Validation during the run on the How to Train Models track. Compute validation on a schedule, in eval mode, without updating weights. Too often and you waste time. Too rarely and you miss the best checkpoint.
This lesson assumes you already worked through Augmentation.
The idea in practice
Use the same preprocessor as training. Do not shuffle validation into the training loader.
A concrete check
goal = {
'track': 'How to Train Models',
'lesson': 'Validation during the run',
}
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 bug where validation batches still call optimizer.step. 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
Describe the flag you set so weights do not change during validation.
Self-check
- Say Validation during the run 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.