Upload 3 files

app.py

@@ -0,0 +1,92 @@
+# import library yang dibutuhkan
+import streamlit as st
+import pandas as pd
+import numpy  as np
+import joblib
+from tensorflow import keras
+
+# load preprocessor
+with open('preprocessor.pkl','rb') as file_1:
+    preprocessor = joblib.load(file_1)
+
+# load ANN model
+model = keras.models.load_model('func_model_tuned.hdf5')
+
+# Construct Data Infer
+# define semua fitur/kolom
+features = [
+    'region_category', 'membership_category', 'joined_through_referral',
+        'preferred_offer_types', 'medium_of_operation', 'days_since_last_login',
+            'avg_time_spent', 'avg_transaction_value', 'avg_frequency_login_days',
+                'points_in_wallet', 'used_special_discount',
+                    'offer_application_preference', 'feedback']
+
+def infer(data_infer):
+    # preprocess input
+    preprocessed_data = preprocessor.transform(data_infer)
+    # predict result with best model
+    pred = model.predict(preprocessed_data)
+    pred = np.where(pred>0.5,1,0)
+    return pred
+
+# header deployment
+st.header("Customer Churn Prediction")
+
+
+# artificial data infer
+region_category_ = ['City', 'Town', 'Village']
+region_category = st.selectbox("Where are you from?", region_category_)
+membership_category_ = ['No Membership','Basic Membership','Premium Membership','Silver Membership','Gold Membership','Platinum Membership']
+membership_category = st.selectbox("What is your membership status?", membership_category_)
+joined_through_referral = st.radio("Did you joined through referral?",('Yes','No'))
+preferred_offer_types_ = ['Gift Vouchers/Coupons', 'Without Offers','Credit/Debit Card Offers'] 
+preferred_offer_types = st.selectbox("Which one is your preferred offer types?", preferred_offer_types_)
+medium_of_operation_ = ['Smartphone', 'Desktop', 'Both']               
+medium_of_operation = st.selectbox("What kind of device do you use to browse our website?", medium_of_operation_)
+days_since_last_login = st.slider("How many days since you last logged in to our website?",0,90)
+avg_time_spent = st.slider("Approximately, how long did you browse for product from our website?",0,90)
+avg_transaction_value = st.slider("How much did you spend from our website? (give us the average)",0,100000)
+avg_frequency_login_days  = st.slider("How frequent did you visits our website in a day?",0,100)    
+points_in_wallet = st.slider("How many points do you have in your wallet?",0,10000)
+used_special_discount = st.radio("Did you used special discount?",('Yes','No'))
+offer_application_preference = st.radio("Did you prefer offers from us?",('Yes','No'))
+feedback_ = ['Quality Customer Care', 'Too many ads', 'User Friendly Website',
+       'Poor Website', 'No reason specified', 'Poor Customer Service',
+       'Poor Product Quality', 'Reasonable Price',
+       'Products always in Stock']
+feedback = st.selectbox("From the following choice, please give us your feedback",feedback_)
+
+
+
+if st.button("Submit"):
+    D ={ 
+    'region_category'              :region_category,              
+    'membership_category'          :membership_category,       
+    'joined_through_referral'      :joined_through_referral,     
+    'preferred_offer_types'        :preferred_offer_types,        
+    'medium_of_operation'          :medium_of_operation,     
+    'days_since_last_login'        :days_since_last_login,        
+    'avg_time_spent'               :avg_time_spent,               
+    'avg_transaction_value'        :avg_transaction_value,        
+    'avg_frequency_login_days'     :avg_frequency_login_days,     
+    'points_in_wallet'             :points_in_wallet,             
+    'used_special_discount'        :used_special_discount,        
+    'offer_application_preference' :offer_application_preference, 
+    'feedback'                     :feedback  
+    }
+    
+    # construct data inference dalam dataframe
+    data_infer = pd.DataFrame(data=D,columns=features,index=[0])
+
+
+    #panggil fungsi inference
+    pred = infer(data_infer)
+    pred_string = ''
+    if pred==1:
+        pred_string = "churn"
+    else:
+        pred_string = "not churn"
+
+    st.header(f"Prediction Result: ")
+    st.write("This customer will " + pred_string)
+

preprocessor.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b22acc3c142b5bf26c8f50063b4b717078bf1424ecd3967f0d67f32f57cde0c
+size 6968