app.py

import streamlit as st
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

# Custom Streamlit styling with sticky navbar
st.markdown(
    """
    <style>
    body {
        background-color: #1E1E1E;
        color: #FFFFFF;
        font-family: 'Arial', sans-serif;
    }
    .stButton>button {
        background-color: #4A90E2;
        color: #FFFFFF;
        border-radius: 15px;
        padding: 12px 24px;
        font-size: 16px;
        font-weight: bold;
    }
    .title {
        color: #64FFDA;
        text-shadow: 1px 1px #FF4C4C;
    }
    .stTabs [data-testid="stHorizontalBlock"] {
        position: sticky;
        top: 0;
        background-color: #1E1E1E;
        z-index: 10;
    }
    </style>
    """,
    unsafe_allow_html=True
)

# Load the Telco Customer Churn dataset
st.title("📲 Telco Customer Churn Prediction")
st.markdown("<h2 class='title'>Predict whether a customer will churn! 🚀</h2>", unsafe_allow_html=True)

# Load dataset
file_path = 'WA_Fn-UseC_-Telco-Customer-Churn.csv'
df = pd.read_csv(file_path)

# Preprocess data and train model (runs once)
df = df[['tenure', 'MonthlyCharges', 'TotalCharges', 'Churn']]
df = df.replace(" ", np.nan).dropna()
df['TotalCharges'] = pd.to_numeric(df['TotalCharges'])
df['Churn'] = df['Churn'].apply(lambda x: 1 if x == 'Yes' else 0)

# Define features and target
X = df.drop('Churn', axis=1)
y = df['Churn']

# Split data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Scale data
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)

# Train Support Vector Machine Model
model = SVC(kernel='linear', probability=True, random_state=42)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

# Top Tabs Navigation
tab1, tab2, tab3 = st.tabs(["📊 Dataset", "📈 Visualization", "🔮 Prediction"])

# Dataset Section
with tab1:
    st.write("### 📊 Dataset Preview")
    st.dataframe(df.head())

# Visualization Section
with tab2:
    # Display model performance
    accuracy = accuracy_score(y_test, y_pred)
    st.write("### 🔥 Model Performance")
    st.write(f"**✅ Model Accuracy:** {accuracy:.2f}")

    # Visualizing performance
    st.write("### 📊 Performance Breakdown")
    conf_matrix = confusion_matrix(y_test, y_pred)
    st.write("Confusion Matrix:")
    fig, ax = plt.subplots()
    sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='coolwarm', ax=ax)
    st.pyplot(fig)

# Prediction Section
with tab3:
    st.write("### 🔮 Predict Customer Churn")
    st.markdown("Adjust the stats below to simulate a customer scenario!")

    tenure = st.slider("Customer Tenure (Months)", min_value=0, max_value=72, value=12)
    monthly_charges = st.slider("Monthly Charges ($)", min_value=0, max_value=200, value=50)
    total_charges = st.slider("Total Charges ($)", min_value=0, max_value=10000, value=600)

    if st.button("✨ Predict Churn"):
        input_data = scaler.transform([[tenure, monthly_charges, total_charges]])
        prediction = model.predict(input_data)[0]
        prediction_proba = model.predict_proba(input_data)[0]

        st.subheader("🔮 Prediction Result")
        result_text = "🚨 Customer is likely to CHURN!" if prediction == 1 else "✅ Customer is likely to STAY."
        st.success(result_text) if prediction == 0 else st.error(result_text)
        st.write(f"Confidence: {prediction_proba[prediction]:.2f}")

        # Churn/Stay Bar Chart
        st.write("### 📊 Churn Probability Breakdown")
        fig, ax = plt.subplots()
        ax.bar(["Stay", "Churn"], [prediction_proba[0], prediction_proba[1]], color=["#64FFDA", "#FF4C4C"])
        ax.set_ylim(0, 1)
        ax.set_ylabel("Probability")
        ax.set_title("Customer Churn Probability")
        st.pyplot(fig)