app.py

import streamlit as st
import pandas as pd
import os
from transformers import T5ForConditionalGeneration, T5Tokenizer
import groq

# Initialize Groq API
groq_client = groq.Client(api_key="your_groq_api_key")

# Load RAG components
retriever_tokenizer = DPRContextEncoderTokenizer.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
retriever_model = DPRContextEncoder.from_pretrained("facebook/dpr-ctx_encoder-single-nq-base")
generator_tokenizer = T5Tokenizer.from_pretrained("google/flan-t5-large")
generator_model = T5ForConditionalGeneration.from_pretrained("google/flan-t5-large")

# Function to process user input and generate financial statements
def generate_financial_statements(file, file_type):
    # Read the file
    if file_type == "csv":
        df = pd.read_csv(file)
    elif file_type == "excel":
        df = pd.read_excel(file)
    else:
        st.error("Unsupported file type. Please upload a CSV or Excel file.")
        return

    # Convert the data into a context string
    context = df.to_string()

    # Define financial statement queries
    queries = [
        "Generate a journal from the following financial data:",
        "Generate a general ledger from the following financial data:",
        "Generate an income statement from the following financial data:",
        "Generate a balance sheet from the following financial data:",
        "Generate a cash flow statement from the following financial data:"
    ]

    # Generate financial statements using RAG
    financial_statements = {}
    for query in queries:
        # Combine query and context
        input_text = f"{query}\n{context}"

        # Retrieve relevant information (optional, if using a retriever)
        input_ids = retriever_tokenizer(input_text, return_tensors="pt").input_ids
        retrieved_context = retriever_model(input_ids)

        # Generate response using the generator model
        input_ids = generator_tokenizer(input_text, return_tensors="pt").input_ids
        output = generator_model.generate(input_ids)
        response = generator_tokenizer.decode(output[0], skip_special_tokens=True)

        # Store the result
        financial_statements[query] = response

    return financial_statements

# Streamlit UI
st.title("Financial Statement Generator")
st.write("Upload your financial data (CSV or Excel) to generate journal, general ledger, income statement, balance sheet, and cash flow statement.")

# File upload
uploaded_file = st.file_uploader("Upload your file", type=["csv", "xlsx"])
if uploaded_file is not None:
    file_type = uploaded_file.name.split(".")[-1]
    financial_statements = generate_financial_statements(uploaded_file, file_type)

    # Display results
    for statement_type, statement in financial_statements.items():
        st.subheader(statement_type)
        st.write(statement)