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

Chunk gpu #11

Merged
merged 4 commits into from
Oct 16, 2023
Merged

Chunk gpu #11

merged 4 commits into from
Oct 16, 2023

Conversation

aRI0U
Copy link
Collaborator

@aRI0U aRI0U commented Oct 16, 2023

A num_chunks option has been added for enabling splitting batch of CQT frames into chunks that will be then processed sequentially. This enable to limit memory usage hence making it possible to deal with longer sequences.

The default (num_chunks=1) exactly corresponds to the previous behaviour.

As a reference, this splitting technique now enables to process one hour of audio in 20s with a single GTX 1080.

@aRI0U aRI0U linked an issue Oct 16, 2023 that may be closed by this pull request
@aRI0U aRI0U merged commit 9cc0ab9 into master Oct 16, 2023
@aRI0U aRI0U deleted the chunk_gpu branch October 16, 2023 14:53
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

Successfully merging this pull request may close these issues.

split inputs into chunks to avoid CUDA OOM
1 participant