diff --git a/demo/slimfacenet/README.md b/demo/slimfacenet/README.md
index f91de459ce970367846aed0009fe92a3b3612e5c..d7e6d273270a326860b5c9fe7ae5d9320b6517db 100755
--- a/demo/slimfacenet/README.md
+++ b/demo/slimfacenet/README.md
@@ -2,6 +2,8 @@
 
 本示例将演示如何训练`slimfacenet`及评测`slimfacenet`量化模型。
 
+本示例依赖：Paddle 1.8 PaddleSlim 1.1.1
+
 当前示例支持以下人脸识别模型：
 
 - `SlimFaceNet_A_x0_60`
diff --git a/demo/slimfacenet/train_eval.py b/demo/slimfacenet/train_eval.py
index 421496f0463ce99a48b19f483f970f44ceb022dd..a779feb390ce70df814a08a8608053a6c0756022 100644
--- a/demo/slimfacenet/train_eval.py
+++ b/demo/slimfacenet/train_eval.py
@@ -233,9 +233,7 @@ def quant_val_reader_batch():
     test_dataset = LFW(nl, nr)
     test_reader = paddle.batch(
         test_dataset.reader, batch_size=1, drop_last=False)
-    shuffle_index = args.seed if args.seed else np.random.randint(1000)
-    print('shuffle_index: {}'.format(shuffle_index))
-    shuffle_reader = fluid.io.shuffle(test_reader, shuffle_index)
+    shuffle_reader = fluid.io.shuffle(test_reader, 3)
 
     def _reader():
         while True:
@@ -289,7 +287,6 @@ def main():
         '--start_epoch', default=0, type=int, help='start_epoch')
     parser.add_argument(
         '--total_epoch', default=80, type=int, help='total_epoch')
-    parser.add_argument('--seed', default=None, type=int, help='shuffle seed')
     parser.add_argument(
         '--save_frequency', default=1, type=int, help='save_frequency')
     parser.add_argument(
@@ -336,8 +333,10 @@ def main():
             sample_generator=quant_val_reader_batch(),
             model_filename=None,  #'model',
             params_filename=None,  #'params',
-            batch_size=100,
-            batch_nums=10)
+            save_model_filename=None,  #'model',
+            save_params_filename=None,  #'params',
+            batch_size=np.random.randint(80, 160),
+            batch_nums=np.random.randint(4, 10))
     elif args.action == 'test':
         [inference_program, feed_target_names,
          fetch_targets] = fluid.io.load_inference_model(