fix en docs of some Apis (gradients, scope_guard, cuda_places, name_scope, device_guard, load_program_state, scale, ParamAttr and WeightNormParamAttr)

python/paddle/fluid/backward.py

@@ -2021,7 +2021,6 @@ def calc_gradient(targets, inputs, target_gradients=None, no_grad_set=None):
 @framework.static_only
 def gradients(targets, inputs, target_gradients=None, no_grad_set=None):
     """
-    :api_attr: Static Graph
 
     Backpropagate the gradients of targets to inputs.
 
@@ -2042,8 +2041,9 @@ def gradients(targets, inputs, target_gradients=None, no_grad_set=None):
         will be None.
 
     Examples:
+
         .. code-block:: python
-
+          :name: code-example
             import paddle
             import paddle.nn.functional as F
 
@@ -2054,7 +2054,7 @@ def gradients(targets, inputs, target_gradients=None, no_grad_set=None):
             y = paddle.static.nn.conv2d(x, 4, 1, bias_attr=False)
             y = F.relu(y)
             z = paddle.static.gradients([y], x)
-            print(z) # [var x@GRAD : fluid.VarType.LOD_TENSOR.shape(-1L, 2L, 8L, 8L).astype(VarType.FP32)]
+            print(z) # [var x@GRAD : LOD_TENSOR.shape(-1, 2, 8, 8).dtype(float32).stop_gradient(False)]
     """
     check_type(targets, 'targets', (framework.Variable, list, tuple),
                'paddle.static.gradients')

python/paddle/fluid/executor.py

@@ -75,7 +75,6 @@ def _switch_scope(scope):
 @signature_safe_contextmanager
 def scope_guard(scope):
     """
-    :api_attr: Static Graph
 
     This function switches scope through python `with` statement.
     Scope records the mapping between variable names and variables ( :ref:`api_guide_Variable` ),
@@ -94,6 +93,7 @@ def scope_guard(scope):
         None
 
     Examples:
+
         .. code-block:: python
 
             import paddle

python/paddle/fluid/framework.py

@@ -690,7 +690,7 @@ def is_compiled_with_rocm():
 
 def cuda_places(device_ids=None):
     """
-    **Note**:
+    Note:
         For multi-card tasks, please use `FLAGS_selected_gpus` environment variable to set the visible GPU device.
         The next version will fix the problem with `CUDA_VISIBLE_DEVICES` environment variable.
 
@@ -715,6 +715,7 @@ def cuda_places(device_ids=None):
         list of paddle.CUDAPlace: Created GPU place list.
 
     Examples:
+
         .. code-block:: python
 
             import paddle
@@ -835,6 +836,7 @@ def cpu_places(device_count=None):
         list of paddle.CPUPlace: Created list of CPU places.
 
     Examples:
+
         .. code-block:: python
 
             import paddle
@@ -954,7 +956,6 @@ def name(self):
 @signature_safe_contextmanager
 def name_scope(prefix=None):
     """
-    :api_attr: Static Graph
 
     Generate hierarchical name prefix for the operators in Static Graph.
 
@@ -967,6 +968,7 @@ def name_scope(prefix=None):
         prefix(str, optional): prefix. Default is none.
 
     Examples:
+
         .. code-block:: python
 
           import paddle
@@ -6877,8 +6879,9 @@ def switch_device(device):
 @signature_safe_contextmanager
 def device_guard(device=None):
     """
-    **Notes**:
-        **The API only supports static mode.**
+
+    Note:
+        The API only supports static mode.
 
     A context manager that specifies the device on which the OP will be placed.
 
@@ -6892,6 +6895,7 @@ def device_guard(device=None):
             assigned devices.
 
