Step Decay Milestone Factor

Problem

Step decay multiplies the learning rate by gamma every step_size steps: lr_t = lr0 * gamma^(floor(t / step_size)).

Given lr0, gamma, step_size, and t, return the learning rate.

Hint

Count completed milestones with integer division.

case 1

in: [1,0.5,10,25]

expected: 0.25

case 2

in: [0.1,0.1,5,5]

expected: 0.010000000000000002

case 3

in: [2,0.9,20,0]

expected: 2

Starter codePython

def step_decay_milestone_factor(lr0, gamma, step_size, t):
    # Your code here
    pass

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