In data science, you will need to learn how to generate random numbers, which in Python, we can do with the random module.
In this tutorial, you’ll learn how to solve the following programming challenge:
⚔️ Challenge: How to do random sampling from a list in Python?
Without further ado, let’s dive right into it!
Method 1: Using Function random.choice()
Description from the help function:
help(choice)
Code snippet:
from random import choice
seq = list(range(0,10))
print(new)
for i in range(5):
print(choice(seq))
This function uses the random module to pick 5 numbers from the sequence in the list. Note that the numbers can repeat. If the list is empty, it will return IndexError.
It can also work for strings:
For example:
Choose a random Beethoven symphony (by opus number) to listen to. This random.choice() command gives me one from the list “sym_opus”
Or from a list:
Here’s another example:
import random
#Create the list of the 9 symphonies:
sym = [1, 2, 3, 4, 5, 6, 7, 8, 9]
print('Select Symphony No. '+str(random.choice(sym)))Output:
Select Symphony No. 9
👉 Recommended Tutorial: Sample a Random Number from a Probability Distribution in Python
Method 2: Using Function random.choices()
If you need to choose a sample with replacement/duplicates
In the example below, we draw five cards (we set k=5) from a 52-card deck to determine the suits. Each card draw has equal probability of being drawn.
choices(self, population, weights=None, *, cum_weights=None, k=1)
Here’s a code example:
import random #Create list of suits suits = ['Spades','Hearts','Clubs','Diamonds'] draw = random.choices(suits, k=5) print(draw)
Here we see that Clubs are drawn three times.
Weighted Outcomes
What happens if there are different weights/probabilities to the outcome?
For example, in a roulette wheel, the outcomes are black, red, and green. But there are 18 spots for black and red, but only 2 for green. So, in the code we can weigh it accordingly.
import random roulette = ['Black', 'Red', 'Green'] spin = random.choices(roulette, weights=[18,18,2], k=1) print(spin)
Bonus Example
Another example:
import random
import pandas as pd
suits = ['Spades','Hearts','Clubs','Diamonds']
fulldeck = (list(range(1,11))+[10]*3)*4
#print(len(fulldeck))
base_name = ['A'] + list(range(2,11)) + ['J', 'Q', 'K']
cards = []
for suit in ['Spades','Hearts','Clubs','Diamonds']:
cards.extend(str(num) + " of "+ suit for num in base_name)
deck = pd.Series(fulldeck, index=cards)
#deck
def draw(deck, n=5):
return deck.sample(n)
draw(deck)Output:
4 of Clubs 4 J of Clubs 10 10 of Clubs 10 9 of Spades 9 10 of Spades 10 dtype: int64
Method 3: Using Function random.sample()
Raffle Ticket Draw:
If the list itself has duplicate elements – like when picking winners in a raffle (in the example below, the number 23, then it could be returned.
import random raffle = [23, 88, 41, 23, 7, 95, 101, 33, 67, 16, 23, 41, 23] random.sample(raffle, k=5)
What about picking unique draws – like cards from a deck?
We could name the cards from 1 to 52, and run the sample() function, to pick out 5 cards:
import random deck = list(range(1,53)) random.sample(deck, k=5)
