看pytorch中文文档摘抄的笔记. class torch.optim.SGD(params, lr=, momentum=0, dampening=0, weight_decay=0, nesterov=False)[source] 实现随机梯度下降算法(momentum可选). Nesterov动量基于On the importance of initialization and momentum in deep learning中的公式. 参数: params (iterable) –