xgboost v0.71 Release Notes
Release Date: 2018-04-11 // almost 6 years ago-
- ๐ This is a minor release, mainly motivated by issues concerning
pip install
, e.g. #2426, #3189, #3118, and #3194. With this release, users of Linux and MacOS will be able to runpip install
for the most part. - ๐จ Refactored linear booster class (
gblinear
), so as to support multiple coordinate descent updaters (#3103, #3134). See BREAKING CHANGES below. - ๐ Fix slow training for multiclass classification with high number of classes (#3109)
- ๐ Fix a corner case in approximate quantile sketch (#3167). Applicable for 'hist' and 'gpu_hist' algorithms
- ๐ Fix memory leak in DMatrix (#3182)
- ๐ New functionality
- ๐ GPU support
- ๐ฆ Python package:
- Correctly handle parameter
verbose_eval=0
(#3115)
- Correctly handle parameter
- ๐ฆ R package:
- Eliminate segmentation fault on 32-bit Windows platform (#2994)
- ๐ฆ JVM packages
- ๐ฅ BREAKING CHANGES: Updated linear modelling algorithms. In particular L1/L2 regularisation penalties are now normalised to number of training examples. This makes the implementation consistent with sklearn/glmnet. L2 regularisation has also been removed from the intercept. To produce linear models with the old regularisation behaviour, the alpha/lambda regularisation parameters can be manually scaled by dividing them by the number of training examples.
- ๐ This is a minor release, mainly motivated by issues concerning