Extension: AI/ML

The OpenVX EdgeAI Vendor Extension. More...

Files
file	all.hpp
	CoreVX single-include header for C++ development.
file	circular_queue.hpp
	Circular queue implementation.
file	execution_queue.hpp
	Execution queue implementation.

Classes
class	CircularQueue< T, MaxDepth >
	Circular queue implementation. More...
class	ExecutionQueue< T, MaxDepth >
	Execution queue implementation. More...

Enumerations
enum	vx_kernel_ext_e {   VX_KERNEL_ORT_CPU_INF = VX_KERNEL_BASE(VX_ID_EDGE_AI, VX_LIBRARY_KHR_BASE) + 0x1 ,   VX_KERNEL_AIS_CHATBOT = VX_KERNEL_BASE(VX_ID_EDGE_AI, VX_LIBRARY_KHR_BASE) + 0x2 ,   VX_KERNEL_LITERT_CPU_INF = VX_KERNEL_BASE(VX_ID_EDGE_AI, VX_LIBRARY_KHR_BASE) + 0x3 ,   VX_KERNEL_TORCH_CPU_INF = VX_KERNEL_BASE(VX_ID_EDGE_AI, VX_LIBRARY_KHR_BASE) + 0x4 }
	Define Edge AI Kernels. More...

Detailed Description

The OpenVX EdgeAI Vendor Extension.

Enumeration Type Documentation

vx_kernel_ext_e

enum vx_kernel_ext_e

#include <vx_corevx_ext.h>

Define Edge AI Kernels.

Enumerator
VX_KERNEL_ORT_CPU_INF	The ONNX Runtime CPU Inference kernel. Parameters [in] vx_array The input char array. [in] vx_object_array The input tensor object array. [out] vx_object_array The output tensor object array. See also Kernel: ORT Inference
VX_KERNEL_AIS_CHATBOT	The AI Model Server Chatbot kernel. Parameters [in] vx_array The input char array. [out] vx_array The output char array. See also Kernel: AI Chatbot
VX_KERNEL_LITERT_CPU_INF	The LiteRT CPU Inference kernel. Parameters [in] vx_array The input char array. [in] vx_object_array The input tensor object array. [out] vx_object_array The output tensor object array. See also Kernel: LiteRT Inference
VX_KERNEL_TORCH_CPU_INF	The Torch CPU Inference kernel. Parameters [in] vx_array The input char array. [in] vx_object_array The input tensor object array. [out] vx_object_array The output tensor object array. See also Kernel: Executorch Inference