simple VQA

app.py

@@ -12,11 +12,26 @@ def vqa(image, text):
     idx = logits.argmax(-1).item()
     return f"{text}: {model.config.id2label[idx]}"
 
+with gr.Blocks() as demo:
+
+    txt = gr.Textbox(label="Insert a question..", lines=2)
+    txt_3 = gr.Textbox(value="", label="Your answer is here..")
+    btn = gr.Button(value="Submit")
+
+    dogs = os.path.join(os.path.dirname(__file__), "617.jpg")
+    image = gr.Image(type="pil", value=dogs)
+    btn.click(vqa, inputs=[image, txt], outputs=[txt_3])
+
+    btn = gr.Button(value="Submit")
+
+if __name__ == "__main__":
+    demo.launch()
+
+
 
-dogs = os.path.join(os.path.dirname(__file__), "617.jpg")
 text = "What are the dogs riding?"
 
-demo = gr.Interface(vqa, gr.Image(type="pil", value=dogs), text, "text")
+demo = gr.Interface(vqa, , text, "text")
 
 if __name__ == "__main__":
     demo.launch()