Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Adds ATen fallback for scaled_dot_product_attention #21107

Open
wants to merge 18 commits into
base: main
Choose a base branch
from
Open

Adds ATen fallback for scaled_dot_product_attention #21107

Changes from 1 commit
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
f22b8dc
attn aten fallback
Jun 19, 2024
612e425
use correct operator names
Jun 20, 2024
bdcfebb
formatting
Jun 20, 2024
80c3107
add unit test
Jun 20, 2024
2b29b4c
formatting
Jun 20, 2024
d2b8566
use pytorch sdpa kernel
Jun 26, 2024
0ca8fa0
bug fix
Jun 26, 2024
8999ff2
lint
Jun 26, 2024
6bf3018
use different kernel
Jun 27, 2024
35bd07a
formatting
Jun 27, 2024
dd1849a
include Peng's & Vincent's editS
Jul 2, 2024
8219ec9
adjust test and comments
prathikr Jul 2, 2024
65c2cb7
move import inside test
prathikr Jul 2, 2024
b5f5863
merge with master
prathikr Jul 2, 2024
18648ad
feature flag
prathikr Jul 2, 2024
be9ce0a
add documentation
prathikr Jul 2, 2024
e269e89
minor fixes
prathikr Jul 2, 2024
d3cc487
doc update
prathikr Jul 2, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
bug fix
  • Loading branch information
root committed Jun 26, 2024
commit 0ca8fa0019b1cd4b62e56fc545ecd073c1124ed5
1 change: 0 additions & 1 deletion orttraining/orttraining/test/python/orttraining_test_ortmodule_api.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6962,4 +6962,3 @@ def run_step(model, inputs):

_test_helpers.assert_values_are_close(ort_prediction, pt_prediction)
_test_helpers.assert_values_are_close(ort_input.grad, pt_input.grad)
test_aten_attention()
Loading