[object Object]

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

• Description:

  Dequantizes the input Tensor, normalizes the Tensor based on the input weight, bias, and epsilon, and then quantizes the Tensor based on the outputScale and outputZeroPoint.

• Formula:

  1. Dequantization:
  $$x' = (x - inputZeroPoint) * inputScale$$
  2. Normalization:
  $$y =\frac{x' - mean}{\sqrt{var + epsilon}} * weight + bias$$
  3. Quantization:
  $$output = round(\frac{y}{outputScale} + outputZeroPoint)$$

[object Object]

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

[object Object]

• Parameters

  [object Object]

• 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:
  • aclnnQuantizedBatchNorm defaults to a deterministic implementation.

[object Object]

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