fix video bug

app.py

@@ -64,7 +64,7 @@ def video_text_zeroshot(video, text_list):
     labels = [label.strip(" ") for label in text_list.strip(" ").split("|")]
     inputs = {
         ModalityType.TEXT: data.load_and_transform_text(labels, device),
-        ModalityType.VIDEO: data.load_and_transform_video_data(video_paths, device),
+        ModalityType.VISION: data.load_and_transform_video_data(video_paths, device),
     }
 
     with torch.no_grad():
@@ -72,7 +72,7 @@ def video_text_zeroshot(video, text_list):
 
     scores = (
         torch.softmax(
-            embeddings[ModalityType.AUDIO] @ embeddings[ModalityType.TEXT].T, dim=-1
+            embeddings[ModalityType.VISION] @ embeddings[ModalityType.TEXT].T, dim=-1
         )
         .squeeze(0)
         .tolist()