[Docs][en] adjust code example format

python/paddle/distribution/transform.py

@@ -590,7 +590,7 @@ def _codomain(self):
 class ExpTransform(Transform):
     r"""Exponent transformation with mapping :math:`y = \exp(x)`.
 
-    Exapmles:
+    Examples:
 
         .. code-block:: python
 
@@ -1169,7 +1169,7 @@ def _codomain(self):
 class TanhTransform(Transform):
     r"""Tanh transformation with mapping :math:`y = \tanh(x)`.
 
-    Examples
+    Examples:
 
         .. code-block:: python
 

python/paddle/fluid/dataloader/dataset.py

@@ -413,7 +413,7 @@ class Subset(Dataset):
     Returns:
         Dataset: A Dataset which is the subset of the original dataset.
 
-    Example code:
+    Examples:
 
         .. code-block:: python
 
@@ -452,10 +452,10 @@ def random_split(dataset, lengths, generator=None):
         lengths (sequence): lengths of splits to be produced
         generator (Generator, optional): Generator used for the random permutation. Default is None then the DefaultGenerator is used in manual_seed().
 
-     Returns:
+    Returns:
         Datasets: A list of subset Datasets, which are the non-overlapping subsets of the original Dataset.
 
-    Example code:
+    Examples:
 
         .. code-block:: python
 

python/paddle/fluid/dygraph/base.py

@@ -485,8 +485,9 @@ def grad(outputs,
         inside `inputs`, and the i-th returned Tensor is the sum of gradients of 
         `outputs` with respect to the i-th `inputs`.
 
-    Examples 1:
+    Examples:
         .. code-block:: python
+            :name: code-example-1
 
             import paddle
 
@@ -519,8 +520,8 @@ def test_dygraph_grad(create_graph):
             print(test_dygraph_grad(create_graph=False)) # [2.]
             print(test_dygraph_grad(create_graph=True)) # [4.]
 
-    Examples 2:
         .. code-block:: python
+            :name: code-example-2
 
             import paddle
 

python/paddle/fluid/dygraph/layers.py

@@ -98,6 +98,26 @@ class Layer(object):
 
     Returns:
         None
+
+    Examples:
+        .. code-block:: python
+
+            import paddle
+            class MyLayer(paddle.nn.Layer):
+                def __init__(self):
+                    super(MyLayer, self).__init__()
+                    self._linear = paddle.nn.Linear(1, 1)
+                    self._dropout = paddle.nn.Dropout(p=0.5)
+                def forward(self, input):
+                    temp = self._linear(input)
+                    temp = self._dropout(temp)
+                    return temp
+            x = paddle.randn([10, 1], 'float32')
+            mylayer = MyLayer()
+            mylayer.eval()  # set mylayer._dropout to eval mode
+            out = mylayer(x)
+            mylayer.train()  # set mylayer._dropout to train mode
+            out = mylayer(x)
     """
 
     def __init__(self, name_scope=None, dtype="float32"):

python/paddle/fluid/executor.py

@@ -1187,8 +1187,9 @@ def run(self,
                results are spliced together in dimension 0 for the same Tensor values
                (Tensors in fetch_list) on different devices.
 
-        Examples 1:
+        Examples:
             .. code-block:: python
+                :name: code-example-1
 
                 import paddle
                 import numpy
@@ -1215,9 +1216,10 @@ def run(self,
                 print(array_val)
                 # [array([0.02153828], dtype=float32)]
 
-        Examples 2:
             .. code-block:: python
+                :name: code-example-2
 
+                # required: gpu
                 import paddle
                 import numpy as np
 
@@ -1265,7 +1267,7 @@ def run(self,
                 print("The merged prediction shape: {}".format(
                     np.array(merged_prediction).shape))
                 print(merged_prediction)
- 
+
                 # Out:
                 # The unmerged prediction shape: (2, 3, 2)
                 # [array([[-0.37620035, -0.19752218],

python/paddle/fluid/initializer.py

@@ -1177,7 +1177,7 @@ def calculate_gain(nonlinearity, param=None):
 
     Examples:
         .. code-block:: python
-          :name: code-example1
+
             import paddle
             gain = paddle.nn.initializer.calculate_gain('tanh') # 5.0 / 3
             gain = paddle.nn.initializer.calculate_gain('leaky_relu', param=1.0) # 1.0 = math.sqrt(2.0 / (1+param^2))

python/paddle/incubate/autotune.py

@@ -54,7 +54,6 @@ def set_config(config=None):
 
