Convert to TorchOptimizer

Converts an object to a TorchOptimizer.

Usage

as_torch_optimizer(x, clone = FALSE, ...)

Arguments

x

(any)
Object to convert to a TorchOptimizer.

clone

(logical(1))
Whether to make a deep clone. Default is FALSE.

...

(any)
Additional arguments. Currently used to pass additional constructor arguments to TorchOptimizer for objects of type torch_optimizer_generator.

Value

TorchOptimizer

See also

Other Torch Descriptor: TorchCallback, TorchDescriptor, TorchLoss, TorchOptimizer, as_torch_callbacks(), as_torch_loss(), mlr3torch_losses, mlr3torch_optimizers, t_clbk(), t_loss(), t_opt()

Examples

# convert a `torch::torch_optimizer_generator`
as_torch_optimizer(optim_adamw)
#> <TorchOptimizer:optim_adamw> optim_adamw
#> * Generator: optim_adamw
#> * Parameters: list()
#> * Packages: torch,mlr3torch
# the id defaults to the name of the generator, but can be overwritten
as_torch_optimizer(optim_adamw, id = "my_adamw", label = "My AdamW")
#> <TorchOptimizer:my_adamw> My AdamW
#> * Generator: optim_adamw
#> * Parameters: list()
#> * Packages: torch,mlr3torch

# convert a key of mlr3torch_optimizers, this is the same as t_opt("adamw")
as_torch_optimizer("adamw")
#> <TorchOptimizer:adamw> Decoupled Weight Decay Regularization
#> * Generator: optim_ignite_adamw
#> * Parameters: list()
#> * Packages: torch,mlr3torch

# TorchOptimizers are returned as-is, unless clone is TRUE
opt = t_opt("adamw")
identical(as_torch_optimizer(opt), opt)
#> [1] TRUE
identical(as_torch_optimizer(opt, clone = TRUE), opt)
#> [1] FALSE