24.02.1
|
Go to the documentation of this file.
24 #ifndef ARM_COMPUTE_NEFUSEBATCHNORMALIZATIONKERNEL_H
25 #define ARM_COMPUTE_NEFUSEBATCHNORMALIZATIONKERNEL_H
38 const char *
name()
const override
40 return "NEFuseBatchNormalizationKernel";
74 const ITensor *input_bias =
nullptr,
75 const ITensor *bn_beta =
nullptr,
76 const ITensor *bn_gamma =
nullptr,
120 bool _run_in_place_weights;
121 bool _run_in_place_bias;
123 using FuseBatchNormFunction = void(
const ITensor *input_weights,
134 FuseBatchNormFunction *_func;
FuseBatchNormalizationType fbn_type
NEFuseBatchNormalizationKernel()
Default constructor.
static Status validate(const ITensorInfo *input_weights, const ITensorInfo *bn_mean, const ITensorInfo *bn_var, const ITensorInfo *fused_weights, const ITensorInfo *fused_bias, const ITensorInfo *input_bias=nullptr, const ITensorInfo *bn_beta=nullptr, const ITensorInfo *bn_gamma=nullptr, float epsilon=0.001f, FuseBatchNormalizationType fbn_type=FuseBatchNormalizationType::CONVOLUTION)
Static function to check if given info will lead to a valid configuration of NEFuseBatchNormalization...
Interface for CPU tensor.
NEFuseBatchNormalizationKernel & operator=(const NEFuseBatchNormalizationKernel &)=delete
Prevent instances of this class from being copied (As this class contains pointers)
FuseBatchNormalizationType
Available FuseBatchNormalizationType.
const char * name() const override
Name of the kernel.
void configure(const ITensor *input_weights, const ITensor *bn_mean, const ITensor *bn_var, ITensor *fused_weights, ITensor *fused_bias, const ITensor *input_bias=nullptr, const ITensor *bn_beta=nullptr, const ITensor *bn_gamma=nullptr, float epsilon=0.001f, FuseBatchNormalizationType fbn_type=FuseBatchNormalizationType::CONVOLUTION)
Set the source, destination of the kernel.
void run(const Window &window, const ThreadInfo &info) override
Execute the kernel on the passed window.
OpenNE kernel to fuse the batch normalization node to a preceding convolution node.
@ CONVOLUTION
For Convolution weights.
Common interface for all kernels implemented in C++.
const Window & window() const
The maximum window the kernel can be executed on.
Information about executing thread and CPU.
Describe a multidimensional execution window.
Copyright (c) 2017-2024 Arm Limited.
Store the tensor's metadata.
ScaleKernelInfo info(interpolation_policy, default_border_mode, PixelValue(), sampling_policy, false)
~NEFuseBatchNormalizationKernel()=default
Default destructor.