Update app.py

app.py

@@ -39,9 +39,6 @@ tokenizer, model = None, None
 # Initialize the data list
 data = []
 
-# Initialize interaction count
-interaction_count = 0
-
 # Load the model and tokenizer once at the beginning
 def load_model(model_name):
     global tokenizer, model, selected_model
@@ -73,8 +70,8 @@ chat_history = []
 
 # Function to handle interaction with model
 @spaces.GPU
-def interact(user_input, history):
-    global tokenizer, model, interaction_count
+def interact(user_input, history, interaction_count):
+    global tokenizer, model
     try:
         if tokenizer is None or model is None:
             raise ValueError("Tokenizer or model is not initialized.")
@@ -88,7 +85,7 @@ def interact(user_input, history):
             farewell_message = "Thank you for the conversation! Have a great day!"
             history.append({"role": "assistant", "content": farewell_message})
             formatted_history = [(entry["content"], None) if entry["role"] == "user" else (None, entry["content"]) for entry in history if entry["role"] in ["user", "assistant"]]
-            return "", formatted_history, history
+            return "", formatted_history, history, interaction_count
         
         messages = history + [{"role": "user", "content": user_input}]
         
@@ -109,7 +106,7 @@ def interact(user_input, history):
         history.append({"role": "assistant", "content": response})
         
         formatted_history = [(entry["content"], None) if entry["role"] == "user" else (None, entry["content"]) for entry in history if entry["role"] in ["user", "assistant"]]
-        return "", formatted_history, history
+        return "", formatted_history, history, interaction_count
     except Exception as e:
         if torch.cuda.is_available():
             torch.cuda.empty_cache()
@@ -118,9 +115,8 @@ def interact(user_input, history):
 
 # Function to send selected story and initial message
 def send_selected_story(title, model_name, system_prompt):
-    global chat_history, selected_story, data, interaction_count
+    global chat_history, selected_story, data
     data = []  # Reset data for new story
-    interaction_count = 0  # Reset interaction count
     tokenizer, model = load_model(model_name)
     selected_story = title
     story_text = ""
@@ -141,9 +137,9 @@ Here is the story:
 
                 # Generate the first question based on the story
                 question_prompt = "Please ask a simple question about the story to encourage interaction."
-                _, formatted_history, chat_history = interact(question_prompt, chat_history)
+                _, formatted_history, chat_history, interaction_count = interact(question_prompt, chat_history, 0)
 
-                return formatted_history, chat_history, gr.update(value=[]), gr.update(value=story_text)  # Reset the data table and update the selected story textbox
+                return formatted_history, chat_history, gr.update(value=[]), gr.update(value=story_text), interaction_count  # Reset the data table and update the selected story textbox
             else:
                 print("Combined message is empty.")
         else:
@@ -220,9 +216,10 @@ with gr.Blocks() as demo:
     data_table = gr.DataFrame(headers=["User Input", "Chat Response", "Score", "Comment"])
 
     chat_history_json = gr.JSON(value=[], visible=False)
+    interaction_state = gr.State(0)
 
-    send_story_button.click(fn=send_selected_story, inputs=[story_dropdown, model_dropdown, system_prompt_dropdown], outputs=[chatbot_output, chat_history_json, data_table, selected_story_textbox])
-    send_message_button.click(fn=interact, inputs=[chatbot_input, chat_history_json], outputs=[chatbot_input, chatbot_output, chat_history_json])
+    send_story_button.click(fn=send_selected_story, inputs=[story_dropdown, model_dropdown, system_prompt_dropdown], outputs=[chatbot_output, chat_history_json, data_table, selected_story_textbox, interaction_state])
+    send_message_button.click(fn=interact, inputs=[chatbot_input, chat_history_json, interaction_state], outputs=[chatbot_input, chatbot_output, chat_history_json, interaction_state])
     save_button.click(fn=save_comment_score, inputs=[chatbot_output, score_input, comment_input, story_dropdown, user_dropdown, system_prompt_dropdown], outputs=[data_table, comment_input])
 
 demo.launch()