Evaluate after you stop on the How to Train Models track. After you choose a checkpoint on validation, score the test set once. Slice the score. Read errors. Do not go back and tune on that test number.
This lesson assumes you already worked through Debugging a run that will not learn.
The idea in practice
If the test score shocks you, collect a new test later. Do not edit the model tonight against this one.
A concrete check
goal = {
'track': 'How to Train Models',
'lesson': 'Evaluate after you stop',
}
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
A loop of test-set tuning. 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 the sentence you will put under the test number to stop yourself retuning.
Self-check
- Say Evaluate after you stop 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.