app.py

import streamlit as st
from transformers import pipeline
from datasets import load_dataset

# Initialize text-generation pipeline with the model
model_name = "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF"
pipe = pipeline("text-generation", model=model_name)

# Load the dataset
ds = load_dataset("refugee-law-lab/canadian-legal-data", "default", split="train")

# Streamlit interface
st.title("Canadian Legal Text Generator")
st.write("Enter a prompt related to Canadian legal data and generate text using Llama-3.1.")

# Show dataset sample
st.subheader("Sample Data from Canadian Legal Dataset:")
st.write(ds[:5])  # Displaying the first 5 rows of the dataset

# Prompt input
prompt = st.text_area("Enter your prompt:", placeholder="Type something...")

if st.button("Generate Response"):
    if prompt:
        # Generate text based on the prompt
        with st.spinner("Generating response..."):
            generated_text = pipe(prompt, max_length=100, do_sample=True, temperature=0.7)[0]["generated_text"]
            st.write("**Generated Text:**")
            st.write(generated_text)
    else:
        st.write("Please enter a prompt to generate a response.")