aclSetTensorShapeRange
Applicability
Product |
Supported |
|---|---|
Atlas 350 Accelerator Card |
√ |
√ |
|
√ |
|
√ |
|
√ |
|
√ |
Function Usage
Sets the size range of each dimension of a tensor after a tensor description is created by the aclCreateTensorDesc call.
Application scenario: For a dynamic-shape operator, a dynamic dimension is represented by –1. However, the value range of each dynamic dimension is different and needs to be set explicitly. For example, if the tensor shape is [16, –1, 20, –1], the corresponding shape range can be [[16, 16], [1, 128], [20, 20], [1, 10]]. This indicates that the first dimension is fixed at 16, the second dimension is sized 1–128, the third dimension is fixed at 20, and the fourth dimension is sized 1–10.
Prototype
1 | aclError aclSetTensorShapeRange(aclTensorDesc* desc, size_t dimsCount, int64_t dimsRange[][ACL_TENSOR_SHAPE_RANGE_NUM]) |
Parameters
Parameter |
Input/Output |
Description |
|---|---|---|
desc |
Output |
Pointer to data of the aclTensorDesc type. Call aclCreateTensorDesc to create data of the aclTensorDesc type in advance. |
dimsCount |
Input |
Number of dimensions of the tensor. |
dimsRange |
Input |
A 2D array for the size range of each dimension. A dimension represented by a value other than –1 is a fixed dimension. The maximum size and minimum size of the dimension are the same. A dimension represented by –1 is a dynamic dimension. The corresponding array specifies the minimum size and maximum size of the dimension, respectively. #define ACL_TENSOR_SHAPE_RANGE_NUM 2 |
Returns
0 on success; else, failure. For details, see aclError.