torch.optim.functional.nadam

torch.optim.functional.nadam(params, grads, exp_avgs, exp_avg_sqs, mu_products, state_steps, decoupled_weight_decay=False, foreach=None, capturable=False, differentiable=False, has_complex=False, maximize=False, *, beta1, beta2, lr, weight_decay, momentum_decay, eps)

Functional API that performs NAdam 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 NAdam for algorithm details.