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.

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

Will there be precision errors after switching to qnn after Aimet pipline? #3268

Open
shifeiwen opened this issue Aug 26, 2024 · 1 comment

Comments

@shifeiwen
Copy link

Hi ,aimet team
I want to know if when I use aimet to generate the encoding file and then use the --quantization_overrides flag when converting to qnn, will the generated NPU backend model lose accuracy again? For example, aimet loss = 1.1 qnn htp backend may be 1.3~1.4
The reason for this is that I was using qc's genAI notebook to call aimet quantization model, and I found that the loss of the actual generated model was higher than that in aimet.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants
@shifeiwen and others