fix bug after high get_places and get_dims

fluid/DeepASR/data_utils/util.py

@@ -28,7 +28,7 @@ def to_lodtensor(data, place):
 def lodtensor_to_ndarray(lod_tensor):
     """conver lodtensor to ndarray
     """
-    dims = lod_tensor.get_dims()
+    dims = lod_tensor._get_dims()
     ret = np.zeros(shape=dims).astype('float32')
     for i in xrange(np.product(dims)):
         ret.ravel()[i] = lod_tensor.get_float_element(i)

fluid/DeepASR/model_utils/model.py

@@ -88,7 +88,7 @@ def stacked_lstmp_model(frame_dim,
         # When the execution place is specified to CUDAPlace, the program will
         # run on all $CUDA_VISIBLE_DEVICES GPUs. Otherwise the program will 
         # run on all CPU devices.
-        places = fluid.layers.get_places()
+        places = fluid.layers.device.get_places()
         pd = fluid.layers.ParallelDo(places)
         with pd.do():
             feat_ = pd.read_input(feature)

fluid/language_model/train_on_cloud.py

@@ -187,7 +187,7 @@ def do_train(train_reader,
                        init_low_bound, init_high_bound)
         avg_cost = fluid.layers.mean(x=cost)
     else:
-        places = fluid.layers.get_places()
+        places = fluid.layers.device.get_places()
         pd = fluid.layers.ParallelDo(places)
         with pd.do():
             cost = network(

fluid/sequence_tagging_for_ner/network_conf.py

@@ -109,7 +109,7 @@ def ner_net(word_dict_len, label_dict_len, parallel, stack_num=2):
         name="target", shape=[1], dtype='int64', lod_level=1)
 
     if parallel:
-        places = fluid.layers.get_places()
+        places = fluid.layers.device.get_places()
         pd = fluid.layers.ParallelDo(places)
         with pd.do():
             word_ = pd.read_input(word)

fluid/text_classification/train.py

@@ -34,7 +34,7 @@ def train(train_reader,
     if not parallel:
         cost, acc, prediction = network(data, label, len(word_dict))
     else:
-        places = fluid.layers.get_places(device_count=2)
+        places = fluid.layers.device.get_places(device_count=2)
         pd = fluid.layers.ParallelDo(places)
         with pd.do():
             cost, acc, prediction = network(