app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.ensemble import RandomForestClassifier
from xgboost import XGBClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
import numpy as np

# Function to process data and return feature importances
def calculate_importances(file):
    # Read uploaded file
    heart_df = pd.read_csv(file)
    
    # Set X and y
    X = heart_df.drop('target', axis=1)
    y = heart_df['target']
    
    # Split the data
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
    
    # Initialize models
    rf_model = RandomForestClassifier(random_state=42)
    xgb_model = XGBClassifier(use_label_encoder=False, eval_metric='logloss', random_state=42)
    cart_model = DecisionTreeClassifier(random_state=42)
    
    # Train models
    rf_model.fit(X_train, y_train)
    xgb_model.fit(X_train, y_train)
    cart_model.fit(X_train, y_train)
    
    # Get feature importances
    rf_importances = rf_model.feature_importances_
    xgb_importances = xgb_model.feature_importances_
    cart_importances = cart_model.feature_importances_
    
    feature_names = X.columns
    
    # Prepare DataFrame
    rf_importance = {'Feature': feature_names, 'Random Forest': rf_importances}
    xgb_importance = {'Feature': feature_names, 'XGBoost': xgb_importances}
    cart_importance = {'Feature': feature_names, 'CART': cart_importances}
    
    # Create DataFrames
    rf_df = pd.DataFrame(rf_importance)
    xgb_df = pd.DataFrame(xgb_importance)
    cart_df = pd.DataFrame(cart_importance)
    
    # Merge DataFrames
    importance_df = rf_df.merge(xgb_df, on='Feature').merge(cart_df, on='Feature')
    
    # Save to Excel
    file_name = 'feature_importances.xlsx'
    importance_df.to_excel(file_name, index=False)
    
    return file_name, importance_df.head()

# Streamlit interface
st.title("Feature Importance Calculation")

# File upload
uploaded_file = st.file_uploader("Upload heart.csv file", type=['csv'])

if uploaded_file is not None:
    # Process the file and get results
    excel_file, preview_df = calculate_importances(uploaded_file)
    
    # Display a preview of the DataFrame
    st.write("Feature Importances (Preview):")
    st.dataframe(preview_df)
    
    # Provide a link to download the Excel file
    with open(excel_file, "rb") as file:
        btn = st.download_button(
            label="Download Excel File",
            data=file,
            file_name=excel_file,
            mime="application/vnd.ms-excel"
        )