[object Object]

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

• Description: Applies the HardSigmoid function and returns a new tensor based on the formula. The shape of the output tensor is identical to that of the input tensor.
• Formula:$$Hardsigmoid(self)=\begin{cases} &1, &if(self\gt3) \\ &0, &if(self\le-3) \\ &\frac{self}{6} + \frac{1}{2}, &otherwise \end{cases}$$

[object Object]

• aclnnHardsigmoid and aclnnInplaceHardsigmoid implement the same function in different ways. Select a proper operator based on your requirements.

  • aclnnHardsigmoid: An output tensor object needs to be created to store the computation result.
  • aclnnInplaceHardsigmoid: No output tensor object needs to be created, and the computation result is stored in the memory of the input tensor.

• Each operator has calls. First, aclnnHardsigmoidGetWorkspaceSize or aclnnInplaceHardsigmoidGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnHardsigmoid or aclnnInplaceHardsigmoid is called to perform computation.

[object Object]

• Parameters:

  [object Object]
  • [object Object]Atlas training products[object Object]: The data type can be FLOAT, FLOAT16, or INT32.

• 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]

• Parameters:

  [object Object]
  • [object Object]Atlas training products[object Object]: The data type can be FLOAT, FLOAT16, or INT32.

• 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 computation:
  • aclnnHardsigmoid and aclnnInplaceHardsigmoid default to a deterministic implementation.

[object Object]

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

aclnnHardsigmoid

aclnnInplaceHardsigmoid