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

k must have shape (batch_size, seqlen_k, num_heads_k, head_size_og) #5

Open
rainsoulsrx opened this issue May 21, 2024 · 3 comments
Open

Comments

@rainsoulsrx
Copy link

When I set --infer_sr_scale 1, there is an error, but when set infer_sr_scale=2, run correct, why?

@rainsoulsrx
Copy link
Author

and in the code, collect_attention is set False, the following set is not taking effect."You can change the "inference_type"(line 27 in generate_sr_big_cli.sh) to "ar"(parallel size=1), "ar2"(parallel size = block_batch(line 28)) or "full"(generate the entire image in one forward)."

@yzy-thu
Copy link
Contributor

yzy-thu commented May 22, 2024

collect_attention is a debug parameter, irrelevant to inference. Why you need infer_sr_scale==1? I haven't tested this.

@Dorniwang
Copy link

yeah, it doesn't work, when i specify the inference_type=ar2 manually, it will give this error

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

3 participants