Running Variance Update

Problem

Update a running variance estimate with exponential moving averaging: new_var = (1 - momentum) * running_var + momentum * batch_var.

The signature is running_var_update(running_var, batch_var, momentum).

Hint

Blend the old statistic with the batch statistic using the momentum coefficient.

case 1

in: [0,1,0.1]

expected: 0.1

case 2

in: [1,0,0.1]

expected: 0.9

case 3

in: [0.5,0.5,0.9]

expected: 0.5

case 4

in: [0,4,0.25]

expected: 1

Starter codePython

def running_var_update(running_var, batch_var, momentum):
    # Your code here
    pass

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