Update app.py

app.py

@@ -17,14 +17,14 @@ InferenceClient(models[2]),
 InferenceClient(models[3]),
 ]
 
-def compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p):
+def compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p,chat_mem):
     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)}"
+    formatted_prompt=f"{COMPRESS_HISTORY.format(history=history[0-chat_mem:])}"
     generate_kwargs = dict(
         temperature=temp,
         max_new_tokens=1024,
@@ -72,7 +72,7 @@ def chat_inf(system_prompt,prompt,history,client_choice,seed,temp,tokens,top_p,r
     print("\n######### TOKENS "+str(tokens))
     if (in_len+tokens) > 8000:
         yield [(prompt,"Wait. I need to compress our Chat history...")]
-        hist=compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p)
+        hist=compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p,chat_mem)
         yield [(prompt,"History has been compressed, processing request...")]
         history = [(prompt,hist)]
     generate_kwargs = dict(
@@ -86,13 +86,8 @@ def chat_inf(system_prompt,prompt,history,client_choice,seed,temp,tokens,top_p,r
     #formatted_prompt=prompt   
     formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history[0-chat_mem:])
     print("\n######### PROMPT "+str(len(formatted_prompt)))
-
-
-
-    
     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
         yield [(prompt,output)]