Now，I woulde like to use te.extern to add an extern op level library ，like
cuDNN. For complex ops, like conv2d or dense , we can add new dispatch rules
in op strategy,like @softmax_strategy.register(["cuda", "gpu"]), and then
redefine the compute with te.extern. But for some injective or broadc
So I want to konw what's the benefit about BYOC. I list some advantanges:
1: if the new backend doesn't support all ops, we can back to cpu
2: use rely to support different frameworks
3: some graph level optimizes(can we use all graph level optimizes?)
But tvm maybe too heavy for newbackend.
I have a question why you choose BYOC. We can't use AutoTVM in BYOC and I think
AutoTVM is the very import feature of TVM.
---
[Visit
Topic](https://discuss.tvm.ai/t/rfc-ethosn-arm-ethos-n-integration/6680/13) to
respond.
You are receiving this because you enabled mailing list mode.
To
This is good, If we can hold it in Xi'an Shaanxi, that would be great.
---
[Visit
Topic](http://tracking.discuss.tvm.ai/tracking/click?d=9jKHkhwD63XhqM8qY5JBgCHNG8FOJbnhOitm1UK7Rehy-8IJ1dUnqcWTbPvjQ7zrYQrZQM0NDO-4byxMBctil8cWIgwR_vcai-NmYJN4Moh8RmVRaETi56FXMtoNIiNygy8txMZ4DxiaRh8blTPcIFc1)
