diff --git a/fastNLP/core/tester.py b/fastNLP/core/tester.py
index b22f32ef..e45f1017 100644
--- a/fastNLP/core/tester.py
+++ b/fastNLP/core/tester.py
@@ -64,7 +64,7 @@ class BaseTester(Action):
         :return save_dev_data: list. Each entry is a sample, which is also a list of features and label(s).
         """
         if self.save_dev_data is None:
-            data_dev = _pickle.load(open(data_path + "/data_train.pkl", "rb"))
+            data_dev = _pickle.load(open(data_path + "/data_dev.pkl", "rb"))
             self.save_dev_data = data_dev
         return self.save_dev_data
 
diff --git a/fastNLP/loader/preprocess.py b/fastNLP/loader/preprocess.py
index fd378ba0..2c972ddd 100644
--- a/fastNLP/loader/preprocess.py
+++ b/fastNLP/loader/preprocess.py
@@ -95,8 +95,11 @@ class POSPreprocess(BasePreprocess):
         if not pickle_exist(pickle_path, "data_train.pkl"):
             data_train = self.to_index(data)
             if train_dev_split > 0 and not pickle_exist(pickle_path, "data_dev.pkl"):
-                data_dev = data_train[: int(len(data_train) * train_dev_split)]
+                split = int(len(data_train) * train_dev_split)
+                data_dev = data_train[: split]
+                data_train = data_train[split:]
                 save_pickle(data_dev, self.pickle_path, "data_dev.pkl")
+                print("{} of the training data is split for validation. ".format(train_dev_split))
             save_pickle(data_train, self.pickle_path, "data_train.pkl")
 
     def build_dict(self, data):
diff --git a/reproduction/chinese_word_seg/cws.cfg b/reproduction/chinese_word_seg/cws.cfg
index ded4f623..cdcb4496 100644
--- a/reproduction/chinese_word_seg/cws.cfg
+++ b/reproduction/chinese_word_seg/cws.cfg
@@ -1,5 +1,5 @@
 [train]
-epochs = 2
+epochs = 10
 batch_size = 32
 pickle_path = "./save/"
 validate = true
diff --git a/reproduction/chinese_word_seg/cws_train.py b/reproduction/chinese_word_seg/cws_train.py
index 6616ff5f..ff549eb9 100644
--- a/reproduction/chinese_word_seg/cws_train.py
+++ b/reproduction/chinese_word_seg/cws_train.py
@@ -15,7 +15,7 @@ from fastNLP.core.inference import Inference
 data_name = "pku_training.utf8"
 cws_data_path = "/home/zyfeng/data/pku_training.utf8"
 pickle_path = "./save/"
-data_infer_path = "data_for_tests/people_infer.txt"
+data_infer_path = "/home/zyfeng/data/pku_test.utf8"
 
 
 def infer():
@@ -59,7 +59,7 @@ def train():
     train_data = loader.load_pku()
 
     # Preprocessor
-    p = POSPreprocess(train_data, pickle_path)
+    p = POSPreprocess(train_data, pickle_path, train_dev_split=0.3)
     train_args["vocab_size"] = p.vocab_size
     train_args["num_classes"] = p.num_classes
 
diff --git a/test/test_cws.py b/test/test_cws.py
index 9d8f973c..8f6c1211 100644
--- a/test/test_cws.py
+++ b/test/test_cws.py
@@ -113,4 +113,4 @@ def train_test():
 
 if __name__ == "__main__":
     train_test()
-    infer()
+    #infer()