GitLab

This reverts commit 656d2676.

* Turn matrix to storage mode "double"

* added "test_Predictor.R" to R tests with check for integer storage mode during prediction

* Explicit integers in test_Preditor to avoid TravisCI failure

* properly aligning comment on storage.mode
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* split double assignment into two lines
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* update number of features attribute

Fixes issue related to https://github.com/scikit-learn/scikit-learn/issues/17353 (see SLEP010 https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep010/proposal.html

).

* Update sklearn.py

* set public attribute in fit method

Reverted ```n_features``` property, and inserted the public attribute ```n_features_in_```.

* Update documentation

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

* [ci] make MiKTeX downloads more reliable

* switch to Powershell

* remove  extra slash

* fixing stuff

* revert install change

* Update .ci/test_r_package_windows.ps1
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* GitHub Actions

* ok

* fixing on list

* stuff

* directories

* directories

* things

* env variables

* working dir

* running a bunch of tasks

* more builds

* PATH

* actually use R task

* TASK

* be right, often

* doing stuff

* trying stuff

* more paths

* conda activate

* updating PATH

* trying bash

* where the hell is activate

* WHERE IS ACTIVATE

* set up conda

* more conda

* PLEASE WORK

* installing cpplint

* try r-package

* R version

* try windows job

* make windows work

* use powershell

* exe

* use conda

* conda init powershell

* different conda approach

* make it work

* cleaning up

* init powershell

* fixing windows

* more windows

* build directory

* no way right

* maybe it will work

* trying Visual Studio

* do this

* Windows is interesting

* put back check-output

* set compiler

* stuff

* more fixes

* fix the broken things

* updating jobs

* continuing

* poweshell is bad

* ok so maybe not powershell

* cmon now

* ok so

* fixing env variables

* maybe this

* MINGW job

* cleaning up

* conda init powershell

* moving more R stuff into GitHub Actions

* everything else

* use powershell

* cmon now powershell

* ttry to Continue

* override powershell

* peg MiKTeX URL

* what is happening

* try powershell -File

* trying stuff

* path

* more testing of output

* Matches uppercase

* more regex stuff

* this is getting ridiculous

* back to powershell I guess

* more commands

* this might work

* adding more reliable miktex download

* trying to download miktex

* installing httr

* fix error in MiKTeX script

* remove comments

* redirect output

* redirect output

* move linting back to Travis

* change redirection

* switch back to just mac and linux

* put linting exclude back

* renamed R_TRAVIS_LINUX

* revert changes to non-R tasks and update events

* simplify

* memory corruption fix for distributed data parallel version before SyncUpGlobalBestSplit

* updated based on comments

* updated voting and feature parallel based on comments

* fixing macos failure

* rename variable

* [R-package] fix issue where early stopping thinks higher MAPE is desirable (fixes #3099)

* fix linting

* only use main metrics

* fix tests

* [R-package] move R source files into R-package

* fix linting warning

* stuff

* [R-package] fix MSVC warning about control paths (fixes #3067)

* linting

* simplify

* Fix loss computation

* fix test

* Optimize ranking loss computation.

* redir log to python console

* fix pylint

* Apply suggestions from code review

* Update basic.py

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

* Update c_api.h

* Apply suggestions from code review

* Apply suggestions from code review

* super-minor: better wording
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: StrikerRUS <nekit94-12@hotmail.com>

* [R-package] [docs] replace build_r_site.R with pkgdown::build_site()

* fix paths

* fix continuation

* new_process arg

* bump XCode version at Travis

* try new AppleClang

* get back to gcc

* [R-package] [docs] fix warnings in pkgdown sitte  building

* trying to get better Travis logs

* trying to get better Travis logs

* undo testing fix

* undo testing fix

* empty commit

* [R-package] use processx to speed up builds

* different fix

* echoing more frequently

* fix it

* more stdout

* use print() instead of message()

* shell tricks

* fix message

* travis

* Update .ci/test_r_package.sh
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* [R-package] [docs] remove devtools dependency for docs builds

* linting

* updating docs conda env

* empty roxygenize()

* env for R libraries

* get logs

* remove comment

* added pkgdown URLs

* more paths

* fix incorrect YAML keys

* update DESCRIPTION URL link

* more URL changes

* revert DESCRIPTION changes

* remove ca-certificates

* empty commit

* Update docs/conf.py
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* empty commit

* remove outdated line in build_r.R
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* fix conflicts

* Update R-package/src/install.libs.R

* empty commit