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

Initial GPU support for the approx tree method. (#9414)

Browse Source
This commit is contained in:
Jiaming Yuan
2023-07-31 15:50:28 +08:00
committed by GitHub
parent 8f0efb4ab3
commit 912e341d57
23 changed files with 639 additions and 360 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
doc/parameter.rst
View File

@@ -162,7 +162,8 @@ Parameters for Tree Booster
- ``grow_colmaker``: non-distributed column-based construction of trees.
- ``grow_histmaker``: distributed tree construction with row-based data splitting based on global proposal of histogram counting.
- ``grow_quantile_histmaker``: Grow tree using quantized histogram.
- ``grow_gpu_hist``: Grow tree with GPU. Enabled when ``tree_method`` is set to ``hist`` along with ``device=cuda``.
- ``grow_gpu_hist``: Enabled when ``tree_method`` is set to ``hist`` along with ``device=cuda``.
- ``grow_gpu_approx``: Enabled when ``tree_method`` is set to ``approx`` along with ``device=cuda``.
- ``sync``: synchronizes trees in all distributed nodes.
- ``refresh``: refreshes tree's statistics and/or leaf values based on the current data. Note that no random subsampling of data rows is performed.
- ``prune``: prunes the splits where loss < min_split_loss (or gamma) and nodes that have depth greater than ``max_depth``.

32
doc/treemethod.rst
View File

@@ -123,23 +123,23 @@ Feature Matrix
Following table summarizes some differences in supported features between 4 tree methods,
`T` means supported while `F` means unsupported.
+------------------+-----------+---------------------+---------------------+------------------------+
| | Exact | Approx | Hist | Hist (GPU) |
+==================+===========+=====================+=====================+========================+
| grow_policy | Depthwise | depthwise/lossguide | depthwise/lossguide | depthwise/lossguide |
+------------------+-----------+---------------------+---------------------+------------------------+
| max_leaves | F | T | T | T |
+------------------+-----------+---------------------+---------------------+------------------------+
| sampling method | uniform | uniform | uniform | gradient_based/uniform |
+------------------+-----------+---------------------+---------------------+------------------------+
| categorical data | F | T | T | T |
+------------------+-----------+---------------------+---------------------+------------------------+
| External memory | F | T | T | P |
+------------------+-----------+---------------------+---------------------+------------------------+
| Distributed | F | T | T | T |
+------------------+-----------+---------------------+---------------------+------------------------+
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| | Exact | Approx | Approx (GPU) | Hist | Hist (GPU) |
+==================+===========+=====================+========================+=====================+========================+
| grow_policy | Depthwise | depthwise/lossguide | depthwise/lossguide | depthwise/lossguide | depthwise/lossguide |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| max_leaves | F | T | T | T | T |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| sampling method | uniform | uniform | gradient_based/uniform | uniform | gradient_based/uniform |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| categorical data | F | T | T | T | T |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| External memory | F | T | P | T | P |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
| Distributed | F | T | T | T | T |
+------------------+-----------+---------------------+------------------------+---------------------+------------------------+
Features/parameters that are not mentioned here are universally supported for all 4 tree
Features/parameters that are not mentioned here are universally supported for all 3 tree
methods (for instance, column sampling and constraints). The `P` in external memory means
special handling. Please note that both categorical data and external memory are
experimental.