Learning-rate schedules on the How to Train Models track. A schedule changes the step size during training. Warmup then decay is common for large models. A schedule cannot rescue a split that leaks.
This lesson assumes you already worked through Batch size and memory.
The idea in practice
Plot learning rate and loss together. Decay only after loss has started to fall.
A concrete check
goal = {
'track': 'How to Train Models',
'lesson': 'Learning-rate schedules',
}
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
Decaying from step 0 so the model never moves. 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
Sketch a warmup-then-decay schedule and say when decay starts.
Self-check
- Say Learning-rate schedules 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.