Update app.py

app.py

@@ -17,7 +17,40 @@ InferenceClient(models[2]),
 InferenceClient(models[3]),
 ]
 
+ou are a Game Master for a Role Playing Game.
+Your duty is to narate the game and create challenges for users to overcome.
+Instructions
+- Describe the location, and scenario, in great detail
+- Create a situation where the user must make a choice of how to proceed
+- Return the scenario, and wait for the user to repond before proceeding
+Progress:
+{history}
+Compress the timeline of progress above into a single summary of events and progress
+Include all important milestones, the current challenges, and implementation details necessary to procee
 
+
+def compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p):
+    client=clients[int(client_choice)-1]
+    COMPRESS_HISTORY="""You are an Information Summarizer Agent. Your duty is to summarize the following information into a more concise format with far less words.
+    Retain all the main points and provide a brief and concise summary of the conversation.
+    Converstion:
+    {history}"""
+    print("COMPRESSING")
+    formatted_prompt=f"{COMPRESS_HISTORY.format(history=history)}"
+    generate_kwargs = dict(
+        temperature=temp,
+        max_new_tokens=1024,
+        top_p=top_p,
+        repetition_penalty=rep_p,
+        do_sample=True,
+        seed=seed,
+    )
+    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
+    output = ""
+    for response in stream:
+        output += response.token.text
+    return output
+    
 def format_prompt(message, history):
     prompt = ""
     if history:
@@ -42,8 +75,8 @@ def chat_inf(system_prompt,prompt,history,client_choice,seed,temp,tokens,top_p,r
     if history:
         hist_len=len(history)
         print(hist_len)
-        
-    #seed = random.randint(1,1111111111111111)
+    if hist_len>4000:
+        history=compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p)
     generate_kwargs = dict(
         temperature=temp,
         max_new_tokens=tokens,