problem of "trace_command.sh"

[cocktailplusHCL]

Dear authors :
I want to inquire about an issue encountered while running "inference.py". The situation is as follows: I want to execute the simulation of cifar-10 on ResNet34. I have already saved the trained model using "train.py" and specified the model_path in the "inference.py". "layer_record_ResNet34" and "NetWork_ResNet31.csv" are prepared. When running the "inference.py", a large number of "Floating point exception" occur in the last line of the code"call (........./trace_command.sh)", causing it to terminate. I would like to ask the authors about the reasons for this situation and possible solutions.
Below is the execution interface.

[siatzjs]

I have the same problem