Calculator.py

# coding: utf-8

#Import dependency
from ML import ML_model,ML_df
classifier = ML_model()
df = ML_df()

#Define Functions
def calculate(array):
    result = round(100*classifier.predict_proba(array)[0][1],4)
    return result

def change_data(array,index,value):
    temp_arr = array.copy()
    temp_arr[index] = value
    return(temp_arr)

def change_to_YN(value):
    if value == 0:
        return("N")
    if value == 1:
        return("Y")


#Calculate Original %

def origin_prob(customer_id):
    output = calculate(df.loc[[customer_id]])
    return(output)
    

def compare_scenario(customer_id):
    _input = df.loc[[customer_id]]
    columns = df.columns
    scenario = {}
    for i in range(23):
        _input = df.loc[[customer_id]]
        _input.iloc[:,i] = 1-_input.iloc[:,i]
        scenario[columns[i]+" to "+change_to_YN(_input.iloc[:,i].values[0])] = str(round(origin_prob(customer_id) - calculate(_input),4))+"%"
    print(scenario)

origin_prob("6713-OKOMC")
compare_scenario("6713-OKOMC")

#importing dependencies to render #announcements calculations to HTML
# from flask import Flask, render_template, jsonify

# app = Flask(__name__)

# @app.route('/')
# def index():
#    return render_template('index.html')

# @app.route('/_get_data/', methods=['POST'])
# def origin_prob(customer_id):
#    output = calculate(df.loc[[customer_id]])
#    return jsonify({'data': render_template('response.html', output=output)})

# if __name__ == "__main__":

#    app.run(debug=True)