Skip to content

Conversation

@Looong01
Copy link
Owner

Summary

  • embed Python model converter inside C++ and generate ONNX via runtime virtualenv
  • invoke embedded converter from MIGraphX backend before parsing ONNX
  • adjust build scripts to include new converter source and drop Python linkage
  • initialize converter's network layers from model config and install required Python packages
  • remove external config JSON dependency by inlining minimal model config values in converter

Testing

  • cmake ../cpp -DUSE_BACKEND=MGX (fails: Could not find a package configuration file provided by "migraphx" with any of the following names: migraphxConfig.cmake, migraphx-config.cmake)

https://chatgpt.com/codex/tasks/task_e_689c294cafac8327b49f2c439e3dc7a0

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

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants