m

app.py

@@ -98,6 +98,7 @@ def main(image, text):
     vqa_answer = vqa(image, cleaned_sentence)
     llm_answer = llm(cleaned_sentence, vqa_answer)
     final_answer, _ = google_translate(llm_answer, dest=dest_lang)
+    print("Final answer: ", final_answer)
     return vqa_answer, final_answer
     
    
@@ -106,7 +107,9 @@ question = gr.Textbox(label="Question")
 answer = gr.Textbox(label="Predicted answer")
 examples = [
     ["monkeys.jpg", "What are the monkeys doing in French?"],
-    ["apple.jpg", "Qu'est-ce que c'est dans ma main en anglais?"]
+    ["apple.jpg", "Qu'est-ce que c'est dans ma main en anglais?"],
+    ["monkeys.jpg", "In Chinese, tell me how many monkeys are here?"],
+    ["apple.jpg", "What color is this in Uyghur?"]
     ]
 
 title = "Cross-lingual VQA"