[object Object]

[object Object][object Object]undefined
[object Object]
  • Description: Computes the symbolic value of each tensor in the input tensor list.

  • Formula:

    x=[x0,x1,...xn1]y=[y0,y1,...yn1]x = [{x_0}, {x_1}, ... {x_{n-1}}]\\ y = [{y_0}, {y_1}, ... {y_{n-1}}]\\ yi={1,xi>00,xi=01,xi<0,(i=0,1,...n1)y_i = \left\{ \begin{aligned} 1,\quad x_i > 0\\ 0,\quad x_i = 0\\ -1,\quad x_i < 0 \end{aligned} \right. \quad {,(i=0,1,...n-1)}
[object Object]

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

[object Object]
[object Object]
[object Object]
  • Parameters

    [object Object]
  • Returns

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

    [object Object]
[object Object]
  • Parameters

    [object Object]
  • Returns

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

[object Object]
  • Deterministic computation:
    • aclnnForeachSign defaults to a deterministic implementation.
[object Object]

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

[object Object]