update 1.5.1 cn API (#1022)

* update 1.5.1 cn API

* add url

doc/fluid/api_cn/average_cn.rst

@@ -19,6 +19,7 @@ WeightedAverage
 
 .. code-block:: python
 
+            import paddle.fluid as fluid
             avg = fluid.average.WeightedAverage()
             avg.add(value=2.0, weight=1)
             avg.add(value=4.0, weight=2)

doc/fluid/api_cn/backward_cn.rst

@@ -36,6 +36,7 @@ append_backward
 
         # 网络配置
         # 损失计算
+        import paddle.fluid as fluid
         x = fluid.layers.data(name='x', shape=[13], dtype='float32')
         y = fluid.layers.data(name='y', shape=[1], dtype='float32') 
               

doc/fluid/api_cn/executor_cn.rst

@@ -267,6 +267,7 @@ scope_guard
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import numpy
 
     new_scope = fluid.Scope()

doc/fluid/api_cn/fluid_cn.rst

@@ -249,7 +249,7 @@ cpu_places
 
 创建 ``fluid.CPUPlace`` 对象列表。
 
-如果 ``device_count`` 为None，则设备数目将由环境变量 ``CPU_NUM`` 确定。如果未设置 ``CPU_NUM`` ，则设备数目将由 ``multiprocessing.cpu_count()`` 确定。
+如果 ``device_count`` 为None，则设备数目将由环境变量 ``CPU_NUM`` 确定。如果未设置 ``CPU_NUM`` ，则设备数目默认为1，也就是说， ``CPU_NUM`` =1。
 
 参数：
   - **device_count** (None|int) - 设备数目
@@ -262,6 +262,7 @@ cpu_places
 
 .. code-block:: python
 
+           import paddle.fluid as fluid
            cpu_places = fluid.cpu_places()
 
 
@@ -279,6 +280,7 @@ CPUPlace是设备的描述符。它代表一个CPU，可以访问CPUPlace对应
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         cpu_place = fluid.CPUPlace()
 
 
@@ -397,6 +399,7 @@ cuda_pinned_places
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         cuda_pinned_places_cpu_num = fluid.cuda_pinned_places()
         # 或者
         cuda_pinned_places = fluid.cuda_pinned_places(1)
@@ -428,6 +431,7 @@ cuda_places
 
 .. code-block:: python
 
+      import paddle.fluid as fluid
       cuda_places = fluid.cuda_places()
 
 .. _cn_api_fluid_CUDAPinnedPlace:
@@ -443,6 +447,7 @@ CUDAPinnedPlace是一个设备描述符，它所指代的存储空间可以被GP
 
 .. code-block:: python
 
+      import paddle.fluid as fluid
       place = fluid.CUDAPinnedPlace()
 
 .. _cn_api_fluid_CUDAPlace:
@@ -458,6 +463,7 @@ CUDAPlace是一个设备描述符，它代表一个GPU，并且每个CUDAPlace
 
 .. code-block:: python
 
+       import paddle.fluid as fluid
        gpu_place = fluid.CUDAPlace(0)
 
 
@@ -482,6 +488,7 @@ DataFeedDesc应由来自磁盘的有效protobuf消息初始化。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     f = open("data.proto", "w")
     print >> f, 'name: "MultiSlotDataFeed"'
     print >> f, 'batch_size: 2'
@@ -508,6 +515,7 @@ DataFeedDesc也可以在运行时更改。一旦你熟悉了每个字段的含
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data_feed = fluid.DataFeedDesc('data.proto')
     data_feed.set_batch_size(128)
     data_feed.set_dense_slots('wd')  # 名为'wd'的slot将被设置为密集的
@@ -534,6 +542,7 @@ DataFeedDesc也可以在运行时更改。一旦你熟悉了每个字段的含
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     f = open("data.proto", "w")
     print >> f, 'name: "MultiSlotDataFeed"'
     print >> f, 'batch_size: 2'
@@ -569,6 +578,7 @@ DataFeedDesc也可以在运行时更改。一旦你熟悉了每个字段的含
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     f = open("data.proto", "w")
     print >> f, 'name: "MultiSlotDataFeed"'
     print >> f, 'batch_size: 2'
@@ -606,6 +616,7 @@ DataFeedDesc也可以在运行时更改。一旦你熟悉了每个字段的含
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     f = open("data.proto", "w")
     print >> f, 'name: "MultiSlotDataFeed"'
     print >> f, 'batch_size: 2'
@@ -642,6 +653,7 @@ DataFeedDesc也可以在运行时更改。一旦你熟悉了每个字段的含
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     f = open("data.proto", "w")
     print >> f, 'name: "MultiSlotDataFeed"'
     print >> f, 'batch_size: 2'
@@ -993,6 +1005,7 @@ DistributeTranspiler
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   x = fluid.layers.data(name='x', shape=[13], dtype='float32')
   y = fluid.layers.data(name='y', shape=[1], dtype='float32')
   y_predict = fluid.layers.fc(input=x, size=1, act=None)
@@ -1053,6 +1066,7 @@ DistributeTranspiler
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     transpiler = fluid.DistributeTranspiler()
     t.transpile(
         trainer_id=0,
@@ -1162,6 +1176,7 @@ DistributeTranspiler
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     pserver_endpoints = "192.168.0.1:6174,192.168.0.2:6174"
     trainer_endpoints = "192.168.0.1:6174,192.168.0.2:6174"
     current_endpoint = "192.168.0.1:6174"
@@ -1207,6 +1222,7 @@ block中分割(split)出的元素个数的最小值。
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     config = fluid.DistributeTranspilerConfig()
     config.slice_var_up = True
 
@@ -1226,6 +1242,7 @@ ExecutionStrategy
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[13], dtype='float32')
     y = fluid.layers.data(name='y', shape=[1], dtype='float32')
     y_predict = fluid.layers.fc(input=x, size=1, act=None)
@@ -1578,6 +1595,7 @@ in_dygraph_mode
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     if fluid.in_dygraph_mode():
         pass
 
@@ -1875,6 +1893,7 @@ name_scope
 
 .. code-block:: python
           
+     import paddle.fluid as fluid
      with fluid.name_scope("s1"):
         a = fluid.layers.data(name='data', shape=[1], dtype='int32')
         b = a + 1
@@ -2043,6 +2062,7 @@ ParallelExecutor
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         pe = fluid.ParallelExecutor(use_cuda=use_cuda,
                                     loss_name=avg_cost.name,
                                     main_program=fluid.default_main_program())
@@ -2211,6 +2231,7 @@ Program
 
     .. code-block:: python
 
+          import paddle.fluid as fluid
           test_program = fluid.default_main_program().clone(for_test=True)
           optimizer = fluid.optimizer.Momentum(learning_rate=0.01, momentum=0.9)
           optimizer.minimize()
@@ -2538,6 +2559,7 @@ scope_guard
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   import numpy
   
   new_scope = fluid.Scope()

doc/fluid/api_cn/initializer_cn.rst

@@ -26,6 +26,7 @@ BilinearInitializer
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     factor = 2
     C = 2
     w_attr = fluid.initializer.ParamAttr(
@@ -77,6 +78,7 @@ ConstantInitializer
 
