Stopping on validation on the How to Train Models track. Stop when validation has not improved for a patience window. Restore the best checkpoint. Patience too small stops on noise. Patience too large wastes compute and can overfit.
This lesson assumes you already worked through Checkpoints.
The idea in practice
Set patience in evaluations, not vibes. Log the step you restored.
A concrete check
goal = {
'track': 'How to Train Models',
'lesson': 'Stopping on validation',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
Early stopping on training loss, which never asks you to stop. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Write patience and which metric it watches.
Self-check
- Say Stopping on validation in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.