app.py

import streamlit as st
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

st.title("TinyLLaMA Python Tutor (LoRA)")

@st.cache_resource
def load_model():
    base_model = AutoModelForCausalLM.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
    tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
    model = PeftModel.from_pretrained(base_model, "lora_adapter")
    return tokenizer, model

tokenizer, model = load_model()

user_input = st.text_area("Ask me a Python coding question:")

if st.button("Generate Answer"):
    inputs = tokenizer(user_input, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=150)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    st.write("**Answer:**", response)