Elastic Net Coordinate Step

Problem

Compute one coordinate-descent update for the elastic net.

Return soft_threshold(rho_j, lam1) / (z_j + lam2), applying the usual three-branch soft-threshold rule with threshold lam1. Return 0.0 when the denominator is not positive.

Hint

lam2 shrinks the denominator while lam1 soft-thresholds the numerator.

Test cases

case 1

in: [1,2,0.25,0.5]

expected: 0.3

case 2

in: [0.1,2,0.25,0.5]

expected: 0

case 3

in: [-1,4,0.5,1]

expected: -0.1

case 4

in: [0.3,0,0.1,0]

expected: 0

Starter codePython

def elastic_net_coordinate_step(rho_j, z_j, lam1, lam2):
    # Your code here
    pass

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