当前位置: 首页>>代码示例>>Python>>正文


Python NervanaGPU.var方法代码示例

本文整理汇总了Python中nervanagpu.NervanaGPU.var方法的典型用法代码示例。如果您正苦于以下问题:Python NervanaGPU.var方法的具体用法?Python NervanaGPU.var怎么用?Python NervanaGPU.var使用的例子?那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在nervanagpu.NervanaGPU的用法示例。


在下文中一共展示了NervanaGPU.var方法的1个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Python代码示例。

示例1: GPU

# 需要导入模块: from nervanagpu import NervanaGPU [as 别名]
# 或者: from nervanagpu.NervanaGPU import var [as 别名]

#.........这里部分代码省略.........
        Arguments:
            tsr (GPUTensor): Input tensor
            axes (int): Axis along which the reduction is performed. If axes
                        is None,  the tensor is flattened and reduced over
                        both dimensions.
            out (GPUTensor): Output tensor
        """
        if axes is None:
            sze = tsr.shape[0]*tsr.shape[1]
            self.ng.min(tsr.reshape(sze, 1), axis=0, out=out)
        else:
            self.ng.min(tsr, axis=axes, out=out)
        return out

    def max(self, tsr, axes, out):
        """
        Calculates the maximum of the elements along the specified
        axes.

        Arguments:
            tsr (GPUTensor): Input tensor
            axes (int): Axis along which the reduction is performed. If axes
                        is None,  the tensor is flattened and reduced over
                        both dimensions.
            out (GPUTensor): Output tensor
        """
        if axes is None:
            sze = tsr.shape[0]*tsr.shape[1]
            self.ng.max(tsr.reshape(sze, 1), axis=0, out=out)
        else:
            self.ng.max(tsr, axis=axes, out=out)
        return out

    def variance(self, tsr, axes, out, mean=None):
        """
        Calculates the variance of the elements along the specified
        axes.

        Arguments:
            tsr (GPUTensor): the tensor on which to compute the variance
            axes (int, list, optional): the dimension(s) along which to
                                        variance.  If set to None, we will
                                        variance over all dimensions.
            out (GPUTensor): where the result will be stored.
            mean (GPUTensor): the tensor containing mean of tsr

        Returns:
            GPUTensor: reference to out
        """
        if mean is None:
            logger.error("GPUTensor requires mean to be specified.")
            raise ValueError("mean not specified")
        self.ng.mean(self.ng.square(tsr-mean),  axis=axes, out=out)
        return out

    def fabs(self, x, out):
        """
        Calculates absolute value of the elements in a tensor

        Arguments:
            x (GPUTensor): Input tensor
            out (GPUTensor): Output tensor

        Returns:
            GPUTensor: reference to out
        """
开发者ID:YouVentures,项目名称:neon,代码行数:70,代码来源:gpu.py


注:本文中的nervanagpu.NervanaGPU.var方法示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。