當前位置: 首頁>>代碼示例>>Python>>正文


Python NervanaGPU.ones方法代碼示例

本文整理匯總了Python中nervanagpu.NervanaGPU.ones方法的典型用法代碼示例。如果您正苦於以下問題:Python NervanaGPU.ones方法的具體用法?Python NervanaGPU.ones怎麽用?Python NervanaGPU.ones使用的例子?那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在nervanagpu.NervanaGPU的用法示例。


在下文中一共展示了NervanaGPU.ones方法的1個代碼示例,這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚,您的評價將有助於係統推薦出更棒的Python代碼示例。

示例1: GPU

# 需要導入模塊: from nervanagpu import NervanaGPU [as 別名]
# 或者: from nervanagpu.NervanaGPU import ones [as 別名]

#.........這裏部分代碼省略.........
        """
        Calculates square root of the elements in a tensor

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

        Returns:
            GPUTensor: reference to out
        """
        self.ng.sqrt(x, out=out)
        return out

    def zeros(self, shape, dtype=default_dtype, persist_values=True):
        """
        Allocate a new GPUTensor and fill it with zeros.

        Arguments:
            shape (tupel): Shape of the desired GPUTensor
            dtype (dtype): Optional datatype
            persist_values (bool, optional): If set to True (the default), the
                                             values assigned to this Tensor
                                             will persist across multiple begin
                                             and end calls.  Setting to False
                                             may provide a performance increase
                                             if values do not need to be
                                             maintained across such calls

        Returns:
            GPUTensor: output
        """
        return self.ng.zeros(shape, dtype=dtype)

    def ones(self, shape, dtype=default_dtype, persist_values=True):
        """
        Allocate a new GPUTensor and fill it with ones.

        Arguments:
            shape (tupel): Shape of the desired GPUTensor
            dtype (dtype): Optional datatype
            persist_values (bool, optional): If set to True (the default), the
                                             values assigned to this Tensor
                                             will persist across multiple begin
                                             and end calls.  Setting to False
                                             may provide a performance increase
                                             if values do not need to be
                                             maintained across such calls

        Returns:
            GPUTensor: output
        """
        return self.ng.ones(shape, dtype=dtype)

    def empty(self, shape, dtype=default_dtype, persist_values=True):
        """
        Allocate a new GPUTensor.

        Arguments:
            shape (tupel): Shape of the desired GPUTensor
            dtype (dtype): Optional datatype
            persist_values (bool, optional): If set to True (the default), the
                                             values assigned to this Tensor
                                             will persist across multiple begin
                                             and end calls.  Setting to False
                                             may provide a performance increase
                                             if values do not need to be
開發者ID:YouVentures,項目名稱:neon,代碼行數:70,代碼來源:gpu.py


注:本文中的nervanagpu.NervanaGPU.ones方法示例由純淨天空整理自Github/MSDocs等開源代碼及文檔管理平台,相關代碼片段篩選自各路編程大神貢獻的開源項目,源碼版權歸原作者所有,傳播和使用請參考對應項目的License;未經允許,請勿轉載。