SavedModel CPU performance much worse than HF Transformers pytorch version #839
Unanswered
marton-avrios
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
Uh oh!
There was an error while loading. Please reload this page.
-
Even after exporting T5 into saved model format, raw pytorch inference speed on CPU is only a fraction of the saved model's (after I import checkpoint weights there). Any steps I might have missed to make the mesh tensorflow model faster?
Beta Was this translation helpful? Give feedback.
All reactions