DML_LP_NORMALIZATION_OPERATOR_DESC structure (directml.h)

Performs an Lp-normalization function along the specified axis of the input tensor.

Syntax

struct DML_LP_NORMALIZATION_OPERATOR_DESC {
  const DML_TENSOR_DESC *InputTensor;
  const DML_TENSOR_DESC *OutputTensor;
  UINT                  Axis;
  FLOAT                 Epsilon;
  UINT                  P;
};

Members

InputTensor

Type: const DML_TENSOR_DESC*

The tensor containing the input data.

OutputTensor

Type: const DML_TENSOR_DESC*

The tensor to write the results to. This tensor's Sizes should match the InputTensor.

Axis

Type: UINT

The axis on which to apply normalization.

Epsilon

Type: FLOAT

The epsilon value to use to avoid division by zero. A value of 0.00001 is recommended as default.

P

Type: UINT

The order of the normalization (either 1 or 2).

Availability

This operator was introduced in DML_FEATURE_LEVEL_1_0.

Tensor constraints

InputTensor and OutputTensor must have the same DataType, DimensionCount, and Sizes.

Tensor support

DML_FEATURE_LEVEL_3_1 and above

DML_FEATURE_LEVEL_1_0 and above

Requirements