未验证 提交 a2a0251d 编写于 作者: T Tao Luo 提交者: GitHub

Merge pull request #15681 from chengduoZH/refine_api_doc

Refine the doc of reshape, stack
......@@ -5936,13 +5936,10 @@ def reshape(x, shape, actual_shape=None, act=None, inplace=False, name=None):
than :attr:`shape`.
act (str): The non-linear activation to be applied to the reshaped tensor
variable.
inplace(bool): Must use :attr:`False` if :attr:`x` is used in multiple
operators. If this flag is set :attr:`True`, reuse input
:attr:`x` to reshape, which will change the shape of
tensor variable :attr:`x` and might cause errors when
:attr:`x` is used in multiple operators. If :attr:`False`,
preserve the shape :attr:`x` and create a new output tensor
variable whose data is copied from input x but reshaped.
inplace(bool): If ``inplace`` is `True`, the input and output of ``layers.reshape``
are the same variable, otherwise, the input and output of
``layers.reshape`` are different variables. Note that if :attr:`x`
is more than one layer's input, ``inplace`` must be :attr:`False`.
name (str): The name of this layer. It is optional.
Returns:
......@@ -8335,6 +8332,46 @@ def stack(x, axis=0):
If :code:`axis` < 0, it would be replaced with :code:`axis+rank(x[0])+1`.
If :code:`axis` is None, it would be replaced with 0.
For Example:
.. code-block:: text
Case 1:
Input:
x[0].data = [ [1.0 , 2.0 ] ]
x[0].dims = [1, 2]
x[1].data = [ [3.0 , 4.0 ] ]
x[1].dims = [1, 2]
x[2].data = [ [5.0 , 6.0 ] ]
x[2].dims = [1, 2]
Attrs:
axis = 0
Output:
Out.data =[ [ [1.0, 2.0] ],
[ [3.0, 4.0] ],
[ [5.0, 6.0] ] ]
Out.dims = [3, 1, 2]
Case 2:
Given
x[0].data = [ [1.0 , 2.0 ] ]
x[0].dims = [1, 2]
x[1].data = [ [3.0 , 4.0 ] ]
x[1].dims = [1, 2]
x[2].data = [ [5.0 , 6.0 ] ]
x[2].dims = [1, 2]
Attrs:
axis = 1 or axis = -2
Output:
Out.data =[ [ [1.0, 2.0]
[3.0, 4.0]
[5.0, 6.0] ] ]
Out.dims = [1, 3, 2]
Args:
x (Variable|list(Variable)|tuple(Variable)): Input variables.
axis (int|None): The axis along which all inputs are stacked.
......
......@@ -567,7 +567,7 @@ def ones(shape, dtype, force_cpu=False):
It also sets *stop_gradient* to True.
Args:
shape(tuple|list|None): Shape of output tensor
shape(tuple|list): Shape of output tensor
dtype(np.dtype|core.VarDesc.VarType|str): Data type of output tensor
Returns:
......@@ -578,6 +578,10 @@ def ones(shape, dtype, force_cpu=False):
data = fluid.layers.ones(shape=[1], dtype='int64')
"""
assert isinstance(shape, list) or isinstance(
shape, tuple), "The shape's type should be list or tuple."
assert reduce(lambda x, y: x * y,
shape) > 0, "The shape is invalid: %s." % (str(shape))
return fill_constant(value=1.0, **locals())
......
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