GitLab

The result of a right-shift of a signed negative number is
implementation-dependent.
cf. https://msdn.microsoft.com/en-us//library/336xbhcz.aspx

Fix #519
Fix #521

* Add external links and some news

* Add interactive parameters link

* fix GPU Windows for Read The Docs

* GPU Targets read the docs fix

* Update to R 3.4 and Performance Note

* Add benchmark docs

* Readme.md does not exist fix

https://docs.pytest.org/

http://www.sphinx-doc.org/en/stable/ext/autodoc.html#confval-autoclass_content

Homebrew updated gcc to 7.1.
https://github.com/Homebrew/homebrew-core/commit/0387fe46e9ce9fe4a14dbce9d31d70f5824ff830
https://github.com/Homebrew/homebrew-core/commit/0ab90d39e3ca786c9f0b2fb44c2fd29880336cd2

* make test fail

* change default best_iteration to 0

* fix test

* change data_splitter to folds in cv

* update docs

If scikit-learn isn't installed (and Read the Docs doesn't have it),
LGBMDeprecated aren't created and can't be imported.

* Update lgb.Booster.R

Added method to call LGBM_BoosterLoadModelFromString_R if model_str is provided for initialization, added option to load from model_str in lgb.load

* Update lightgbm_R.cpp

Adding LGBM_BoosterLoadModelFromString_R

* Update lightgbm_R.cpp

Added LGBM_BoosterSaveModelToString_R

* Update lightgbm_R.cpp

* Update lgb.Booster.R

Added save_model_to_string method

* Update lgb.Booster.R

Implemented @Laurae2 comments

* Update lgb.Booster.R

* Update lightgbm_R.h

Added load and save model from/to string exports

* docs: move files into general

* docs: initial commit for Read the Docs

http://docs.readthedocs.io/en/latest/getting_started.html

just running `sphinx-quickstart`

* docs: update conf.py

* docs: add requirements.txt

* docs: add README.md

* fix tests

* docs: put original docs back

* auto replace gbdt::prediction

* add test if_else_prediction

* not override gbdt_prediction.cpp

* close ifstream

* re-order .travis.yml

* translate model to if-else

* support multiclass and predictleaf

* remove java option for now

* support multi-thread

* add task:convert_model