diff --git a/dataset/README.txt b/dataset/README.txt
index 73bd43b..1f428ca 100644
--- a/dataset/README.txt
+++ b/dataset/README.txt
@@ -5,3 +5,5 @@ Utilities in this folder are for preparing data from
 PartA of the ShanghaiTech dataset (Y. Zhang et. al,
 Single-image crowd counting via multi-column 
 convolutional neural network. CVPR 2016.)
+
+The link to the ShanghaiTech dataset can be found [here](https://www.dropbox.com/s/fipgjqxl7uj8hd5/ShanghaiTech.zip?dl=0).