app.py

import streamlit as st
from datasets import load_dataset

# Load your dataset
dataset = load_dataset('csv', data_files='CCI_Details_Structured_Full.csv')

# Title and description
st.title('Contributor Search App')
st.write('Enter the name of a Contributor to search in the dataset.')

# Streamlit widget for user input
contributor_query = st.text_input('Enter Contributor to search:')

# Run the query based on the widget input
if contributor_query:  # Check if the user has entered a query
    results = [
        example for example in dataset['train'] 
        if example.get('Contributor') and example['Contributor'].lower() == contributor_query.lower()
    ]
    st.write(results)  # Display the results in the app

from transformers import T5ForConditionalGeneration, T5Tokenizer
from datasets import load_dataset

# Load the T5-small model and tokenizer (or your custom model)
model_name = "Lexim011/NISTER"  # Ensure this is the correct string identifier for your model
model = T5ForConditionalGeneration.from_pretrained(model_name)
tokenizer = T5Tokenizer.from_pretrained(model_name)

# Load your dataset
dataset = load_dataset("Lexim011/Compliance")

# Load your dataset (replace with your dataset path or identifier)
dataset = load_dataset("Lexim011/Compliance")

# Example: Encode and generate a response
def generate_answer(question, context):
    input_text = f"question: {question} context: {context} </s>"
    input_ids = tokenizer.encode(input_text, return_tensors="pt")

    outputs = model.generate(input_ids)
    answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return answer

# Example usage with your dataset
# Assuming your dataset has 'question' and 'context' columns
for example in dataset['train']:
    # Use 'Definition' as the question
    definition = example['Definition']
    
    # Use 'References' or another field as context (if needed)
    context = example['References']

    # Generate an answer based on the definition and context
    answer = generate_answer(definition, context)
    
    # Print out the results
    print(f"Definition: {definition}")
    print(f"References: {context}")
    print(f"Answer: {answer}")
    print("---")