[object Object]

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

• API function: Calculates the size of the weight to be allocated when the Matmul operator is in ND format. This API is used only to determine the size required for preprocessing the weight tensor to achieve optimal Matmul operator execution performance. For example, if the input is [510, 510], this function changes the shape to [512, 512] for better performance. Therefore, the reference input is changed into 262144.

• Formula:

  $$Float16/Bfloat16: result=\prod_{i \in(0, 3]}Align(tensorShape[i], 16)$$ $$INT8: result = Align(Shapesize[0], 16) * Align(Shapesize[1], 32)$$

[object Object][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]

• Deterministic computation:
  • [object Object] defaults to a deterministic implementation.

[object Object]

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