use TVM backend error

[earlyEpiphyte]

I want to use TVM backend to compare runtime of TASO`s like above code.

def evaluate_runtime(onnx_model):
    mod, params = relay.frontend.from_onnx(onnx_model, shape_dict)

    lib = relay.build_module.build(mod, target=target, params=params)
    dev = tvm.device(str(target), 0)
    module = graph_executor.GraphModule(lib["default"](dev))
    module.set_input(input_name, input_data)
    #module.run()
    use_tvm_time = (
        np.array(timeit.Timer(lambda: module.run()).repeat(repeat=timing_repeat, number=timing_number))
        * 1000
        / timing_number
    )
    use_tvm_time = {
        "mean": np.mean(use_tvm_time),
        "median": np.median(use_tvm_time),
        "std": np.std(use_tvm_time),
    }
    print(use_tvm_time)

Take resnet50 as an example. I get the original resnet onnx model and optimized model by TASO.
This function works fine when i put the original model in, but get error to optimized model.
Error message:

The Relay type checker is unable to show the following types match:
  Tensor[(64, 160, 3, 3), float32]
  Tensor[(64, 64, 3, 3), float32]
In particular:
  dimension 1 conflicts: 160 does not match 64.
The Relay type checker is unable to show the following types match.
In particular `Tensor[(64, 64, 3, 3), float32]` does not match `Tensor[(64, 160, 3, 3), float32]`

I am confused and don`t konw how to do.

[earlyEpiphyte]

@jiazhihao Can you help me?