Get operator fusion result

Questions
#1

Example code url: https://tvm.apache.org/docs/how_to/compile_models/from_onnx.html

And code here:

with tvm.transform.PassContext(opt_level=1):
    executor = relay.build_module.create_executor(
        "graph", mod, tvm.cpu(0), target, params
    ).evaluate()

Can i get the optimized result(best in onnx model) after operaor fusion? By the way, run error when i replace tvm.cpu(0) with gpu(0), error stack message was

assert device.device_type == raw_targets[0].get_target_device_type()

I really appreciate it if somebody can solve my confusion.