Square Root Inverse Schedule

Problem

The square root schedule scales the learning rate as lr_t = lr0 / sqrt(t + 1).

Given lr0 and the step t, return the learning rate.

Hint

The offset keeps the first step at the base rate.

case 1

in: [0.1,0]

expected: 0.1

case 2

in: [0.1,99]

expected: 0.01

case 3

in: [1,15]

expected: 0.25

Starter codePython

def square_root_inverse_schedule(lr0, t):
    # Your code here
    pass

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