Description: Creates a weight tensor for computation performance affinity of the Convolution operator. This API must be used in conjunction with .
Each operator has calls. First, aclnnTransConvolutionWeightGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnTransConvolutionWeight is called to perform computation.
Parameters
[object Object]Returns
aclnnStatus: status code. For details, see [object Object]aclnn Return Codes[object Object].
The first-phase API implements input parameter validation. The following error codes may be returned.
[object Object]
Parameters
[object Object]Returns
aclnnStatus: status code. For details, see [object Object]aclnn Return Codes[object Object].
Deterministic computation:
[object Object]defaults to a deterministic implementation.
Only forward Conv2D is supported.
Transposed convolution is not supported.
Cache is not supported.
Empty tensor is supported: When both
[object Object]and[object Object]are empty tensors, the actual conversion is not performed and[object Object]returns[object Object].
The following is the sample code, which is for reference only. For details about the compilation and execution procedures, see [object Object]Compile and Run Sample[object Object].