From 7a2fbacf722eec6ce04042a161b259f67af75379 Mon Sep 17 00:00:00 2001
From: eneren <engin.eren@cern.ch>
Date: Sun, 2 Oct 2022 14:47:36 +0000
Subject: [PATCH] checkpoint path error
---
pytorch_job_wganSingleGen_ncc.yaml | 4 ++--
wganSingleGen.py | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/pytorch_job_wganSingleGen_ncc.yaml b/pytorch_job_wganSingleGen_ncc.yaml
index e1c5db6..e0fe6f2 100644
--- a/pytorch_job_wganSingleGen_ncc.yaml
+++ b/pytorch_job_wganSingleGen_ncc.yaml
@@ -35,7 +35,7 @@ spec:
value: "1"
command: [sh, -c]
args:
- - python -u wganSingleGen.py --backend nccl --epochs 50 --exp wganSingleGenV1 --batch-size 64 --ncrit 4
+ - python -u wganSingleGen.py --backend nccl --epochs 50 --exp wganSingleGenV1 --lrGen_E_H 0.00001 --chpt --chpt_eph 50 --batch-size 64 --ncrit 4
resources:
limits:
nvidia.com/gpu: 1
@@ -70,7 +70,7 @@ spec:
value: "1"
command: [sh, -c]
args:
- - python -u wganSingleGen.py --backend nccl --epochs 50 --exp wganSingleGenV1 --batch-size 64 --ncrit 4
+ - python -u wganSingleGen.py --backend nccl --epochs 50 --exp wganSingleGenV1 --lrGen_E_H 0.00001 --chpt --chpt_eph 50 --batch-size 64 --ncrit 4
resources:
limits:
nvidia.com/gpu: 1
diff --git a/wganSingleGen.py b/wganSingleGen.py
index 96e11b2..4641cc6 100644
--- a/wganSingleGen.py
+++ b/wganSingleGen.py
@@ -292,7 +292,7 @@ def run(args):
# load from checkpoint if desired
if (args.chpt):
- critic_E_H_checkpoint = torch.load(args.chpt_base + args.exp + "_criticE_"+ str(args.chpt_eph) + ".pt")
+ critic_E_H_checkpoint = torch.load(args.chpt_base + args.exp + "_critic_E_H_"+ str(args.chpt_eph) + ".pt")
gen_E_H_checkpoint = torch.load(args.chpt_base + args.exp + "_generatorE_H_"+ str(args.chpt_eph) + ".pt")
Crit_E_H.load_state_dict(critic_E_H_checkpoint['model_state_dict'])
--
GitLab