-
Notifications
You must be signed in to change notification settings - Fork 521
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.
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
ChooseQuantizationParams is not checking for min/max validity like Pytorch does #1590
Comments
zhengwy888
changed the title
ChooseQuantizationParams is not checking for min/max validity
ChooseQuantizationParams is not checking for min/max validity like Pytorch does
Feb 10, 2023
cc @jspark1105 could you help ;-) |
jspark1105
added a commit
to jspark1105/FBGEMM
that referenced
this issue
Feb 24, 2023
Summary: pytorch#1590 Differential Revision: D43383474 fbshipit-source-id: ff27b44b589f8dbde54f3a2c830d7d489f7a5138
jspark1105
added a commit
to jspark1105/FBGEMM
that referenced
this issue
Feb 26, 2023
Summary: X-link: pytorch/pytorch#95529 Pull Request resolved: pytorch#1612 pytorch#1590 Reviewed By: summerdengfb Differential Revision: D43383474 fbshipit-source-id: 7b61eda2e80e5ca9c34cfe6b26f85adfd20ddc6e
jspark1105
added a commit
to jspark1105/pytorch
that referenced
this issue
Feb 26, 2023
Summary: Pull Request resolved: pytorch#95529 X-link: pytorch/FBGEMM#1612 pytorch/FBGEMM#1590 Test Plan: CI Reviewed By: summerdengfb Differential Revision: D43383474 fbshipit-source-id: 337c82fa89b14b3633a2d61f233c98c1994bc71e
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi, in Pytorch's ChooseQuantizationParams they check for
min <= max
, but FBGEMM doesn't. This causes discrepancy in the behavior of pytorch code when compiled with USE_FBGEMM=1 and USE_FBGEMM=0. Can FBGEMM port this check over to make the behavior consistent?The text was updated successfully, but these errors were encountered: