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

Fix spelling in documents (#6948)

Browse Source 
* Update roxygen2 doc.

Co-authored-by: fis <jm.yuan@outlook.com>
This commit is contained in:
Andrew Ziem
2021-05-11 06:44:36 -06:00
committed by GitHub
parent 2a9979e256
commit 3e7e426b36
100 changed files with 284 additions and 284 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
doc/tutorials/dask.rst
View File

@@ -127,7 +127,7 @@ In previous example we used ``DaskDMatrix`` as input to ``predict`` function. I
practice, it's also possible to call ``predict`` function directly on dask collections
like ``Array`` and ``DataFrame`` and might have better prediction performance. When
``DataFrame`` is used as prediction input, the result is a dask ``Series`` instead of
array. Also, there's inplace predict support on dask interface, which can help reducing
array. Also, there's in-place predict support on dask interface, which can help reducing
both memory usage and prediction time.
.. code-block:: python
@@ -479,7 +479,7 @@ Here are some pratices on reducing memory usage with dask and xgboost.
``xgboost.dask.DaskDeviceQuantileDMatrix`` as a drop in replacement for ``DaskDMatrix``
to reduce overall memory usage. See ``demo/dask/gpu_training.py`` for an example.
- Use inplace prediction when possible.
- Use in-place prediction when possible.
References: