Replay Buffer Warmup

Problem

Count how many full batches the replay buffer can already provide: buffer_size - batch_size + 1, floored at 0. A non-positive batch size returns 0.

Hint

Learning usually starts only after the buffer holds at least one full batch.

case 1

in: [100,32]

expected: 69

case 2

in: [10,32]

expected: 0

case 3

in: [32,32]

expected: 1

case 4

in: [50,0]

expected: 0

Starter codePython

def replay_buffer_warmup(buffer_size, batch_size):
    # Your code here
    pass

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