CosineAnnealingLR

class torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max, eta_min=0.0, last_epoch=-1, verbose='deprecated')

使用余弦退火计划设置每个参数组的学习率。

$\eta_{max}$ 被设置为初始学习率，$T_{cur}$ 是 SGDR 中自上次重启以来的 epoch 数

$$\begin{aligned} \eta_t & = \eta_{min} + \frac{1}{2}(\eta_{max} - \eta_{min})\left(1 + \cos\left(\frac{T_{cur}}{T_{max}}\pi\right)\right), & T_{cur} \neq (2k+1)T_{max}; \\ \eta_{t+1} & = \eta_{t} + \frac{1}{2}(\eta_{max} - \eta_{min}) \left(1 - \cos\left(\frac{1}{T_{max}}\pi\right)\right), & T_{cur} = (2k+1)T_{max}. \end{aligned}$$

当 last_epoch=-1 时，将初始学习率设置为 lr。请注意，由于该 schedule 是递归定义的，学习率可以同时被此 scheduler 之外的其他操作符修改。如果学习率仅由此 scheduler 设置，则每个步骤的学习率变为

$$\eta_t = \eta_{min} + \frac{1}{2}(\eta_{max} - \eta_{min})\left(1 + \cos\left(\frac{T_{cur}}{T_{max}}\pi\right)\right)$$

It has been proposed in SGDR: Stochastic Gradient Descent with Warm Restarts. Note that this only implements the cosine annealing part of SGDR, and not the restarts.

Parameters

• optimizer (Optimizer) – Wrapped optimizer.

• T_max (int) – Maximum number of iterations.

• eta_min (float) – Minimum learning rate. Default: 0.

• last_epoch (int) – The index of last epoch. Default: -1.

• verbose (bool | str) –

  If True, prints a message to stdout for each update. Default: False.

  Deprecated since version 2.2: verbose is deprecated. Please use get_last_lr() to access the learning rate.

get_last_lr()

Return last computed learning rate by current scheduler.

Return type

List[float]

get_lr()

Retrieve the learning rate of each parameter group.

load_state_dict(state_dict)

Load the scheduler’s state.

Parameters

state_dict (dict) – scheduler state. Should be an object returned from a call to state_dict().

print_lr(is_verbose, group, lr, epoch=None)

Display the current learning rate.

Deprecated since version 2.4: print_lr() is deprecated. Please use get_last_lr() to access the learning rate.

state_dict()

Return the state of the scheduler as a dict.

It contains an entry for every variable in self.__dict__ which is not the optimizer.

step(epoch=None)

Perform a step.