Observation Scaling

Problem

Map observations from their raw range into [0, 1]: (x - low[i]) / (high[i] - low[i]) elementwise.

Hint

Bounded observations stabilize neural network training.

case 1

in: [[5,0],[0,0],[10,10]]

expected: [0.5,0]

case 2

in: [[10,10],[0,0],[10,10]]

expected: [1,1]

case 3

in: [[2],[1],[5]]

expected: [0.25]

Starter codePython

def observation_scaling(observation, low, high):
    # Your code here
    pass

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