Update app.py

app.py

@@ -1,60 +1,46 @@
+import streamlit as st
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
-from peft import PeftModel
-import streamlit as st
-
-# Load tokenizer and model (on CPU)
-base_model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
-adapter_path = "lora_adapter"
 
-device = torch.device("cpu")
+# Load model and tokenizer
+model_name = "path/to/your/lora_adapter_or_model"  # Update this to your LoRA model path
 
-tokenizer = AutoTokenizer.from_pretrained(base_model_name, use_fast=True)
-base_model = AutoModelForCausalLM.from_pretrained(base_model_name).to(device)
-model = PeftModel.from_pretrained(base_model, adapter_path).to(device)
-
-# Streamlit UI setup
-st.set_page_config(page_title="Python Tutor Chatbot", page_icon="🐍", layout="centered")
-st.title("🐍 Python Tutor Chatbot")
-st.markdown("Ask me anything about Python programming!")
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
 
-# 🔁 Prompt template with instruction to ignore unrelated queries
-def create_prompt(user_input):
-    return f"""You are a helpful and expert AI Python tutor.
-
-Your job is to only answer questions strictly related to Python programming (syntax, concepts, libraries, frameworks, tools, errors, etc.).
-
-If the question is unrelated to Python, politely respond:
-"Sorry, I can only answer Python programming questions."
+# Chat function with prompt-based filtering
+def chat(instruction):
+    prompt = f"""You are a helpful and expert Python programming tutor.
+If the question is about Python, explain clearly with examples.
+If the question is unrelated to Python, respond with "Sorry, I can only answer Python-related questions."
 
 ### Instruction:
-{user_input}
+{instruction}
 
 ### Response:
 """
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=150,
+            temperature=0.7,
+            top_p=0.95,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id
+        )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response.split("### Response:")[-1].strip()
+
+# Streamlit UI
+st.set_page_config(page_title="Python Tutor Chatbot", page_icon="🐍")
+st.title("🐍 Python Tutor Chatbot")
+st.write("Ask me Python programming questions!")
 
-# 🔎 User Input
-user_input = st.text_input("Your Python Question:")
+user_input = st.text_input("Your question:")
 
-# 🔄 Inference
 if user_input:
     with st.spinner("Generating response..."):
-        prompt = create_prompt(user_input)
-        inputs = tokenizer(prompt, return_tensors="pt").to(device)
-
-        with torch.no_grad():
-            outputs = model.generate(
-                **inputs,
-                max_new_tokens=200,
-                temperature=0.7,
-                top_p=0.9,
-                top_k=50,
-                do_sample=True,
-                pad_token_id=tokenizer.eos_token_id
-            )
-
-        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        final_response = response.split("### Response:")[-1].strip()
-
+        response = chat(user_input)
         st.markdown("**Answer:**")
-        st.write(final_response)
+        st.markdown(response)