     Examples:
+
         .. code-block:: python
 
             import paddle

python/paddle/fluid/io.py

@@ -2154,7 +2154,6 @@ def set_var(var, ndarray):
 
 def load_program_state(model_path, var_list=None):
     """
-    :api_attr: Static Graph
 
     Load program state from local file
 
@@ -2169,6 +2168,7 @@ def load_program_state(model_path, var_list=None):
         state_dict(dict): the dict store Parameter and optimizer information
 
     Examples:
+
         .. code-block:: python
 
             import paddle

python/paddle/fluid/layers/nn.py

@@ -11803,8 +11803,7 @@ def _elementwise_op(helper):
 
 def scale(x, scale=1.0, bias=0.0, bias_after_scale=True, act=None, name=None):
     """
-    Scale operator.
-
+
     Putting scale and bias to the input Tensor as following:
 
     ``bias_after_scale`` is True:
@@ -11829,6 +11828,7 @@ def scale(x, scale=1.0, bias=0.0, bias_after_scale=True, act=None, name=None):
         Tensor: Output tensor of scale operator, with shape and data type same as input.
 
     Examples:
+
         .. code-block:: python
 
             # scale as a float32 number

python/paddle/fluid/param_attr.py

@@ -30,16 +30,17 @@
 
 class ParamAttr(object):
     """
-    Create a object to represent the attribute of parameter. The attributes are:
-    name, initializer, learning rate, regularizer, trainable, gradient clip,
-    and model average.
-
+
     Note:
         ``gradient_clip`` of ``ParamAttr`` HAS BEEN DEPRECATED since 2.0. 
         Please use ``need_clip`` in ``ParamAttr`` to speficiy the clip scope.
         There are three clipping strategies: :ref:`api_paddle_nn_ClipGradByGlobalNorm` , 
         :ref:`api_paddle_nn_ClipGradByNorm` , :ref:`api_paddle_nn_ClipGradByValue` .
 
+    Create a object to represent the attribute of parameter. The attributes are:
+    name, initializer, learning rate, regularizer, trainable, gradient clip,
+    and model average.
+
     Parameters:
         name (str, optional): The parameter's name. Default None, meaning that the name
                 would be created automatically.
@@ -63,6 +64,7 @@ class ParamAttr(object):
        ParamAttr Object.
 
     Examples:
+
         .. code-block:: python
 
             import paddle
@@ -213,24 +215,22 @@ def _to_kwargs(self, with_initializer=False):
 
 class WeightNormParamAttr(ParamAttr):
     r"""
-	:api_attr: Static Graph
 
     Note:
         Please use 'paddle.nn.utils.weight_norm' in dygraph mode.
-
+
+    Note:
+        ``gradient_clip`` of ``ParamAttr`` HAS BEEN DEPRECATED since 2.0. 
+        Please use ``need_clip`` in ``ParamAttr`` to speficiy the clip scope.
+        There are three clipping strategies: :ref:`api_paddle_nn_ClipGradByGlobalNorm` , 
+        :ref:`api_paddle_nn_ClipGradByNorm` , :ref:`api_paddle_nn_ClipGradByValue` .
+
     Parameter of weight Norm. Weight Norm is a reparameterization of the weight vectors
     in a neural network that decouples the magnitude of those weight vectors from
     their direction. Weight Norm has been implemented as discussed in this
     paper: `Weight Normalization: A Simple Reparameterization to Accelerate
     Training of Deep Neural Networks
     <https://arxiv.org/pdf/1602.07868.pdf>`_.
-
-    Note:
-        ``gradient_clip`` of ``ParamAttr`` HAS BEEN DEPRECATED since 2.0. 
-        Please use ``need_clip`` in ``ParamAttr`` to speficiy the clip scope.
-        There are three clipping strategies: :ref:`api_paddle_nn_ClipGradByGlobalNorm` , 
-        :ref:`api_paddle_nn_ClipGradByNorm` , :ref:`api_paddle_nn_ClipGradByValue` .
-
 
     Args:
         dim(int, optional): Dimension over which to compute the norm. Dim is a non-negative
@@ -258,6 +258,7 @@ class WeightNormParamAttr(ParamAttr):
         need_clip (bool, optional): Whether the parameter gradient need to be cliped in optimizer. Default is True.
 
     Examples:
+
         .. code-block:: python
 
             import paddle