From 5d14b97aa6e66587fee8fdecb8c24b57187b64f6 Mon Sep 17 00:00:00 2001 From: William Falcon Date: Fri, 12 Jul 2019 12:57:54 -0400 Subject: [PATCH] testing file init --- pytorch_lightning/models/trainer.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pytorch_lightning/models/trainer.py b/pytorch_lightning/models/trainer.py index 83ea60da..f42357c2 100644 --- a/pytorch_lightning/models/trainer.py +++ b/pytorch_lightning/models/trainer.py @@ -352,7 +352,7 @@ class Trainer(TrainerIO): print('SLURM ROOT NODE: ', root_node) print('-'*100) - dist.init_process_group("nccl", init_method=f'env://{root_node}', rank=self.proc_rank, + dist.init_process_group("nccl", init_method=f'env://{root_node}:12007', rank=self.proc_rank, world_size=self.world_size) # self.__init_tcp_connection(ip_file_dir)