mlx.optimizers.step_decay#

step_decay(init: float, decay_rate: float, step_size: int) → Callable#

Make a step decay scheduler.

Parameters:

  • init (float) – Initial value.

  • decay_rate (float) – Multiplicative factor to decay by.

  • step_size (int) – Decay every step_size steps.

Example

>>> lr_schedule = optim.step_decay(1e-1, 0.9, 10)
>>> optimizer = optim.SGD(learning_rate=lr_schedule)
>>> optimizer.learning_rate
array(0.1, dtype=float32)
>>>
>>> for _ in range(21): optimizer.update({}, {})
...
>>> optimizer.learning_rate
array(0.081, dtype=float32)