Epochs and passes over the data on the AI Learning track. An epoch is one pass over the training set. More epochs are not more learning if the model has already fit the noise. Count epochs, but decide with the validation curve.
This lesson assumes you already worked through Learning rate.
The idea in practice
Shuffle training rows each epoch unless order is the signal, as in a sequence. Do not shuffle away time order inside a single series.
A concrete check
goal = {
'track': 'AI Learning',
'lesson': 'Epochs and passes over the data',
}
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
Running 200 epochs because the tutorial did, with no validation curve. 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
State a stopping rule in one sentence that mentions validation.
Self-check
- Say Epochs and passes over the data 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.