TensorDataTypeTraits.ixx File Reference

Compile-time traits for the abstract TensorDataType enumeration. More...

#include <cstdint>
#include <type_traits>
#include <string_view>
import Compute.MemoryResource;
import Compute.DeviceType;
import Dnn.TensorDataType;

Classes
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::BF16 >
	Traits specialization for 16-bit brain floating point. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP16 >
	Traits specialization for 16-bit half precision floating point. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP32 >
	Traits specialization for 32-bit IEEE 754 floating point. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP4_E2M1 >
	Traits specialization for 4-bit floating point with E2M1 format. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP4_E3M0 >
	Traits specialization for 4-bit floating point with E3M0 format. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP8_E4M3 >
	Traits specialization for 8-bit floating point with E4M3 format. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::FP8_E5M2 >
	Traits specialization for 8-bit floating point with E5M2 format. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::INT16 >
	Traits specialization for 16-bit signed integer. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::INT32 >
	Traits specialization for 32-bit signed integer. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::INT8 >
	Traits specialization for 8-bit signed integer. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::UINT16 >
	Traits specialization for 16-bit unsigned integer. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::UINT32 >
	Traits specialization for 32-bit unsigned integer. More...
struct	Mila::Dnn::TensorDataTypeTraits< TensorDataType::UINT8 >
	Traits specialization for 8-bit unsigned integer. More...

Namespaces
namespace	Mila
	Mila main API namespace.
namespace	Mila::Dnn

Concepts
concept	Mila::Dnn::DeviceOnlyTensorDataType
	Concept identifying device-only abstract data types.
concept	Mila::Dnn::HostCompatibleTensorDataType
	Concept identifying host-compatible abstract data types.
concept	Mila::Dnn::isValidTensor
	Primary tensor configuration validation concept.
concept	Mila::Dnn::PrecisionSupportedOnDevice
	Concept to validate precision is supported on a device at compile-time.
concept	Mila::Dnn::ValidFloatTensorDataType
	Concept constraining abstract data types to floating-point formats.
concept	Mila::Dnn::ValidIntegerTensorDataType
	Concept constraining abstract data types to integer formats.

Typedefs
template<TensorDataType TDataType>
using	Mila::Dnn::host_value_t = std::conditional_t<TensorDataTypeTraits<TDataType>::is_integer_type, int32_t, float>
	Host value type for given abstract tensor data type.

Detailed Description

Compile-time traits for the abstract TensorDataType enumeration.

Provides per-type properties used throughout the tensor system:

• size and alignment
• host/device classification flags
• human-readable type name
• host conversion type used when converting values to/from host representations