From 406c70ba0e831babce4855d48793df6924e21cbf Mon Sep 17 00:00:00 2001
From: Jiaming Yuan <jm.yuan@outlook.com>
Date: Tue, 12 Oct 2021 19:10:18 +0800
Subject: [PATCH] [doc] Fix typo. [skip ci] (#7311)

---
 doc/tutorials/categorical.rst | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/doc/tutorials/categorical.rst b/doc/tutorials/categorical.rst
index 6ee724d45b6a..9598512e34d0 100644
--- a/doc/tutorials/categorical.rst
+++ b/doc/tutorials/categorical.rst
@@ -50,7 +50,7 @@ can plot the model and calculate the global feature importance:
   # Get a graph
   graph = xgb.to_graphviz(clf, num_trees=1)
   # Or get a matplotlib axis
-  ax = xgb.plot_tree(reg, num_trees=1)
+  ax = xgb.plot_tree(clf, num_trees=1)
   # Get feature importances
   clf.feature_importances_
 
@@ -60,8 +60,8 @@ idea is create dataframe with category feature type, and tell XGBoost to use ``g
 with parameter ``enable_categorical``.  See `this demo
 <https://github.com/dmlc/xgboost/blob/master/demo/guide-python/categorical.py>`_ for a
 worked example using categorical data with ``scikit-learn`` interface.  For using it with
-the Kaggle tutorial dataset, see `<this demo
-https://github.com/dmlc/xgboost/blob/master/demo/guide-python/cat_in_the_dat.py>`_
+the Kaggle tutorial dataset, see `this demo
+<https://github.com/dmlc/xgboost/blob/master/demo/guide-python/cat_in_the_dat.py>`_
 
 
 **********************
@@ -114,5 +114,5 @@ Next Steps
 **********
 
 As of XGBoost 1.5, the feature is highly experimental and have limited features like CPU
-training is not yet supported.  Please see `<this issue>
-https://github.com/dmlc/xgboost/issues/6503`_ for progress.
+training is not yet supported.  Please see `this issue
+<https://github.com/dmlc/xgboost/issues/6503>`_ for progress.