diff --git a/python/paddle/trainer/config_parser.py b/python/paddle/trainer/config_parser.py
index fc112f1327f5ad5f1bdd04873394b1fa0e761e29..5477158ecb8646992ebdded0b15cce50720ebf36 100644
--- a/python/paddle/trainer/config_parser.py
+++ b/python/paddle/trainer/config_parser.py
@@ -2055,8 +2055,7 @@ class BatchNormLayer(LayerBase):
         # Automatically select cudnn_batch_norm for GPU and batch_norm for CPU.
         # Also based on cudnn version.
         use_cudnn = use_gpu and batch_norm_type != "batch_norm" and \
-            ((not parallel_nn) or self.config.device > -1) and \
-            cudnn_version >= 4007
+                ((not parallel_nn) or self.config.device > -1)
         self.layer_type = "cudnn_batch_norm" if use_cudnn else "batch_norm"
         super(BatchNormLayer, self).__init__(
             name, self.layer_type, 0, inputs=inputs, **xargs)
diff --git a/python/paddle/v2/__init__.py b/python/paddle/v2/__init__.py
index 07ab2c9b1898f0ec7a5ca168912f2f03597b094a..5bea980611904b37a4a5d4e2cbbee13503a61ff0 100644
--- a/python/paddle/v2/__init__.py
+++ b/python/paddle/v2/__init__.py
@@ -34,6 +34,7 @@ import minibatch
 import plot
 import image
 import model
+import paddle.trainer.config_parser as cp
 
 __all__ = [
     'optimizer',
@@ -58,6 +59,8 @@ __all__ = [
     'model',
 ]
 
+cp.begin_parse()
+
 
 def init(**kwargs):
     import py_paddle.swig_paddle as api
@@ -73,6 +76,11 @@ def init(**kwargs):
     for key in args_dict.keys():
         args.append('--%s=%s' % (key, str(args_dict[key])))
 
+    if 'use_gpu' in kwargs:
+        cp.g_command_config_args['use_gpu'] = kwargs['use_gpu']
+    assert 'parallel_nn' not in kwargs, ("currently 'parallel_nn' is not "
+                                         "supported in v2 APIs.")
+
     api.initPaddle(*args)
 
 
diff --git a/python/paddle/v2/layer.py b/python/paddle/v2/layer.py
index 4ade1c6f329ae39769279963af6809f938807bdd..6a2bb8d337b7667aa2b1e3ef0815bb80f6e38d6a 100644
--- a/python/paddle/v2/layer.py
+++ b/python/paddle/v2/layer.py
@@ -324,6 +324,3 @@ def parse_network(output_layers, extra_layers=None):
 
 def get_layer(name):
     return config_base.__layer_map__.get(name)
-
-
-cp.begin_parse()