mirror of
https://github.com/osmarks/nanogpt-experiments.git
synced 2024-11-10 20:09:58 +00:00
clarify that these should be run on two separate machines
This commit is contained in:
parent
9352df23de
commit
684800dd87
2
train.py
2
train.py
@ -9,7 +9,9 @@ To run with DDP on 4 gpus on 1 node, example:
|
||||
$ torchrun --standalone --nproc_per_node=4 train.py
|
||||
|
||||
To run with DDP on 4 gpus across 2 nodes, example:
|
||||
- Run on the first (master) node:
|
||||
$ torchrun --nproc_per_node=8 --nnodes=2 --node_rank=0 --master_addr=123.456.123.456 --master_port=1234 train.py
|
||||
- Run on the worker node:
|
||||
$ torchrun --nproc_per_node=8 --nnodes=2 --node_rank=1 --master_addr=123.456.123.456 --master_port=1234 train.py
|
||||
"""
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user