Feature Store Consistency

Problem

Compare offline and online feature values and return the sorted indices where they differ, up to the shorter length.

Empty input returns [].

Hint

Mismatches between offline and online features cause training-serving skew.

case 1

in: [[1,2,3],[1,2.5,3]]

expected: [1]

case 2

in: [[1,2],[1,2]]

expected: []

case 3

in: [[],[]]

expected: []

Starter codePython

def feature_store_mismatches(offline, online):
    # Your code here
    pass

3 test cases · run your code against them in the editor