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

Column sampling at individual nodes (splits). (#3971)

Browse Source 
* Column sampling at individual nodes (splits).

* Documented colsample_bynode parameter.

- also updated documentation for colsample_by* parameters

* Updated documentation.

* GetFeatureSet() returns shared pointer to std::vector.

* Sync sampled columns across multiple processes.
This commit is contained in:
Andy Adinets
2018-12-14 15:37:35 +01:00
committed by Jiaming Yuan
parent e0a279114e
commit 42bf90eb8f
8 changed files with 140 additions and 80 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
doc/parameter.rst
View File

@@ -82,15 +82,22 @@ Parameters for Tree Booster
- Subsample ratio of the training instances. Setting it to 0.5 means that XGBoost would randomly sample half of the training data prior to growing trees. and this will prevent overfitting. Subsampling will occur once in every boosting iteration.
- range: (0,1]
* ``colsample_bytree`` [default=1]
- Subsample ratio of columns when constructing each tree. Subsampling will occur once in every boosting iteration.
- range: (0,1]
* ``colsample_bylevel`` [default=1]
- Subsample ratio of columns for each split, in each level. Subsampling will occur each time a new split is made.
- range: (0,1]
* ``colsample_bytree``, ``colsample_bylevel``, ``colsample_bynode`` [default=1]
- This is a family of parameters for subsampling of columns.
- All ``colsample_by*`` parameters have a range of (0, 1], the default value of 1, and
specify the fraction of columns to be subsampled.
- ``colsample_bytree`` is the subsample ratio of columns when constructing each
tree. Subsampling occurs once for every tree constructed.
- ``colsample_bylevel`` is the subsample ratio of columns for each level. Subsampling
occurs once for every new depth level reached in a tree. Columns are subsampled from
the set of columns chosen for the current tree.
- ``colsample_bynode`` is the subsample ratio of columns for each node
(split). Subsampling occurs once every time a new split is evaluated. Columns are
subsampled from the set of columns chosen for the current level.
- ``colsample_by*`` parameters work cumulatively. For instance,
the combination ``{'colsample_bytree':0.5, 'colsample_bylevel':0.5,
'colsample_bynode':0.5}`` with 64 features will leave 4 features to choose from at
each split.
* ``lambda`` [default=1, alias: ``reg_lambda``]