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

Ego4D subset #8

Open
G-JWLee opened this issue Dec 18, 2022 · 0 comments
Open

Ego4D subset #8

G-JWLee opened this issue Dec 18, 2022 · 0 comments

Comments

@G-JWLee
Copy link

G-JWLee commented Dec 18, 2022

Hello, I read the paper, and it says that you used FHO subset of Ego4D dataset.
However, it seems that there are only about 1700 video clips in FHO subset, and your training batch size is 512.
Plus, it seems that there are some video clips that do not contain audio modality.
Hence, the code repeatedly search for videoes that contain audio and gain single sample.
Therefore, it takes so long time to consist single batch.
Is there something wrong in my understanding? or you just rely on high num_workers?
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
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@G-JWLee