未验证 提交 b89cea33 编写于 作者: Y yuchen202 提交者: GitHub

修改了API文档的相关内容 (#49055)

* 修改了API文档的相关内容

对weight_norm进行修改

* Update python/paddle/profiler/utils.py

* Update python/paddle/utils/cpp_extension/cpp_extension.py

* Update python/paddle/device/__init__.py

* Update python/paddle/device/__init__.py

* test=document_fix

* for Hyperlink; test=document_fix

* Update dlpack.py

* test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>
上级 2190ea09
...@@ -53,7 +53,8 @@ def is_compiled_with_npu(): ...@@ -53,7 +53,8 @@ def is_compiled_with_npu():
""" """
Whether paddle was built with WITH_ASCEND_CL=ON to support Ascend NPU. Whether paddle was built with WITH_ASCEND_CL=ON to support Ascend NPU.
Returns (bool): `True` if NPU is supported, otherwise `False`. Return:
bool, ``True`` if NPU is supported, otherwise ``False``.
Examples: Examples:
.. code-block:: python .. code-block:: python
......
...@@ -208,7 +208,7 @@ def max_memory_allocated(device=None): ...@@ -208,7 +208,7 @@ def max_memory_allocated(device=None):
For instance, a float32 tensor with shape [1] in GPU will take up 256 bytes memory, even though storing a float32 data requires only 4 bytes. For instance, a float32 tensor with shape [1] in GPU will take up 256 bytes memory, even though storing a float32 data requires only 4 bytes.
Args: Args:
device(paddle.CUDAPlace or int or str): The device, the id of the device or device(paddle.CUDAPlace or int or str, optional): The device, the id of the device or
the string name of device like 'gpu:x'. If device is None, the device is the current device. the string name of device like 'gpu:x'. If device is None, the device is the current device.
Default: None. Default: None.
...@@ -239,7 +239,7 @@ def max_memory_reserved(device=None): ...@@ -239,7 +239,7 @@ def max_memory_reserved(device=None):
Return the peak size of GPU memory that is held by the allocator of the given device. Return the peak size of GPU memory that is held by the allocator of the given device.
Args: Args:
device(paddle.CUDAPlace or int or str): The device, the id of the device or device(paddle.CUDAPlace or int or str, optional): The device, the id of the device or
the string name of device like 'gpu:x'. If device is None, the device is the current device. the string name of device like 'gpu:x'. If device is None, the device is the current device.
Default: None. Default: None.
...@@ -396,7 +396,7 @@ def get_device_properties(device=None): ...@@ -396,7 +396,7 @@ def get_device_properties(device=None):
Return the properties of given device. Return the properties of given device.
Args: Args:
device(paddle.CUDAPlace or int or str): The device, the id of the device or device(paddle.CUDAPlace or int or str, optional): The device, the id of the device or
the string name of device like 'gpu:x' which to get the properties of the the string name of device like 'gpu:x' which to get the properties of the
device from. If device is None, the device is the current device. device from. If device is None, the device is the current device.
Default: None. Default: None.
......
...@@ -170,9 +170,10 @@ def weight_norm(layer, name='weight', dim=0): ...@@ -170,9 +170,10 @@ def weight_norm(layer, name='weight', dim=0):
Weight normalization is a reparameterization of the weight vectors in a neural network that Weight normalization is a reparameterization of the weight vectors in a neural network that
decouples the magnitude of those weight vectors from their direction. Weight normalization decouples the magnitude of those weight vectors from their direction. Weight normalization
replaces the parameter specified by `name`(eg: 'weight') with two parameters: one parameter replaces the parameter specified by ``name`` (eg: 'weight') with two parameters: one parameter
specifying the magnitude (eg: 'weight_g') and one parameter specifying the direction specifying the magnitude (eg: 'weight_g') and one parameter specifying the direction
(eg: 'weight_v'). Weight normalization has been implemented as discussed in this paper: (eg: 'weight_v'). Weight normalization has been implemented as discussed in this paper:
`Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks `Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
<https://arxiv.org/pdf/1602.07868.pdf>`_. <https://arxiv.org/pdf/1602.07868.pdf>`_.
......
...@@ -63,7 +63,7 @@ class RecordEvent(ContextDecorator): ...@@ -63,7 +63,7 @@ class RecordEvent(ContextDecorator):
result = data1 + data2 result = data1 + data2
record_event.end() record_event.end()
**Note**: Note:
RecordEvent will take effect only when :ref:`Profiler <api_paddle_profiler_Profiler>` is on and at the state of `RECORD`. RecordEvent will take effect only when :ref:`Profiler <api_paddle_profiler_Profiler>` is on and at the state of `RECORD`.
""" """
......
...@@ -69,7 +69,7 @@ def from_dlpack(dlpack): ...@@ -69,7 +69,7 @@ def from_dlpack(dlpack):
dlpack (PyCapsule): a PyCapsule object with the dltensor. dlpack (PyCapsule): a PyCapsule object with the dltensor.
Returns: Returns:
out (Tensor): a tensor decoded from DLPack. One thing to be noted, if we get out (Tensor), a tensor decoded from DLPack. One thing to be noted, if we get
an input dltensor with data type as `bool`, we return the decoded an input dltensor with data type as `bool`, we return the decoded
tensor as `uint8`. tensor as `uint8`.
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册