torch.optim.functional.rmsprop

torch.optim.functional.rmsprop(params, grads, square_avgs, grad_avgs, momentum_buffer_list, state_steps, foreach=None, maximize=False, differentiable=False, capturable=False, has_complex=False, *, lr, alpha, eps, weight_decay, momentum, centered)

Functional API that performs RMSprop 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 RMSprop for algorithm details.