     Examples:
         .. code-block:: python
-            :name: auto-tuning
 
             import paddle
             import json

python/paddle/nn/functional/activation.py

@@ -601,7 +601,6 @@ def rrelu(x, lower=1. / 8., upper=1. / 3., training=True, name=None):
 
     Examples:
         .. code-block:: python
-            :name: rrelu-example
 
             import paddle
             import paddle.nn.functional as F

python/paddle/nn/functional/loss.py

@@ -2106,7 +2106,7 @@ def cross_entropy(input,
 
                    Return the average value of the previous results
 
-             .. math::
+            .. math::
                 \\loss=\sum_{j}loss_j/N
 
                   where, N is the number of samples and C is the number of categories.
@@ -2115,29 +2115,29 @@ def cross_entropy(input,
 
             1. Hard labels (soft_label = False)
 
-             .. math::
+            .. math::
                 \\loss=\sum_{j}loss_j/\sum_{j}weight[label_j] 
 
             2. Soft labels (soft_label = True)
 
-             .. math::
+            .. math::
                 \\loss=\sum_{j}loss_j/\sum_{j}\left(\sum_{i}weight[label_i]\right)
- 
- 
+
+
     Parameters:
 
         - **input** (Tensor)
 
             Input tensor, the data type is float32, float64. Shape is
-	    :math:`[N_1, N_2, ..., N_k, C]`, where C is number of classes ,  ``k >= 1`` . 
+        :math:`[N_1, N_2, ..., N_k, C]`, where C is number of classes ,  ``k >= 1`` . 
 
             Note: 
 
                 1. when use_softmax=True, it expects unscaled logits. This operator should not be used with the 
                 output of softmax operator, which will produce incorrect results.
 
                 2. when use_softmax=False, it expects the output of softmax operator.
- 
+
         - **label** (Tensor)
 
             1. If soft_label=False, the shape is
@@ -2205,10 +2205,11 @@ def cross_entropy(input,
         2. if soft_label = True, the dimension of return value is :math:`[N_1, N_2, ..., N_k, 1]` . 
 
 
-     Example1(hard labels):
+    Examples:
 
         .. code-block:: python
-
+
+            # hard labels
             import paddle
             paddle.seed(99999)
             N=100
@@ -2225,11 +2226,9 @@ def cross_entropy(input,
                                        label)
             print(dy_ret.numpy()) #[5.41993642]
 
-
-    Example2(soft labels):
-
         .. code-block:: python
-
+
+            # soft labels
             import paddle
             paddle.seed(99999)
             axis = -1
@@ -2896,7 +2895,6 @@ def cosine_embedding_loss(input1,
 
     Examples:
         .. code-block:: python
-          :name: code-example1
 
             import paddle
 

python/paddle/nn/functional/pooling.py

@@ -1286,26 +1286,25 @@ def adaptive_avg_pool1d(x, output_size, name=None):
         Tensor: The result of 1D adaptive average pooling. Its data type is same as input.
     Examples:
         .. code-block:: python
-          :name: adaptive_avg_pool1d-example
 
-              # average adaptive pool1d
-              # suppose input data in shape of [N, C, L], `output_size` is m or [m],
-              # output shape is [N, C, m], adaptive pool divide L dimension
-              # of input data into m grids averagely and performs poolings in each
-              # grid to get output.
-              # adaptive max pool performs calculations as follow:
-              #
-              #     for i in range(m):
-              #         lstart = floor(i * L / m)
-              #         lend = ceil((i + 1) * L / m)
-              #         output[:, :, i] = sum(input[:, :, lstart: lend])/(lstart - lend)
-              #
-              import paddle
-              import paddle.nn.functional as F
+            # average adaptive pool1d
+            # suppose input data in shape of [N, C, L], `output_size` is m or [m],
+            # output shape is [N, C, m], adaptive pool divide L dimension
+            # of input data into m grids averagely and performs poolings in each
+            # grid to get output.
+            # adaptive max pool performs calculations as follow:
+            #
+            #     for i in range(m):
+            #         lstart = floor(i * L / m)
+            #         lend = ceil((i + 1) * L / m)
+            #         output[:, :, i] = sum(input[:, :, lstart: lend])/(lstart - lend)
+            #
+            import paddle
+            import paddle.nn.functional as F
 
