generates dummy data for cohorts project

generate_dummy_data.py

def generate_dummy_dataframe(
    dummy_products,
    dummy_customers,
    dummy_customer_types = ['company','private','government'],
    first_date=datetime.datetime(2014,1,1),
    last_date=datetime.datetime(2018,12,31),
    data_points=1000):
    
    customer_type = {customer:np.random.choice(['company','private','government']) for customer in dummy_customers}
    product_prices = {product:np.random.randint(100,10000) for product in dummy_products}

    df = pd.DataFrame({
        'order_id' : [generate_dummy_order_id() for i in range(data_points)],
        'order_date' : [np.random.choice(pd.date_range(first_date,last_date)) for i in range(data_points)],
        'customer' : [np.random.choice(dummy_customers) for i in range(data_points)],
        'product' : [np.random.choice(dummy_products) for i in range(data_points)],
        'order_size': [np.random.randint(1,5) for i in range(data_points)]
    })
    df['customer_type'] = df['customer'].map(customer_type)
    df['product_price'] = df['product'].map(product_prices)
    df['basket_size'] = df['order_size']*df['product_price']
    
    return df