midas_loan_generator.ipynb

import pandas as pd
import numpy as np
import faker
import datetime
import random
from six.moves import reduce
from dateutil.relativedelta import relativedelta
import uuid



# initialize a generator
fake = faker.Faker()
status_list = ['ACTIVE', 'PENDING', 'COMPLETED', "REJECTED"]
last_name_list = ['Alexandre', 'Victor', 'John', "Emmanuel", "Peter", "Kanamugire", "Joyeuse", "Claire", "Violette", "Henrietter", "Emerthe", "Rachel", "Clementine", "Phoebe", "Jean d'amour", "Joseph", 
                 "Rene", "Julius", "Israel", "Benoit", "Pius", "Anita", "Rosine","Rosine", "Julia", "Daria", "Doreen", "Emery",
                  "Esperance", "Patricie", "Nathan", "Samuel"
                 ]
first_name_list = ['Kayonga', 'Abizeyimana', 'Ooon', "Davis", "Byiringiro", "Kwizera", "Uwamahoro", "Masengesho", 
                  "Mahoro", "Rukundo", "Cyuzuzo", "Keza", "Kaliza", "Mucyo", "Mizero", "Gaju", 
                 "Manzi", "Juru", "Mutesi", "Sheja", "Izabayo", "Tuyisenge", "Tuyizere","Akimana", "Uwimana", "Tuyishime"
                  , "Iradukunda", "Ntakirutimana", "munzezero", "Mbanzabigwi", "Byiringiro", "Bapfakurera", "Hakizimana",
                  "Habanabakize", "Habanabashaka", "Nziyomaze", "Bazindyiki", "Mahirwe", "Yehovayire", "Uwamariya",
                  "Uwimbabazi", "Dushime", "Shimirwa", "Aganze", "Ganza"
                 ]

employee_id = []
for i in range(217):
    employee_id.append(str(uuid.uuid4()))

print(len(employee_id))

month_list = [
    {
        'start': datetime.date(2021, 4, 1), 
        'end': datetime.date(2021, 6, 30)
    }, 
    {
        'start': datetime.date(2021, 7, 1), 
        'end':  datetime.date(2021, 7, 31)
    },
    {
        'start': datetime.date(2021, 8, 1), 
        'end': 'today'
    }
]

def Rand(start, end, num):
    res = []
 
    for j in range(num):
        
        requestedAmount = np.random.randint(start, end)
        month = np.random.choice(month_list, p=[0.05,0.21,0.74])
        
        status = np.random.choice(status_list, p=[0.66,0.14,0.07,0.13])
        
        outstandingAmount = None
        approvedAmount = None
        if status == 'COMPLETED':
            outstandingAmount = 0
            approvedAmount = requestedAmount
            
        if status == 'ACTIVE':
            outstandingAmount = requestedAmount - (requestedAmount * 30/100)
            approvedAmount = requestedAmount
        
        maxLoanAmount = requestedAmount + (requestedAmount * 6/100)
        createdAt = fake.date_between(start_date=month['start'], end_date=month['end'])
        employeeId = np.random.choice(employee_id)
        
        approvedDate=None
        disbursedDate=None
        dueDate=None
        if status not in ['PENDING', 'REJECTED']:
            approval_days = [2,3,4,5,6]
            dys_to_approve = np.random.choice(approval_days, p=[0.25,0.3,0.25,0.1,0.1])
            
            approvedDate = createdAt + datetime.timedelta(days=int(dys_to_approve))
            disbursedDate = approvedDate
            dueDate = disbursedDate + relativedelta(months=+3)
       
        
        res.append(
            {
                'firstName': np.random.choice(first_name_list),
                'lastName': np.random.choice(last_name_list),
                'approvedDate':approvedDate,
                'createdAt':createdAt,
                'dueDate': dueDate,
                'disbursedDate': disbursedDate,
                "creditScore": np.random.randint(321,876),
                'status': status, 
                'requestedAmount': requestedAmount,
                'approvedAmount': approvedAmount,
                'outstandingAmount': outstandingAmount,
                'paymentFrequency': 'MONTHLY',
                'maxLoanAmount': maxLoanAmount,
                'isPaused': 'FALSE',
                'interestRate': 2,
                "employeeId": employeeId
            }
        )

 
    return res
 

loans = 413
minimum_amount = 88000
maximum_amount = 237600

loans_df = pd.DataFrame(Rand(minimum_amount, maximum_amount, loans))

# loans_df['createdAt'] = pd.DatetimeIndex(loans_df['createdAt']).month_name()
# g= loans_df.resample(rule='M', on='createdAt')
# g = loans_df.groupby(['createdAt'])["requestedAmount"].sum()
# g
# g.plot(x="createdAt", y=["requestedAmount"])

loans_df.to_csv('loans.csv', index=False)