vart_npu_tensor.hpp File Reference

Tensor data types, memory descriptors, and the NpuTensor handle for VART ML. More...

#include <cstdint>
#include <memory>
#include <string>
#include <unordered_map>
#include <vector>

Include dependency graph for vart_npu_tensor.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	vart::NpuTensorInfo
	Metadata structure describing a tensor used in VART. More...
class	vart::NpuTensor
	This class represents a tensor in the VART API. More...

Namespaces
	vart
	VART (Vitis AI Runtime) ML inference API namespace.

Enumerations
enum class	vart::DataType {   vart::UNKNOWN , vart::BOOLEAN , vart::INT8 , vart::UINT8 ,   vart::INT16 , vart::UINT16 , vart::BF16 , vart::FP16 ,   vart::INT32 , vart::UINT32 , vart::FLOAT32 , vart::INT64 ,   vart::UINT64 }
	Enumerates the supported data types for tensors in the VART API. More...
enum class	vart::MemoryLayout {   vart::UNKNOWN , vart::NC , vart::NCH , vart::NHC ,   vart::NHW , vart::NWC , vart::NHWC , vart::NCHW ,   vart::NHWC4 , vart::NHWC8 , vart::NC4HW4 , vart::NC8HW8 ,   vart::HCWNC4 , vart::HCWNC8 , vart::HCWNC16 , vart::NHW16C4WC ,   vart::NHW16WC4C , vart::GENERIC }
	Enumerates the supported memory layouts for tensors in the VART API. More...
enum class	vart::MemoryType {   vart::UNKNOWN , vart::XRT_BO , vart::DMA_FD , vart::USER_POINTER_CMA ,   vart::USER_POINTER_NON_CMA }
	Enumerates the various memory types utilized for tensors in the VART API. More...
enum class	vart::TensorDirection { vart::INPUT , vart::OUTPUT }
	Enumerates the supported tensor directions in the VART API. More...
enum class	vart::TensorType { vart::CPU , vart::HW }
	Specifies the tensor types supported in the VART API. More...

Detailed Description

Tensor data types, memory descriptors, and the NpuTensor handle for VART ML.

Defines the enums, structs, and classes used to describe and manage tensor data in VART ML:

• DataType, MemoryLayout, MemoryType, TensorDirection, TensorType enums
• NpuTensorInfo metadata struct
• NpuTensor class for wrapping and accessing tensor buffers

Include this header directly only when working with tensors without the Runner interface. For full API access, include vart_runner_factory.hpp which includes this header internally.