@areusch for step 2, if you also wanted to use TVM for the accelerator codegen
as well, would VTA provide a guide there? And would you still use BYOC there
(just calling out to TVM in BYOC for compilation as well)?
---
[Visit
Topic](https://discuss.tvm.apache.org/t/tvm-and-custom-accelera
@asdlalala make sure and check out the [new GPU auto-scheduler tutorial
PR](https://github.com/apache/incubator-tvm/pull/6512) that @merrymercy just
put up. It still doesn't support full models yet to my knowledge (full
integration with Relay is coming next) but this should enable you to optim
