From 12f717ad4a7212bf703f764e57079812461a0047 Mon Sep 17 00:00:00 2001 From: William Falcon Date: Fri, 26 Jul 2019 18:52:02 -0400 Subject: [PATCH 1/3] added global rank var name --- pytorch_lightning/models/trainer.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/pytorch_lightning/models/trainer.py b/pytorch_lightning/models/trainer.py index bbfafd89..b2881db7 100644 --- a/pytorch_lightning/models/trainer.py +++ b/pytorch_lightning/models/trainer.py @@ -500,6 +500,9 @@ class Trainer(TrainerIO): self.proc_rank = self.node_rank * len(self.data_parallel_device_ids) + gpu_nb self.world_size = self.nb_gpu_nodes * len(self.data_parallel_device_ids) + # let the exp know the rank to avoid overwriting logs + self.experiment.rank = self.global_rank + # set up server using proc 0's ip address # try to init for 20 times at max in case ports are taken # where to store ip_table From d6bfb94215ac8a74f8c16a62a635250335719160 Mon Sep 17 00:00:00 2001 From: William Falcon Date: Fri, 26 Jul 2019 18:52:38 -0400 Subject: [PATCH 2/3] added global rank var name --- 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 b2881db7..44cafb0c 100644 --- a/pytorch_lightning/models/trainer.py +++ b/pytorch_lightning/models/trainer.py @@ -501,7 +501,7 @@ class Trainer(TrainerIO): self.world_size = self.nb_gpu_nodes * len(self.data_parallel_device_ids) # let the exp know the rank to avoid overwriting logs - self.experiment.rank = self.global_rank + self.experiment.rank = self.proc_rank # set up server using proc 0's ip address # try to init for 20 times at max in case ports are taken From df37c8418a1b895ed067a706c4afcf577d027721 Mon Sep 17 00:00:00 2001 From: William Falcon Date: Fri, 26 Jul 2019 18:57:18 -0400 Subject: [PATCH 3/3] updated test-tube dep number --- setup.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/setup.py b/setup.py index 5edadfa1..89ac6f29 100755 --- a/setup.py +++ b/setup.py @@ -19,7 +19,7 @@ setup( install_requires=[ "torch>=1.1.0", "tqdm", - "test-tube>=0.6.7.1", + "test-tube>=0.6.7.4", ], packages=find_packages(), long_description=open("README.md", encoding="utf-8").read(),