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

[Forward Call] fake quant fix #145

Merged
merged 4 commits into from
Sep 3, 2024
Merged

[Forward Call] fake quant fix #145

merged 4 commits into from
Sep 3, 2024

Conversation

horheynm
Copy link
Member

Pass in updated scale or zp.

Fixed off by one error - Previously, the params gets registered by the module and will use that instead of the updated.

bfineran
bfineran previously approved these changes Aug 29, 2024
View reviewed changes
@kylesayrs
Copy link
Contributor

Doesn't this cause a RuntimeError?

a = torch.zeros(10)
b = torch.ones(10)
c = a or b

RuntimeError: Boolean value of Tensor with more than one value is ambiguous

@horheynm horheynm changed the title fake quant fix [Forward Call] fake quant fix Aug 30, 2024
rahul-tuli
rahul-tuli previously approved these changes Aug 30, 2024
View reviewed changes
src/compressed_tensors/quantization/lifecycle/forward.py Outdated Show resolved Hide resolved
@kylesayrs kylesayrs self-requested a review September 3, 2024 14:41
@rahul-tuli rahul-tuli merged commit 4328b41 into main Sep 3, 2024
1 check passed
@rahul-tuli rahul-tuli deleted the fake_quant-fix branch September 3, 2024 19:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@rahul-tuli rahul-tuli rahul-tuli approved these changes

@kylesayrs kylesayrs kylesayrs approved these changes

@bfineran bfineran bfineran left review comments

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@horheynm @kylesayrs @bfineran @rahul-tuli