 .. code-block:: python
         
+        import paddle.fluid as fluid
         x = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
         fc = fluid.layers.fc(input=x, size=10,
             param_attr=fluid.initializer.Constant(value=2.0))
@@ -104,6 +106,7 @@ force_init_on_cpu
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     if fluid.initializer.force_init_on_cpu():
         step = fluid.layers.create_global_var(shape=[2,3], value=1.0, dtype='float32')
 
@@ -130,6 +133,7 @@ init_on_cpu
 
 .. code-block:: python
         
+        import paddle.fluid as fluid
         with fluid.initializer.init_on_cpu():
             step = fluid.layers.create_global_var(shape=[2,3], value=1.0, dtype='float32')
 
@@ -183,6 +187,7 @@ MSRAInitializer
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
     fc = fluid.layers.fc(input=x, size=10, param_attr=fluid.initializer.MSRA(uniform=False))
 
@@ -219,6 +224,7 @@ NormalInitializer
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         x = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
         fc = fluid.layers.fc(input=x, size=10,
             param_attr=fluid.initializer.Normal(loc=0.0, scale=2.0)
@@ -240,6 +246,7 @@ NumpyArrayInitializer
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[5], dtype='float32')
     fc = fluid.layers.fc(input=x, size=10,
         param_attr=fluid.initializer.NumpyArrayInitializer(numpy.array([1,2])))

doc/fluid/api_cn/io_cn.rst

@@ -96,6 +96,7 @@ load_params
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     exe = fluid.Executor(fluid.CPUPlace())
     param_path = "./my_paddle_model"
     prog = fluid.default_main_program()
@@ -131,6 +132,7 @@ load_persistables
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     exe = fluid.Executor(fluid.CPUPlace())
     param_path = "./my_paddle_model"
     prog = fluid.default_main_program()
@@ -237,6 +239,7 @@ PyReader
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     EPOCH_NUM = 3
     ITER_NUM = 5
     BATCH_SIZE = 3
@@ -278,6 +281,7 @@ PyReader
 
 .. code-block:: python
 
+   import paddle.fluid as fluid
    EPOCH_NUM = 3
    ITER_NUM = 5
    BATCH_SIZE = 10
@@ -346,6 +350,7 @@ PyReader
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   BATCH_SIZE = 10
      
   def generator():
@@ -376,6 +381,7 @@ PyReader
 
 .. code-block:: python
 
+            import paddle.fluid as fluid
             BATCH_SIZE = 10
      
             def generator():
@@ -418,6 +424,7 @@ PyReader
 
 .. code-block:: python
      
+            import paddle.fluid as fluid
             EPOCH_NUM = 3
             ITER_NUM = 15
             BATCH_SIZE = 3
@@ -464,6 +471,7 @@ PyReader
 
 .. code-block:: python
             
+            import paddle.fluid as fluid
             EPOCH_NUM = 3
             ITER_NUM = 15
             BATCH_SIZE = 3
@@ -510,6 +518,7 @@ PyReader
 
 .. code-block:: python
 
+            import paddle.fluid as fluid
             EPOCH_NUM = 3
             ITER_NUM = 15
             BATCH_SIZE = 3
@@ -636,6 +645,7 @@ save_params
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     exe = fluid.Executor(fluid.CPUPlace())
     param_path = "./my_paddle_model"
     prog = fluid.default_main_program()

doc/fluid/api_cn/layers_cn/control_flow_cn.rst

@@ -149,6 +149,7 @@ create_array
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   data = fluid.layers.create_array(dtype='float32')
 
 
@@ -391,6 +392,7 @@ greater_equal
 
 .. code-block:: python
 
+     import paddle.fluid as fluid
      out = fluid.layers.greater_equal(x=label, y=limit)
 
 
@@ -417,6 +419,7 @@ greater_than
 
 .. code-block:: python
 
+     import paddle.fluid as fluid
      out = fluid.layers.greater_than(x=label, y=limit)
 
 
@@ -564,6 +567,7 @@ less_equal
 
 .. code-block:: python
 
+     import paddle.fluid as fluid
      out = fluid.layers.less_equal(x=label, y=limit)
 
 
@@ -594,6 +598,7 @@ less_than
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     label = fluid.layers.data(name='y', shape=[1], dtype='int64')
     limit = fluid.layers.fill_constant(shape=[1], dtype='int64', value=5)
     cond = fluid.layers.less_than(x=label, y=limit)
@@ -621,6 +626,7 @@ not_equal
 
 .. code-block:: python
 
+     import paddle.fluid as fluid
      out = fluid.layers.not_equal(x=label, y=limit)
 
 

doc/fluid/api_cn/layers_cn/detection_cn.rst

@@ -34,6 +34,7 @@ anchor_generator
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     conv1 = fluid.layers.data(name='conv1', shape=[48, 16, 16], dtype='float32')
     anchor, var = fluid.layers.anchor_generator(
     input=conv1,
@@ -86,6 +87,7 @@ bipartite_match
 
 ..  code-block:: python
 
+         import paddle.fluid as fluid
          x = fluid.layers.data(name='x', shape=[4], dtype='float32')
          y = fluid.layers.data(name='y', shape=[4], dtype='float32')
          iou = fluid.layers.iou_similarity(x=x, y=y)
@@ -131,6 +133,7 @@ box_clip
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     boxes = fluid.layers.data(
         name='boxes', shape=[8, 4], dtype='float32', lod_level=1)
     im_info = fluid.layers.data(name='im_info', shape=[3])
@@ -210,6 +213,7 @@ Bounding Box Coder
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     prior_box = fluid.layers.data(name='prior_box',
                                   shape=[512, 4],
                                   dtype='float32',
@@ -278,6 +282,7 @@ box decode过程得出decode_box，然后分配方案如下所述：
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     pb = fluid.layers.data(
         name='prior_box', shape=[4], dtype='float32')
     pbv = fluid.layers.data(
@@ -322,6 +327,7 @@ collect_fpn_proposals
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     multi_rois = []
     multi_scores = []
     for i in range(4):
@@ -402,6 +408,7 @@ density prior box的量由fixed_sizes and fixed_ratios决定。显然地，fixed
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[3,6,9])
     images = fluid.layers.data(name="images", shape=[3,9,12])
     box, var = fluid.layers.density_prior_box(
@@ -468,6 +475,8 @@ detection_map
 
 ..  code-block:: python
 
+        import paddle.fluid as fluid
+        from fluid.layers import detection
         detect_res = fluid.layers.data(
             name='detect_res',
             shape=[10, 6],
@@ -581,6 +590,7 @@ distribute_fpn_proposals
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     fpn_rois = fluid.layers.data(
         name='data', shape=[4], dtype='float32', lod_level=1)
     multi_rois, restore_ind = fluid.layers.distribute_fpn_proposals(
@@ -962,6 +972,7 @@ multiclass_nms
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     boxes = fluid.layers.data(name='bboxes', shape=[81, 4],
                               dtype='float32', lod_level=1)
     scores = fluid.layers.data(name='scores', shape=[81],
@@ -1047,6 +1058,7 @@ prior_box
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[3,6,9])
     images = fluid.layers.data(name="images", shape=[3,9,12])
     box, var = fluid.layers.prior_box(
@@ -1422,6 +1434,7 @@ ssd_loss
 
 ..  code-block:: python
 
+         import paddle.fluid as fluid
          pb = fluid.layers.data(
                            name='prior_box',
                            shape=[10, 4],
@@ -1669,6 +1682,7 @@ yolov3_loss
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[255, 13, 13], dtype='float32')
     gt_box = fluid.layers.data(name='gtbox', shape=[6, 4], dtype='float32')
     gt_label = fluid.layers.data(name='gtlabel', shape=[6], dtype='int32')

doc/fluid/api_cn/layers_cn/io_cn.rst

@@ -24,6 +24,7 @@ batch
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     raw_reader = fluid.layers.io.open_files(filenames=['./data1.recordio',
                                                './data2.recordio'],
                                         shapes=[(3,224,224), (1,)],
@@ -153,6 +154,7 @@ data
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='x', shape=[784], dtype='float32')
 
 
@@ -299,6 +301,7 @@ reader变量中数据预处理块。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     reader = fluid.layers.io.open_files(
         filenames=['./data1.recordio', './data2.recordio'],
         shapes=[(3, 224, 224), (1, )],
@@ -485,6 +488,7 @@ random_data_generator
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     reader = fluid.layers.random_data_generator(
                                  low=0.0,
                                  high=1.0,
@@ -563,6 +567,7 @@ shuffle
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     raw_reader = fluid.layers.io.open_files(filenames=['./data1.recordio',
                                                    './data2.recordio'],
                                             shapes=[(3,224,224), (1,)],

doc/fluid/api_cn/layers_cn/learning_rate_scheduler_cn.rst

@@ -29,6 +29,7 @@ cosine_decay
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     base_lr = 0.1
     lr = fluid.layers.cosine_decay( learning_rate = base_lr, step_each_epoch=10000, epochs=120)
 
@@ -156,6 +157,7 @@ linear_lr_warmup
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         boundaries = [100, 200]
         lr_steps = [0.1, 0.01, 0.001]
         warmup_steps = 50
@@ -225,6 +227,7 @@ Noam衰减方法。noam衰减的numpy实现如下。
 
 .. code-block:: python
 
+    import padde.fluid as fluid
     import numpy as np
     # 设置超参数
     d_model = 2

doc/fluid/api_cn/layers_cn/nn_cn.rst

@@ -60,6 +60,7 @@ pooling2d操作根据输入 ``input`` ， ``pool_size`` ， ``pool_type`` 参数
     #             wend = ceil((i + 1) * W / n)
     #             output[:, :, i, j] = avg(input[:, :, hstart: hend, wstart: wend])
     #
+    import paddle.fluid as fluid
     data = fluid.layers.data(
         name='data', shape=[3, 32, 32], dtype='float32')
     pool_out = fluid.layers.adaptive_pool2d(
@@ -377,6 +378,7 @@ autoincreased_step_counter
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     global_step = fluid.layers.autoincreased_step_counter(
         counter_name='@LR_DECAY_COUNTER@', begin=0, step=1)
 
@@ -447,6 +449,7 @@ batch_norm
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[3, 7, 3, 7], dtype='float32', append_batch_size=False)
     hidden1 = fluid.layers.fc(input=x, size=200, param_attr='fc1.w')
     hidden2 = fluid.layers.batch_norm(input=hidden1)
@@ -621,6 +624,7 @@ bilinear_tensor_product
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   layer1 = fluid.layers.data("t1", shape=[-1, 5], dtype="float32")
   layer2 = fluid.layers.data("t2", shape=[-1, 4], dtype="float32")
   tensor = fluid.layers.bilinear_tensor_product(x=layer1, y=layer2, size=1000)
@@ -693,6 +697,7 @@ BRelu 激活函数
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[2,3,16,16], dtype=”float32”)
     y = fluid.layers.brelu(x, t_min=1.0, t_max=20.0)
 
@@ -876,6 +881,7 @@ ClipByNorm算子
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(
         name='data', shape=[1], dtype='float32')
     reward = fluid.layers.clip_by_norm(x=input, max_norm=1.0)
@@ -912,6 +918,7 @@ continuous_value_model
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[-1, 1], lod_level=1, append_batch_size=False, dtype="int64")#, stop_gradient=False)
     label = fluid.layers.data(name="label", shape=[-1, 1], append_batch_size=False, dtype="int64")
     embed = fluid.layers.embedding(
@@ -997,6 +1004,7 @@ conv2d
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[3, 32, 32], dtype='float32')
     conv2d = fluid.layers.conv2d(input=data, num_filters=2, filter_size=3, act="relu")
 
@@ -1094,6 +1102,7 @@ conv2d_transpose
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[3, 32, 32], dtype='float32')
     conv2d_transpose = fluid.layers.conv2d_transpose(input=data, num_filters=2, filter_size=3)
 
@@ -1174,6 +1183,7 @@ conv3d
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[3, 12, 32, 32], dtype='float32')
     conv3d = fluid.layers.conv3d(input=data, num_filters=2, filter_size=3, act="relu")
 
@@ -1278,6 +1288,7 @@ conv3d_transpose
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[3, 12, 32, 32], dtype='float32')
     conv3d_transpose = fluid.layers.conv3d_transpose(input=data, num_filters=2, filter_size=3)
 
@@ -1319,6 +1330,7 @@ cos_sim
 
 ..  code-block:: python
 
+     import paddle.fluid as fluid
      x = fluid.layers.data(name='x', shape=[3, 7], dtype='float32', append_batch_size=False)
      y = fluid.layers.data(name='y', shape=[1, 7], dtype='float32', append_batch_size=False)
      out = fluid.layers.cos_sim(x, y)
@@ -1359,6 +1371,7 @@ crf_decoding
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     images = fluid.layers.data(name='pixel', shape=[784], dtype='float32')
     label = fluid.layers.data(name='label', shape=[1], dtype='int32')
     hidden = fluid.layers.fc(input=images, size=2)
@@ -1506,6 +1519,7 @@ cross_entropy
 
 ..  code-block:: python
 
+        import paddle.fluid as fluid
         classdim = 7
         x = fluid.layers.data(name='x', shape=[3, 7], dtype='float32', append_batch_size=False)
         label = fluid.layers.data(name='label', shape=[3, 1], dtype='float32', append_batch_size=False)
@@ -1705,6 +1719,7 @@ deformable_conv
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[3, 32, 32], dtype='float32')
     offset = fluid.layers.data(name='offset', shape=[18, 32, 32], dtype='float32')
     mask = fluid.layers.data(name='mask', shape=[9, 32, 32], dtype='float32')
@@ -1747,6 +1762,7 @@ deformable_roi_pooling
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input",
                               shape=[2, 192, 64, 64],
                               dtype='float32',
@@ -1866,6 +1882,7 @@ dropout操作符可以从程序中移除，程序变得高效。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
     droped = fluid.layers.dropout(x, dropout_prob=0.5)
 
