diff --git a/pytorch_lightning/trainer/trainer.py b/pytorch_lightning/trainer/trainer.py index 25ecd543..eefbfe1a 100644 --- a/pytorch_lightning/trainer/trainer.py +++ b/pytorch_lightning/trainer/trainer.py @@ -842,7 +842,10 @@ class Trainer( # route to appropriate start method # when using multi-node or DDP within a node start each module in a separate process if self.use_ddp2: - task = int(os.environ['SLURM_LOCALID']) + if self.is_slurm_managing_tasks: + task = int(os.environ['SLURM_LOCALID']) + elif 'WORLD_SIZE' in os.environ and 'GROUP_RANK' in os.environ: + task = int(os.environ['LOCAL_RANK']) self.ddp_train(task, model) elif self.use_ddp: if self.is_slurm_managing_tasks: