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

wieght no found when mkl gemm #313

Open
WuShengHan opened this issue Dec 31, 2024 · 1 comment
Open

wieght no found when mkl gemm #313

WuShengHan opened this issue Dec 31, 2024 · 1 comment

Comments

@WuShengHan
Copy link

Hi,

I want to speed up cpu-only execution.

And I see ctranslate2 by os.environ["CT2_USE_EXPERIMENTAL_PACKED_GEMM"] = "1"

So I install Intel® oneAPI Math Kernel Library: Offline 2025.0.1

Original whisper-live is ok for me, but when i turn on
os.environ["CT2_USE_EXPERIMENTAL_PACKED_GEMM"] = "1"
I get message such as
ERROR:root:Error during new connection initialization: variable encoder/conv1/weight not found

How to solve? Thank you...
@makaveli10
Copy link
Collaborator

@WuShengHan I'd recomment first using your custom model in faster-whisper. If it already works there please check the version of faster-whisper with which custom model works. If it doesnt work there please raise an issue in faster-whisper.

We use faster-whisper==1.1.0 in our latest release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@makaveli10 @WuShengHan