2.10 random

Important

This lesson is still under development.

import random

random

We can use random function to generate a random number between 0 and 1.

print(random.random())

0.5005484333139992

every time we call random.random(), the result will be different and therefore unpredictable

print(random.random())

0.0661104417162296

print(random.random())

0.38796794455387107

randint

If we want to generate a random integer between two ranges, we can make use of randint.

print(random.randint(1, 10))

8

choice()

If we want to select a value randomly from a given sequence, we can make use of random.choice function.

random.choice([1,2,3,4])

1

random.choice([1,2,3,4])

4

choices()

However, if we want to select more than one value, but randomly, from a sequence, we can then make use of random.choices.

random.choices([1,2,3,4,5, 6, 7])

[7]

random.choices([1,2,3,4,5], k=2)

[2, 3]

sample()

The sample function is very much similar to choices function.

random.sample([1,2,3,4,5], k=2)

dakoos = ['generals', 'siyasatdan', 'bureaucracy', 'saiths', 'anchors']
random.sample(dakoos, k=2)

['bureaucracy', 'saiths']

seed()

Suppose someone is making use of random.random() function in his script and getting some results. If you use the same script, your result will be different because we have seen that random() function generates a different number upon every call.

How can we make use of random() function and still get reproducible results?

The seed function from random module is used to generate reproducible results. Let’s generate five random numbers using random.random() function inside a for loop.

for i in range(5):
  print(random.random())

0.5866402292282346
0.2502512383618066
0.4722123289821849
0.641841067289597
0.9979048767107404

The above five numbers are completely random in the sense that we can not predict, what would be the next number. If we call the random.random() again, the newly generated random numbers will again be random and unpredictable.

for i in range(5):
  print(random.random())

0.613522915464018
0.26318101423084694
0.052186725936064216
0.9383054274343333
0.818001763144731

Now, Let’s set the random seed and then generate the random numbers using random.random()

random.seed(313)

for i in range(5):
  print(random.random())

0.5646032330638283
0.9693332662214504
0.42674278078547345
0.8722737879866462
0.19282451151232616

We have generated 5 random numbers so far after setting the ranomd seed with 313. If we reset the random seed again, the next five random numbers that will be generated will be exactly same as the first five random numbers that were generated (above) when we had set the seed to 313 initially.

random.seed(313)

for i in range(5):
  print(random.random())

0.5646032330638283
0.9693332662214504
0.42674278078547345
0.8722737879866462
0.19282451151232616

This means, every time we set the random seed, the generated numbers will be random but reproducible.

Question: (Can you) Predict the random numbers generated as a result of following code?

# random.seed(92)
# for i in range(5):
#   print(random.random())

Consider for example a function which uses random numbers in it using random.random(). Now every time we run the function, the results will be different. This means the output of our function will not be reproducible. We can set the random seed either globally (at the start of script) or locally (inside the function) in order to make our results reproducible.

Question: What will be the last value printed by the following code?

# random.seed(313)
# for i in range(7):
#   print(random.random())

Setting the random seed does not only affect random.random() function, but if affects all functions of random module. Consider for example random.choice function.

for _ in range(3):
  print(random.choice(dakoos))

siyasatdan
generals
bureaucracy

random.seed(313)

for _ in range(3):
  print(random.choice(dakoos))

anchors
siyasatdan
saiths

random.seed(313)

for _ in range(3):
  print(random.choice(dakoos))

anchors
siyasatdan
saiths

shuffle()

print(dakoos)

random.shuffle(dakoos)

print(dakoos)

['generals', 'siyasatdan', 'bureaucracy', 'saiths', 'anchors']
['siyasatdan', 'bureaucracy', 'anchors', 'generals', 'saiths']

random.shuffle(dakoos)

print(dakoos)

['generals', 'saiths', 'bureaucracy', 'anchors', 'siyasatdan']

Question: Write a function called pseudo_shuffle, which suffles the dakoos list, but the order of values in the returned/shuffled list is always same. This means calling pseudo_shuffle(dakoos) multile times return same order in the list. It should be noted that the function pseudo_shuffle must make use of random.suffle function inside it.

randrange

random.randrange(1, 10)

3

random.randrange(1, 10, step=2)

7

getstate()

setstate()

distributions