[object Object]

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

• Description: Performs backpropagation of . It is used to calculate the gradient of the input tensor so that the model parameters can be updated during backpropagation.
• Formula:$$res\_for\_gamma = (input - mean) \times rstd$$ $$dy\_g = gradOut \times weightOptional$$ $$temp_1 = 1/N \times \sum_{reduce\_axis\_1} gradOut \times weightOptional$$ $$temp_2 = 1/N \times (input - mean) \times rstd \times \sum_{reduce\_axis\_1}(gradOut \times weightOptional \times (input - mean) \times rstd)$$ $$gradInputOut = (gradOut \times weightOptional - (temp_1 + temp_2)) \times rstd$$ $$gradWeightOut = \sum_{reduce\_axis\_0}gradOut \times (input - mean) \times rstd$$ $$gradBiasOut = \sum_{reduce\_axis\_0}gradOut$$ N indicates the dimension of the axis on which normalization computation is performed, that is, the size of the normalized axis dimension.

[object Object]

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

[object Object]

• Parameters:

  [object Object]

  • [object Object]Atlas inference series products[object Object] and [object Object]Atlas training series products[object Object]:

    The data type of [object Object], [object Object], [object Object], [object Object], [object Object], [object Object], [object Object], [object Object], and [object Object] does not support BFLOAT16.

• Returns:

  aclnnStatus: 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:

  aclnnStatus: status code. For details, see .

[object Object]

• Shape constraints:

  The shape of gradOut, input, mean, rstd, weightOptional (non-empty), biasOptional (non-empty), gradInputOut (non-empty), gradWeightOut (non-empty), or gradBiasOut (non-empty) supports one to eight dimensions.

• Deterministic compute:

  aclnnLayerNormBackward defaults to a deterministic implementation.

[object Object]

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