Update app.py

app.py

@@ -21,7 +21,7 @@ conversations = {}
 device = "cuda" # the device to load the model onto
 
 def mistral_model():
-        """
+    """
     Loads the GPT-3.5 model and tokenizer.
     
     Returns:
@@ -40,21 +40,6 @@ def mistral_model():
     
 model, tokenizer = mistral_model()
 
-def mistral_generated_response(msg_prompt, persona_desc_prompt):
-    user_prompt = f'{msg_prompt} [/INST]'
-    persona_prompt = f'{persona_desc_prompt} [/INST]'
-    prompt_template = f'''### [INST] Instruction:{persona_prompt} [INST] {user_prompt}'''
-    
-    encodeds = tokenizer.apply_chat_template(prompt_template, return_tensors="pt")
-    
-    model_inputs = encodeds.to(device)
-    model.to(device)
-    generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
-    decoded = tokenizer.batch_decode(generated_ids)
-    response = (decoded[0])
-    return response
-
-
 
 def load_model_norm():
     """
@@ -75,6 +60,19 @@ def load_model_norm():
 
 #model, tokenizer = load_model_norm()
 
+def mistral_generated_response(msg_prompt, persona_desc_prompt):
+    user_prompt = f'{msg_prompt} [/INST]'
+    persona_prompt = f'{persona_desc_prompt} [/INST]'
+    prompt_template = f'''### [INST] Instruction:{persona_prompt} [INST] {user_prompt}'''
+    
+    encodeds = tokenizer.apply_chat_template(prompt_template, return_tensors="pt")
+    
+    model_inputs = encodeds.to(device)
+    model.to(device)
+    generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
+    decoded = tokenizer.batch_decode(generated_ids)
+    response = (decoded[0])
+    return response
 
 def generate_response(msg_prompt: str) -> dict:
     """