diff --git a/03.image_classification/README.cn.md b/03.image_classification/README.cn.md
index d4fcef9cf0dcf44f8cab19916d36ee06483fce01..9a66a037279d1e275f48639ac5577c4dfeca6699 100644
--- a/03.image_classification/README.cn.md
+++ b/03.image_classification/README.cn.md
@@ -14,6 +14,7 @@
 3.文档和脚本中代码的一致性问题：
 请注意：为使本文更加易读易用，我们拆分、调整了train.py的代码并放入本文。本文中代码与train.py的运行结果一致，可直接运行[train.py](https://github.com/PaddlePaddle/book/blob/develop/03.image_classification/train.py)进行验证。
 
+4. PaddlePaddle版本：PaddlePaddle 1.6及以上版本或适当的develop版本。
 
 ## 背景介绍
 
diff --git a/03.image_classification/README.md b/03.image_classification/README.md
index 4704a4baea4b87f846253de4001e5494e702c34a..bdeea442a277aea218ca6bf2702016b8f8332f58 100644
--- a/03.image_classification/README.md
+++ b/03.image_classification/README.md
@@ -15,6 +15,10 @@ With Deep learning, image classification can be framed as a supervised or unsupe
 
 In this chapter, we introduce deep-learning-based image classification methods and explain how to train a CNN model using PaddlePaddle.
 
+## Requirement
+
+1. PaddlePaddle version 1.6 or higher, or suitable develop version.
+
 ## Result Demo
 
 Image Classification can be divided into general image classification and fine-grained image classification.
diff --git a/03.image_classification/index.cn.html b/03.image_classification/index.cn.html
index a0d6d2231e04f0ff2b4431c1e151a8d9172b1225..c41ceca4ba48139a40f4b6973659269c57f784a6 100644
--- a/03.image_classification/index.cn.html
+++ b/03.image_classification/index.cn.html
@@ -56,6 +56,7 @@
 3.文档和脚本中代码的一致性问题：
 请注意：为使本文更加易读易用，我们拆分、调整了train.py的代码并放入本文。本文中代码与train.py的运行结果一致，可直接运行[train.py](https://github.com/PaddlePaddle/book/blob/develop/03.image_classification/train.py)进行验证。
 
+4. PaddlePaddle版本：PaddlePaddle 1.6及以上版本或适当的develop版本。
 
 ## 背景介绍
 
diff --git a/03.image_classification/index.html b/03.image_classification/index.html
index 50ed16241ce0268ddb633eb883611845c51efcc7..e30ad0e9d2e381f9e797702ec8a442c809bc01cf 100644
--- a/03.image_classification/index.html
+++ b/03.image_classification/index.html
@@ -57,6 +57,10 @@ With Deep learning, image classification can be framed as a supervised or unsupe
 
 In this chapter, we introduce deep-learning-based image classification methods and explain how to train a CNN model using PaddlePaddle.
 
+## Requirement
+
+1. PaddlePaddle version 1.6 or higher, or suitable develop version.
+
 ## Result Demo
 
 Image Classification can be divided into general image classification and fine-grained image classification.