Prompt Cache Hit Ratio

Problem

Compute the fraction of requests whose full prompt already exists in the cache (exact token-sequence match).

The signature is prompt_cache_hit_ratio(requests, cached). Return a float between 0 and 1.

Hint

Prompt caching trades memory for latency by skipping repeated prefill work.

case 1

in: [[[1,2],[3,4],[1,2]],[[1,2]]]

expected: 0.6666666667

case 2

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

expected: 0

case 3

in: [[[1,2],[3,4]],[[1,2],[3,4]]]

expected: 1

case 4

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

expected: 0

Starter codePython

def prompt_cache_hit_ratio(requests, cached):
    # Your code here
    pass

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