pypose.randn_like¶

class pypose.randn_like(input, sigma=1.0, **kwargs)[source]¶

Returns a LieTensor with the same size as input that is filled with random LieTensor with the corresponding input.ltype.

The corresponding random generator can be

List of available random LieTensor generator of input `ltype`.¶
Name	ltype	randn function	Manifold	randn function
Orthogonal Group	`SO3_type`	`randn_SO3()`	`so3_type`	`randn_so3()`
Euclidean Group	`SE3_type`	`randn_SE3()`	`se3_type`	`randn_se3()`
Similarity Group	`Sim3_type`	`randn_Sim3()`	`sim3_type`	`randn_sim3()`
Scaling Orthogonal	`RxSO3_type`	`randn_RxSO3()`	`rxso3_type`	`randn_rxso3()`

Parameters

input (LieTensor) – the size of input will determine size of the output tensor.
sigma (float or (float...), optional) – standard deviation for generating random LieTensors. Default: 1.0.
dtype (torch.dtype, optional) – the desired data type of returned Tensor. Default: if None, defaults to the dtype of input.
layout (torch.layout, optional) – the desired layout of returned tensor. Default: if None, defaults to the layout of input.
device (torch.device, optional) – the desired device of returned tensor. Default: if None, defaults to the device of input.
requires_grad (bool, optional) – If autograd should record operations on the returned tensor. Default: False.
memory_format (torch.memory_format, optional) – the desired memory format of returned Tensor. Default: torch.preserve_format.

Note

The parameter sigma (\(\sigma\)) can either be:

a single float – in which all the elements in the LieType share the same sigma.
a tuple of a number of floats – in which case, the specific sigma for each element can be assigned independently.

Note

If we have:

import pypose as pp
x = pp.SO3([0, 0, 0, 1])

Then the following two usages are equivalent:

pp.randn_like(x)
pp.randn_SO3(x.lshape, dtype=x.dtype, layout=x.layout, device=x.device)

Example

>>> x = pp.so3(torch.tensor([[0, 0, 0]]))
>>> pp.randn_like(x)
so3Type LieTensor:
tensor([[ 0.5162, -0.4600, -0.9085]])

Docs