Skip to content

If you REALLY want to run this code #95

Description

@amoazeni75

Hello all,

I want to recount my frustrating experience with this source code. After dedicating almost four days to struggling with the complexities of building and running this code across four different machines, I'm compelled to share the necessary configurations and the myriad of errors you'll likely encounter during the build process. I intend to spare you unnecessary frustration and compensate for the potential lack of helpful responses in the repository's issue section.

Okay, let's start with the Anaconda Environment:

You must use GGC V9 and CUDA-ToolKit 11.3

  • Use "install_pytorch_source.sh"

  • For this issue: static const unsigned kSigStackSize = std::max(16384L, SIGSTKSZ);
    See here: exception_handler.cc  #92

  • For this issue: GLIBCXX_3.4.30 not found
    See here: `GLIBCXX_3.4.30' not found #93

  • For this issue: /saiga/src/saiga/core/image/templatedImage.h:81 Function: bool Saiga::TemplatedImage<T>::load(const string&) [with T = Eigen::Matrix<unsigned char, 3, 1, 0>; std::string = std::__cxx11::basic_string<char>]
    See here: Loading images failed #94

  • Convert all .png files to .jpg before converting colmap to ADOP.

  • You need to change the colmap to adop c++ file to replace .png file extension with .jpg

  • Use: python3 -m pip install setuptools==59.5.0

Good luck with this source code :)

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions