mindsponge.common.rots_scale

mindsponge.common.rots_scale(rot, scale)[源代码]

对旋转矩阵进行缩放。

\[\begin{split}\begin{split} &rot=(xx,xy,xz,yx,yy,yz,zx,zy,zz) \\ &scaled\_{rots} = (scale*xx,scale*xy,scale*xz,scale*yx,scale*yy,scale*yz,scale*zx,scale*zy,scale*zz) \end{split}\end{split}\]
参数:
  • rot (Tuple) - 旋转矩阵,tuple长度为9,\((xx,xy,xz,yx,yy,yz,zx,zy,zz)\),其中每个元素是标量或者Tensor。 若为Tensor其shape相同。

  • scale (float) - 缩放值。

返回:
  • scaled_rots (Tuple) -缩放后的旋转矩阵,tuple长度为9,shape与rot相同。

支持平台:

Ascend GPU CPU

样例:

>>> import numpy as np
>>> from mindspore import Tensor
>>> from mindspore import dtype as mstype
>>> from mindsponge.common.geometry import rots_scale
>>> x = Tensor(np.ones(256), mstype.float32)
>>> result = rots_scale((x, x, x, x, x, x, x, x, x),10)
>>> print(len(result))
>>> print(result[0].shape)
>>> print(result[1].shape)
>>> print(result[2].shape)
>>> print(result[3].shape)
>>> print(result[4].shape)
>>> print(result[5].shape)
>>> print(result[6].shape)
>>> print(result[7].shape)
>>> print(result[8].shape)
3
(256,)
(256,)
(256,)
(256,)
(256,)
(256,)
(256,)
(256,)
(256,)