GC3_Deployment_Evan_Derin_Ihsanudin

app.py

@@ -0,0 +1,12 @@
+import streamlit as st
+import eda
+import prediction
+
+#Membuat navigasi
+navigation = st.sidebar.selectbox('Pilih Halaman : ', ('Death Event Prediction','Exploratory Data Analysis'))
+
+#Run modul dengan if else
+if navigation == 'Death Event Prediction' :
+    prediction.run()
+else :
+    eda.run()

eda.py

@@ -0,0 +1,424 @@
+import streamlit as st
+import pandas as pd
+import seaborn as sns
+import matplotlib.pyplot as plt
+from PIL import Image
+
+#Set Config
+st.set_page_config(
+        page_title='Heart Failure - EDA',
+        layout='wide',
+        initial_sidebar_state='expanded' 
+        )
+
+def run() :
+    # Membuat Title 
+    st.markdown("<h1 style='text-align: center; color: black;'>Exploratory Data Analysis</h1>", unsafe_allow_html=True)
+    st.write('Berikut adalah EDA dari setiap feature')
+
+    # Import DF
+    df_eda = pd.read_csv('eda.csv')
+
+    # Membuat Sub Header
+    st.subheader('**EDA Feature Age**')
+    st.write('1. Range `age` pasien dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `age` terhadap klasifikasi `DEATH_EVENT`')
+
+    col1, col2 = st.columns(2)
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'AgeBin', hue="DEATH_EVENT", palette = 'winter', order = ['(40, 50]', '(50, 60]', '(60, 70]', '(70, 80]', '(80, 90]', '(90, 100]'])
+    plt.title('Distribusi Age', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Age", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+1), ha='center', va='center',fontsize = 11) 
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'AgeBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(40, 50]', '(50, 60]', '(60, 70]', '(70, 80]', '(80, 90]', '(90, 100]'], errorbar=None)
+    plt.xlabel("Range Age", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Age', fontsize=18, fontweight='bold')
+    plt.ylim(0,1.3)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+
+    # Membuat Sub Header
+    st.subheader('**EDA Feature `creatinine_phosphokinase`**')
+    st.write('1. Range `creatinine_phosphokinase` dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `creatinine_phosphokinase` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'CPBin', hue="DEATH_EVENT", palette = 'winter', order = ['(0, 200]', '(200, 400]', '(400, 600]', '(600, 800]', '(800, 1000]', '(1000, 2000]', '(2000, 5000]', '(5000, 7000]', '(7000, 9000]'])
+    plt.title('Distribusi Creatinine Phosphokinase', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Creatinine Phosphokinase (mcg/L)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+2), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,100)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'CPBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(0, 200]', '(200, 400]', '(400, 600]', '(600, 800]', '(800, 1000]', '(1000, 2000]', '(2000, 5000]', '(5000, 7000]', '(7000, 9000]'], errorbar=None)
+    plt.xlabel("Range Creatinine Phosphokinase (mcg/L)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Creatinine Phosphokinase', fontsize=18, fontweight='bold')
+    plt.ylim(0,1.1)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+    # Membuat Sub Header 
+    st.subheader('**EDA Feature `ejection_fraction`**')
+    st.write('1. Range `ejection_fraction` dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `ejection_fraction` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'EfBin', hue="DEATH_EVENT", palette = 'winter', order = ['(10, 20]', '(20, 30]', '(30, 40]', '(40, 50]', '(50, 60]', '(60, 70]', '(70, 80]'])
+    plt.title('Distribusi Ejection Fraction', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Ejection Fraction (%)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+2), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,110)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'EfBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order =  ['(10, 20]', '(20, 30]', '(30, 40]', '(40, 50]', '(50, 60]', '(60, 70]', '(70, 80]'], errorbar=None)
+
+    plt.xlabel("Range Ejection Fraction (%)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Ejection Fraction', fontsize=18, fontweight='bold')
+    plt.ylim(0,1)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+    # Membuat Sub Header
+    st.subheader('**EDA Feature `platelets`**')
+    st.write('1. Range `platelets` dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `platelets` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'PlatBin', hue="DEATH_EVENT", palette = 'winter', order = ['(200000, 300000]', '(300000, 400000]', '(400000, 500000]', '(500000, 600000]', '(600000, 700000]', '(700000, 800000]', '(800000, 900000]'])
