Create Fake Data with Faker

Data engineering

How to create fake data using the Python library, Faker

Author

Kedar Dabhadkar

Published

May 23, 2021

Generate Fake Data with Faker

Having the ability to create dummy/ fake dataset is useful for a variety of reasons. Some use-cases that come to mind are 1. Testing python code, 2. Developing sample analysis, which can be replaced by real data later, 3. Writing tutorials, and 4. Proving a point. Yes, generating POCs, or Proof of Concepts, is a very important skill in data science, which helps to get stakeholder buy-in.

# !pip install Faker
from faker import Faker
fake = Faker()

Some types of fake data

# Fake name
fake.name()

# Fake job
fake.job()

# Fake color
fake.color_name()

'YellowGreen'

# Fake city
fake.city()

'New Richardbury'

# Fake text
fake.text()

'Edge reflect under simple above. Security trial Democrat entire since something. Crime bag risk throw yard.\nSize sell hot have rise consider wife. Century play answer election clear ahead.'

# Fake Boolean
fake.pybool()

False

# Fake list
fake.pylist(nb_elements=5, variable_nb_elements=False)

['http://williams-peterson.net/',
 Decimal('82966.369131538'),
 2841,
 'kathleen35@yahoo.com',
 'DiIHCibASmTFzRrDqZyQ']

# Fake decimal
fake.pydecimal(left_digits=5, right_digits=6, positive=False, min_value=None, max_value=None)

Decimal('42545.554972')

## Some other providers:
fake.address()
fake.license_plate()
fake.swift()
fake.color()
fake.company()
fake.credit_card_full()
fake.currency()
fake.date_time()
fake.file_path()
fake.local_latlng()
fake.ascii_safe_email()
fake.job()
fake.paragraph(nb_sentences=5)
fake.fixed_width(data_columns=[(20, 'name'), (3, 'pyint', {'min_value':50, 'max_value':100})], align='right', num_rows=2)
fake.phone_number()
fake.profile()
fake.ssn()
fake.user_agent()

'Mozilla/5.0 (iPod; U; CPU iPhone OS 4_1 like Mac OS X; szl-PL) AppleWebKit/532.39.2 (KHTML, like Gecko) Version/3.0.5 Mobile/8B113 Safari/6532.39.2'

Create fake data from a specific locale

# Some locale options:
# 'it_IT','ja_JP', 'zh_CN', 'de_DE','en_US' and many more

fake = Faker('it_IT')
fake.address()

'Rotonda Lisa 18\nPapetti salentino, 31326 Medio Campidano (PO)'

Create an entire profile at once

fake = Faker()
fake.profile()

{'job': 'Drilling engineer',
 'company': 'Trevino, Campbell and Young',
 'ssn': '608-83-8424',
 'residence': '321 Brian Valleys Suite 374\nWest Wendyville, CA 36613',
 'current_location': (Decimal('82.340949'), Decimal('-80.920579')),
 'blood_group': 'B-',
 'website': ['http://www.anderson-richardson.com/',
  'https://www.smith.net/',
  'http://fisher-romero.com/'],
 'username': 'gileslaura',
 'name': 'Ryan Miller',
 'sex': 'M',
 'address': '53066 Carr Centers\nWest Brianland, IA 06589',
 'mail': 'crystalgarrett@gmail.com',
 'birthdate': datetime.date(2012, 12, 8)}

[1] https://faker.readthedocs.io/en/master/
[2] https://towardsdatascience.com/how-to-create-fake-data-with-faker-a835e5b7a9d9

Generate Fake Data with Faker

Some types of fake data

Create fake data from a specific locale

Create an entire profile at once

Read more