Update app.py

app.py

@@ -13,14 +13,13 @@ def format_prompt(message, history):
     global system_prompt_sent
     prompt = "<s>"
 
-    if history is not None and isinstance(history, list):
-        if not any(f"[INST] {system_prompt} [/INST]" in user_prompt for user_prompt, _ in history):
-            prompt += f"[INST] {system_prompt} [/INST]"
-            system_prompt_sent = True
+    if not any(f"[INST] {system_prompt} [/INST]" in user_prompt for user_prompt, _ in history):
+        prompt += f"[INST] {system_prompt} [/INST]"
+        system_prompt_sent = True
 
-        for user_prompt, bot_response in history:
-            prompt += f"[INST] {user_prompt} [/INST]"
-            prompt += f" {bot_response}</s> "
+    for user_prompt, bot_response in history:
+        prompt += f"[INST] {user_prompt} [/INST]"
+        prompt += f" {bot_response}</s> "
 
     prompt += f"[INST] {message} [/INST]"
     return prompt
@@ -37,6 +36,7 @@ def text_to_speech(text, speed=1.3):
     return audio_file_path
 
 def generate(user_input, history, temperature=None, max_new_tokens=1024, top_p=0.95, repetition_penalty=1.0):
+    
     global system_prompt_sent
     temperature = float(temperature) if temperature is not None else 0.9
     if temperature < 1e-2:
@@ -54,8 +54,8 @@ def generate(user_input, history, temperature=None, max_new_tokens=1024, top_p=0
 
     formatted_prompt = format_prompt(user_input, history)
     stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=True)
-
     response = ""
+
     for response_token in stream:
         response += response_token.token.text