【寒武纪硬件产品型号】必填*：
mlu270

【使用操作系统】必填*：
ubuntu18.04

【使用驱动版本】必填*：
4.9.2

【出错信息】必填*：

cast INT64 data into INT32 data on cpu.

unequal new numel and old numel is not supported

as_strided Op cannot run on MLU device, start running on CPU!

empty_strided Op cannot run on MLU device, start running on CPU!

cast INT64 data into INT32 data on cpu

(toTraceableIValue at /pytorch/torch/csrc/jit/pybind_utils.h:286)

  File "/torch/venv3/pytorch/lib/python3.6/site-packages/torch/jit/__init__.py", line 858, in trace

    check_tolerance, _force_outplace, _module_class)

  File "/torch/venv3/pytorch/lib/python3.6/site-packages/torch/jit/__init__.py", line 997, in trace_module

    module._c._create_method_from_trace(method_name, func, example_inputs, var_lookup_fn, _force_outplace)

RuntimeError: Tracer cannot infer type of <NULL>

【当前已做了哪些信息确认】选填：
模型转换需要算子支持，

example_mlu = torch.randn(1,1024, 3,dtype=torch.float).to(ct.mlu_device())

    trace_model = torch.jit.trace(quantized_net_mlu, example_mlu,check_trace=False)

运行到这会报错

cast INT64 data into INT32 data on cpu.

unequal new numel and old numel is not supported

as_strided Op cannot run on MLU device, start running on CPU!

empty_strided Op cannot run on MLU device, start running on CPU!

cast INT64 data into INT32 data on cpu

【相关日志文档】选填

【出错代码链接】选填：
github的或gitee的代码的链接，