app.py

import streamlit as st
from dotenv import load_dotenv
import os
import pandas as pd
from transformers import pipeline

# Load environment variables from the .env file
load_dotenv()
HUGGINGFACE_API_KEY = os.getenv("HUGGINGFACE_API_KEY")

st.title("Smart Budget Planner")

# Input for monthly income
income = st.number_input("Enter your monthly income:", min_value=0.0, step=0.01)

st.header("Add Your Expenses")
# Create a form to add expenses
with st.form(key='expense_form'):
    expense_name = st.text_input("Expense Name")
    expense_amount = st.number_input("Expense Amount", min_value=0.0, step=0.01)
    add_expense = st.form_submit_button("Add Expense")

# Initialize expenses in session_state if not present
if "expenses" not in st.session_state:
    st.session_state.expenses = []

# When the form is submitted, add the expense
if add_expense and expense_name and expense_amount:
    st.session_state.expenses.append({"name": expense_name, "amount": expense_amount})
    st.success(f"Added expense: {expense_name} - {expense_amount}")

# Display the list of expenses
if st.session_state.expenses:
    st.subheader("Your Expenses")
    df = pd.DataFrame(st.session_state.expenses)
    st.table(df)

    # Calculate and display budget summary
    total_expenses = sum(item["amount"] for item in st.session_state.expenses)
    remaining_budget = income - total_expenses
    st.subheader("Budget Summary")
    st.write(f"**Total Expenses:** {total_expenses:.2f}")
    st.write(f"**Remaining Budget:** {remaining_budget:.2f}")
else:
    st.info("No expenses added yet.")

# Integration with a Hugging Face model for sentiment analysis
st.header("Budget Sentiment Analysis")
if st.button("Analyze Budget"):
    # Prepare a summary text for analysis
    summary_text = f"My monthly income is {income} and my total expenses are {total_expenses if st.session_state.expenses else 0}."
    st.write("Analyzing the following text:")
    st.write(summary_text)
    
    # Use a sentiment-analysis pipeline (model: distilbert-base-uncased-finetuned-sst-2-english)
    classifier = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english")
    result = classifier(summary_text)
    
    st.write("**Sentiment Analysis Result:**")
    st.write(result)