eda.py

import streamlit as st
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import plotly.express as px
from PIL import Image

st.set_page_config(
    page_title = 'Customer Churn Risk - EDA',
    layout = 'wide',
    initial_sidebar_state = 'expanded'
    )

def run():
    # Membuat Title
    st.title('Customer Churn Prediction')

    # Membuat Sub Header
    st.subheader('EDA for Customer Churn Risk')

    # Menambahkan Gambar
    image = Image.open('churn.jpeg')
    st.image(image, caption='Customer Churn')

    # Menambahkan Deskripsi
    st.write('This page created by **Imam Zarkasie**')
    st.write('### Hello!')
    st.write('#### The competition of product sales in the e-commerce industry in heating up!')
    st.write('##### In this page we can explore customer segmentation, more than that this website provides an ability to predict a customer churn risk.')


    # Membuat Garis Lurus
    st.markdown('---')

    # Magic Syntax
    '''
    On this page, the author will do a simple exploration.
    The dataset used is the churn dataset.
    '''

    # Show DataFrame
    df = pd.read_csv('churn.csv')
    st.dataframe(df)

    #Melihat histogram fitur target
    df['churn_risk_score'].value_counts().plot(kind='bar')

    # Membuat Historgram
    st.write('#### Histogram of Churn')
    fig = plt.figure(figsize=(15, 5))
    sns.histplot(df['churn_risk_score'], bins=30, kde=True)
    st.pyplot(fig)

    # Mengelompokkan data dan menghitung statistik churn risk score
    grouped = df.groupby('membership_category')['churn_risk_score'].value_counts().unstack().fillna(0)

    # Membuat barplot
    fig, ax = plt.subplots(figsize=(8, 3))
    grouped.plot(kind='bar', stacked=False, ax=ax)

    # Menampilkan plot di Streamlit
    st.pyplot(fig)

    # Mengelompokkan data dan menghitung statistik churn risk score
    grouped = df.groupby('gender')['churn_risk_score'].value_counts().unstack().fillna(0)

    # Mengambil nilai churn risk score yang unik
    churn_risk_scores = df['churn_risk_score'].unique()

    # Mengatur lebar barplot
    bar_width = 0.35

    # Mengatur posisi barplot untuk setiap gender
    male_positions = np.arange(len(churn_risk_scores))
    female_positions = male_positions + bar_width

    # Membuat figure dan axes untuk plot
    fig, ax = plt.subplots(figsize=(8, 4))

    # Membuat barplot untuk gender Male dengan warna biru (blue)
    ax.bar(male_positions, grouped.loc['M'], width=bar_width, label='Male', color='lightblue')

    # Membuat barplot untuk gender Female dengan warna merah (red)
    ax.bar(female_positions, grouped.loc['F'], width=bar_width, label='Female', color='pink')

    # Memberikan label pada sumbu x
    ax.set_xticks(male_positions + bar_width/2)
    ax.set_xticklabels(churn_risk_scores)

    # Memberikan judul pada sumbu x dan y
    ax.set_xlabel('Churn Risk Score')
    ax.set_ylabel('Count')

    # Menambahkan judul plot
    ax.set_title('Churn Risk Score by Gender')

    # Menampilkan legenda
    ax.legend()

    # Menampilkan grid
    ax.grid(True, axis='y', linestyle='--', alpha=0.5)

    # Menampilkan plot di Streamlit
    st.pyplot(fig)

    # Membuat Histogram Berdasarkan Input User
    st.write('#### Histogram based on input user')
    pilihan = st.selectbox('Pilih column : ', ('avg_time_spent', 'avg_transaction_value', 'avg_frequency_login_days', 'points_in_wallet'))
    fig = plt.figure(figsize=(15, 5))
    sns.histplot(df[pilihan], bins=30, kde=True)
    st.pyplot(fig)

    # Membuat Plotly Plot
    st.write('#### Plotly Plot - points_in_wallet VS avg_transaction_value')
    fig = px.scatter(df, x='points_in_wallet', y='avg_transaction_value')
    st.plotly_chart(fig)

if __name__=='__main__':
    run()