@@ -2051,6 +2068,7 @@ W 代表了权重矩阵(weight matrix)，例如 :math:`W_{xi}` 是从输入门
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   emb_dim = 256
   vocab_size = 10000
   hidden_dim = 512
@@ -2162,6 +2180,7 @@ LSTMP层(具有循环映射的LSTM)在LSTM层后有一个分离的映射层，
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     dict_dim, emb_dim = 128, 64
     data = fluid.layers.data(name='sequence', shape=[1],
                          dtype='int32', lod_level=1)
@@ -2929,31 +2948,37 @@ elementwise_pow
 ..  code-block:: python
 
     # 例1: shape(x) = (2, 3, 4, 5), shape(y) = (2, 3, 4, 5)
+    import paddle.fluid as fluid
     x0 = fluid.layers.data(name="x0", shape=[2, 3, 4, 5], dtype='float32')
     y0 = fluid.layers.data(name="y0", shape=[2, 3, 4, 5], dtype='float32')
     z0 = fluid.layers.elementwise_pow(x0, y0)
      
     # 例2: shape(X) = (2, 3, 4, 5), shape(Y) = (5)
+    import paddle.fluid as fluid
     x1 = fluid.layers.data(name="x1", shape=[2, 3, 4, 5], dtype='float32')
     y1 = fluid.layers.data(name="y1", shape=[5], dtype='float32')
     z1 = fluid.layers.elementwise_pow(x1, y1)
      
     # 例3: shape(X) = (2, 3, 4, 5), shape(Y) = (4, 5), with axis=-1(default) or axis=2
+    import paddle.fluid as fluid
     x2 = fluid.layers.data(name="x2", shape=[2, 3, 4, 5], dtype='float32')
     y2 = fluid.layers.data(name="y2", shape=[4, 5], dtype='float32')
     z2 = fluid.layers.elementwise_pow(x2, y2, axis=2)
      
     # 例4: shape(X) = (2, 3, 4, 5), shape(Y) = (3, 4), with axis=1
+    import paddle.fluid as fluid
     x3 = fluid.layers.data(name="x3", shape=[2, 3, 4, 5], dtype='float32')
     y3 = fluid.layers.data(name="y3", shape=[3, 4], dtype='float32')
     z3 = fluid.layers.elementwise_pow(x3, y3, axis=1)
      
     # 例5: shape(X) = (2, 3, 4, 5), shape(Y) = (2), with axis=0
+    import paddle.fluid as fluid
     x4 = fluid.layers.data(name="x4", shape=[2, 3, 4, 5], dtype='float32')
     y4 = fluid.layers.data(name="y4", shape=[2], dtype='float32')
     z4 = fluid.layers.elementwise_pow(x4, y4, axis=0)
      
     # 例6: shape(X) = (2, 3, 4, 5), shape(Y) = (2, 1), with axis=0
+    import paddle.fluid as fluid
     x5 = fluid.layers.data(name="x5", shape=[2, 3, 4, 5], dtype='float32')
     y5 = fluid.layers.data(name="y5", shape=[2], dtype='float32')
     z5 = fluid.layers.elementwise_pow(x5, y5, axis=0)
@@ -3079,6 +3104,7 @@ ELU激活层（ELU Activation Operator）
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3,10,32,32], dtype="float32")
     y = fluid.layers.elu(x, alpha=0.2)
 
@@ -3168,6 +3194,7 @@ expand运算会按给定的次数对输入各维度进行复制（tile）运算
 
 ..  code-block:: python
 
+        import paddle.fluid as fluid
         x = fluid.layers.data(name='x', shape=[10], dtype='float32')
         out = fluid.layers.expand(x=x, expand_times=[1, 2, 2])
 
@@ -3253,6 +3280,7 @@ fc
 
 ..  code-block:: python
 
+         import paddle.fluid as fluid
          # 当输入为单个张量时
 
         data = fluid.layers.data(name="data", shape=[32, 32], dtype="float32")
@@ -3323,6 +3351,7 @@ flatten
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[4, 4, 3], dtype="float32")
     out = fluid.layers.flatten(x=x, axis=2)
 
@@ -3411,6 +3440,7 @@ gather
 
 ..  code-block:: python
   
+  import paddle.fluid as fluid
   x = fluid.layers.data(name='x', shape=[-1, 5], dtype='float32')
   index = fluid.layers.data(name='index', shape=[-1, 1], dtype='int32')
   output = fluid.layers.gather(x, index)
@@ -3449,6 +3479,7 @@ gaussian_random算子。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     out = fluid.layers.gaussian_random(shape=[20, 30])
 
@@ -3488,6 +3519,7 @@ gaussian_random_batch_size_like
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[13, 11], dtype='float32')
 
     out = fluid.layers.gaussian_random_batch_size_like(
@@ -3636,6 +3668,7 @@ group_norm
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[8, 32, 32],
                              dtype='float32')
     x = fluid.layers.group_norm(input=data, groups=4)
@@ -4175,6 +4208,7 @@ https://en.wikipedia.org/wiki/Bilinear_interpolation。
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   input = fluid.layers.data(name="input", shape=[3,6,9], dtype="float32")
   out = fluid.layers.image_resize(input, out_shape=[12, 12], resample="NEAREST")
 
@@ -4211,6 +4245,7 @@ image_resize_short
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[3,6,9], dtype="float32")
     out = fluid.layers.image_resize_short(input, out_short_len=3)
 
@@ -4253,6 +4288,7 @@ kL发散损失计算如下：
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[4,2,2], dtype='float32')
     target = fluid.layers.data(name='target', shape=[4,2,2], dtype='float32')
     loss = fluid.layers.kldiv_loss(x=x, target=target, reduction='batchmean')
@@ -4294,6 +4330,7 @@ L2正则（L2 normalize Layer）
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name="data",
                          shape=(3, 17, 13),
                          dtype="float32")
@@ -4343,6 +4380,7 @@ label_smooth
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
 
     label = fluid.layers.data(name="label", shape=[1], dtype="float32")
@@ -4401,6 +4439,7 @@ layer_norm
 
 ..  code-block:: python
 
+   import paddle.fluid as fluid
    data = fluid.layers.data(name='data', shape=[3, 32, 32],
                                            dtype='float32')
    x = fluid.layers.layer_norm(input=data, begin_norm_axis=1)
@@ -4433,6 +4472,7 @@ LeakyRelu 激活函数
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[2,3,16,16], dtype="float32")
     y = fluid.layers.leaky_relu(x, alpha=0.01)
 
@@ -4605,6 +4645,7 @@ lod_reset
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[10])
     y = fluid.layers.data(name='y', shape=[10, 20], lod_level=2)
     out = fluid.layers.lod_reset(x=x, y=y)
@@ -4644,6 +4685,7 @@ log
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   x = fluid.layers.data(name="x", shape=[3, 4], dtype="float32")
   output = fluid.layers.log(x)
 
