Expected Result:
A executable version of my compiled network, that runs and outputs a correct inference result.
Actual Result:
A binary, that crashes with a memory violation if I try to run inferences.
Steps to reproduce:
take simple network (ONNX, correct version for ONNC);
compile it using docker and master branch of C backend;
feed it with input data and weights;
it fails with a segmentation fault;
Environment:
Xubuntu 20.04 LTS;
docker;
x86

• ONNC version: latest commit on master (1.3)

• OS (e.g. from python -c "import platform as p; print(p.platform());print(p.processor())"):
  Linux-5.4.0-42-generic-x86_64-with-Ubuntu-20.04-focal
  x86_64

• Others:
  I suspect it is a problem with the file loading, but I am not sure