Best Epoch from Loss Curve

Problem

Return the index of the epoch with the lowest validation loss (ties go to the earliest epoch).

Empty input returns -1.

Hint

Scan for the minimum with a strict less-than so the first minimum wins.

case 1

in: [[0.9,0.7,0.5,0.6]]

expected: 2

case 2

in: [[1,1,1]]

expected: 0

case 3

in: [[0.4]]

expected: 0

case 4

in: [[]]

expected: -1

Starter codePython

def best_epoch(val_losses):
    # Your code here
    pass

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