Skip to content

Open-Sora: Democratizing Efficient Video Production for All

Python

add filter by flow score function

Shen-Chenhuif3c417a61e4b
diff --git a/tools/datasets/csvutil.py b/tools/datasets/csvutil.py
index 8ad7294..64fc33c 100644
--- a/tools/datasets/csvutil.py
+++ b/tools/datasets/csvutil.py
@@ -305,6 +305,7 @@ def parse_args():
# aesthetic filtering
parser.add_argument("--aesmin", type=float, default=None)
parser.add_argument("--matchmin", type=float, default=None)
+ parser.add_argument("--flowmin", type=float, default=None)
return parser.parse_args()
@@ -352,6 +353,8 @@ def get_output_path(args, input_name):
# clip score filtering
if args.matchmin is not None:
name += f"_matchmin{args.matchmin}"
+ if args.flowmin is not None:
+ name += f"_flowmin{args.flowmin}"
# sort
if args.sort_descending is not None:
assert args.sort_ascending is None
@@ -476,6 +479,9 @@ def main(args):
if args.matchmin is not None:
assert "match" in data.columns
data = data[data["match"] >= args.matchmin]
+ if args.flowmin is not None:
+ assert "flow" in data.columns
+ data = data[data["flow"] >= args.flowmin]
print(f"Filtered number of samples: {len(data)}.")
# sort