Optimization AlgorithmsMedium
Learning Rate from a Range Test
~15 mincode completion
Implement suggest_lr(lrs, losses, beta) returning the suggested learning rate as a float.
Where this shows up
~15 min
••••••••••••••••
8 employers weight this skill
4 big tech firms, 2 quant funds, 1 frontier lab, 1 autonomy company. Top match scores 92.
Python
import numpy as np
def suggest_lr(lrs, losses, beta):
"""
Suggested learning rate from a range test.
Args:
lrs: learning rate at each step, increasing, shape (n,)
losses: loss at each step, shape (n,)
beta: smoothing factor in [0, 1)
Returns:
float: one tenth of the lr where the smoothed loss is lowest
"""
# YOUR CODE HERE
pass