Accomplish the exp scripts v1; Add video recording and trajectory recording of desktop agent; Fix minor bugs

desktop_env/evaluators/metrics/vlc.py

@@ -86,6 +86,9 @@ def compare_images(image1_path, image2_path):
     # score = compare_images('path_to_image1', 'path_to_image2')
     # print("Similarity score:", score)
 
+    if not image1_path or not image2_path:
+        return 0
+
     # Open the images and convert to grayscale
     image1 = Image.open(image1_path).convert('L')
     image2 = Image.open(image2_path).convert('L')
@@ -119,6 +122,9 @@ def compare_audios(audio_path_1, audio_path_2, max_distance=1000):
     # print(f'Similarity Score: {similarity}')
 
     # Convert to common format if necessary and load audio
+    if not audio_path_1 or not audio_path_2:
+        return 0
+
     y1, sr1 = librosa.load(audio_path_1)
     y2, sr2 = librosa.load(audio_path_2)