Linear SHAP Contribution

Problem

Compute linear SHAP contributions for one point: phi_j = w_j * (x_j - background_mean_j).

Return the list of contributions; empty input returns [].

Hint

For a linear model each feature's SHAP value is its weight times its deviation from the background mean.

case 1

in: [[5,10],[2,-1],[3,8]]

expected: [4,-2]

case 2

in: [[1],[3],[2]]

expected: [-3]

case 3

in: [[0.5,1.5],[-2,4],[0.5,0.5]]

expected: [0,4]

case 4

in: [[],[],[]]

expected: []

Starter codePython

def linear_shap_values(x, w, background_means):
    # Your code here
    pass

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