Update app.py

app.py

@@ -1,5 +1,5 @@
-import gradio as gr 
-import torch 
+import gradio as gr
+import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 from peft import PeftModel
 
@@ -10,45 +10,44 @@ base_model = AutoModelForCausalLM.from_pretrained(
     load_in_4bit=True
 )
 
-#tokenizer 
+# tokenizer
 tokenizer = AutoTokenizer.from_pretrained("unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit")
 
-#LoRA adaptors 
+# LoRA adaptors
 model = PeftModel.from_pretrained(base_model, "rezaenayati/RezAi-Model")
 
 def chat_with_rezAi(messages, history):
     conversation = "<|start_header_id|>system<|end_header_id|>\nYou are Reza Enayati, a Computer Science student and entrepreneur from Los Angeles, who is eager to work as a software engineer or machine learning engineer. Answer these questions as if you are in an interview.<|eot_id|>"
-
-     for user_msg, assistant_msg in history:
+    
+    for user_msg, assistant_msg in history:
         conversation += f"<|start_header_id|>user<|end_header_id|>\n{user_msg}<|eot_id|>"
         conversation += f"<|start_header_id|>assistant<|end_header_id|>\n{assistant_msg}<|eot_id|>"
-
-    conversation += f"<|start_header_id|>user<|end_header_id|>\n{message}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n"
-
+    
+    conversation += f"<|start_header_id|>user<|end_header_id|>\n{messages}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n"
+    
     inputs = tokenizer([conversation], return_tensors="pt")
-
+    
     with torch.no_grad():
-    outputs = model.generate(
-        **inputs,
-        max_new_tokens=128,
-        temperature=0.5,
-        do_sample=True,
-        pad_token_id=tokenizer.eos_token_id
-    )
-
-    #get response 
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=128,
+            temperature=0.5,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id
+        )
+    
+    # get response
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
     new_response = response.split("<|start_header_id|>assistant<|end_header_id|>")[-1].strip()
-
     return new_response
 
 demo = gr.ChatInterface(
     fn=chat_with_rezAi,
     title="💬 Chat with RezAI",
-    description="Hi! I'm RezAI. Ask me about his technical background, projects, or experience!",
+    description="Hi! I'm RezAI. Ask me about Reza's technical background, projects, or experience!",
     examples=[
         "Tell me about your background",
-        "What programming languages do you know?", 
+        "What programming languages do you know?",
         "Walk me through your Pizza Guys project",
         "What's your experience with machine learning?",
         "How did you get into computer science?"