diff --git a/mindspore/dataset/engine/datasets.py b/mindspore/dataset/engine/datasets.py
index 3d660d58a8d..4e0b0827347 100644
--- a/mindspore/dataset/engine/datasets.py
+++ b/mindspore/dataset/engine/datasets.py
@@ -394,6 +394,8 @@ class Dataset:
             The order of using repeat and batch reflects the number of batches. Recommend that
             repeat operation should be used after batch operation.
             If dataset_sink_mode is False, here repeat operation is invalid.
+            If dataset_sink_mode is True, repeat count should be euqal to the epoch of training. Otherwise,
+            errors could occur since the amount of data is not the amount training requires.
 
         Args:
             count (int): Number of times the dataset should be repeated (default=None).
diff --git a/mindspore/train/model.py b/mindspore/train/model.py
index a1acec859c3..7604b8ac389 100755
--- a/mindspore/train/model.py
+++ b/mindspore/train/model.py
@@ -362,6 +362,9 @@ class Model:
 
         Note:
             CPU is not supported when dataset_sink_mode is true.
+            If dataset_sink_mode is True, epoch of training should be equal to the count of repeat
+            operation in dataset processing. Otherwise, errors could occur since the amount of data
+            is not the amount training requires.
 
         Args:
             epoch (int): Total number of iterations on the data.