qelk123 commented on issue #15101:
URL: https://github.com/apache/tvm/issues/15101#issuecomment-1727065572
Thank you for your contirbution! I am wondering are there any e2e test case
I could follow? It seems I can't specify more than one target in the
`relax.build` API. How should I compile when nodes in my graph are
heterogeneous to each other? (CPU node and CUDA node for example)
In `relay.vm.compile` I could offer a target dict like: `target={"cpu":
tvm.target.Target("llvm"), "cuda": tvm.target.Target("cuda")}` to support
multi-device compilation. What should I do when I using relax vm?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
