GaussianNoise CV-CUDA Backend #9288
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/9288
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ❌ 5 New FailuresAs of commit c7e0725 with merge base aa35ca1 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
justincdavis
force-pushed
the
feat/gaussian_noise_cvcuda
branch
from
3386083 to
2102fb5
Compare
zy1git
left a comment
zy1git
left a comment
There was a problem hiding this comment.
I left some comments for the first round review. Feel free to let me know your thoughts.
|
|
||
| # per-channel means each channel gets unique random noise, same behavior as torch.randn_like | ||
| # produce a seed with torch RNG, if seed is manually set then this will be deterministic | ||
| # note: clip is not supported in CV-CUDA, so we don't need to clamp the values |
There was a problem hiding this comment.
The clip parameter is accepted but has no effect since CV-CUDA's gaussiannoise always clamps output. Should we raise a warning when clip=False is passed to inform users that their request cannot be honored? Like:
if not clip: warnings.warn("clip=False is not supported for CV-CUDA backend; output will still be clipped") ...
| sigma_tensor = cvcuda.as_tensor(torch.full((batch_size,), sigma, dtype=torch.float32).cuda(), "N") | ||
|
|
||
| # per-channel means each channel gets unique random noise, same behavior as torch.randn_like | ||
| # produce a seed with torch RNG, if seed is manually set then this will be deterministic |
There was a problem hiding this comment.
Do we need to delete these comments?
2 participants
Summary
Adds the CV-CUDA backend for the
gaussian_noisetransformRun tests
python3 -m pytest test/test_transforms_v2.py::TestGaussianNoise