Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Move logic from TorchX CLI -> API, so MVAI can call it #955

Merged
merged 1 commit into from
Sep 11, 2024
Merged

Move logic from TorchX CLI -> API, so MVAI can call it #955

merged 1 commit into from
Sep 11, 2024

Commits on Sep 11, 2024

  1. Move logic from TorchX CLI -> API, so MVAI can call it (pytorch#955) 

    Summary:
Pull Request resolved: pytorch#955

MVAI's "light" is synchronous - you can immediately see the logs for jobs you start. Only "fire" is asynchronous.

TorchX's API, since it's generic, *always* creates jobs that are asynchronous. Therefore, there isn't a built-in interface for "tailing" the stderr of every started process - just for tailing individual replicas of a given role.

The TorchX CLI's `torchx run` command **has** implemented this, but its implementation is coupled with the CLI implementations of `torchx run` and `torchx log`.

This diff extracts the useful logic into a helper function of the TorchX API

Reviewed By: andywag

Differential Revision: D62463211
    @facebook-github-bot
    Julie Ganeshan authored and facebook-github-bot committed Sep 11, 2024
    Configuration menu
    Copy the full SHA
    0bfee30 View commit details
    Browse the repository at this point in the history