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

Add benchmark tool #84

Closed
ggerganov opened this issue Oct 24, 2022 · 1 comment · Fixed by #87
Closed

Add benchmark tool #84

ggerganov opened this issue Oct 24, 2022 · 1 comment · Fixed by #87
Labels
enhancement New feature or request

Comments

@ggerganov
Copy link
Owner

We need a more deterministic way to benchmark the performance, so I'm thinking about adding a tool that runs just the Encoder part of the transformer on random noise input. This has the advantage that one does not require any audio data to run and the number of operations is fixed (as opposed to when running the Decoder).
@ggerganov ggerganov added the enhancement New feature or request label Oct 24, 2022
@ArtyomZemlyak
Copy link

Very good idea for testing performance directly of speed up only computation part.

But in some cases steel need check overal model performance - if quality of model somehow dropped, then its can run slowly (bc decoder trying get good answer). But it depends on decoder type, parameters of decoder...

Therefore good idea!

@ggerganov ggerganov mentioned this issue Oct 25, 2022
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Refactoring and adding bench tool ggerganov/whisper.cpp
2 participants
@ggerganov @ArtyomZemlyak