Which of the following code blocks returns a 15 percent sample of rows from DataFrame storesDF without replacement?

Question

Options

• AstoresDF.sample(True, fraction = 0.15)
• BstoresDF.sample(fraction = 0.15)
• CstoresDF.sampleBy(fraction = 0.15)
• DstoresDF.sample(fraction = 0.10)
• EstoresDF.sample()

• A
  3% (1)
• B
  89% (31)
• C
  6% (2)
• E
  3% (1)

Explanation

The PySpark DataFrame sample() method signature is sample(withReplacement=False, fraction=None, seed=None). The default for withReplacement is False, so calling storesDF.sample(fraction=0.15) returns a 15% sample WITHOUT replacement - exactly what the question asks for. Choice A passes True as the first positional argument, enabling sampling WITH replacement, which is wrong. Choice C uses sampleBy(), which is for stratified sampling and requires a column and fractions dictionary. Choice D uses fraction=0.10, which gives a 10% sample, not 15%. Choice E omits the required fraction argument entirely.

No community discussion yet for this question.