Early Stopping with Patience
~10 mincode completion
Implement early_stopping(val_losses, patience) returning the array [best_epoch, stop_epoch], both 1-based.
Where this shows up
~10 min
••••••••••••••••
8 employers weight this skill
4 frontier labs, 2 big tech firms, 1 autonomy company, 1 enterprise vendor. Top match scores 81.
Python
import numpy as np
def early_stopping(val_losses, patience):
"""
Where early stopping keeps its checkpoint and where it ends the run.
Args:
val_losses: validation loss after each epoch, shape (n,)
patience: epochs without a new best before stopping
Returns:
array [best_epoch, stop_epoch], both counted from 1
"""
# YOUR CODE HERE
pass