[object Object][object Object]

[object Object]undefined[object Object]

Performs 3D adaptive average pooling on tensor self with the specified 3D output shape (outputSize). Unlike aclnnAvgPool3d, aclnnAdaptiveAvgPool3d only needs to specify the output size to automatically deduce the kernel size and corresponding stride.

[object Object]

Each operator has calls. First, aclnnAdaptiveAvgPool3dGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnAdaptiveAvgPool3d is called to perform computation.

• [object Object]
• [object Object]

[object Object]

• Parameters:

  • self (aclTensor*, compute input): target tensor to be computed, aclTensor on the device. are supported, but empty tensors are not supported. The shape supports 4–5 dimensions. The can be NCDHW or ND.

    • [object Object]Atlas inference series products[object Object]: The data type can be FLOAT16 or FLOAT32.
    • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training series products/Atlas A3 inference series products[object Object]: The data type can be BFLOAT16, FLOAT16, or FLOAT32.

  • outputSize (aclIntArray*, compute input): shape size of the output in the D, H, and W dimensions, aclIntArray on the device. The data type can be INT64, and the array length is fixed at 3.

  • out (aclTensor*, output): aclTensor on the device. The data type is the same as that of self. The shape of out must be the same as the shape of self and the shape deduced from outputSize. are supported. The can be NCDHW or ND, and must be the same as that of self.

    • [object Object]Atlas inference series products[object Object]: The data type can be FLOAT16 or FLOAT32.
    • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training series products/Atlas A3 inference series products[object Object]: The data type can be BFLOAT16, FLOAT16, or FLOAT32.

  • workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.

  • executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

• Returns:

  aclnnStatus: status code. For details, see .

[object Object]

• Parameters:

  • workspace (void*, input): address of the workspace to be allocated on the device.
  • workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnAdaptiveAvgPool3dGetWorkspaceSize.
  • executor (aclOpExecutor*, input): operator executor, containing the operator computation process.
  • stream (aclrtStream, input): stream for executing the task.

• Returns:

  aclnnStatus: status code. For details, see .

[object Object]

• Deterministic compute:

  • aclnnAdaptiveAvgPool3d defaults to a deterministic implementation.

• Shape description:

  • self.shape = (N, C, Din, Hin, Win) or ( C, Din, Hin, Win)
  • outputSize = [Dout, Hout, Wout]
  • out.shape = (N, C, Dout, Hout, Wout) or ( C, Dout, Hout, Wout)

[object Object]

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