Continuous Batching Rounds

Problem

Count scheduling rounds for continuous batching. Requests are packed greedily in order into rounds whose total token count must not exceed max_tokens_per_batch; a request that does not fit starts a new round.

The signature is continuous_batching_rounds(request_tokens, max_tokens_per_batch). Return the number of rounds.

Hint

Continuous batching fills each round with as many queued requests as fit.

Test cases

case 1

in: [[100,200,300],400]

expected: 2

case 2

in: [[400,400],400]

expected: 2

case 3

in: [[100,100],400]

expected: 1

case 4

in: [[500],400]

expected: 1

case 5

in: [[],400]

expected: 0

Starter codePython

def continuous_batching_rounds(request_tokens, max_tokens_per_batch):
    # Your code here
    pass

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