24.02.1
|
Go to the documentation of this file.
24 #ifndef ARM_COMPUTE_CL_WINOGRADCONV2D_H
25 #define ARM_COMPUTE_CL_WINOGRADCONV2D_H
36 class CLCompileContext;
42 class ClWinogradInputTransformKernel;
43 class ClWinogradFilterTransformKernel;
44 class ClWinogradOutputTransformKernel;
104 bool enable_fast_math =
false);
117 bool enable_fast_math =
false);
126 std::unique_ptr<kernels::ClWinogradInputTransformKernel> _input_transform;
127 std::unique_ptr<kernels::ClWinogradFilterTransformKernel> _filter_transform;
128 std::unique_ptr<kernels::ClWinogradOutputTransformKernel> _output_transform;
std::vector< MemoryInfo > MemoryRequirements
SimpleTensor< float > src
Basic function to execute Winograd-based convolution on OpenCL.
Interface for filling the border of a kernel.
ClWinogradConv2d()
Default constructor.
experimental::MemoryRequirements workspace() const override
Return the memory requirements required by the workspace.
Activation Layer Information class.
void run(ITensorPack &tensors) override
Run the kernels contained in the function.
Basic interface for functions which have a single async CL kernel.
~ClWinogradConv2d()
Default destructor.
void prepare(ITensorPack &tensors) override
Prepare the function for executing.
Basic function to execute GEMM on OpenCL.
ClWinogradConv2d & operator=(const ClWinogradConv2d &)=delete
Prevent instances of this class from being copied (As this class contains pointers)
Store the tensor's metadata.
static Status validate(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info=ActivationLayerInfo(), bool enable_fast_math=false)
Static function to check if given info will lead to a valid configuration.
Copyright (c) 2017-2024 Arm Limited.
Store the tensor's metadata.
void configure(const ClCompileContext &compile_context, ITensorInfo *src, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *dst, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info=ActivationLayerInfo(), bool enable_fast_math=false)
Set the input and output tensors.