Which of the following code blocks returns a 10 percent sample of rows from DataFrame storesDF with replacement?

Question

Options

• AstoresDF.sample(true)
• BstoresDF.sample(true, fraction = 0.1)
• CstoresDF.sample(true, fraction = 0.15)
• DstoresDF.sampleBy(fraction = 0.1)
• EstoresDF.sample(false, fraction = 0.1)

• A
  2% (1)
• B
  88% (46)
• D
  6% (3)
• E
  4% (2)

Explanation

The PySpark DataFrame.sample(withReplacement, fraction, seed) method takes a boolean as the first argument (True = sample with replacement) and a float fraction as the second argument (0.1 = 10%). Option B - storesDF.sample(True, fraction=0.1) - correctly specifies both with-replacement (True) and 10% (0.1). Option A omits the fraction entirely. Option C uses fraction=0.15, which is 15%, not 10%. Option D uses sampleBy, which is for stratified sampling by column values, not a simple percentage sample. Option E uses False as the first argument, meaning sampling without replacement, which contradicts the requirement.

No community discussion yet for this question.