mxnet Segmentation fault: 11 libmxnet.so(+0x40c6b50)

Segmentation fault: 11

Stack trace:
  [bt] (0) /data/user1/pkgs/conda/envs/drc/lib/python3.7/site-packages/mxnet/libmxnet.so(+0x40c6b50) [0x7f2dfe4f8b50]
  [bt] (1) /lib/x86_64-linux-gnu/libc.so.6(+0x354b0) [0x7f2edc06b4b0]
  [bt] (2) /lib/x86_64-linux-gnu/libc.so.6(strlen+0x26) [0x7f2edc0c1746]
  [bt] (3) /lib/x86_64-linux-gnu/libc.so.6(+0x4c980) [0x7f2edc082980]
  [bt] (4) /lib/x86_64-linux-gnu/libc.so.6(_IO_vfprintf+0x336) [0x7f2edc0834a6]
  [bt] (5) /lib/x86_64-linux-gnu/libc.so.6(vsnprintf+0x79) [0x7f2edc0aca49]
  [bt] (6) /data/user1/pkgs/conda/envs/drc/lib/python3.7/site-packages/torch/lib/libtorch_python.so(+0x542fb4) [0x7f2ed7101fb4]
  [bt] (7) /data/user1/pkgs/conda/envs/drc/lib/python3.7/site-packages/torch/lib/libtorch_python.so(torch::TypeError::TypeError(char const*, ...)+0xb6) [0x7f2ed71022d6]
  [bt] (8) /data/user1/pkgs/conda/envs/drc/lib/python3.7/site-packages/torch/lib/libtorch_python.so(+0x83853e) [0x7f2ed73f753e]

代码其实不调用mxnet进行处理（用的是pytorch），但是因为历史遗留import了它。
在注释掉以下两个分别位于不同位置的内容之后，这个报错消失：

import mxnet as mx

label1 = torch.LongTensor(label1)