Update app.py

app.py

@@ -12,11 +12,8 @@ processor = BlipProcessor.from_pretrained("Salesforce/blip-vqa-capfilt-large")
 model_vqa = BlipForQuestionAnswering.from_pretrained("Salesforce/blip-vqa-capfilt-large").to(device)
 def inference_chat(input_image,input_text):
     inputs = processor(images=input_image, text=input_text,return_tensors="pt")
-
-
     inputs["max_length"] = 20
     inputs["num_beams"] = 5
-
     out = model_vqa.generate(**inputs)
     return processor.batch_decode(out, skip_special_tokens=True)[0]
     
@@ -39,8 +36,8 @@ with gr.Blocks(
             with gr.Row():
                 
                 with gr.Column(scale=1):
-                    caption_output = gr.Textbox(lines=0, label="")
-                    chat_input = gr.Textbox(lines=1, label="VQA Input")
+                    caption_output = gr.Textbox(lines=0, label="VQA Output(模型答案输出)")
+                    chat_input = gr.Textbox(lines=1, label="VQA Input(问题输入)")
                     chat_input.submit(
                         inference_chat,
                         [
@@ -70,7 +67,6 @@ with gr.Blocks(
                             ],
                             [caption_output],
                         )
-                    caption_output = gr.Textbox(lines=1, label="VQA Output")
             
     
             image_input.change(