[object Object]

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

Description: Performs backpropagation of . Formula:

$$output = \begin{cases} gradOutput, &if\ self \gt 0 \\ gradOutput*negativeSlope, &if\ self \le 0 \end{cases}$$ [object Object]

Each operator has calls. First, aclnnLeakyReluBackwardGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnLeakyReluBackward is called to perform computation.

[object Object]

• Parameters:

  [object Object]
  • [object Object]Atlas inference products[object Object] and [object Object]Atlas training products[object Object]: The data type of gradOutput, self, and out can be FLOAT, FLOAT16, or DOUBLE. The data type of negativeSlope can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT8, BOOL, INT16, or UINT8.

• Returns:

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

  The first-phase API implements input parameter verification. The following errors may be thrown:

  [object Object]

[object Object]

• Parameters:

  [object Object]

• Returns:

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

[object Object]

• Deterministic compute:
  • aclnnLeakyReluBackward defaults to a deterministic implementation.

[object Object]

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