How do I get data/files out from a ray job?

Finally got my AWS cluster running, able to submit jobs and see the logs, even can create custom environments and start jobs with --working-dir and upload files.

But how does one get files back from a ray job sandbox? “ray rsync” seems to be more oriented to absolute filenames. It would of course be best if a job read and wrote from eg s3 storage or some other artifact store, but is there a way to do it without? (I’m trying to interface with code that assumes writing to the local filesystem).

1 Like