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

Multi-gpu context is lost before training #4

Open
eric-mc2 opened this issue Jan 13, 2022 · 0 comments
Open

Multi-gpu context is lost before training #4

eric-mc2 opened this issue Jan 13, 2022 · 0 comments

Comments

@eric-mc2
Copy link
Contributor

When the CTC Pipeline fit() methods are called, they compile the model, overwriting the model variable, and clearing the distributed data parallel context. Thus, the model only runs on a single GPU even if multiple GPUs are visible to tensorflow.
eric-mc2 added a commit to eric-mc2/DeepAsr that referenced this issue Jan 13, 2022
@eric-mc2
fix issue scionoftech#4
881fe28
eric-mc2 added a commit to eric-mc2/DeepAsr that referenced this issue Jan 13, 2022
@eric-mc2
fix issue scionoftech#4
bcdf65b 
forgot to update calling function
scionoftech added a commit that referenced this issue Jan 13, 2022
@scionoftech
Merge pull request #5 from eric-mc2/multi-gpu
d7b5ad4 
fix issue #4
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
@eric-mc2