Setting random seed for reproducibility in NumPy - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of numpy random with fixed seed

What is the output of the following code snippet?

NumPy

import numpy as np
np.random.seed(42)
result = np.random.randint(0, 10, size=5)
print(result.tolist())

A ]6 ,4 ,7 ,3 ,6[

B [6, 3, 7, 4, 6]

C 6, 3, 7, 4, 6]

D [6, 3, 7, 4, 6

💡 Hint

Setting the seed fixes the random numbers generated by numpy.

Using np.random.seed(42) ensures the random numbers generated are always the same. The randint function with these parameters produces [6, 3, 7, 4, 6].

❓ data_output

intermediate

Number of unique values with and without seed

Consider the following code. How many unique values are in the array 'a' after running it?

NumPy

import numpy as np
np.random.seed(0)
a = np.random.choice([1, 2, 3, 4, 5], size=10, replace=True)
unique_count = len(np.unique(a))
print(unique_count)

A 2

B 4

C 5

D 3

💡 Hint

The seed fixes the random choices, so the unique count is consistent.

With seed 0, the random choice picks values so that all 5 unique numbers appear at least once in the array.

🔧 Debug

advanced

Identify the error in reproducibility code

What error will this code raise when run?

NumPy

import numpy as np
np.random.seed(123)
result = np.random.randint(0, 5, size='10')
print(result)

A No error, prints an array of 10 integers

B ValueError: size must be positive

C SyntaxError: invalid syntax

D TypeError: 'str' object cannot be interpreted as an integer

💡 Hint

Check the type of the 'size' argument in randint.

The size parameter must be an integer or tuple of integers, not a string. Passing '10' causes a TypeError.

🚀 Application

advanced

Reproducible random sampling from a DataFrame

You have a DataFrame with 100 rows. You want to randomly select 10 rows reproducibly. Which code snippet achieves this?

A df.sample(n=10, random_state=42)

B df.sample(n=10).set_seed(42)

C df.sample(n=10, seed=42)

D df.sample(n=10, random_seed=42)

💡 Hint

Check the correct parameter name for reproducible sampling in pandas.

The pandas sample method uses 'random_state' to set the seed for reproducibility.

🧠 Conceptual

expert

Effect of setting seed on multiple random generators

If you set np.random.seed(0) and then use both numpy and Python's built-in random module to generate random numbers, which statement is true?

A Setting np.random.seed(0) only affects numpy's random functions, not Python's random module.

B Setting np.random.seed(0) also fixes the output of Python's random module.

C Python's random module uses numpy's seed internally, so both are fixed.

D Neither numpy nor Python's random module are affected by np.random.seed(0).

💡 Hint

Consider if numpy and Python's random share the same seed state.

Numpy and Python's random module have separate random number generators and seeds. Setting numpy's seed does not affect Python's random module.