Open BiophiliaSWDA opened 10 months ago
For pytorch, there is a tool https://github.com/pytorch/pytorch/tree/main/tools/code_coverage For tensorflow, just try bazel https://github.com/tensorflow/tensorflow/issues/51091
Hi, I have encountered a similar issue. How obtain the line coverage of the C/C++ code called by TensorFlow when running a Python test script? It seems that there is no mature solution in the link provided.
I have tried using Bazel to build the Python project, but there are many issues with importing TensorFlow. Do I need to compile TensorFlow locally using Bazel, or can it be imported as a third-party library?
I'm looking forward to your reply, any hint would be greatly helpful to me, thank you!! :-*) @Anjiang-Wei @BiophiliaSWDA
I remember there was a version compatibility issue with Bazel and GCOV. I forgot the details. @YangChenyuan may tell you more.
Hi, I am wondering how you relate GCOV to TensorFlow or Pytorch, because I am also currently experimenting with code coverage for deep learning networks, but I can't get operator coverage (or line coverage) for a TensorFlow network model after calling it from a C++ program.
I did this by calling the encapsulated network model from a C++ program and then using GCOV on the C++ program, but that didn't associate it with the source code.