torch.optim.functional.asgd

torch.optim.functional.asgd(params, grads, axs, mus, etas, state_steps, foreach=None, maximize=False, differentiable=False, capturable=False, has_complex=False, *, lambd, lr, t0, alpha, weight_decay)

Functional API that performs ASGD algorithm computation.

This function updates the provided parameters and optimizer state in place. The caller must initialize and retain optimizer state. Unless intentionally constructing a differentiable update with a supported differentiable=True argument, call this function under torch.no_grad. See Functional optimizer API for the common functional optimizer contract and examples, and ASGD for algorithm details.