ftl Namespace Reference

Namespaces
	detail

Classes
struct	DynamicMapper
struct	StaticMapper
	Interface which provides static mapping (uses TMP to determine some of the mapping variables at compile time for improved performance) from indices to a single offset index and from an index offset to a list of indices. More...
class	TensorAddition
	Expression class for calculating the addition of two tensors. More...
struct	TensorContainer
	Container for tensor data depending on if the tensor is static (dimension sizes, and hence the total number of elements, are known at compile time – uses std::array) or dynamic (dimension sizes are not known at compile time – uses std::vector). More...
class	TensorContainer< Dtype >
class	TensorContainer< Dtype, SizeFirst, SizeRest...>
class	TensorExpression
	Defines a general tensor expression so that opertions on tensor expressions can be defined the syntax of the operations to look as they would mathematically. More...
class	TensorExpression< Expression, TensorTraits< DT, CPU, SF, SR...> >
class	TensorExpression< Expression, TensorTraits< Dtype, CPU > >
class	TensorInterface
class	TensorInterface< TensorTraits< DT, CPU > >
class	TensorInterface< TensorTraits< DT, CPU, SF, SR...> >
class	TensorSubtraction
	Expression class for calculating the subtraction of two tensors. More...
struct	TensorTraits
	Traits class which specifies parameters for a tensor, such as what type of container it uses and what type of device the computations should be performed on. More...
struct	TensorTraits< Dtype, DeviceType >
struct	TensorTraits< Dtype, DeviceType, SizeFirst, SizeRest...>

Typedefs
template<typename Dtype , device DeviceType, size_t... DimSizes>
using	Tensor = TensorInterface< TensorTraits< Dtype, DeviceType, DimSizes...>>
template<typename DT >
using	DynamicTensorCpu = TensorInterface< TensorTraits< DT, CPU >>
template<typename DT , size_t SF, size_t... SR>
using	StaticTensorCpu = TensorInterface< TensorTraits< DT, CPU, SF, SR...>>
using	device = short

Typedef Documentation

using ftl::device = typedef short

template<typename DT >

using ftl::DynamicTensorCpu = typedef TensorInterface<TensorTraits<DT, CPU>>

template<typename DT , size_t SF, size_t... SR>

using ftl::StaticTensorCpu = typedef TensorInterface<TensorTraits<DT, CPU, SF, SR...>>

template<typename Dtype , device DeviceType, size_t... DimSizes>

using ftl::Tensor = typedef TensorInterface<TensorTraits<Dtype, DeviceType, DimSizes...>>