optim_lbfgs: LBFGS optimizer

Description

Implements L-BFGS algorithm, heavily inspired by minFunc

Usage

optim_lbfgs(
  params,
  lr = 1,
  max_iter = 20,
  max_eval = NULL,
  tolerance_grad = 1e-07,
  tolerance_change = 1e-09,
  history_size = 100,
  line_search_fn = NULL
)

Arguments

params: (iterable): iterable of parameters to optimize or dicts defining parameter groups
lr: (float): learning rate (default: 1)
max_iter: (int): maximal number of iterations per optimization step (default: 20)
max_eval: (int): maximal number of function evaluations per optimization step (default: max_iter * 1.25).
tolerance_grad: (float): termination tolerance on first order optimality (default: 1e-5).
tolerance_change: (float): termination tolerance on function value/parameter changes (default: 1e-9).
history_size: (int): update history size (default: 100).
line_search_fn: (str): either 'strong_wolfe' or None (default: None).

Warning

This optimizer doesn't support per-parameter options and parameter groups (there can be only one).

Right now all parameters have to be on a single device. This will be improved in the future.

If you need to move a model to GPU via $cuda(), please do so before constructing optimizers for it. Parameters of a model after $cuda() will be different objects from those before the call. In general, you should make sure that the objects pointed to by model parameters subject to optimization remain the same over the whole lifecycle of optimizer creation and usage.