This repository has been archived by the owner on Dec 16, 2022. It is now read-only.
Biggest batch first during training as a sanity check #5615
Unanswered
vikigenius
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
In older versions of allennlp. I distinctly remember an option "biggest_batch_first" that allowed a kind of sanity check where the biggest batch would be put first so that OOM errors are detected early.
Right now I could not find that anymore. Is there a way to replicate that behavior easily now? Can we bring back that feature? I found it very useful.
Beta Was this translation helpful? Give feedback.
All reactions