423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-12-22 07:10:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

remove requires_grad from random rotations

Browse Source 
Summary: Because rotations and (rotation) quaternions live on curved manifolds, it doesn't make sense to optimize them directly. Having a prominent option to require gradient on random ones may cause people to try, and isn't particularly useful.

Reviewed By: theschnitz

Differential Revision: D29160734

fbshipit-source-id: fc9e320672349fe334747c5b214655882a460a62
This commit is contained in:
Jeremy Reizenstein
2021-06-21 11:45:01 -07:00
committed by Facebook GitHub Bot
parent 31c448a95d
commit ce60d4b00e
2 changed files with 12 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
pytorch3d/transforms/rotation_conversions.py
View File

@@ -282,9 +282,7 @@ def matrix_to_euler_angles(matrix, convention: str):
return torch.stack(o, -1)
def random_quaternions(
n: int, dtype: Optional[torch.dtype] = None, device=None, requires_grad=False
):
def random_quaternions(n: int, dtype: Optional[torch.dtype] = None, device=None):
"""
Generate random quaternions representing rotations,
i.e. versors with nonnegative real part.
@@ -294,21 +292,17 @@ def random_quaternions(
dtype: Type to return.
device: Desired device of returned tensor. Default:
uses the current device for the default tensor type.
requires_grad: Whether the resulting tensor should have the gradient
flag set.
Returns:
Quaternions as tensor of shape (N, 4).
"""
o = torch.randn((n, 4), dtype=dtype, device=device, requires_grad=requires_grad)
o = torch.randn((n, 4), dtype=dtype, device=device)
s = (o * o).sum(1)
o = o / _copysign(torch.sqrt(s), o[:, 0])[:, None]
return o
def random_rotations(
n: int, dtype: Optional[torch.dtype] = None, device=None, requires_grad=False
):
def random_rotations(n: int, dtype: Optional[torch.dtype] = None, device=None):
"""
Generate random rotations as 3x3 rotation matrices.
@@ -317,21 +311,15 @@ def random_rotations(
dtype: Type to return.
device: Device of returned tensor. Default: if None,
uses the current device for the default tensor type.
requires_grad: Whether the resulting tensor should have the gradient
flag set.
Returns:
Rotation matrices as tensor of shape (n, 3, 3).
"""
quaternions = random_quaternions(
n, dtype=dtype, device=device, requires_grad=requires_grad
)
quaternions = random_quaternions(n, dtype=dtype, device=device)
return quaternion_to_matrix(quaternions)
def random_rotation(
dtype: Optional[torch.dtype] = None, device=None, requires_grad=False
):
def random_rotation(dtype: Optional[torch.dtype] = None, device=None):
"""
Generate a single random 3x3 rotation matrix.
@@ -339,13 +327,11 @@ def random_rotation(
dtype: Type to return
device: Device of returned tensor. Default: if None,
uses the current device for the default tensor type
requires_grad: Whether the resulting tensor should have the gradient
flag set
Returns:
Rotation matrix as tensor of shape (3, 3).
"""
return random_rotations(1, dtype, device, requires_grad)[0]
return random_rotations(1, dtype, device)[0]
def standardize_quaternion(quaternions):