update

app_regression.py

@@ -47,8 +47,8 @@ then give scores from 5 different dimensions:
 (5) factual consistency, the consistency of the video content with the common-sense and factual knowledge
 
 For each dimension, output a number from [1,2,3,4], 
-in which ‘1’ means ‘Bad’, ‘2’ means ‘Average’, ‘3’ means ‘Good’, 
-'4' means ‘Real’ or ‘Perfect’ (the video is like a real video)
+in which '1' means 'Bad', '2' means 'Average', '3' means 'Good', 
+'4' means 'Real' or 'Perfect' (the video is like a real video)
 Here is an output example:
 visual quality: 4
 temporal consistency: 4
@@ -57,7 +57,7 @@ text-to-video alignment: 1
 factual consistency: 2
 
 For this video, the text prompt is "{text_prompt}",
-all the frames of video are as follows:
+all the frames of video are as follows: 
 
 """