stadaData.py

# colab version : https://colab.research.google.com/drive/1BgPchQi2KjQ9Bae2lAJx6SRaLu0sO9ye?usp=sharing

import pandas as pd

#parameters
columnNames = ['smoking', 'lbw']
a00 = [0,0]
a01 = [0,1]
a10 = [1,0]
a11 = [1,1]
n00 = 75
n01 = 45
n10 = 25
n11 = 95

def createData(ann, num) :
    Ann = []
    for _ in range(num) : 
        Ann.append(ann)
    return Ann

A00 = createData(a00, n00)
for data,num in [(a01,n01),(a10,n10),(a11,n11)] :
    Ann = createData(data, num)
    A00 += Ann
data = A00.copy()

dataFrame = pd.DataFrame(data, columns=columnNames)
print(dataFrame.value_counts())

fileName = f'{columnNames[0]}-{columnNames[1]} data.xls'
dataFrame.to_excel(fileName, index=False)
print('saved to :', fileName)