+    plt.title('Distribusi Platelets', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Platelets (kp/mL)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,140)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'PlatBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(200000, 300000]', '(300000, 400000]', '(400000, 500000]', '(500000, 600000]', '(600000, 700000]', '(700000, 800000]', '(800000, 900000]'], errorbar=None)
+    plt.xlabel("Range Platelets (kp/mL)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Platelets', fontsize=18, fontweight='bold')
+    plt.ylim(0,1.3)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+
+    # Membuat Sub Header
+    st.subheader('**EDA Feature `serum_creatinine`**')
+    st.write('1. Range `serum_creatinine` pasien dengan *deceased* terbanyak')
+    st.write('1. Melihat pengaruh `serum_creatinine` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'SCBin', hue="DEATH_EVENT", palette = 'winter', order = ['(0, 2]', '(2, 4]', '(4, 6]', '(6, 8]', '(8, 10]'])
+    plt.title('Distribusi Serum Creatinine', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Serum Creatinine (mg/dL)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,220)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'SCBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(0, 2]', '(2, 4]', '(4, 6]', '(6, 8]', '(8, 10]'], errorbar=None)
+    plt.xlabel("Range Serum Creatinine (mg/dL)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Serum Creatinine', fontsize=18, fontweight='bold')
+    plt.ylim(0,1.1)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `serum_sodium`**')
+    st.write('1. Range `serum_sodium` dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `serum_sodium` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'SSBin', hue="DEATH_EVENT", palette = 'winter', order = ['(110, 120]', '(120, 130]', '(130, 140]'])
+    plt.title('Distribusi Serum Sodium', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Serum Sodium (mEq/L)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,175)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'SSBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(110, 120]', '(120, 130]', '(130, 140]'], errorbar=None)
+    plt.xlabel("Range Serum Sodium (mEq/L)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Serum Sodium', fontsize=18, fontweight='bold')
+    plt.ylim(0,0.7)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `time`**')
+    st.write('1. Range `time` follow up dengan *deceased* yang terbanyak')
+    st.write('2. Melihat pengaruh `time` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'TimeBin', hue="DEATH_EVENT", palette = 'winter', order = ['(0, 50]', '(50, 100]', '(100, 150]', '(150, 200]', '(200, 250]', '(250, 300]'])
+    plt.title('Distribusi Time Follow Up', fontsize=18, fontweight='bold')
+    plt.xlabel("Range Time Follow Up (days)", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    ax.tick_params(axis="x", labelsize= 9.5)
+    plt.legend(fontsize=10,title='Klasifikasi Death Event', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+2), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,80)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = 'TimeBin', y = 'DEATH_EVENT', data = df_eda, palette = 'winter', order = ['(0, 50]', '(50, 100]', '(100, 150]', '(150, 200]', '(200, 250]', '(250, 300]'], errorbar=None)
+    plt.xlabel("Range Time Follow Up (days)", fontsize= 12)
+    plt.ylabel("% Deceased", fontsize= 12)
+    plt.title('% Deceased berdasarkan Time Follow Up', fontsize=18, fontweight='bold')
+    plt.ylim(0,1)
+    for p in ax.patches:
+        ax.annotate("%.2f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+0.019), ha='center', va='center',fontsize = 11) 
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `DEATH_EVENT`**')
+    st.write('1. Balance dataset antara *class no deceased* dan *class deceased*')
+
+    #EDA 1
+    fig= plt.figure(figsize=(10,3))
+    ax = sns.countplot(x='DEATH_EVENT', data=df_eda, palette="winter")
+    plt.xlabel("Death Status", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1], ['Not Deceased', 'Deceased'], fontsize = 11)
+    plt.title('Not Deceased vs Deceased', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+5), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,230)
+    st.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `anaemia`**')
+    st.write('1. Melihat pengaruh `anaemia` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'anaemia', hue="DEATH_EVENT", palette = 'winter')
+    plt.xlabel("Anaemia", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 11)
+    plt.title('Anaemia vs Death Event', fontsize=18, fontweight='bold')
+    plt.legend(fontsize=10,title='Deceased', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,140)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = "anaemia", y = "DEATH_EVENT", data = df_eda, palette = 'winter', errorbar= None)
