Per-Class Accuracy

Problem

Compute accuracy for each class in a label map.

For class c divide the number of pixels whose prediction equals the target by the number of target pixels of class c. Classes with no target pixels score 0.0; round each value to 4 decimal places.

Test cases

case 1

in: [[[0,1],[2,1]],[[0,1],[2,0]],3]

expected: [0.5,1,1]

case 2

in: [[[0,0]],[[1,1]],2]

expected: [0,0]

case 3

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

expected: [0,1]

Starter codePython

def per_class_accuracy(pred, target, num_classes):
    # Your code here
    pass

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