[Apache TVM Discuss] [Questions] [BYOC] How backwards compatible does the TensorRT partition_for_tensorrt function need to be?

Good to know, will do, thanks. --- [Visit Topic](https://discuss.tvm.apache.org/t/byoc-how-backwards-compatible-does-the-tensorrt-partition-for-tensorrt-function-need-to-be/12957/6) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails

[Apache TVM Discuss] [Questions] [BYOC] How backwards compatible does the TensorRT partition_for_tensorrt function need to be?

Thanks Cody. It looks like the API will be: ``` trt_target = tvm.target.Target("tensorrt -use_fp16=True -implicit_batch_mode=False") mod = partition_for_tensorrt(mod, params=params, target=trt_target) exe = vm.compile(mod, target=["cuda", trt_target], params=params) ``` (and similarly for

[Apache TVM Discuss] [Questions] [BYOC] How backwards compatible does the TensorRT partition_for_tensorrt function need to be?

Hi all, I'm looking to do to the TensorRT BYOC integration what I just did for CUTLASS, namely make sure all compilation configuration is captured within a "tensorrt" Target instance rather than the current combination of PassContext and environment variables. This helps Collage, both because

[Apache TVM Discuss] [Questions] How to add my custom Relay node to pattern matcher? [EthosU example]

cc @mikepapadim, who's reworking FuseOps to be more Composite+patterns like. --- [Visit Topic](https://discuss.tvm.apache.org/t/how-to-add-my-custom-relay-node-to-pattern-matcher-ethosu-example/11498/5) to respond. You are receiving this because you enabled mailing list mode. To unsubscr