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.

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

there are many ... in your script ,please provide complete instruction #108

Open
ouyang11111 opened this issue Dec 13, 2024 · 2 comments
Open

Comments

@ouyang11111
Copy link

torchrun --nproc_per_node=8 --nnodes=... --node_rank=... --master_addr=... --master_port=... train.py
--depth=16 --bs=768 --ep=200 --fp16=1 --alng=1e-3 --wpe=0.1 fail to run it please provide complete instruction

@iFighting
Copy link
Contributor

@ouyang11111
Are you familiar with pytorch torchrun scripts?
you should fill in your training setting(how many nodes, the rank...) in "--nnodes=... --node_rank=... --master_addr=... --master_port=..."
--nodes means how many nodes you use
--node_rank means the node rank

@ouyang11111
Copy link
Author

@ouyang11111 Are you familiar with pytorch torchrun scripts? you should fill in your training setting(how many nodes, the rank...) in "--nnodes=... --node_rank=... --master_addr=... --master_port=..." --nodes means how many nodes you use --node_rank means the node rank

How do I set the "master_addr" and "master_port" for only one server with 4 3090 cards?
when I run:torchrun --nproc_per_node=4 --nnodes=1 --node_rank=3 --master_addr=192.168.1.1 --master_port=12345 train.py --depth=16 --bs=768 --ep=200 --fp16=1 --alng=1e-3 --wpe=0.1
and i stuck for a long time .....

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants