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

Note: When using this API, ensure that the driver firmware package and CANN package are in the 8.0.RC2 version or later. Otherwise, an error, such as BUS ERROR, will be reported.

[object Object]

• API function: Performs MatMul computation on quantized input parameters [object Object] and [object Object], followed by dequantization, then performs addition with [object Object], and finally executes AllReduce computation. Per-tensor and per-channel quantization modes are supported.
• Formula:$$output= AllReduce(dequantScale*(x1_{int8}@x2_{int8} + bias_{int32}) + x3)$$

[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]
  • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]: The [object Object] input can be in ND (2D input only) or FRACTAL_NZ (4D input only) format. When the format of [object Object] is FRACTAL_NZ, [object Object] and [object Object] are used to convert the data format from ND into NZ. Non-contiguous tensors support only the transpose scenario.

• 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:

  aclnnStatus status code. For details, see .

[object Object]

• Deterministic computing:

  • [object Object] defaults to a non-deterministic implementation. You can call [object Object] to enable deterministic compute.

• MC2 is disabled in incremental generation scenarios but enabled in full generation scenarios.

• [object Object] can be 2D (m, k) or 3D (b, s, k). [object Object] must be 2D (k, n), where the k axis meets the input parameter requirements of the MatMul operator. The k axes of [object Object] and [object Object] must be equal. [object Object] or [object Object] cannot be an empty matrix.

• The value of m cannot exceed 2147483647. The size of the last dimension of [object Object] (k) and [object Object] (k when transposed and n when not transposed) cannot exceed 65535. If [object Object] is not empty, its shape is (n). If [object Object] is not empty, its shape is the same as that of [object Object].

• When the shape of input [object Object] is (b, s, k), [object Object] shape is (b, s, n). When the shape of input [object Object] is (m, k), [object Object] shape is (m, n).

• The passed [object Object], [object Object], [object Object], or [object Object] cannot be a null pointer.

• The data types and formats of [object Object], [object Object], [object Object], [object Object], [object Object] (when not empty), and [object Object] (when not empty) must be supported.

• If [object Object] is of FLOAT16 type, the type of [object Object] is INT64 or UINT64. If [object Object] is of BFLOAT16 type, the types of [object Object] and [object Object] are both BFLOAT16.

• Only the all-mesh networking of HCCS links is supported.

  • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]: 1, 2, 4, or 8 ranks are supported.

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]: supports only one communication domain for MC2 operators within a model.

[object Object]

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

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]:

  [object Object]