@@ -4688,6 +4730,7 @@ log_loss
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     label = fluid.layers.data(name='label', shape=[1], dtype='int64')
     prob = fluid.layers.data(name='prob', shape=[10], dtype='float32')
     cost = fluid.layers.log_loss(input=prob, label=label)
@@ -4731,6 +4774,7 @@ logical_and算子
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     left = fluid.layers.data(
         name='left', shape=[1], dtype='int32')
     right = fluid.layers.data(
@@ -4773,6 +4817,7 @@ logical_not算子
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     left = fluid.layers.data(
         name='left', shape=[1], dtype='int32')
     result = fluid.layers.logical_not(x=left)
@@ -4814,6 +4859,7 @@ logical_or算子
 .. code-block:: python
 
 
+    import paddle.fluid as fluid
     left = fluid.layers.data(
         name='left', shape=[1], dtype='int32')
     right = fluid.layers.data(
@@ -4855,6 +4901,7 @@ logical_xor算子
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     left = fluid.layers.data(
         name='left', shape=[1], dtype='int32')
     right = fluid.layers.data(
@@ -4908,6 +4955,7 @@ lrn
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(
         name="data", shape=[3, 112, 112], dtype="float32")
     lrn = fluid.layers.lrn(input=data)
@@ -4980,6 +5028,7 @@ sigmoid的计算公式为： :math:`sigmoid(x) = 1 / (1 + e^{-x})` 。
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   emb_dim = 256
   vocab_size = 10000
   data = fluid.layers.data(name='x', shape=[-1, 100, 1],
@@ -5115,6 +5164,7 @@ margin rank loss（差距排序损失）层。在排序问题中，它可以比
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     label = fluid.layers.data(name="label", shape=[-1, 1], dtype="float32")
     left = fluid.layers.data(name="left", shape=[-1, 1], dtype="float32")
     right = fluid.layers.data(name="right", shape=[-1, 1], dtype="float32")
@@ -5191,6 +5241,7 @@ matmul
     # x: [M], y: [N]
     # fluid.layers.matmul(x, y, True, True)  # out: [M, N]
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[2, 3], dtype='float32')
     y = fluid.layers.data(name='y', shape=[3, 2], dtype='float32')
     out = fluid.layers.matmul(x, y, True, True)
@@ -5237,6 +5288,7 @@ maxout
     
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(
         name='data',
         shape=[256, 32, 32],
@@ -5273,6 +5325,7 @@ mean算子计算X中所有元素的平均值
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(
         name='data', shape=[2, 3], dtype='float32')
     mean = fluid.layers.mean(input)
@@ -5370,6 +5423,7 @@ merge_selected_rows
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   b = fluid.default_main_program().global_block()
   var = b.create_var(
         name="X", dtype="float32", persistable=True,
@@ -5555,6 +5609,7 @@ nce
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     import numpy as np
 
     window_size = 5
@@ -5619,6 +5674,7 @@ NPair损失需要成对的数据。NPair损失分为两部分：第一部分是
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     anchor = fluid.layers.data(
               name = 'anchor', shape = [18, 6], dtype = 'float32', append_batch_size=False)
     positive = fluid.layers.data(
@@ -5654,6 +5710,7 @@ one_hot
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     label = fluid.layers.data(name="label", shape=[1], dtype="int64")
     one_hot_label = fluid.layers.one_hot(input=label, depth=10)
 
@@ -5886,6 +5943,7 @@ pixel shuffle 层（像素重组层）
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[9,4,4])
     output = fluid.layers.pixel_shuffle(x=input, upscale_factor=3)
 
@@ -5975,6 +6033,7 @@ pooling2d操作符根据 ``input`` ， 池化类型 ``pool_type`` ， 池化核
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(
         name='data', shape=[3, 32, 32], dtype='float32')
     pool2d = fluid.layers.pool2d(
@@ -6071,6 +6130,7 @@ pooling3d操作根据input，pool_type，pool_size，strides和paddings参数计
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(
         name='data', shape=[3, 32, 32, 32], dtype='float32')
     pool3d = fluid.layers.pool3d(
@@ -6306,6 +6366,7 @@ random_crop
 
 ..  code-block:: python
 
+   import paddle.fluid as fluid
    img = fluid.layers.data("img", [3, 256, 256])
    cropped_img = fluid.layers.random_crop(img, shape=[3, 224, 224])
 
@@ -6335,6 +6396,7 @@ rank
 
 .. code-block:: python
 
+       import paddle.fluid as fluid
        input = layers.data(
             name="input", shape=[3, 100, 100], dtype="float32")
        rank = layers.rank(input) # 4
@@ -6379,6 +6441,7 @@ P 的取值可为： {0, 1} 或 {0, 0.5, 1}, 其中，0.5表示输入的两文
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     label = fluid.layers.data(name="label", shape=[-1, 1], dtype="float32")
     left = fluid.layers.data(name="left", shape=[-1, 1], dtype="float32")
     right = fluid.layers.data(name="right", shape=[-1, 1], dtype="float32")
@@ -6414,6 +6477,9 @@ reduce_all
         #    [[True, False]
         #     [True, True]]
         # 接下来的示例中，我们在每处函数调用后面都标注出了它的结果张量。
+        import paddle.fluid as fluid
+        import paddle.fluid.layers as layers
+        import numpy as np
         fluid.layers.reduce_all(x)  # False
         fluid.layers.reduce_all(x, dim=0)  # [True, False]
         fluid.layers.reduce_all(x, dim=-1)  # [False, True]
@@ -6448,6 +6514,9 @@ reduce_any
         #    [[True, False]
         #     [False, False]]
         # 接下来的示例中，我们在每处函数调用后面都标注出了它的结果张量。
+        import paddle.fluid as fluid
+        import paddle.fluid.layers as layers
+        import numpy as np
         fluid.layers.reduce_any(x)  # True
         fluid.layers.reduce_any(x, dim=0)  # [True, False]
         fluid.layers.reduce_any(x, dim=-1)  # [True, False]
@@ -6740,6 +6809,7 @@ Relu接受一个输入数据(张量)，输出一个张量。将线性函数y = m
 
 ..  code-block:: python
       
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3, 4], dtype="float32")
     output = fluid.layers.relu(x)
 
@@ -6780,6 +6850,7 @@ relu6激活算子（Relu6 Activation Operator）
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3,10,32,32], dtype="float32")
     y = fluid.layers.relu6(x, threshold=6.0)
 
@@ -6833,6 +6904,7 @@ reshape
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   data = fluid.layers.data(
       name='data', shape=[2, 4, 6], dtype='float32')
   reshaped = fluid.layers.reshape(
@@ -6916,6 +6988,7 @@ align_corners和align_mode是可选参数，插值的计算方法可以由它们
 
 .. code-block:: python
   
+  import paddle.fluid as fluid
   input = fluid.layers.data(name="input", shape=[3,6,9], dtype="float32")
   out = fluid.layers.resize_bilinear(input, out_shape=[12, 12])
 
