Validation for decisions on the AI Learning track. Validation is the set you use to choose models, features, and when to stop. It is not a second test set you peek at casually. Every peek spends some of its honesty.
This lesson assumes you already worked through Train and test.
The idea in practice
Fix the validation set before the search. Compare models on it. Do not add the validation rows back into training and keep the same score.
A concrete check
goal = {
'track': 'AI Learning',
'lesson': 'Validation for decisions',
}
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
Trying fifty ideas on one validation set and reporting the best as if you had planned it. 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 rule: who is allowed to look at validation, and when the number is final.
Self-check
- Say Validation for decisions 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.