diff --git a/doc/BERT_10_MINS.md b/doc/BERT_10_MINS.md
index 08ca0f31c27a26e3615e6e5ce70eb1d3c5c66d88..7a778558b34bba7f4e65ba729155493ea1a3b7c3 100644
--- a/doc/BERT_10_MINS.md
+++ b/doc/BERT_10_MINS.md
@@ -81,4 +81,3 @@ cat data.txt | python bert_client.py
 我们基于V100对基于Padde Serving研发的Bert-As-Service的性能进行测试并与基于Tensorflow实现的Bert-As-Service进行对比，从用户配置的角度，采用相同的batch size和并发数进行压力测试，得到4块V100下的整体吞吐性能数据如下。
 
 ![4v100_bert_as_service_benchmark](4v100_bert_as_service_benchmark.png)
-
diff --git a/doc/CUBE_QUANT.md b/doc/CUBE_QUANT.md
index 1693de79965d98ddcbde5b58dd40c6708a170de2..b191695aed247fcadcf10c4bfe3d72343d6d64d0 100644
--- a/doc/CUBE_QUANT.md
+++ b/doc/CUBE_QUANT.md
@@ -48,4 +48,3 @@ python test_client.py ctr_client_conf/serving_client_conf.prototxt ./raw_data
 ```
 
 Users can compare AUC results after quantization with AUC before quantization. 
-
diff --git a/doc/CUBE_QUANT_CN.md b/doc/CUBE_QUANT_CN.md
index 9f1c009151cd66f312e7d1939fb8ef868ceeacda..023f4d2fe246341688dd69d8978ee42817c7adfd 100644
--- a/doc/CUBE_QUANT_CN.md
+++ b/doc/CUBE_QUANT_CN.md
@@ -48,4 +48,3 @@ python test_client.py ctr_client_conf/serving_client_conf.prototxt ./raw_data
 ```
 
 用户可以将量化后的AUC结果同量化前的AUC做比较
-