update betas

mindee · Dec 20, 2024 · d83beca · d83beca
1 parent a034ed8
commit d83beca
Show file tree

Hide file tree

Showing 6 changed files with 6 additions and 6 deletions.
diff --git a/references/classification/train_pytorch_character.py b/references/classification/train_pytorch_character.py
@@ -283,7 +283,7 @@ def main(args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )

diff --git a/references/classification/train_pytorch_orientation.py b/references/classification/train_pytorch_orientation.py
@@ -289,7 +289,7 @@ def main(args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )

diff --git a/references/detection/train_pytorch.py b/references/detection/train_pytorch.py
@@ -335,7 +335,7 @@ def main(args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )

diff --git a/references/detection/train_pytorch_ddp.py b/references/detection/train_pytorch_ddp.py
@@ -351,7 +351,7 @@ def main(rank: int, world_size: int, args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )

diff --git a/references/recognition/train_pytorch.py b/references/recognition/train_pytorch.py
@@ -341,7 +341,7 @@ def main(args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )

diff --git a/references/recognition/train_pytorch_ddp.py b/references/recognition/train_pytorch_ddp.py
@@ -275,7 +275,7 @@ def main(rank: int, world_size: int, args):
         optimizer = torch.optim.Adam(
             [p for p in model.parameters() if p.requires_grad],
             args.lr,
-            betas=(0.95, 0.99),
+            betas=(0.95, 0.999),
             eps=1e-6,
             weight_decay=args.weight_decay,
         )