Update app.py

app.py

@@ -350,16 +350,14 @@ def get_single_prompt(user_input):
         f"User input: \"{user_input}\""
     )
     
-    response = openai.Completion.create(
-        engine="text-davinci-003",
-        prompt=prompt_instruction,
-        max_tokens=50,
-        n=1,
-        stop=None,
+    response = openai.ChatCompletion.create(
+        model="gpt-4o",  # adjust model name if needed
+        messages=[{"role": "user", "content": prompt_instruction}],
         temperature=0.3,
+        max_tokens=50,
     )
-    generated_prompt = response.choices[0].text.strip()
-    # Ensure no question marks exist in the prompt.
+    generated_prompt = response.choices[0].message.content.strip()
+    # Ensure no question marks remain.
     generated_prompt = generated_prompt.replace("?", "")
     return generated_prompt
 
@@ -379,12 +377,12 @@ def process_question_and_detect(user_input, image):
     1. Uses OpenAI to generate a single, concise prompt (without question marks) from the user's input.
     2. Feeds that prompt to the VisionAgent detection function.
     3. Overlays the detection bounding boxes on the image.
-    4. If the user's input is a counting query, it also returns the count of detected objects.
+    4. If the user's input implies a counting request, it also returns the count of detected objects.
     """
     if image is None:
         return None, "Please upload an image."
     
-    # Generate a concise prompt from the user's input.
+    # Generate the concise prompt from the user's input.
     generated_prompt = get_single_prompt(user_input)
     
     # Run object detection using the generated prompt.
@@ -393,7 +391,7 @@ def process_question_and_detect(user_input, image):
     # Overlay bounding boxes on the image.
     viz = T.overlay_bounding_boxes(image, dets)
     
-    # Check if the user's input implies a counting request.
+    # If the user's input implies a counting request, include the count.
     count_text = ""
     if is_count_query(user_input):
         count = len(dets)
@@ -402,7 +400,6 @@ def process_question_and_detect(user_input, image):
     output_text = f"Generated prompt: {generated_prompt}\n{count_text}"
     return viz, output_text
 
-# Build the Gradio interface.
 with gr.Blocks() as demo:
     gr.Markdown("# VisionAgent Object Detection and Counting App")
     gr.Markdown(
@@ -429,6 +426,7 @@ with gr.Blocks() as demo:
     
     submit_btn.click(fn=process_question_and_detect, inputs=[user_input, image_input], outputs=[output_image, output_text])
     
-    demo.launch()
+    demo.launch(share=True)
+