+    plt.ylabel("% Deceased", fontsize= 14)
+    plt.xlabel("Anaemia", fontsize= 14)
+    plt.ylim(0,0.5)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 14)
+    plt.title('% Deceased vs Anaemia', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.2f" %(p.get_height()), (p.get_x()+0.35, p.get_height()+0.01),fontsize=13)
+    col2.pyplot(fig)
+
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `diabetes`**')
+    st.write('1. Melihat pengaruh `diabetes` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'diabetes', hue="DEATH_EVENT", palette = 'winter')
+    plt.xlabel("Diabetes", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 11)
+    plt.title('Diabetes vs Death Event', fontsize=18, fontweight='bold')
+    plt.legend(fontsize=10,title='Deceased', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,140)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = "diabetes", y = "DEATH_EVENT", data = df_eda, palette = 'winter', errorbar= None)
+    plt.ylabel("% Deceased", fontsize= 14)
+    plt.xlabel("Diabetes", fontsize= 14)
+    plt.ylim(0,0.5)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 14)
+    plt.title('% Deceased vs Diabetes', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.2f" %(p.get_height()), (p.get_x()+0.35, p.get_height()+0.01),fontsize=13)
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `high_blood_pressure`**')
+    st.write('1. Melihat pengaruh `high_blood_pressure` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'high_blood_pressure', hue="DEATH_EVENT", palette = 'winter')
+    plt.xlabel("Hypertension", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 11)
+    plt.title('Hypertension vs Death Event', fontsize=18, fontweight='bold')
+    plt.legend(fontsize=10,title='Deceased', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,160)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = "high_blood_pressure", y = "DEATH_EVENT", data = df_eda, palette = 'winter', errorbar= None)
+    plt.ylabel("% Deceased", fontsize= 14)
+    plt.xlabel("Hypertension", fontsize= 14)
+    plt.ylim(0,0.5)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 14)
+    plt.title('% Deceased vs Hypertension', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.2f" %(p.get_height()), (p.get_x()+0.35, p.get_height()+0.01),fontsize=13)
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `sex`**')
+    st.write('1. Melihat pengaruh `sex` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'sex', hue="DEATH_EVENT", palette = 'winter')
+    plt.xlabel("Gender", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1],['Woman', 'Men'], fontsize = 11)
+    plt.title('Gender vs Death Event', fontsize=18, fontweight='bold')
+    plt.legend(fontsize=10,title='Deceased', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,140)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = "sex", y = "DEATH_EVENT", data = df_eda, palette = 'winter', errorbar= None)
+    plt.ylabel("% Deceased", fontsize= 14)
+    plt.xlabel("Gender", fontsize= 14)
+    plt.ylim(0,0.5)
+    plt.xticks([0,1],['Woman', 'Man'], fontsize = 14)
+    plt.title('% Deceased vs Gender', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.2f" %(p.get_height()), (p.get_x()+0.35, p.get_height()+0.01),fontsize=13)
+    col2.pyplot(fig)
+
+    # Membuat Sub Header EDA
+    st.subheader('**EDA Feature `smoking`**')
+    st.write('1. Melihat pengaruh `smoking` terhadap klasifikasi `DEATH_EVENT`')
+    col1, col2 = st.columns(2)
+
+    #EDA 1
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.countplot(data = df_eda, x = 'smoking', hue="DEATH_EVENT", palette = 'winter')
+    plt.xlabel("Smoking", fontsize= 12)
+    plt.ylabel("# of Patient", fontsize= 12)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 11)
+    plt.title('Smoking vs Death Event', fontsize=18, fontweight='bold')
+    plt.legend(fontsize=10,title='Deceased', loc='upper right', labels=['Not Deceased', 'Deceased'])
+    for p in ax.patches:
+        ax.annotate("%.0f"%(p.get_height()), (p.get_x() + p.get_width() / 2,
+                        p.get_height()+3), ha='center', va='center',fontsize = 11) 
+    plt.ylim(0,160)
+    col1.pyplot(fig)
+
+    #EDA 2
+    fig= plt.figure(figsize=(15,6))
+    ax = sns.barplot(x = "smoking", y = "DEATH_EVENT", data = df_eda, palette = 'winter', errorbar= None)
+    plt.ylabel("% Deceased", fontsize= 14)
+    plt.xlabel("Smoking", fontsize= 14)
+    plt.ylim(0,0.5)
+    plt.xticks([0,1],['No', 'Yes'], fontsize = 14)
+    plt.title('% Deceased vs Smoking', fontsize=18, fontweight='bold')
+    for p in ax.patches:
+        ax.annotate("%.2f" %(p.get_height()), (p.get_x()+0.35, p.get_height()+0.01),fontsize=13)
+
+    col2.pyplot(fig)
+
+if __name__ == '__main__':
+    run()