@@ -6988,6 +7061,7 @@ resize_nearest
 
 ..  code-block:: python
     
+    import paddle.fluid as fluid
     input = fluid.layers.data(name="input", shape=[3,6,9], dtype="float32")
     out = fluid.layers.resize_nearest(input, out_shape=[12, 12])
 
@@ -7032,6 +7106,7 @@ Region of Interests align(直译：有意义、有价值选区对齐) 用于实
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(
             name='data', shape=[256, 32, 32], dtype='float32')
     rois = fluid.layers.data(
@@ -7229,6 +7304,7 @@ sampling_id算子。用于从输入的多项分布中对id进行采样的图层
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(
     name="X",
     shape=[13, 11],
@@ -7500,6 +7576,7 @@ sequence_enumerate
 
 ..  code-block:: python
 
+      import paddle.fluid as fluid
       x = fluid.layers.data(shape[-1, 1], dtype='int32', lod_level=1)
       out = fluid.layers.sequence_enumerate(input=x, win_size=3, pad_value=0)
 
@@ -7570,6 +7647,7 @@ sequence_expand
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     x = fluid.layers.data(name='x', shape=[10], dtype='float32')
     y = fluid.layers.data(name='y', shape=[10, 20],
@@ -7642,9 +7720,14 @@ Sequence Expand As Layer
 
 .. code-block:: python
 
-    x = fluid.layers.data(name='x', shape=[7, 1],
+    import paddle.fluid as fluid
+    import paddle.fluid.layers as layers
+
+    x = fluid.layers.data(name='x', shape=[10], dtype='float32')
+    y = fluid.layers.data(name='y', shape=[10, 20],
                      dtype='float32', lod_level=1)
-    x_first_step = fluid.layers.sequence_first_step(input=x)
+    out = layers.sequence_expand_as(x=x, y=y)
+
 
 
 
@@ -7684,6 +7767,7 @@ sequence_first_step
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[7, 1],
                  dtype='float32', lod_level=1)
     x_first_step = fluid.layers.sequence_first_step(input=x)
@@ -7731,6 +7815,7 @@ sequence_last_step
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[7, 1],
                  dtype='float32', lod_level=1)
     x_last_step = fluid.layers.sequence_last_step(input=x)
@@ -7772,6 +7857,7 @@ sequence_mask
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     
     x = fluid.layers.data(name='x', shape=[10], dtype='float32', lod_level=1)
@@ -7869,6 +7955,7 @@ sequence_pad
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import numpy
 
     x = fluid.layers.data(name='y', shape=[10, 5],
@@ -8110,6 +8197,7 @@ sequence_scatter
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
      
     input = layers.data( name="x", shape=[3, 6], append_batch_size=False, dtype='float32' )
@@ -8171,6 +8259,7 @@ sequence_slice
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   import numpy as np
   seqs = fluid.layers.data(name='x', shape=[10, 5],
        dtype='float32', lod_level=1)
@@ -8220,6 +8309,7 @@ sequence_softmax
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[7, 1],
                  dtype='float32', lod_level=1)
     x_sequence_softmax = fluid.layers.sequence_softmax(input=x)
@@ -8277,6 +8367,7 @@ sequence_unpad
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', shape=[10, 5], dtype='float32')
     len = fluid.layers.data(name='length', shape=[1], dtype='int64')
     out = fluid.layers.sequence_unpad(x=x, length=len)
@@ -8382,6 +8473,7 @@ shuffle_channel
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name='input', shape=[4,2,2], dtype='float32')
     out = fluid.layers.shuffle_channel(x=input, group=2)
 
@@ -8443,6 +8535,7 @@ sigmoid_cross_entropy_with_logits
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(
         name='data', shape=[10], dtype='float32')
     label = fluid.layers.data(
@@ -8481,6 +8574,7 @@ sign
 ..  code-block:: python
 
     # [1, 0, -1]
+    import paddle.fluid as fluid
     data = fluid.layers.sign(np.array([3, 0, -2]))
 
 
@@ -8571,6 +8665,7 @@ similarity_focus
 
 ..  code-block:: python
 
+            import paddle.fluid as fluid
             data = fluid.layers.data(
               name='data', shape=[-1, 3, 2, 2], dtype='float32')
             fluid.layers.similarity_focus(input=data, axis=1, indexes=[0])
@@ -8673,6 +8768,7 @@ smooth_l1
 
 ..  code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='data', shape=[128], dtype='float32')
     label = fluid.layers.data(
         name='label', shape=[100], dtype='float32')
@@ -8824,6 +8920,7 @@ softmax_with_cross_entropy
 
 ..  code-block:: python
 
+  import paddle.fluid as fluid
   data = fluid.layers.data(name='data', shape=[128], dtype='float32')
         label = fluid.layers.data(name='label', shape=[1], dtype='int64')
         fc = fluid.layers.fc(input=data, size=100)
@@ -9028,6 +9125,7 @@ square_error_cost
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     y = fluid.layers.data(name='y', shape=[1], dtype='float32')
     y_predict = fluid.layers.data(name='y_predict', shape=[1], dtype='float32')
     cost = fluid.layers.square_error_cost(input=y_predict, label=y)
@@ -9080,6 +9178,7 @@ squeeze
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     x = fluid.layers.data(name='x', shape=[5, 1, 10])
     y = fluid.layers.sequeeze(input=x, axes=[1])
@@ -9159,6 +9258,7 @@ stack
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     x1 = layers.data(name='x1', shape=[1, 2], dtype='int32')
     x2 = layers.data(name='x2', shape=[1, 2], dtype='int32')
@@ -9196,6 +9296,7 @@ STanh 激活算子（STanh Activation Operator.）
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3,10,32,32], dtype="float32")
     y = fluid.layers.stanh(x, scale_a=0.67, scale_b=1.72)
 
@@ -9228,6 +9329,7 @@ sum算子。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     input0 = fluid.layers.data(name="input0", shape=[13, 11], dtype='float32')
     input1 = layers.data(name="input1", shape=[13, 11], dtype='float32')
@@ -9265,6 +9367,7 @@ Swish 激活函数
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   x = fluid.layers.data(name="x", shape=[3,10,32,32], dtype="float32")
   y = fluid.layers.swish(x, beta=2.0)
 
@@ -9358,6 +9461,7 @@ temporal_shift
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     input = fluid.layers.data(name='input', shape=[4,2,2], dtype='float32')
     out = fluid.layers.temporal_shift(x=input, seg_num=2, shift_ratio=0.2)
 
@@ -9411,6 +9515,7 @@ topk
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     input = layers.data(name="input", shape=[13, 11], dtype='float32')
     top5_values, top5_indices = fluid.layers.topk(input, k=5)
@@ -9488,6 +9593,7 @@ tree_conv
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     # 10 代表数据集的最大节点大小max_node_size，5 代表向量宽度
     nodes_vector = fluid.layers.data(name='vectors', shape=[10, 5], dtype='float32')
     # 10 代表数据集的最大节点大小max_node_size, 2 代表每条边连接两个节点
@@ -9542,6 +9648,7 @@ uniform_random_batch_size_like算子。
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
 
     input = fluid.layers.data(name="input", shape=[13, 11], dtype='float32')
@@ -9685,6 +9792,9 @@ where
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
+        import paddle.fluid.layers as layers
+        import numpy as np
         # condition为张量[True, False, True]
         out = fluid.layers.where(condition) # [[0], [2]]
      

doc/fluid/api_cn/layers_cn/ops_cn.rst

@@ -26,6 +26,7 @@ abs
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.abs(data)
 
