提交 b3975685 编写于 作者: oiouou123's avatar oiouou123 提交者: MegChai

docs(mge/functional): update functional.tensor.reshape docstring

无相关合并请求
...@@ -851,34 +851,33 @@ def transpose(inp: Tensor, pattern: Iterable[int]) -> Tensor: ...@@ -851,34 +851,33 @@ def transpose(inp: Tensor, pattern: Iterable[int]) -> Tensor:
return inp.transpose(list(-1 if _ == "x" else _ for _ in pattern)) return inp.transpose(list(-1 if _ == "x" else _ for _ in pattern))
def reshape(inp: Tensor, target_shape: Iterable[int]) -> Tensor: def reshape(inp: Tensor, shape: Iterable[int]) -> Tensor:
r"""Reshapes a tensor to given target shape; total number of logical elements must r"""Reshapes a tensor without changing its data.
remain unchanged
Args: Args:
inp: input tensor. inp (Tensor): input tensor to reshape.
target_shape: target shape, it can contain an element of -1 representing ``unspec_axis``. shape (sequence of ints): target shape compatible with the original shape. One shape dimension is allowed
to be ``-1``. When a shape dimension is ``-1``, the corresponding output tensor shape dimension
Examples: must be inferred from the length of the tensor and the remaining dimensions.
.. testcode::
import numpy as np
from megengine import tensor
import megengine.functional as F
x = tensor(np.arange(12, dtype=np.int32))
out = F.reshape(x, (3, 4))
print(out.numpy())
Outputs: Returns:
an output tensor having the same data type, elements, and underlying element order as ``inp``.
.. testoutput:: Examples:
[[ 0 1 2 3] >>> x = F.arange(12)
[ 4 5 6 7] >>> x
[ 8 9 10 11]] Tensor([ 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.], device=xpux:0)
>>> F.reshape(x, (3, 4))
Tensor([[ 0. 1. 2. 3.]
[ 4. 5. 6. 7.]
[ 8. 9. 10. 11.]], device=xpux:0)
>>> F.reshape(x, (2, -1))
Tensor([[ 0. 1. 2. 3. 4. 5.]
[ 6. 7. 8. 9. 10. 11.]], device=xpux:0)
""" """
return inp.reshape(target_shape) return inp.reshape(shape)
def flatten(inp: Tensor, start_axis: int = 0, end_axis: int = -1) -> Tensor: def flatten(inp: Tensor, start_axis: int = 0, end_axis: int = -1) -> Tensor:
......
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
反馈
建议
客服 返回
顶部