-              data = paddle.uniform([1, 3, 32])
-              pool_out = F.adaptive_avg_pool1d(data, output_size=16)
-              # pool_out shape: [1, 3, 16])
+            data = paddle.uniform([1, 3, 32])
+            pool_out = F.adaptive_avg_pool1d(data, output_size=16)
+            # pool_out shape: [1, 3, 16])
     """
     pool_type = 'avg'
     if not in_dynamic_mode():

python/paddle/nn/functional/vision.py

@@ -366,7 +366,6 @@ def pixel_unshuffle(x, downscale_factor, data_format="NCHW", name=None):
 
     Examples:
         .. code-block:: python
-            :name: pixel_unshuffle-example
 
             import paddle
             import paddle.nn.functional as F
@@ -423,7 +422,6 @@ def channel_shuffle(x, groups, data_format="NCHW", name=None):
 
     Examples:
         .. code-block:: python
-            :name: channel_shuffle-example
 
             import paddle
             import paddle.nn.functional as F

python/paddle/nn/initializer/constant.py

@@ -26,7 +26,7 @@ class Constant(ConstantInitializer):
 
     Examples:
         .. code-block:: python
-          :name: code-example1
+
             import paddle
             import paddle.nn as nn
 

python/paddle/nn/initializer/normal.py

@@ -72,7 +72,6 @@ class TruncatedNormal(TruncatedNormalInitializer):
 
     Examples:
         .. code-block:: python
-            :name: initializer_TruncatedNormal-example
 
             import paddle
 

python/paddle/nn/initializer/uniform.py

@@ -30,7 +30,6 @@ class Uniform(UniformInitializer):
 
     Examples:
         .. code-block:: python
-            :name: initializer_Uniform-example
 
             import paddle
 

python/paddle/nn/initializer/xavier.py

@@ -41,7 +41,6 @@ class XavierNormal(XavierInitializer):
 
     Examples:
         .. code-block:: python
-            :name: initializer_XavierNormal-example
 
             import paddle
 
@@ -97,7 +96,6 @@ class XavierUniform(XavierInitializer):
 
     Examples:
         .. code-block:: python
-            :name: initializer_XavierUniform-example
 
             import paddle
 

python/paddle/nn/layer/activation.py

@@ -486,7 +486,6 @@ class RReLU(Layer):
 
     Examples:
         .. code-block:: python
-            :name: RReLU-example
 
             import paddle
 

python/paddle/nn/layer/loss.py

@@ -295,15 +295,15 @@ class CrossEntropyLoss(Layer):
         - **input** (Tensor)
 
             Input tensor, the data type is float32, float64. Shape is
-	    :math:`[N_1, N_2, ..., N_k, C]`, where C is number of classes ,  ``k >= 1`` . 
+        :math:`[N_1, N_2, ..., N_k, C]`, where C is number of classes ,  ``k >= 1`` . 
 
             Note: 
 
                 1. when use_softmax=True, it expects unscaled logits. This operator should not be used with the 
                 output of softmax operator, which will produce incorrect results.
 
                 2. when use_softmax=False, it expects the output of softmax operator.
- 
+
 
         - **label** (Tensor)
 
@@ -313,7 +313,7 @@ class CrossEntropyLoss(Layer):
 
             2. If soft_label=True, the shape and data type should be same with ``input`` , 
             and the sum of the labels for each sample should be 1.
- 
+
         - **output** (Tensor)
 
             Return the softmax cross_entropy loss of ``input`` and ``label``.
@@ -328,10 +328,11 @@ class CrossEntropyLoss(Layer):
 
             2. if soft_label = True, the dimension of return value is :math:`[N_1, N_2, ..., N_k, 1]` . 
 
-     Example1(hard labels):
+    Examples:
 
         .. code-block:: python
 
+            # hard labels
             import paddle
             paddle.seed(99999)
             N=100
@@ -348,11 +349,9 @@ class CrossEntropyLoss(Layer):
                                        label)
             print(dy_ret.numpy()) #[5.41993642]
 
-
-    Example2(soft labels):
-
         .. code-block:: python
-
+
+            # soft labels
             import paddle
             paddle.seed(99999)
             axis = -1
@@ -1435,7 +1434,6 @@ class CosineEmbeddingLoss(Layer):
 
     Examples:
         .. code-block:: python
-          :name: code-example1
 
             import paddle