@@ -51,6 +52,7 @@ arccosine激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.acos(data)
 
@@ -76,6 +78,7 @@ arcsine激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.asin(data)
 
@@ -102,6 +105,7 @@ arctanh激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.atan(data)
 
@@ -134,6 +138,7 @@ ceil
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.ceil(data)
 
@@ -172,6 +177,7 @@ Cosine余弦激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.cos(data)
 
@@ -203,6 +209,7 @@ cumsum
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name="input", shape=[32, 784])
     result = fluid.layers.cumsum(data, axis=0)
 
@@ -238,6 +245,7 @@ Exp激活函数(Exp指以自然常数e为底的指数运算)。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.exp(data)
 
@@ -275,6 +283,7 @@ floor
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.floor(data)
 
@@ -315,6 +324,7 @@ HardShrink激活函数(HardShrink activation operator)
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name="input", shape=[784])
     result = fluid.layers.hard_shrink(x=data, threshold=0.3)
 
@@ -351,6 +361,7 @@ Logsigmoid激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.logsigmoid(data)
 
@@ -386,6 +397,7 @@ Reciprocal（取倒数）激活函数
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.reciprocal(data)
 
@@ -425,6 +437,7 @@ Round取整激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.round(data)
 
@@ -456,6 +469,7 @@ rsqrt激活函数
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.rsqrt(data)
 
@@ -485,6 +499,7 @@ sigmoid激活函数
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.sigmoid(data)
 
@@ -524,6 +539,7 @@ sin
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.sin(data)
 
@@ -561,6 +577,7 @@ softplus激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.softplus(data)
 
@@ -600,6 +617,7 @@ Softshrink激活算子
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.softshrink(data)
 
@@ -638,6 +656,7 @@ softsign激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.softsign(data)
 
@@ -677,6 +696,7 @@ sqrt
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.sqrt(data)
 
@@ -714,6 +734,7 @@ square
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.square(data)
 
@@ -754,6 +775,7 @@ tanh 激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.tanh(data)
 
@@ -792,6 +814,7 @@ tanh_shrink激活函数。
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         data = fluid.layers.data(name="input", shape=[32, 784])
         result = fluid.layers.tanh_shrink(data)
 
@@ -829,6 +852,7 @@ ThresholdedRelu激活函数
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   data = fluid.layers.data(name="input", shape=[1])
   result = fluid.layers.thresholded_relu(data, threshold=0.4)
 

doc/fluid/api_cn/layers_cn/tensor_cn.rst

@@ -26,6 +26,7 @@ argmax
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3, 4], dtype="float32")
     out = fluid.layers.argmax(x=in, axis=0)
     out = fluid.layers.argmax(x=in, axis=-1)
@@ -61,6 +62,7 @@ argmin
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3, 4], dtype="float32")
     out = fluid.layers.argmin(x=in, axis=0)
     out = fluid.layers.argmin(x=in, axis=-1)
@@ -113,6 +115,7 @@ argsort
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name="x", shape=[3, 4], dtype="float32")
     out, indices = fluid.layers.argsort(input=x, axis=0)
 
@@ -180,6 +183,7 @@ cast
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(name='x', shape=[13], dtype='float32')
     result = fluid.layers.cast(x=data, dtype='float64')
 
@@ -215,6 +219,7 @@ concat
 
 .. code-block:: python
     
+    import paddle.fluid as fluid
     a = fluid.layers.data(name='a', shape=[2, 13], dtype='float32')
     b = fluid.layers.data(name='b', shape=[2, 3], dtype='float32')
     c = fluid.layers.data(name='c', shape=[2, 2], dtype='float32')
@@ -254,6 +259,7 @@ create_global_var
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     var = layers.create_global_var(shape=[2,3], value=1.0, dtype='float32',
                      persistable=True, force_cpu=True, name='new_var')
@@ -290,6 +296,7 @@ create_parameter
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.layers as layers
     W = fluid.layers.create_parameter(shape=[784, 200], dtype='float32')
 
@@ -323,6 +330,7 @@ create_tensor
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     tensor = fluid.layers.create_tensor(dtype='float32')
 
 
@@ -350,6 +358,7 @@ diag
         #  [3, 0, 0]
         #  [0, 4, 0]
         #  [0, 0, 5]
+        import paddle.fluid as fluid
         data = fluid.layers.diag(np.arange(3, 6))
 
 
@@ -508,6 +517,7 @@ isfinite
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     var = fluid.layers.data(name="data",
                             shape=(4, 6),
                             dtype="float32")
@@ -540,6 +550,7 @@ linspace
 
 .. code-block:: python
 
+      import paddle.fluid as fluid
       data = fluid.layers.linspace(0, 10, 5, 'float32') # [0.0,  2.5,  5.0,  7.5, 10.0]
       data = fluid.layers.linspace(0, 10, 1, 'float32') # [0.0]
 
@@ -602,6 +613,7 @@ range
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.range(0, 10, 2, 'int32')
 
 
@@ -808,6 +820,7 @@ zeros_like
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     x = fluid.layers.data(name='x', dtype='float32', shape=[3], append_batch_size=False)
     data = fluid.layers.zeros_like(x) # [0.0, 0.0, 0.0]
 

doc/fluid/api_cn/metrics_cn.rst

@@ -21,6 +21,7 @@ https://en.wikipedia.org/wiki/Accuracy_and_precision
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     # 假设有batch_size = 128
     batch_size=128
     accuracy_manager = fluid.metrics.Accuracy()
@@ -77,6 +78,7 @@ auc函数创建四个局部变量true_positives, true_negatives, false_positives
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import numpy as np
     # 初始化auc度量
     auc_metric = fluid.metrics.Auc("ROC")
@@ -134,6 +136,7 @@ ChunkEvaluator
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         # 初始化chunck-level的评价管理。
         metric = fluid.metrics.ChunkEvaluator()
         
@@ -185,6 +188,7 @@ CompositeMetric
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         import numpy as np
         preds = [[0.1], [0.7], [0.8], [0.9], [0.2],
                  [0.2], [0.3], [0.5], [0.8], [0.6]]
@@ -271,6 +275,7 @@ DetectionMAP
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         import paddle.fluid.layers as layers
          
         batch_size = -1 # 可以为任意大小
@@ -339,6 +344,7 @@ EditDistance
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import numpy as np
     
