Calculates the weight size to be requested under the NCHW input of the Convolution operator. Only the Float16 type is supported. This API is only used to determine the size required for preprocessing the weight tensor, in order to achieve the optimal execution performance of the Convolution operator.
For example, for performance optimization, when the input shape is [2, 4, 8, 8], the function reshapes it to [64, 1, 16, 16]. As a result, the input is interpreted as having 16,384 elements.
Parameters:
[object Object]Returns:
[object Object]: 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]
Deterministic computation:
[object Object]defaults to a deterministic implementation.
Only forward Conv2D is supported.
Transposed convolution is not supported.
Empty tensors are supported. The returned
[object Object]is[object Object].
The following example is for reference only. For details, see [object Object]Compile and Run Sample[object Object].