install the huggingface spaces

app.py

@@ -56,15 +56,6 @@ def image_parser(args):
     return out
 
 
-
-def load_images(image_files):
-    out = []
-    for image_file in image_files:
-        image = load_image(image_file)
-        out.append(image)
-    return out
-
-
 def generate_video_caption(video_path):
     video_id = video_path.split('/')[-1].strip().split('.')[0]
 
@@ -95,7 +86,7 @@ iface = gr.Interface(
     inputs=gr.File(file_types=["video"]),
     outputs="text",
     title="MAMA Video-Text Generation Pipeline",
-    description="Upload a video and get the transcribed text"
+    description="Upload a video and get the description. Due to high financial cost, we can only use TinyLLaVA on CPUs. Please only try videos which are less than 1MB. Thank you so much and Welcome to MAMA!"
 )
 
 iface.launch(share=True)