根据00_Yolov3_example，将yolov3网络换成其他网络，可以成功量化，但运行到融合模式时

model = torch.jit.trace(model, trace_input.to(ct.mlu_device()), check_trace = False)

，报错如下：

[WARNING][/workspace/volume/private/sdk/cambricon_pytorch/pytorch/src/catch/torch_mlu/csrc/aten/core/tensor_impl.cpp][line:779][raw_mutable_data][thread:140348009772800][process:56894]: 

cast INT64 data into INT32 data on cpu.

*** Error in `/workspace/volume/private/sdk/venv/pytorch/bin/python': malloc(): memory corruption (fast): 0x000000001b900d90 ***