TransData
Availability
Function Usage
Converts the tensor's format without changing its value.
Specifically, this API converts the format of an input tensor into the specified dstPrimaryFormat.
Prototype
const aclTensor *TransData(const aclTensor *x, op::Format dstPrimaryFormat, int64_t groups, aclOpExecutor *executor)
Parameters
Parameter |
Input/Output |
Description |
|---|---|---|
x |
Input |
Tensor to be converted. The data type can be FLOAT16, FLOAT32, INT32, UINT32, INT8, or UINT8. |
dstPrimaryFormat |
Input |
Destination format of the input tensor. |
groups |
Input |
Group parameter, which is used for conversion by group. The data type can be INT64. |
executor |
Input |
Operator executor, containing the operator computation process. |
Returns
A tensor of dstPrimaryFormat.
Constraints
When the input tensor is of type FLOAT32, INT32, or UINT32, C0 can only be processed based on 8.
Example
1 2 3 | // Convert the output format from NC1HWC0 to NCHW. auto transGradInput = l0op::TransData(gradInputNC1HWC0, Format::FORMAT_NCHW, params.groups, executor); CHECK_RET(transGradInput != nullptr, ACLNN_ERR_INNER_NULLPTR); |