[object Object]

[object Object][object Object]undefined[object Object]

• Description: Computes the norm of each tensor in the input tensor list.
• Formula:

  • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]:

    $$x = [{x_0}, {x_1}, ... {x_{n-1}}]\\$$ $$y = \left(\sum_{i=0}^{n-1}|x_i|^{p}\right)^{\frac{1}{{p}}} (i=0,1,...n-1)$$

  • Ascend 950PR/Ascend 950DT:

    $$x = [{x_0}, {x_1}, ... {x_{n-1}}]\\$$

    • scalar = +Inf:

      $$y = \max|x_i|$$

    • int(scalar) = 1:

      $$y = \sum_{i=0}^{n-1}|x_i|$$

    • other:

      $$y = \left(\sum_{i=0}^{n-1}|x_i|^{2}\right)^{\frac{1}{{2}}}$$

[object Object]

Each operator has calls. First, aclnnForeachNormGetWorkspaceSize is called to obtain the input parameters and compute the required workspace size based on the process. Then, aclnnForeachNorm is called to perform computation.

[object Object]

• Parameters

  [object Object]
  • Ascend 950PR/Ascend 950DT: The maximum length of the [object Object] and [object Object] parameters is 256 characters.

• Returns

  [object Object]: status code. For details, see .

  The first-phase API implements input parameter validation. The following error codes may be returned.

  [object Object]

[object Object]

• Parameters

  [object Object]

• Returns

  [object Object]: status code. For details, see .

[object Object]

• Deterministic computation:
  • aclnnForeachNorm defaults to a deterministic implementation.

[object Object]

The following example is for reference only. For details, see .