Implement _fft_r2c core ATen op #8277
Conversation
|
Stack from ghstack (oldest at bottom): |
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8277
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e23ad51 with merge base 8ec08f9 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Mostly following suit on PyTorch impl here. (For your reference: https://github.com/pytorch/pytorch/blob/main/aten/src/ATen/native/mkl/SpectralOps.cpp#L281) ghstack-source-id: d05d5db7f08d98baec9a37d25c4561128bd8a625 ghstack-comment-id: 2640451211 Pull Request resolved: #8277
swolchok
added
the
release notes: ops & kernels
|
update coming soon to fix internal build issues |
Mostly following suit on PyTorch impl here. (For your reference: https://github.com/pytorch/pytorch/blob/main/aten/src/ATen/native/mkl/SpectralOps.cpp#L281) ghstack-source-id: 5caee518c2de5c48e233b442510007a66a40ee2b ghstack-comment-id: 2640451211 Pull Request resolved: #8277
swolchok
requested review from
kirklandsign,
GregoryComer and
manuelcandales
and removed request for
kirklandsign
|
no code sharing for this one? |
What's to share? The interesting bits are pocketfft, and we use that, same as PyTorch core. |
I havent looked in depth but shouldnt this file kernels/optimized/cpu/op_fft_r2c.cpp be subset of https://github.com/pytorch/pytorch/blob/main/aten/src/ATen/native/mkl/SpectralOps.cpp#L281 |
Mostly following suit on PyTorch impl here. (For your reference: https://github.com/pytorch/pytorch/blob/main/aten/src/ATen/native/mkl/SpectralOps.cpp#L281)