Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Document GPU objectives in NEWS. (#3865)

Browse Source
This commit is contained in:
Jiaming Yuan 2018-11-05 14:46:45 +13:00 committed by GitHub
parent 91cc14ea70
commit 1698fe64bb
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
NEWS.md
View File

@ -23,6 +23,10 @@ This file records the changes in xgboost library in reverse chronological order.
* Mitigate tracker "thundering herd" issue on large cluster. Add exponential backoff retry when workers connect to tracker. * Mitigate tracker "thundering herd" issue on large cluster. Add exponential backoff retry when workers connect to tracker.
* With this change, we were able to scale to 1.5k executors on a 12 billion row dataset after some tweaks here and there. * With this change, we were able to scale to 1.5k executors on a 12 billion row dataset after some tweaks here and there.
### New feature: Additional objective functions for GPUs
* New objective functions ported to GPU: `hinge`, `multi:softmax`, `multi:softprob`, `count:poisson`, `reg:gamma`, `"reg:tweedie`.
* With supported objectives, XGBoost will select the correct devices based on your system and `n_gpus` parameter.
### Major bug fix: learning to rank with XGBoost4J-Spark ### Major bug fix: learning to rank with XGBoost4J-Spark
* Previously, `repartitionForData` would shuffle data and lose ordering necessary for ranking task. * Previously, `repartitionForData` would shuffle data and lose ordering necessary for ranking task.
* To fix this issue, data points within each RDD partition is explicitly group by their group (query session) IDs (#3654). Also handle empty RDD partition carefully (#3750). * To fix this issue, data points within each RDD partition is explicitly group by their group (query session) IDs (#3654). Also handle empty RDD partition carefully (#3750).
@ -33,6 +37,7 @@ This file records the changes in xgboost library in reverse chronological order.
### API changes ### API changes
* Column sampling by level (`colsample_bylevel`) is now functional for `hist` algorithm (#3635, #3862) * Column sampling by level (`colsample_bylevel`) is now functional for `hist` algorithm (#3635, #3862)
* GPU tag `gpu:` for regression objectives are now deprecated. XGBoost will select the correct devices automatically (#3643)
* Add `disable_default_eval_metric` parameter to disable default metric (#3606) * Add `disable_default_eval_metric` parameter to disable default metric (#3606)
* Experimental AVX support for gradient computation is removed (#3752) * Experimental AVX support for gradient computation is removed (#3752)
* XGBoost4J-Spark * XGBoost4J-Spark
@ -334,7 +339,7 @@ This version is only applicable for the Python package. The content is identical
- Compatibility fix for Python 2.6 - Compatibility fix for Python 2.6
- Call `print_evaluation` callback at last iteration - Call `print_evaluation` callback at last iteration
- Use appropriate integer types when calling native code, to prevent truncation and memory error - Use appropriate integer types when calling native code, to prevent truncation and memory error
- Fix shared library loading on Mac OS X - Fix shared library loading on Mac OS X
* R package: * R package:
- New parameters: - New parameters:
- `silent` in `xgb.DMatrix()` - `silent` in `xgb.DMatrix()`
@ -375,7 +380,7 @@ This version is only applicable for the Python package. The content is identical
- Support instance weights - Support instance weights
- Use `SparkParallelismTracker` to prevent jobs from hanging forever - Use `SparkParallelismTracker` to prevent jobs from hanging forever
- Expose train-time evaluation metrics via `XGBoostModel.summary` - Expose train-time evaluation metrics via `XGBoostModel.summary`
- Option to specify `host-ip` explicitly in the Rabit tracker - Option to specify `host-ip` explicitly in the Rabit tracker
* Documentation * Documentation
- Better math notation for gradient boosting - Better math notation for gradient boosting
- Updated build instructions for Mac OS X - Updated build instructions for Mac OS X