Basic function to execute mean and standard deviation normalization by calling NEMeanStdDevNormalizationKernel.
More...
#include <NEMeanStdDevNormalizationLayer.h>
Basic function to execute mean and standard deviation normalization by calling NEMeanStdDevNormalizationKernel.
Definition at line 36 of file NEMeanStdDevNormalizationLayer.h.
◆ NEMeanStdDevNormalizationLayer() [1/3]
◆ NEMeanStdDevNormalizationLayer() [2/3]
Prevent instances of this class from being copied (As this class contains pointers)
◆ NEMeanStdDevNormalizationLayer() [3/3]
Prevent instances of this class from being moved (As this class contains non movable objects)
◆ ~NEMeanStdDevNormalizationLayer()
◆ configure()
void configure |
( |
ITensor * |
input, |
|
|
ITensor * |
output = nullptr , |
|
|
float |
epsilon = 1e-8f |
|
) |
| |
Initialise the function's input and outputs.
- Note
- If the output tensor is a nullptr, the normalization will be performed in-place.
- Parameters
-
[in,out] | input | Input tensor with 2 dimensions. Data types supported: F16/F32. |
[out] | output | (Optional) Destination tensor. It can be nullptr in case of in-place computation. Data type supported: same as input |
[in] | epsilon | (Optional) Small float to avoid division by zero in case of zero standard deviation. Defaults to 1e-8. |
Definition at line 32 of file NEMeanStdDevNormalizationLayer.cpp.
Referenced by NELSTMLayer::configure().
34 auto k = std::make_unique<NEMeanStdDevNormalizationKernel>();
36 _kernel = std::move(k);
◆ operator=() [1/2]
Prevent instances of this class from being copied (As this class contains pointers)
◆ operator=() [2/2]
Prevent instances of this class from being moved (As this class contains non movable objects)
◆ validate()
Static function to check if given info will lead to a valid configuration of NEMeanStdDevNormalizationKernel.
- Parameters
-
[in] | input | Source tensor info with 2 dimensions. In case of output tensor info = nullptr, this tensor will store the result of the normalization. Data types supported: F16/F32. |
[in] | output | (Optional) Destination tensor info. It can be nullptr in case of in-place computation. Data type supported: same as input |
[in] | epsilon | (Optional) Small float to avoid division by zero in case of zero standard deviation. Defaults to 1e-8. |
- Returns
- a status
Definition at line 39 of file NEMeanStdDevNormalizationLayer.cpp.
References NEMeanStdDevNormalizationKernel::validate().
Referenced by NELSTMLayer::validate().
static Status validate(const ITensorInfo *input, const ITensorInfo *output=nullptr, float epsilon=1e-8f)
Static function to check if given info will lead to a valid configuration of NEMeanStdDevNormalizatio...
The documentation for this class was generated from the following files: