frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Revert "[PyTorch] Add native fast path for transformer encoder inference"

Browse Source 
This reverts commit b369b89f235f54bc9de85d768fb62ac4579681dc.

This has internal changes and should not have been landed via mergebot.

Ref: https://github.com/pytorch/pytorch/pull/75809#issuecomment-1108717166
This commit is contained in:
Jon Janzen
2022-04-25 11:40:02 -04:00
parent 1c35f37c9f
commit 2387efd356
16 changed files with 30 additions and 1815 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
torch/utils/hipify/hipify_python.py
View File

@ -811,7 +811,6 @@ def preprocessor(
or f.startswith("ATen/native/nested/cuda")
or f.startswith("ATen/native/quantized/cuda")
or f.startswith("ATen/native/sparse/cuda")
or f.startswith("ATen/native/transformers/cuda")
or f.startswith("THC/")
or (f.startswith("THC") and not f.startswith("THCP"))
):