Siddhartha Menon
|
e1e6417d4c
|
Add SVE implementation of embedding_lookup_idx (#133995)
Adds an accelerated version of the embedding_lookup_idx perfkernels. This is done via a python codegen file similarly to `caffe2/perfkernels/hp_emblookup_codegen.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133995
Approved by: https://github.com/malfet, https://github.com/huydhn
|
2024-10-15 18:52:44 +00:00 |
|
PyTorch MergeBot
|
dac0b4e62b
|
Revert "Add SVE implementation of embedding_lookup_idx (#133995)"
This reverts commit 770c134998.
Reverted https://github.com/pytorch/pytorch/pull/133995 on behalf of https://github.com/clee2000 due to breaking internal tests, I wondering if this just needs a targets change for buck? ([comment](https://github.com/pytorch/pytorch/pull/133995#issuecomment-2414596554))
|
2024-10-15 17:23:50 +00:00 |
|
Siddhartha Menon
|
770c134998
|
Add SVE implementation of embedding_lookup_idx (#133995)
Adds an accelerated version of the embedding_lookup_idx perfkernels. This is done via a python codegen file similarly to `caffe2/perfkernels/hp_emblookup_codegen.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133995
Approved by: https://github.com/malfet, https://github.com/huydhn
|
2024-10-14 10:17:27 +00:00 |
|