mirror of
https://github.com/wassname/pytorch-lightning.git
synced 2026-09-11 12:31:23 +08:00
fixed nccl init
This commit is contained in:
@@ -379,6 +379,7 @@ class Trainer(TrainerIO):
|
||||
:param tries:
|
||||
:return:
|
||||
"""
|
||||
sleep(self.proc_rank*2)
|
||||
root_node = os.environ['SLURM_NODELIST'].split(' ')[0]
|
||||
os.environ['MASTER_ADDR'] = root_node
|
||||
os.environ['MASTER_PORT'] = f'{port}'
|
||||
|
||||
Reference in New Issue
Block a user