Replies: 1 comment 9 replies
-
With To answer you question correctly it would help you share the command line / Python API call. |
Beta Was this translation helpful? Give feedback.
9 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi all,
What are the differences between submitting a job with "horovodrun" command vs. with regular Python command?
When I tried to run a simple PyTorch multi-node job (2 nodes, with 2 GPUs on each) through Slurm (with a single process per GPU) using only Python it surprisingly worked and recognized 4 GPUs, so why should I use horovodrun anyway?
Also, how did it know to expect 2 nodes and 2 GPUs on each without me telling him? Is it thanks to Slurm?
Thanks,
Assaf.
Beta Was this translation helpful? Give feedback.
All reactions