App.py modified to append "Please describe this image." text if image only is received as input

app.py

@@ -185,6 +185,13 @@ def gradio_get_answers_fn(image=None, audio=None, text=None):
     if text:
         text_tokens = tokenizer.encode(text)
 
+    if image:
+        if audio or text:
+            pass
+        else:
+            text = "Please describe this image."
+            text_tokens = tokenizer.encode(text)        
+
     if image or audio or text:
         input_embed = prepare_input_embed(image, audio_tokens, text_tokens)
         with torch.no_grad():