ConstantLR
ConstantLR¶
class modelzoo.common.pytorch.optim.lr_scheduler.ConstantLR (optimizer: torch.optim.optimizer.Optimizer, val: float, decay_steps: int = None, disable_lr_steps_reset: bool = False)
Maintains a constant learning rate for each parameter group (no decaying).
- Parameters:
optimizer – The optimizer to schedule
val – The learning_rate value to maintain
decay_steps – The number of steps to decay for