GitLab

* bring consistency and clearness into early_stopping_rounds desc, metric desc and implementation

* hotfix

* hotfix

* used NDCG as default metric for lambdarank task

* fixed missed methods at ReadTheDocs and changed default eval_metric

* leaved only unique metrics

* fixed comment

* added new aliases for params

* run helper/parameter_generator.py

* removed useless test

* Fix lgb.Dataset construct issue with 1:n sequence in R v 3.5

* Add tests for feature penalties

The document of `early_stopping_rounds` says it will check all of
eval_set. But, this is not true. It doesn't check the dataset
specified as the training data.

This change appends an extra phrase "except the training data" to all
of the sentences "If there's more than one, will check all of them" in
documents.

* Bump R version

* Bump CLI/Py version

* Roll back to 2.1.1 in R

* Fix Microsoft/LightGBM#1280

* Update lgb.Booster.R

* Update lgb.Booster.R

* Fixed misc CRAN issues

* Added additional details to R-package DESCRIPTION

* added code to expose c_api_pred_contrib in the R package

* removed Rprintf

* reverted to previous version of install.libs.R

objective funciton in reset_parameter()

Fixes error algb.params2str to lgb.params2str

* fix the bug

* fix R

* fix early stopping edge case

* fix message.

* fix tests

* fix GPU tests.

* use int64 for string

* [R] Fatal when CSC exceed int32.max

* Fix Microsoft/LightGBM#1001

* fixed typos and hotfixes

* converted gcc-tips.Rmd; added ref to gcc-tips

* renamed files

* renamed Advanced-Topics

* renamed README

* renamed Parameters-Tuning

* renamed FAQ

* fixed refs to FAQ

* fixed undecodable source characters

* renamed Features

* renamed Quick-Start

* fixed undecodable source characters in Features

* renamed Python-Intro

* renamed GPU-Tutorial

* renamed GPU-Windows

* fixed markdown

* fixed undecodable source characters in GPU-Windows

* renamed Parameters

* fixed markdown

* removed recommonmark dependence

* hotfixes

* added anchors to links

* fixed 404

* fixed typos

* added more anchors

* removed sphinxcontrib-napoleon dependence

* removed outdated line in Travis config

* fixed max-width of the ReadTheDocs theme

* added horizontal align to images

* Silent parameter does not exist anymore

* Fix Microsoft/LightGBM#913

* Generate Rd files for R documentation

fix typos in callback.R and lgb.cv.R, closing #946 and #947

* minor changes for robustness and efficiency

* fix charcter typo

* don't use lgb.is.* in filter

* revert to cat instead of message; don't use mget

* fix bugs in previous commits

* should be good to go now

* Add efficient demo for LightGBM

* Add bank dataset and more examples

* Add bank manual

Add Depth Column
Refine Code Structure

* Prepare merging parameters

* Add parameters from params for lgb.train

* Usage of params in lgb.cv for rounds/early stop

* Double bracket (1)

* Double brackets (2)

* Revert "[R-package] Prevent remembering parameters (#796)"

This reverts commit c795e2c8e12e8fe989bde0ab4ee59c2301325814.

* Use model load/unload trick for free-ing up memory

* Missing comma

* Hand copy manual update (1/2)

* Hand copy manual update (2/2)

* Enable simple cloning of dataset

* Add remodel logical to Dataset

* Allow remodeling of data

* Missing comma

* Use name instead of assignment for R6 class

* There should be no need to deep copy ALL objects

* Force fork

* Better random CV + Group + Info handling

* Trailing comma missing

* Switch group fold to nrow

* Switch comparison from non-atomic to atomic

* fix r random crash problem.

* fix R error msg.

* Try to store best scores in R package Microsoft/LightGBM#686

* Add binding for best_score (unlock environment)

* Fix double comma

* Add best_score to lgb.Booster