Create app.py

app.py

@@ -0,0 +1,68 @@
+import streamlit as st
+import numpy as np
+import pandas as pd
+import pickle
+import joblib
+import sklearn
+
+model = joblib.load("model_Rf-3.pkl",'rb')
+
+st.title("Customer Churn Prediction")
+st.write("Predict whether a customer will churn based on their details")
+
+# ['CreditScore', 'Age', 'Tenure', 'Balance', 'EstimatedSalary']
+
+credit_score = st.number_input("Credit Score",min_value=300,max_value=900)
+credit_score =  credit_score/900
+
+age = st.slider("Age",min_value=18,max_value=100)
+age = age/92
+
+tenure = st.slider("Tenure",min_value=0,max_value=10)
+tenure = tenure/10
+
+balance = st.number_input("Balance",min_value=0.0,step=1000.0)
+balance = balance/250898.090000
+
+num_of_prods = st.slider("Number of Products",min_value=1,max_value=4)
+num_of_prods =  num_of_prods/4
+
+has_cr_card = st.selectbox("Has Credit Card",[0,1],format_func = lambda x:"YES" if x==1 else "NO")
+
+is_activemember = st.selectbox("Are you an Active Member",[0,1],format_func = lambda x:"YES" if x==1 else "NO")
+
+salary = st.number_input("Estimated Salary",min_value=0.0,step=5000.0)
+salary = salary/199992.480000
+
+geography = st.selectbox("Please Ennter your Country",["France","Germany","Spain"])
+france,germany,spain = 0,0,0
+if geography=="France":
+    france = 1
+    germany = 0
+    spain = 0
+elif geography == "Germany":
+    france = 0
+    germany = 1
+    spain = 0
+else:
+    france = 0
+    germany = 0
+    spain = 1
+
+gender = st.selectbox("Please Ennter your Gender",["Male","Female"])
+gender_male , gender_female = 0,0
+if gender=="Male":
+    gender_male = 1
+    gender_female = 0
+else:
+    gender_male = 0
+    gender_female = 1
+
+inputs = np.array([[credit_score,age,tenure,balance,num_of_prods,has_cr_card,is_activemember,salary,france,germany,spain,gender_male,gender_female]])
+
+if st.button("--PREDICT--"):
+    prediction = model.predict(inputs)
+    if prediction[0] == 1:
+        st.error("The customer is likely to churn")
+    else:
+        st.success("The customer is not likely to churn")