Lower default batch size for VideoGPT training args

wilson1yan · Jun 10, 2021 · 76a7c24 · 76a7c24
1 parent ca8891c
commit 76a7c24
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/scripts/train_videogpt.py b/scripts/train_videogpt.py
@@ -13,7 +13,7 @@ def main():
     parser.add_argument('--data_path', type=str, required=True)
     parser.add_argument('--resolution', type=int, default=128)
     parser.add_argument('--sequence_length', type=int, default=16)
-    parser.add_argument('--batch_size', type=int, default=32)
+    parser.add_argument('--batch_size', type=int, default=8)
     parser.add_argument('--num_workers', type=int, default=8)
     args = parser.parse_args()