inferno.extensions.optimizers package¶

Submodules¶

inferno.extensions.optimizers.adam module¶

class inferno.extensions.optimizers.adam.Adam(params, lr=0.001, betas=(0.9, 0.999), eps=1e-08, lambda_l1=0, weight_decay=0, **kwargs)[source]¶

Bases: torch.optim.optimizer.Optimizer

Implements Adam algorithm with the option of adding a L1 penalty.

It has been proposed in Adam: A Method for Stochastic Optimization.

Parameters:

params (iterable) – iterable of parameters to optimize or dicts defining parameter groups
lr (float, optional) – learning rate (default: 1e-3)
betas (Tuple[float, float], optional) – coefficients used for computing running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional) – term added to the denominator to improve numerical stability (default: 1e-8)
weight_decay (float, optional) – weight decay (L2 penalty) (default: 0)

step(closure=None)[source]¶

Performs a single optimization step.

Parameters:	closure (callable, optional) – A closure that reevaluates the model and returns the loss.

inferno.extensions.optimizers.annealed_adam module¶

class inferno.extensions.optimizers.annealed_adam.AnnealedAdam(params, lr=0.001, betas=(0.9, 0.999), eps=1e-08, lambda_l1=0, weight_decay=0, lr_decay=1.0)[source]¶

Bases: inferno.extensions.optimizers.adam.Adam

Implements Adam algorithm with learning rate annealing and optional L1 penalty.