adadelta.py 文件源码

def step(self, closure=None):
        """Performs a single optimization step.

        Arguments:
            closure (callable, optional): A closure that reevaluates the model
                and returns the loss.
        """
        loss = None
        if closure is not None:
            loss = closure()

        for group in self.param_groups:
            for p in group['params']:
                if p.grad is None:
                    continue
                grad = p.grad.data
                if grad.is_sparse:
                    raise RuntimeError('Adadelta does not support sparse gradients')
                state = self.state[p]

                # State initialization
                if len(state) == 0:
                    state['step'] = 0
                    state['square_avg'] = torch.zeros_like(p.data)
                    state['acc_delta'] = torch.zeros_like(p.data)

                square_avg, acc_delta = state['square_avg'], state['acc_delta']
                rho, eps = group['rho'], group['eps']

                state['step'] += 1

                if group['weight_decay'] != 0:
                    grad = grad.add(group['weight_decay'], p.data)

                square_avg.mul_(rho).addcmul_(1 - rho, grad, grad)
                std = square_avg.add(eps).sqrt_()
                delta = acc_delta.add(eps).sqrt_().div_(std).mul_(grad)
                p.data.add_(-group['lr'], delta)
                acc_delta.mul_(rho).addcmul_(1 - rho, delta, delta)

        return loss