diff --git a/python/paddle/dataset/flowers.py b/python/paddle/dataset/flowers.py
index fee127dfb35956bd582f25e2ead913cb7ebf0e03..832eeacff9b57ccbd9b8533404c3d48ce4d06f17 100644
--- a/python/paddle/dataset/flowers.py
+++ b/python/paddle/dataset/flowers.py
@@ -43,10 +43,10 @@ from multiprocessing import cpu_count
 __all__ = ['train', 'test', 'valid']
 
 
-DATA_URL = 'http://www.robots.ox.ac.uk/~vgg/data/flowers/102/102flowers.tgz'
-LABEL_URL = 'http://www.robots.ox.ac.uk/~vgg/data/flowers/102/imagelabels.mat'
-SETID_URL = 'http://www.robots.ox.ac.uk/~vgg/data/flowers/102/setid.mat'
-DATA_MD5 = '33bfc11892f1e405ca193ae9a9f2a118'
+DATA_URL = 'http://paddlemodels.cdn.bcebos.com/flowers/102flowers.tgz'
+LABEL_URL = 'http://paddlemodels.cdn.bcebos.com/flowers/imagelabels.mat'
+SETID_URL = 'http://paddlemodels.cdn.bcebos.com/flowers/setid.mat'
+DATA_MD5 = '52808999861908f626f3c1f4e79d11fa'
 LABEL_MD5 = 'e0620be6f572b9609742df49c70aed4d'
 SETID_MD5 = 'a5357ecc9cb78c4bef273ce3793fc85c'
 # In official 'readme', tstid is the flag of test data