gpu: nvidia: ip: respect acc_mode for sum post-op #2479
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
0f0897a to
383b1a2
Compare
mgouicem
reviewed
Jan 22, 2025
Rbiessy
reviewed
Jan 22, 2025
AD2605
approved these changes
Jan 22, 2025
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
383b1a2 to
441a27d
Compare
github-actions
bot
added
documentation
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
441a27d to
7841ac5
Compare
Rbiessy
approved these changes
Feb 4, 2025
ShanoToni
approved these changes
Feb 5, 2025
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
7841ac5 to
d1255da
Compare
Rbiessy
approved these changes
Feb 6, 2025
|
I think this PR still needs a review from @oneapi-src/onednn-doc |
ranukund
reviewed
Feb 6, 2025
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
d1255da to
a8d99a7
Compare
|
Thanks @ranukund , I added the suggested changes. |
sgeor255
changed the title
sgeor255
changed the title
dzarukin
reviewed
Feb 19, 2025
sgeor255
force-pushed
the
svet/nvdia-ip-precision
branch
from
a8d99a7 to
90190e7
Compare
|
It looks like the precommit check failures are in files unrelated to this PR. |
dzarukin
approved these changes
Feb 24, 2025
manaalmj
pushed a commit
to manaalmj/oneDNN
that referenced
this pull request
Mar 4, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Currently the inner product error threshold in benchdnn is set to 0. In some cases for large shapes on nvidia backend there are some precision issues (e.g. the cases reported in MFDNN-12610). This PR adjusts the error threshold so that such cases are not reported as failures.
Fixes MFDNN-12610.