Home » Oreps » Light Gbm » All Releases » v2.0.5 Release
microsoft/lightgbm
A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.
Watch
0
Commit Activity
Directory Browser
Velocity History
Releasesv2.0.5
Basic Details Issues Addressed Top Contributors Directory Browser
Light Gbm v2.0.5

Light Gbm: v2.0.5 Release

Release date:
August 18, 2017
Previous version:
v2.0.4 (released June 30, 2017)
Magnitude:
7,158 Diff Delta
Contributors:
17 total committers
Data confidence:
Commits:
84 impacting commits

84 Commits in this Release

Ordered by the degree to which they evolved the repo in this version.

Optimal split for categorical feature (#762)
Authored August 2, 2017
Better missing value handle (#747)
Authored July 30, 2017
Add Random Forest Mode (#678)
Authored July 10, 2017
cross entropy metrics and objective (#685)
Authored July 17, 2017
clean code for tree learner.
Authored June 30, 2017
clean code for Boosting::ResetTrainingData.
Authored July 4, 2017
clean code for config.
Authored July 9, 2017
Update docs (#776)
Authored August 3, 2017
[R-package] CV Improvements, fix #390 (#791)
Authored August 7, 2017
fix R crash problem. (#784)
Authored August 5, 2017
Force checking the label used in lambdarank.
Authored July 21, 2017
add more documents.
Authored August 2, 2017
fix bug for csc prediction.
Authored July 6, 2017
[R-package] Fallback to MinGW when VS build fails (#664)
Authored July 13, 2017
[python] Dataset params back up before training (#786)
Authored August 5, 2017
[python] refine: solve several trivial issues (#753)
Authored August 3, 2017
update the calculation of cat_smooth .
Authored August 2, 2017
[R] update build commands.
Authored July 14, 2017
[python] add keep_training_booster (#673)
Authored July 7, 2017
Customized average fixes (nullptr and network) (#739)
Authored July 28, 2017
Added method for customizable "boost-from-average" (#731)
Authored July 27, 2017
Update key events to 15th July 2017 (#693)
Authored July 16, 2017
update documents.
Authored August 2, 2017
[python] update build command in setup.py
Authored July 14, 2017
fix a bug for the label_index
Authored July 27, 2017
fix bug in parallel bin finding .
Authored July 23, 2017
Remove CMAKE_CXX_COMPILER and CMAKE_C_COMPILER for OSX (#719)
Authored July 22, 2017
Update R's readme
Authored July 5, 2017
[python package] added get_ref_chain function to Dataset object, used it in train() to… (#745)
Authored July 31, 2017
add docs for if-else (#785)
Authored August 4, 2017
fix a potential bug in tree prediction.
Authored August 4, 2017
add reference to GPU acceleration paper (#756)
Authored July 31, 2017
[python] fix categorical_feature parameter (#792)
Authored August 7, 2017
Cannot use negative number for categorical feature.
Authored August 4, 2017
Update CMakeLists.txt
Authored July 22, 2017
added test for training when both train and valid are subsets of a si… (#759)
Authored August 2, 2017
fix model feature importances (#755)
Authored August 1, 2017
Fatal if set is_unbalance and scale_pos_weight at the same time
Authored July 28, 2017
fix subset params (#676)
Authored July 7, 2017
Update gbdt.cpp
Authored August 1, 2017

Browse Other Releases

Latest Pending Unreleased 😎
v2.0.11 Released November 25, 2017
3,107 Ξ”
v2.0.10 Released October 18, 2017
514 Ξ”
v2.0.8 Released October 14, 2017
7,937 Ξ”
v2.0.7 Released September 28, 2017
10,618 Ξ”
v2.0.6 Released August 18, 2017
1,870 Ξ”
v2.0.5 Released August 18, 2017
7,158 Ξ”
v2.0.4 Released June 30, 2017
703 Ξ”
v2.0.3 Released June 25, 2017
22,538 Ξ”
v2.0 Released May 3, 2017
15,592 Ξ”
v1 Released April 24, 2017
0 Ξ”

Top Contributors in v2.0.5

guolinke
olofer
Laurae2
wxchan
henry0312
smooth-brain-mark
chivee
huanzhang12
aamster
i3v

Directory Browser for v2.0.5

We haven't yet finished calculating and confirming the files and directories changed in this release. Please check back soon.

GitClear uses cookies to ensure you get the best experience on our website. Learn more