model_forest_opt.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a925f02e2d0276f0cdfb2f39223ecc339cc977fc8fd17c4c1ea490a9a62b573
+size 146064

prediction.py

@@ -0,0 +1,73 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import pickle 
+
+# Load All Files
+with open('windsorizer.pkl', 'rb') as file_1:
+  winsorizer_inf = pickle.load(file_1)
+
+with open('scaler_final.pkl', 'rb') as file_2:
+  scaler_final = pickle.load(file_2)
+
+with open('model_forest_opt.pkl', 'rb') as file_3:
+  model_forest_opt = pickle.load(file_3)
+
+
+def run() :
+  # Membuat Title 
+  st.markdown("<h1 style='text-align: center; color: black;'>Death Event Prediction</h1>", unsafe_allow_html=True)
+
+  # Menambahkan Deskripsi
+  st.write('Page ini berisi model untuk memprediksi *Death Event* (potensi meninggalnya pasien) dari pasien')
+  st.write('Mohon persiapkan **data Age, Creatinine Phosphokinase (mcg/L), Ejection Fraction (%), Platelets (kp/L), Serum Creatinine (mg/dl), Serum Sodium (mEq/L) dan Time (days) sesuai satuan yang tertera sebelum melakukan prediksi**')
+
+  #Membuat Form
+  with st.form(key= 'form_health_parameter'):
+      #Deklasrasi input
+      age = st.number_input('Age', min_value=40, max_value=100, value=40 ,step=1, help='Usia Pasien')
+      creatinine_phosphokinase = st.number_input('Creatinine Phospokinase (mcg/L)', min_value=20, max_value=8000, value=250 ,step=1, help='Kadar creatine kinase dalam darah (mcg/L)')
+      ejection_fraction = st.slider('Ejection Fraction (%)', min_value=10, max_value=100, value=40 ,step=1, help='Kemampuan jantung untuk memompa darah dengan oksigen ke seluruh tubuh (%)')
+      platelets = st.number_input('Platelets (kp/L)', min_value=25000, max_value=850000, value=250000 ,step=1, help='Jumlah trombosit pada darah (kp/L)')
+      serum_creatinine = st.number_input('Serum Creatinine (mg/dl)', min_value=1, max_value=10, value=1 ,step=1, help='Kadar Kreatinin (Zat Limbah) pada darah (mg/dl)')
+      serum_sodium = st.slider('Serum Sodium (mEq/L)', min_value=100, max_value=150, value=130 ,step=1, help='Kadar natrium serum pada darah (mEq/L)')
+      time = st.number_input('Time Follow Up (days)', min_value=1, max_value=300, value=100 ,step=1, help='Waktu Follow Up Pasien (days)')
+      #Membuat Button
+      submitted = st.form_submit_button('Predict')
+
+  #Membuat Data Inference    
+  data_inf = {
+      'age': age, 
+      'creatinine_phosphokinase': creatinine_phosphokinase, 
+      'ejection_fraction': ejection_fraction, 
+      'platelets': platelets, 
+      'serum_creatinine': serum_creatinine, 
+      'serum_sodium': serum_sodium, 
+      'time': time
+  }
+
+  #Membuat Dataframe
+  data_inf = pd.DataFrame([data_inf])
+  data_inf
+
+  #Prediksi
+  if submitted :
+    #Windsoriser pada data inferencing dengan modul windsoriser
+    data_inf = winsorizer_inf.transform(data_inf)
+
+    #Feature Scaling 
+    data_inf_final = scaler_final.transform(data_inf)
+
+    #Predict using Random Forest Parameter Tuning
+    y_pred_inf = model_forest_opt.predict(data_inf_final)
+
+    #Print hasil prediksi
+    if y_pred_inf == 1:
+      prediction = 'Deceased (Class 1)'
+    else:
+      prediction = 'Not Deceasead (Class 0)'
+      
+    st.write('# Death Event : ', prediction)
+
+if __name__ == '__main__':
+    run()

requirements.txt

@@ -0,0 +1,6 @@
+streamlit
+pandas
+seaborn
+matplotlib
+scikit-learn == 1.1.3
+numpy

scaler_final.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4a53a10c9ffcb9a36702e63ae8c960885f785dc238a85ae8f734d20d98fc802
+size 698

windsorizer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1fb9553f7240b00c14b34cd1339b813ff7dc96d723b8fbe870c3137753e40a7
+size 480