diff --git a/propeller/README.en.md b/propeller/README.en.md
index a8b9767f8ff900647d61b951776fd80fae9f0cce..a9a0b8185693fa06cc398c5973e229472919e9f6 100644
--- a/propeller/README.en.md
+++ b/propeller/README.en.md
@@ -18,11 +18,6 @@ Propeller provide the following benefits:
     -   create checkpoint files and recover from failures
     -   save visualizable results
 
-## install
-
-```script
-pip install --user .
-```
 
 ## Getting Started
 ```python
@@ -71,6 +66,7 @@ pip install --user .
     # Start training!
     propeller.train_and_eval(BowModel, hparams, run_config, train_ds, eval_ds)
 ```
+More detail see example/toy/
 
 ## Main Feature
 1. train_and_eval
diff --git a/propeller/README.md b/propeller/README.md
index 004edadc172e578ab7f3a0381c90c207d1313545..6a6175b29c360871eac00d741a6600c3a937fab2 100644
--- a/propeller/README.md
+++ b/propeller/README.md
@@ -19,10 +19,6 @@ Propeller 具有下列优势：
     -   创建检查点文件并从故障中恢复
     -   保存可视化的摘要结果
 
-## install|安装
-
-pip install --user .
-
 ## Getting Started|快速开始
 ```python
 
@@ -70,6 +66,7 @@ pip install --user .
     # 开始训练！
     propeller.train_and_eval(BowModel, hparams, run_config, train_ds, eval_ds)
 ```
+详细详细请见example/toy/
 
 ## 主要构件
 1. train_and_eval