DML_ELEMENT_WISE_BIT_XOR_OPERATOR_DESC structure (directml.h)

Computes the bitwise XOR (eXclusive OR) between each corresponding element of the input tensors, and writes the result into the output tensor.

The bitwise operation is applied to tensor data in its native encoding. Therefore, the tensor data type is ignored except for determining the width of each element.

This operator supports in-place execution, meaning that the output tensor is permitted to alias one or more of the input tensors during binding.

Syntax

struct DML_ELEMENT_WISE_BIT_XOR_OPERATOR_DESC {
  const DML_TENSOR_DESC *ATensor;
  const DML_TENSOR_DESC *BTensor;
  const DML_TENSOR_DESC *OutputTensor;
};

Members

ATensor

Type: const DML_TENSOR_DESC*

A tensor containing the left-hand side inputs.

BTensor

Type: const DML_TENSOR_DESC*

A tensor containing the right-hand side inputs.

OutputTensor

Type: const DML_TENSOR_DESC*

The output tensor to write the results to.

Example

InputTensor: (Sizes:{2,2}, DataType:UINT8)
[[0,  128], // 0b00000000, 0b10000000
 [42, 255]] // 0b00101010, 0b11111111

OutputTensor: (Sizes:{2,2}, DataType:UINT8)
[[255, 127], // 0b11111111, 0b01111111
 [213, 0  ]] // 0b11010101, 0b00000000

Availability

This operator was introduced in DML_FEATURE_LEVEL_3_0.

Tensor constraints

ATensor, BTensor, and OutputTensor must have the same DataType, DimensionCount, and Sizes.

Tensor support

DML_FEATURE_LEVEL_4_1 and above

DML_FEATURE_LEVEL_3_0 and above

Requirements