Logistic Coordinate Step

Problem

Take one gradient step for logistic regression on a single feature.

With p = 1 / (1 + exp(-w * x)) and grad = (p - y) * x, return w - lr * grad. The label y is 0 or 1.

Hint

The logistic gradient is (p - y) * x.

case 1

in: [0,1,1,0.5]

expected: 0.25

case 2

in: [1,2,0,0.1]

expected: 0.8238405844044235

case 3

in: [-1,1,1,0.2]

expected: -0.853788284273999

case 4

in: [0.5,0,1,0.5]

expected: 0.5

Starter codePython

def logistic_coordinate_step(w, x, y, lr):
    # Your code here
    pass

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