Replies: 1 comment
-
Pinging @sharadmv who may know more: I believe that callbacks are non-blocking on TPU, but ultimately the details will be up to the XLA scheduler. |
Beta Was this translation helpful? Give feedback.
0 replies
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.
Uh oh!
There was an error while loading. Please reload this page.
-
We want to have a callback inside jitted code to run in parallel with other TPU computations. To illustrate, take the following as an example. We hope the first and second line could run in parallel, one on CPU and one TPU. Then, run the third line when both are ready. Would this be possible with a single device setup?
Beta Was this translation helpful? Give feedback.
All reactions