaborator Author

jameslamb commented Jul 13, 2023

Uploaded v4.0.0 wheels and sdist to PyPI 😁

https://pypi.org/project/lightgbm/4.0.0/

Screen Shot 2023-07-13 at 4 20 06 PM

https://pypi.org/project/lightgbm/4.0.0/#files

image

@jameslamb
Copy link
Collaborator Author

CRAN is accepting submissions again 🎉

I just submitted the R package there.

image

@shiyu1994 please check your email, they will send you a link to click on to confirm the submission.

@jameslamb
Copy link
Collaborator Author

Homebrew recipe was updated in Homebrew/homebrew-core#136562.

@jameslamb
Copy link
Collaborator Author

@shiyu1994 please check your email for a message from CRAN.

Sorry to @ you again, but we only have a few more days until CRAN stops all work from July 21-August 7 for their summer break.

@jameslamb
Copy link
Collaborator Author

For those subscribed here waiting for v4.0.0 of the R package to make it to CRAN... our 2 attempted submissions this week were not accepted. CRAN is on its summer break starting today (see notice at https://cran.r-project.org/) and ending August 7. We will try to resolve the issues (#5987) and submit to CRAN again when they return.

If you'd like to test out v4.0.0 of the package prior to that, you can follow the instructions at https://github.com/microsoft/LightGBM/tree/master/R-package#installation or install from the source distribution hosted here on GitHub.

install.packages("https://github.com/microsoft/LightGBM/releases/download/v4.0.0/lightgbm-4.0.0-r-cran.tar.gz")

Sorry for the inconvenience, we're doing the best we can 😅

@jameslamb jameslamb mentioned this pull request Jul 21, 2023
60 tasks
@jameslamb
Copy link
Collaborator Author

✅ uploaded LightGBM.4.0.0.nupkg to NuGet: https://www.nuget.org/packages/LightGBM/4.0.0

@jameslamb jameslamb mentioned this pull request Sep 5, 2023
19 tasks
@joeycouse
Copy link

Any idea on when 4.0 will make it to CRAN? Thanks for the amazing work!

@jameslamb
Copy link
Collaborator Author

As it says a few comments up (#5952 (comment)), you can subscribe to #5987 to track the progress.

@jameslamb jameslamb mentioned this pull request Nov 14, 2023
24 tasks
Copy link

This pull request has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.

@github-actions github-actions bot locked as resolved and limited conversation to collaborators Dec 27, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[RFC] 4.0.0 Release [python-package] LightGBM wheel on PyPI seems to have the wrong platform tag
5 participants