app.py

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

# Load the model and tokenizer
model_id = "google/gemma-7b"  # Replace with "google/gemma-7b-it" for instruction tuning
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# Function to generate responses based on user messages
def generate_response(messages):
    input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
    outputs = model.generate(input_ids, max_new_tokens=100)
    generated_response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return generated_response

st.title("Gemma Chatbot")
messages = []
user_input = st.text_input("You:", "")
if st.button("Send"):
    if user_input:
        messages.append({"role": "user", "content": user_input})
        bot_response = generate_response(messages)
        messages.append({"role": "assistant", "content": bot_response})
    else:
        st.warning("Please enter a message to process.")

# Display conversation
for message in messages:
    if message["role"] == "user":
        st.text_input("You:", value=message["content"], disabled=True)
    elif message["role"] == "assistant":
        st.text_area("Gemma:", value=message["content"], disabled=True)