Fix minor problems when aggragating the results (#106)

desktop_env/evaluators/metrics/vlc.py

@@ -186,10 +186,10 @@ def compare_videos(video_path1, video_path2, max_frames_to_check=100, threshold=
             mismatch_count += 1
             # If there's a significant difference, the frames are not the same
             if mismatch_count > threshold:
-                return False
+                return 0.
 
     # If we reach here, the content appears to be the same
-    return True
+    return 1.
 
 
 def are_audio_files_similar(mp3_file_path, mp4_file_path):
@@ -203,7 +203,7 @@ def are_audio_files_similar(mp3_file_path, mp4_file_path):
                         mp4_audio_path], check=True)
     except subprocess.CalledProcessError as e:
         print(f"An error occurred during audio extraction from MP4: {e}")
-        return False
+        return 0.
 
     # Extract audio fingerprint from the extracted audio
     mp4_fingerprint, mp4_duration = acoustid.fingerprint_file(mp4_audio_path)
@@ -213,9 +213,9 @@ def are_audio_files_similar(mp3_file_path, mp4_file_path):
 
     # Compare fingerprints (rudimentary comparison)
     if mp3_duration >= mp4_duration and mp3_fingerprint == mp4_fingerprint:
-        return True
+        return 1.
 
-    return False
+    return 0.
 
 
 def check_qt_bgcone(actual_config_path, rule):