Home » Oreps » Light Gbm » All Releases » v2.0 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
Basic Details Issues Addressed Top Contributors Directory Browser Release Notes
Light Gbm v2.0

Light Gbm: v2.0 Release

Release date:
May 3, 2017
Previous version:
v1 (released April 24, 2017)
Magnitude:
15,592 Diff Delta
Contributors:
13 total committers
Data confidence:
Commits:
112 impacting commits

112 Commits in this Release

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

update to v2.
Authored February 20, 2017
[R-package] Improvements, readability, and bug fixes (#378)
Authored April 3, 2017
remove some files.
Authored April 24, 2017
fix bugs for parallel learning.
Authored March 13, 2017
add time tags (#322)
Authored February 27, 2017
support OVA multi-classification.
Authored March 29, 2017
refine api of constructing from sampling data.
Authored March 21, 2017
slight speed up the iterator of sparse bin.
Authored February 28, 2017
add 4bits_bin
Authored February 22, 2017
support multi-threading exceptions.
Authored March 28, 2017
add FitByExistingTree.
Authored March 25, 2017
change to boost from average output.
Authored March 22, 2017
add data_splitter to cv (#298)
Authored February 18, 2017
fix subset bug.
Authored April 26, 2017
skip the training of empty class in classification.
Authored March 30, 2017
Add feature contribution computing (#323)
Authored February 27, 2017
fix a bug in multi-class
Authored March 7, 2017
revert the multi-class loss.
Authored April 17, 2017
[R-package] Add $raw to LightGBM (#347)
Authored March 17, 2017
eliminate all VS2017 warnings (#367)
Authored March 28, 2017
only support boost_from_average in regression.
Authored March 30, 2017
fix #200 (#312)
Authored February 23, 2017
refine early stopping and add a test case (#369)
Authored March 28, 2017
unrolling loop of histogram buliding in order_bin.
Authored February 26, 2017
fix bug in finding best split for categorical feature.
Authored March 3, 2017
fix bug in data/voting parallel.
Authored March 3, 2017
Switch RMSE to MSE (true L2 loss) (#408)
Authored April 13, 2017
boost_from_average seems doesn't work well in binary classification. remove it.
Authored March 24, 2017
add un_balance to multi class (#314)
Authored March 1, 2017
fix #287
Authored February 24, 2017
Support build self-contained R package. (#340)
Authored March 10, 2017
fix bug in filter bin
Authored March 22, 2017
fix prediction of ova multi-class.
Authored March 31, 2017
Add Feature Contribution Plot (#330)
Authored March 3, 2017
remove the useless training.
Authored March 7, 2017
robust tree model loading.
Authored March 24, 2017
not use exception in command line due to catch-miss of multi-threading exceptions.
Authored March 13, 2017
Squashed commit of the following:
Authored February 22, 2017
add LGBM_DatasetCreateFromSampledMat api.
Authored March 14, 2017
limit the max tree output. change hessian in multi-class objective.
Authored March 24, 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
7,933 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
22,534 Ξ”
v2.0 Released May 3, 2017
15,592 Ξ”
v1 Released April 24, 2017
0 Ξ”

Top Contributors in v2.0

guolinke
Laurae2
wxchan
yanyachen
zhangyafeikimi
henry0312
JayveeHe
chivee
Xivid
alexander-rakhlin

Directory Browser for v2.0

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

Release Notes Published

A release before GPU support.

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