Dippgray/FDAII

1import streamlit as st2from transformers import AutoTokenizer, AutoModelForCausalLM3 4# Load LLaMA 2 model5@st.cache_resource6def load_model():7    model_name = "meta-llama/Llama-2-7b-chat-hf"  # or Llama-2-13b-chat-hf8    tokenizer = AutoTokenizer.from_pretrained(model_name)9    model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")10    return tokenizer, model11 12tokenizer, model = load_model()13 14# Streamlit app15st.title("LLaMA 2 Chatbot")16 17user_input = st.text_area("Enter your question:")18if st.button("Generate Response"):19    inputs = tokenizer(user_input, return_tensors="pt").to("cuda")20    outputs = model.generate(inputs.input_ids, max_length=200)21    response = tokenizer.decode(outputs[0], skip_special_tokens=True)22    st.text_area("LLaMA 2's Response", response)