mindspore.nn.LARS ================== .. image:: https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source.svg :target: https://gitee.com/mindspore/mindspore/blob/master/docs/api/api_python/nn/mindspore.nn.LARS.rst :alt: 查看源文件 .. py:class:: mindspore.nn.LARS(optimizer, epsilon=1e-05, coefficient=0.001, use_clip=False, lars_filter=lambda x: 'LayerNorm' not in x.name and 'bias' not in x.name) LARS算法的实现。 LARS算法采用大量的优化技术。详见论文 `LARGE BATCH TRAINING OF CONVOLUTIONAL NETWORKS `_。 更新公式如下: .. math:: \begin{array}{ll} \\ &\newline &\hline \\ &\textbf{Parameters}: \text{base learning rate } \gamma_{0} , \text{ momentum m}, \text{ weight decay } \lambda , \\ &\hspace{5mm}\text{ LARS coefficient } \eta , \text{ number of steps } T \\ &\textbf{Init}: \text{ t=0, v=0, init weight } w_{0}^{l} \text{ for each layer } l \\[-1.ex] &\newline &\hline \\ &\textbf{while} \text{ t