From fd4335d0bfb5795abeedf6f166bf3292d74fe5ed Mon Sep 17 00:00:00 2001 From: Jiaming Yuan Date: Sun, 13 Aug 2023 23:42:27 +0800 Subject: [PATCH] [doc] Document the current status of some features. (#9469) --- demo/guide-python/quantile_regression.py | 5 +++++ doc/tutorials/categorical.rst | 19 ++++++++++--------- doc/tutorials/multioutput.rst | 7 ++++++- 3 files changed, 21 insertions(+), 10 deletions(-) diff --git a/demo/guide-python/quantile_regression.py b/demo/guide-python/quantile_regression.py index 6d3e08df5..4f69a8c80 100644 --- a/demo/guide-python/quantile_regression.py +++ b/demo/guide-python/quantile_regression.py @@ -7,6 +7,11 @@ Quantile Regression The script is inspired by this awesome example in sklearn: https://scikit-learn.org/stable/auto_examples/ensemble/plot_gradient_boosting_quantile.html +.. note:: + + The feature is only supported using the Python package. In addition, quantile + crossing can happen due to limitation in the algorithm. + """ import argparse from typing import Dict diff --git a/doc/tutorials/categorical.rst b/doc/tutorials/categorical.rst index 2a84080cf..fb612bca1 100644 --- a/doc/tutorials/categorical.rst +++ b/doc/tutorials/categorical.rst @@ -4,16 +4,17 @@ Categorical Data .. note:: - As of XGBoost 1.6, the feature is experimental and has limited features + As of XGBoost 1.6, the feature is experimental and has limited features. Only the + Python package is fully supported. -Starting from version 1.5, XGBoost has experimental support for categorical data available -for public testing. For numerical data, the split condition is defined as :math:`value < -threshold`, while for categorical data the split is defined depending on whether -partitioning or onehot encoding is used. For partition-based splits, the splits are -specified as :math:`value \in categories`, where ``categories`` is the set of categories -in one feature. If onehot encoding is used instead, then the split is defined as -:math:`value == category`. More advanced categorical split strategy is planned for future -releases and this tutorial details how to inform XGBoost about the data type. +Starting from version 1.5, the XGBoost Python package has experimental support for +categorical data available for public testing. For numerical data, the split condition is +defined as :math:`value < threshold`, while for categorical data the split is defined +depending on whether partitioning or onehot encoding is used. For partition-based splits, +the splits are specified as :math:`value \in categories`, where ``categories`` is the set +of categories in one feature. If onehot encoding is used instead, then the split is +defined as :math:`value == category`. More advanced categorical split strategy is planned +for future releases and this tutorial details how to inform XGBoost about the data type. ************************************ Training with scikit-learn Interface diff --git a/doc/tutorials/multioutput.rst b/doc/tutorials/multioutput.rst index 983002aed..73e89fe03 100644 --- a/doc/tutorials/multioutput.rst +++ b/doc/tutorials/multioutput.rst @@ -11,6 +11,11 @@ can be simultaneously classified as both sci-fi and comedy. For detailed explan terminologies related to different multi-output models please refer to the :doc:`scikit-learn user guide `. +.. note:: + + As of XGBoost 2.0, the feature is experimental and has limited features. Only the + Python package is tested. + ********************************** Training with One-Model-Per-Target ********************************** @@ -49,7 +54,7 @@ Training with Vector Leaf .. note:: - This is still working-in-progress, and many features are missing. + This is still working-in-progress, and most features are missing. XGBoost can optionally build multi-output trees with the size of leaf equals to the number of targets when the tree method `hist` is used. The behavior can be controlled by the