wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-12 12:40:20 +08:00
Code Issues Packages Projects Releases Wiki Activity

added fallback local init

Browse Source
This commit is contained in:
William Falcon
2019-07-13 22:16:10 -04:00
parent 6d55adb0d8
commit eb05fa316f
1 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pytorch_lightning/models/trainer.py
+6 -2
View File
@@ -382,10 +382,14 @@ class Trainer(TrainerIO):
port = 12910 port = 12910
os.environ['MASTER_PORT'] = f'{port}' os.environ['MASTER_PORT'] = f'{port}'
sleep(self.proc_rank*0.5) try:
root_node = os.environ['SLURM_NODELIST'].split(' ')[0]
except Exception as e:
root_node = '127.0.0.2'
root_node = os.environ['SLURM_NODELIST'].split(' ')[0]
os.environ['MASTER_ADDR'] = root_node os.environ['MASTER_ADDR'] = root_node
sleep(self.proc_rank*0.5)
dist.init_process_group("nccl", rank=self.proc_rank, world_size=self.world_size) dist.init_process_group("nccl", rank=self.proc_rank, world_size=self.world_size)
def __run_pretrain_routine(self, model): def __run_pretrain_routine(self, model):