[object Object][object Object]

[object Object]undefined[object Object]

• Description: Computes the size of each kernel based on the input outputSize, performs 3D max pooling on the input self, and outputs the pooled value outputOut and index indicesOut. The difference between aclnnAdaptiveMaxPool3d and aclnnMaxPool3d is that you only need to specify the outputSize size and divide the pooling region based on the outputSize size.

• Formula: For each element at position $(l,m,n)$ on the DHW axis in the outputOut tensor, the calculation formula is as follows:

  $$D^{l}_{left} = floor((l*D)/D_o)$$ $$D^{l}_{right} = ceil(((l+1)*D)/D_o)$$ $$H^{m}_{left} = floor((m*H)/H_o)$$ $$H^{m}_{right} = ceil(((m+1)*H)/H_o)$$ $$W^{n}_{left} = floor((n*W)/W_o)$$ $$W^{n}_{right} = ceil(((n+1)*W)/W_o)$$ $$outputOut(N,C,l,m,n)=\underset {i \in [D^{l}_{left}, D^{l}_{right}],j\in [H^m_{left},H^m_{right}], k \in [W^n_{left},W^n_{right}] }{max} input(N,C,i,j,k)$$ $$indicesOut(N,C,l,m,n)=\underset {i \in [D^{l}_{left}, D^{l}_{right}],j\in [H^m_{left},H^m_{right}], k \in [W^n_{left},W^n_{right}] }{argmax} input(N,C,i,j,k)$$

[object Object]

Each operator has calls. First, aclnnAdaptiveMaxPool3dGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnAdaptiveMaxPool3d is called to perform computation.

[object Object]

• Parameters:

  [object Object]

• 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]

• Deterministic compute:

  • aclnnAdaptiveMaxPool3d defaults to a deterministic implementation.

• Shape description:

  • self.shape = (N, C, Din, Hin, Win)
  • outputSize = [Din, Hout, Wout]
  • outputOut.shape = (N, C, Din, Hout, Wout)
  • indicesOut.shape = (N, C, Din, Hout, Wout)

[object Object]

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