fix loop size error when exec train after model init

mindspore/train/dataset_helper.py

@@ -83,12 +83,12 @@ class DatasetHelper:
 class _DatasetIter:
     """Base iter for dataset help"""
     def __init__(self, dataset):
-        self.loop_size = 1
+        if not hasattr(dataset, '__loop_size__'):
+            self.loop_size = dataset.get_dataset_size()
+        else:
+            self.loop_size = dataset.__loop_size__
+
         if not hasattr(dataset, '__ME_INITED__'):
-            if not hasattr(dataset, '__loop_size__'):
-                self.loop_size = dataset.get_dataset_size()
-            else:
-                self.loop_size = dataset.__loop_size__
             dataset.__TRANSFER_DATASET__ = _exec_datagraph(dataset, self.loop_size)
             dataset.__ME_INITED__ = dataset.__TRANSFER_DATASET__.queue_name