Enum LearningRateSchedulerType

Namespace

AiDotNet.LearningRateSchedulers

Assembly

AiDotNet.dll

Enumeration of available learning rate scheduler types.

public enum LearningRateSchedulerType

Fields

Constant = 0

Constant learning rate (no decay).

CosineAnnealing = 5

Cosine annealing: smooth cosine-shaped decay.

CosineAnnealingWarmRestarts = 6

Cosine annealing with warm restarts (SGDR).

Cyclic = 9

Cyclic learning rate: oscillate between bounds.

Exponential = 3

Exponential decay: multiply LR by gamma every epoch.

Lambda = 11

Custom lambda function scheduler.

LinearWarmup = 8

Linear warmup followed by optional decay.

MultiStep = 2

Multi-step decay: multiply LR by gamma at specified milestones.

OneCycle = 7

One cycle policy: warmup then annealing.

Polynomial = 4

Polynomial decay: LR follows polynomial curve to end value.

ReduceOnPlateau = 10

Reduce on plateau: decrease when metric stops improving.

Sequential = 12

Sequential composition of multiple schedulers.

Step = 1

Step decay: multiply LR by gamma every step_size epochs.

Remarks

Use this enum with the LearningRateSchedulerFactory to create schedulers by type without having to reference the concrete classes directly.