     # 假设batch_size为128
@@ -366,6 +372,7 @@ EditDistance
 
 .. code-block:: python
 
+  import paddle.fluid as fluid
   edit_distances_batch2 = np.random.randint(low = 0, high = 10, size = (batch_size, 1))
   seq_num_batch2 = batch_size
   distance_evaluator.update(edit_distances_batch2, seq_num_batch2)
@@ -456,6 +463,7 @@ https://en.wikipedia.org/wiki/Evaluation_of_binary_classifiers
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import numpy as np
 
     metric = fluid.metrics.Precision() 
@@ -499,6 +507,7 @@ https://en.wikipedia.org/wiki/Precision_and_recall
 
 .. code-block:: python
 
+        import paddle.fluid as fluid
         import numpy as np
 
         metric = fluid.metrics.Recall()

doc/fluid/api_cn/nets_cn.rst

@@ -30,6 +30,7 @@ he Gated Linear Units(GLU)由切分（split），sigmoid激活函数和按元素
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     data = fluid.layers.data(
         name="words", shape=[-1, 6, 3, 9], dtype="float32")
     # 输出的形状为[-1, 3, 3, 9]
@@ -74,6 +75,7 @@ Image Convolution Group由Convolution2d，BatchNorm，DropOut和Pool2d组成。
 
 .. code-block:: python
 
+          import paddle.fluid as fluid
           img = fluid.layers.data(name='img', shape=[1, 28, 28], dtype='float32')
           conv_pool = fluid.nets.img_conv_group(input=img,
                                                 conv_padding=1,

doc/fluid/api_cn/optimizer_cn.rst

@@ -340,6 +340,7 @@ DGC还使用动量因子掩藏(momentum factor masking)和预训练(warm-up)来
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     optimizer = fluid.optimizer.DGCMomentumOptimizer(
                                         learning_rate=0.0001,
                                         momentum=0.9,
@@ -546,15 +547,16 @@ FTRL 原始论文: ( `https://www.eecs.tufts.edu/~dsculley/papers/ad-click-predi
 LambOptimizer
 -------------------------------
 
-.. py:class:: paddle.fluid.optimizer.LambOptimizer(learning_rate=0.001, lamb_weight_decay=0.01, beta1=0.9, beta2=0.999, epsilon=1e-06, regularization=None, name=None)
+.. py:class:: paddle.fluid.optimizer.LambOptimizer(learning_rate=0.001, lamb_weight_decay=0.01, beta1=0.9, beta2=0.999, epsilon=1e-06, regularization=None, exclude_from_weight_decay_fn=None, name=None)
 
 LAMB（Layer-wise Adaptive Moments optimizer for Batching training）优化器
-LAMB优化器旨在不降低准确性的条件下扩大训练的批量大小，支持自适应元素更新和精确的分层校正。 更多信息请参考Reducing BERT Pre-Training Time from 3 Days to 76 Minutes。
+LAMB优化器旨在不降低准确性的条件下扩大训练的批量大小，支持自适应元素更新和精确的分层校正。 更多信息请参考 `Large Batch Optimization for
+Deep Learning: Training BERT in 76 minutes <https://arxiv.org/pdf/1904.00962.pdf>`_ 。
 参数更新如下：
 
 .. math::
 
-    \begin{align}\begin{aligned}m_t^l & = \beta_1 m_{t - 1}^l + (1 - \beta_1)g_t^l\\v_t^l & = \beta_2 v_{t - 1}^l + (1 - \beta_2)g_t^l \odot g_t^l\\\widehat{m}_t^l & = m_t^l/(1 - \beta_1^t)\\\widehat{v}_t^l & = v_t^l/(1 - \beta_2^t)\\r_1 & = \left \| w_{t-1}^l \right \|_2\\r_2 & = \left \|  \frac{\widehat{m}_t^l}{\sqrt{\widehat{v}_t^l+\epsilon}} + \lambda w_{t-1}^l \right \|_2\\r & = r_1 / r_2\\\eta^l & = r \times \eta\\w_t^l & = w_{t-1}^l -\eta ^l \times (\frac{\widehat{m}_t^l}{\sqrt{\widehat{v}_t^l+\epsilon}} + \lambda w_{t-1}^l)\end{aligned}\end{align}
+    \begin{align}\begin{aligned}m_t &= \beta_1 m_{t - 1}+ (1 - \beta_1)g_t \\\v_t &= \beta_2 v_{t - 1}  + (1 - \beta_2)g_t^2 \\\r_t &= \frac{m_t}{\sqrt{v_t}+\epsilon} \\\w_t &= w_{t-1} -\eta_t \frac{\left \| w_{t-1}\right \|}{\left \| r_t + \lambda w_{t-1}\right \|} (r_t + \lambda w_{t-1})\end{aligned}\end{align}
 
 其中 :math:`m` 为第一个时刻，:math:`v` 为第二个时刻，:math:`\eta` 为学习率，:math:`\lambda` 为LAMB权重衰减率。
 
@@ -564,7 +566,8 @@ LAMB优化器旨在不降低准确性的条件下扩大训练的批量大小，
     - **beta1** (float) – 第一个时刻估计的指数衰减率。
     - **beta2** (float) – 第二个时刻估计的指数衰减率。
     - **epsilon** (float) – 一个小的浮点值，目的是维持数值稳定性。
-    - **regularization** – 一个正则化器，如fluid.regularizer.L1DecayRegularizer。
+    - **regularization** (Regularizer) – 一个正则化器，如fluid.regularizer.L1DecayRegularizer。
+    - **exclude_from_weight_decay_fn** (function) – 当返回值为True时从权重衰减中去除某个参数。 
     - **name** (str|None) – 名字前缀（可选项）。
 
 **代码示例**
@@ -576,8 +579,12 @@ LAMB优化器旨在不降低准确性的条件下扩大训练的批量大小，
     data = fluid.layers.data(name='x', shape=[5], dtype='float32')
     hidden = fluid.layers.fc(input=data, size=10)
     cost = fluid.layers.mean(hidden)
+
+    def exclude_fn(param):
+        return param.name.endswith('.b_0')
      
-    optimizer = fluid.optimizer.Lamb(learning_rate=0.002)
+    optimizer = fluid.optimizer.Lamb(learning_rate=0.002,
+                                     exclude_from_weight_decay_fn=exclude_fn)
     optimizer.minimize(cost)
 
 
@@ -625,6 +632,7 @@ LARS支持的Momentum优化器
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     optimizer = fluid.optimizer.LarsMomentum(learning_rate=0.2, momentum=0.1, lars_weight_decay=0.001)
     optimizer.minimize(cost)
 

doc/fluid/api_cn/profiler_cn.rst

@@ -82,6 +82,7 @@ profile interface 。与cuda_profiler不同，此profiler可用于分析CPU和GP
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.profiler as profiler
     import numpy as np
 
@@ -118,6 +119,7 @@ reset_profiler
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.profiler as profiler
     with profiler.profiler('CPU', 'total', '/tmp/profile'):
     for iter in range(10):
@@ -155,6 +157,7 @@ start_profiler
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.profiler as profiler
 
     profiler.start_profiler('GPU')
@@ -195,6 +198,7 @@ stop_profiler
 
 .. code-block:: python
 
+    import paddle.fluid as fluid
     import paddle.fluid.profiler as profiler
 
     profiler.start_profiler('GPU')