the data that to the distributed training should be load all in every node or apart? or just load the main node? #1231
-
as the title |
Beta Was this translation helpful? Give feedback.
Answered by
bghira
Dec 19, 2024
Replies: 1 comment 2 replies
-
they need shared storage. |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
please follow "DISTRIBUTED.md" document.