423A35C7/pytorch3d
1
0
Fork 0
mirror of https://github.com/facebookresearch/pytorch3d.git synced 2025-08-02 11:52:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

run lint

Browse Source 
Summary: Run `/dev/linter.sh` to fix linting

Reviewed By: nikhilaravi

Differential Revision: D20584037

fbshipit-source-id: 69e45b33d22e3d54b6d37c3c35580bb3e9dc50a5
This commit is contained in:
Georgia Gkioxari 2020-03-21 17:55:21 -07:00 committed by Facebook GitHub Bot
parent 6d34e1c60d
commit 03f441e7ca
4 changed files with 18 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
pytorch3d/renderer/cameras.py
View File

@ -2,7 +2,7 @@
import math import math
import numpy as np import numpy as np
from typing import Tuple, Optional, Sequence from typing import Optional, Sequence, Tuple
import torch import torch
import torch.nn.functional as F import torch.nn.functional as F
@ -1039,15 +1039,18 @@ def look_at_view_transform(
if eye is not None: if eye is not None:
broadcasted_args = convert_to_tensors_and_broadcast( broadcasted_args = convert_to_tensors_and_broadcast(
eye, at, up, device=device) eye, at, up, device=device
)
eye, at, up = broadcasted_args eye, at, up = broadcasted_args
C = eye C = eye
else: else:
broadcasted_args = convert_to_tensors_and_broadcast( broadcasted_args = convert_to_tensors_and_broadcast(
dist, elev, azim, at, up, device=device) dist, elev, azim, at, up, device=device
)
dist, elev, azim, at, up = broadcasted_args dist, elev, azim, at, up = broadcasted_args
C = camera_position_from_spherical_angles( C = camera_position_from_spherical_angles(
dist, elev, azim, degrees=degrees, device=device) dist, elev, azim, degrees=degrees, device=device
)
R = look_at_rotation(C, at, up, device=device) R = look_at_rotation(C, at, up, device=device)
T = -torch.bmm(R.transpose(1, 2), C[:, :, None])[:, :, 0] T = -torch.bmm(R.transpose(1, 2), C[:, :, None])[:, :, 0]

2
tests/test_cameras.py
View File

@ -121,7 +121,7 @@ class TestCameraHelpers(unittest.TestCase):
dist = math.sqrt(2) dist = math.sqrt(2)
elev = math.pi / 4 elev = math.pi / 4
azim = 0.0 azim = 0.0
eye = ((0.0, 1.0, 1.0), ) eye = ((0.0, 1.0, 1.0),)
# using passed values for dist, elev, azim # using passed values for dist, elev, azim
R, t = look_at_view_transform(dist, elev, azim, degrees=False) R, t = look_at_view_transform(dist, elev, azim, degrees=False)
# using other values for dist, elev, azim - eye overrides # using other values for dist, elev, azim - eye overrides

2
tests/test_compositing.py
View File

@ -188,7 +188,7 @@ class TestAccumulatePoints(unittest.TestCase):
): ):
res1 = fn1(*args1) res1 = fn1(*args1)
res2 = fn2(*args2) res2 = fn2(*args2)
self.assertTrue(torch.allclose(res1.cpu(), res2.cpu(), atol=1e-6)) self.assertTrue(torch.allclose(res1.cpu(), res2.cpu(), atol=1e-6))
if not compare_grads: if not compare_grads:

13
tests/test_sample_points_from_meshes.py
View File

@ -291,7 +291,7 @@ class TestSamplePoints(unittest.TestCase):
if sampled_weights.min() <= 0: if sampled_weights.min() <= 0:
return False return False
return True return True
def test_verts_nan(self): def test_verts_nan(self):
num_verts = 30 num_verts = 30
num_faces = 50 num_faces = 50
@ -300,14 +300,19 @@ class TestSamplePoints(unittest.TestCase):
verts = torch.rand( verts = torch.rand(
(num_verts, 3), dtype=torch.float32, device=device (num_verts, 3), dtype=torch.float32, device=device
) )
# randomly assign an invalid type # randomly assign an invalid type
verts[torch.randperm(num_verts)[:10]] = float(invalid) verts[torch.randperm(num_verts)[:10]] = float(invalid)
faces = torch.randint( faces = torch.randint(
num_verts, size=(num_faces, 3), dtype=torch.int64, device=device num_verts,
size=(num_faces, 3),
dtype=torch.int64,
device=device,
) )
meshes = Meshes(verts=[verts], faces=[faces]) meshes = Meshes(verts=[verts], faces=[faces])
with self.assertRaisesRegex(ValueError, "Meshes contain nan or inf."): with self.assertRaisesRegex(
ValueError, "Meshes contain nan or inf."
):
sample_points_from_meshes( sample_points_from_meshes(
meshes, num_samples=100, return_normals=True meshes, num_samples=100, return_normals=True
) )