Non-Contiguous Tensors

Currently, the input aclTensors of most operator APIs support non-contiguous tensors. That is, a tensor can be represented by (shape, strides, offset).

Note: You can create an aclTensor by calling the aclCreateTensor API in "Common APIs" in Operator Library API Reference.

Example 1

For a tensor with shape=(6, 5), strides=(10, 1), and offset=22, its memory layout is as follows:

a_0,0 , a_0,1 , a_0,2 , a_0,3 , a_0,4 , a_0,5 , a_0,6 , a_0,7 , a_0,8 , a_0,9
a_1,0 , a_1,1 , a_1,2 , a_1,3 , a_1,4 , a_1,5 , a_1,6 , a_1,7 , a_1,8 , a_1,9
a_2,0 , a_2,1 , a_2,2 , a_2,3 , a_2,4 , a_2,5 , a_2,6 , a_2,7 , a_2,8 , a_2,9
a_3,0 , a_3,1 , a_3,2 , a_3,3 , a_3,4 , a_3,5 , a_3,6 , a_3,7 , a_3,8 , a_3,9
a_4,0 , a_4,1 , a_4,2 , a_4,3 , a_4,4 , a_4,5 , a_4,6 , a_4,7 , a_4,8 , a_4,9
a_5,0 , a_5,1 , a_5,2 , a_5,3 , a_5,4 , a_5,5 , a_5,6 , a_5,7 , a_5,8 , a_5,9
a_6,0 , a_6,1 , a_6,2 , a_6,3 , a_6,4 , a_6,5 , a_6,6 , a_6,7 , a_6,8 , a_6,9
a_7,0 , a_7,1 , a_7,2 , a_7,3 , a_7,4 , a_7,5 , a_7,6 , a_7,7 , a_7,8 , a_7,9
a_8,0 , a_8,1 , a_8,2 , a_8,3 , a_8,4 , a_8,5 , a_8,6 , a_8,7 , a_8,8 , a_8,9
a_9,0 , a_9,1 , a_9,2 , a_9,3 , a_9,4 , a_9,5 , a_9,6 , a_9,7 , a_9,8 , a_9,9

The boldfaced values represent the memory layout of the tensor. The complete tensor is discontiguous in memory layout. strides indicates the interval between two adjacent elements on a tensor dimension. If the stride on dimension 1 is 1, the dimension is contiguous. If the stride on dimension 0 is 10, adjacent elements are separated by 10 elements, that is, they are discontiguous elements. offset indicates the offset of the first element of the tensor relative to the address.

Example 2

For a tensor with shape=(4, 3), strides=(20, 2), and offset=22, its memory layout is as follows:

a_0,0 , a_0,1 , a_0,2 , a_0,3 , a_0,4 , a_0,5 , a_0,6 , a_0,7 , a_0,8 , a_0,9
a_1,0 , a_1,1 , a_1,2 , a_1,3 , a_1,4 , a_1,5 , a_1,6 , a_1,7 , a_1,8 , a_1,9
a_2,0 , a_2,1 , a_2,2 , a_2,3 , a_2,4 , a_2,5 , a_2,6 , a_2,7 , a_2,8 , a_2,9
a_3,0 , a_3,1 , a_3,2 , a_3,3 , a_3,4 , a_3,5 , a_3,6 , a_3,7 , a_3,8 , a_3,9
a_4,0 , a_4,1 , a_4,2 , a_4,3 , a_4,4 , a_4,5 , a_4,6 , a_4,7 , a_4,8 , a_4,9
a_5,0 , a_5,1 , a_5,2 , a_5,3 , a_5,4 , a_5,5 , a_5,6 , a_5,7 , a_5,8 , a_5,9
a_6,0 , a_6,1 , a_6,2 , a_6,3 , a_6,4 , a_6,5 , a_6,6 , a_6,7 , a_6,8 , a_6,9
a_7,0 , a_7,1 , a_7,2 , a_7,3 , a_7,4 , a_7,5 , a_7,6 , a_7,7 , a_7,8 , a_7,9
a_8,0 , a_8,1 , a_8,2 , a_8,3 , a_8,4 , a_8,5 , a_8,6 , a_8,7 , a_8,8 , a_8,9
a_9,0 , a_9,1 , a_9,2 , a_9,3 , a_9,4 , a_9,5 , a_9,6 , a_9,7 , a_9,8 , a_9,9

The boldfaced values represent the memory layout of the tensor. The whole tensor is not contiguous in memory. strides indicates the interval between two adjacent elements on a tensor dimension. If the stride on dimension 1 is 2, there is an interval of 1 element on this dimension. If the stride on dimension 0 is 20, adjacent elements are separated by 20 elements, that is, they are discontiguous elements. offset indicates the offset of the first element of the tensor relative to the address.