diff --git a/README.md b/README.md
index 251fc58..585230d 100644
--- a/README.md
+++ b/README.md
@@ -4,6 +4,11 @@ A pytorch implementation of the "Selective Convolutional Descriptor Aggregation"
 ## NOTE
  cpu-only version
 
+ no in [1]
+ ```
+ train_data_L31a(i,:) = train_data_L31a(i,:) ./ norm(train_data_L31a(i,:));
+ ```
+
 
 [largestConnectComponent](https://blog.csdn.net/xuyangcao123/article/details/81023732)
 
@@ -15,13 +20,13 @@ install requirements
   pip install -r requirements.txt;
 ```
 
-On CUB and split dataset in 'CUB_200.py'.
+On CUB and split dataset in `CUB_200.py`.
 
 random split CUB-200-2011 results:
 ||top1|top5
 ---|:--:|---:
-CUB|-|-
+CUB|0.46|0.71
 
 
 
->Wei X S , Luo J H , Wu J , et al. Selective Convolutional Descriptor Aggregation for Fine-Grained Image Retrieval[J]. IEEE Transactions on Image Processing, 2017, 26(6):2868-2881.
\ No newline at end of file
+>[1] Wei X S , Luo J H , Wu J , et al. Selective Convolutional Descriptor Aggregation for Fine-Grained Image Retrieval[J]. IEEE Transactions on Image Processing, 2017, 26(6):2868-2881.
\ No newline at end of file