[object Object]

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

• Description: Obtains the input data of permutedTokens based on sortedIndices. If probs data exists, permutedTokens is multiplied by probs. Then, computes the cumulative sum and outputs the computation result.

• Formulas:

  • If [object Object] is not set to [object Object], the formula is as follows:

    $$T[k] = T[S[k]]$$ $$T[k] = T[k] * P[i][j]$$ $$O[i] = \sum_{k=i*topK}^{(i+1)*topK - 1 } T[k]$$

    where $i \in {0,1,...,tokens-1}$; $j \in {0,1,...,topK-1}$; $k \in {0,1,...,tokens*topK-1}$; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object].

  • If [object Object] is set to [object Object], [object Object] is [object Object]. The formula is as follows:

    $$T[i] = T[S[i]]$$ $$O[i] = T[i]$$

    where $i \in {0,1,...,tokens-1}$; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object]; [object Object] indicates [object Object].

[object Object]

Each operator has calls. First, [object Object] is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, [object Object] 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 computation:

  • [object Object] defaults to deterministic implementation.

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]: The value of [object Object] is less than or equal to [object Object].

• Ascend 950PR/Ascend 950DT: When this API is called, the framework internally calls the API. If a parameter error message is displayed, see the following parameter mapping:

  • The permutedTokens input is equivalent to the expandedX input of the aclnnMoeFinalizeRoutingV2 API.
  • The sortedIndices input is equivalent to the expandedRowIdx input of the aclnnMoeFinalizeRoutingV2 API.
  • The probsOptional input is equivalent to the scalesOptional input of the aclnnMoeFinalizeRoutingV2 API.
  • The paddedMode input is equivalent to the dropPadMode input of the aclnnMoeFinalizeRoutingV2 API.
  • The out output is equivalent to the out output of the aclnnMoeFinalizeRoutingV2 API.

• [object Object]Atlas inference products[object Object]:

  • The data types supported by [object Object] and [object Object] are FLOAT16 and FLOAT32.
  • The value of [object Object] is less than or equal to [object Object].
  • hidden_size is a multiple of 128 and is less than 10240.

[object Object]

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