Tensor class with arbitrary dimensions. More...

#include <tensor.h>

Detailed Description

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>
class o2scl::tensor< data_t, vec_t, vec_size_t >

The elements of a tensor are typically specified as a list of size_t numbers with length equal to the tensor rank. For a rank-4 tensor named t, the element t[1][2][0][3] can be obtained with something similar to

size_t ix[4]={1,2,0,3};

double x=t.get(ix);

Empty tensors have zero rank.

The type vec_t can be any vector type with operator[], size() and resize() methods. The type vec_size_t can be any integer-like vector type with operator[], size() and resize() methods.

For I/O with tensors, see o2scl_hdf::hdf_file::setd_ten() and o2scl_hdf::hdf_file::getd_ten() . See also the discussion in the sections Tensors and I/O and contiguous storage of the user's guide.

The storage pattern is a generalization of row-major order. In the case of a 4-rank tensor, the location of a generic element is

$\left( \left( i_0 s_1 + i_1 \right) s_2 + i_2 \right) s_3 + i_3 \, .$

In this case the distance between two elements $(i_0,i_1, i_2,i_3)$ and $ (i_0+1,i_1,i_2,i_3) $ is $ s_1 s_2 s_3 $ , the distance between two elements $(i_0,i_1,i_2, i_3)$ and $ (i_0,i_1+1,i_2,i_3) $ is $ s_2 s_3 $ , and the elements $(i_0,i_1,i_2,i_3)$ and $ (i_0,i_1,i_2,i_3+1) $ are adjacent.

Note: Slices of tensors are subsets obtained from fixing the index of several dimensions, while letting other dimensions vary. For a slice with one dimension not fixed, see vector_slice(). The o2scl::tensor::vector_slice() function should clearly work for uBlas vectors, and seems to work with std::vector objects also, but latter use has not been fully tested.

Idea for Future:: Create an operator[] for tensor and not just tensor1?

Idea for Future:: Could implement arithmetic operators + and - and some different products.

Idea for Future:: Implement copies to and from vector and matrices

Idea for Future:: Implement tensor contractions, i.e. tensor = tensor * tensor

Idea for Future:: Could be interesting to write an iterator for this class.

Definition at line 226 of file tensor.h.

Public Member Functions
	tensor ()
	Create an empty tensor with zero rank.

template<class size_vec_t >
	tensor (size_t rank, const size_vec_t &dim)
	Create a tensor of rank `rank` with sizes given in `dim`. More...

Method to check for valid object
void	is_valid () const
	Check that the o2scl::tensor object is valid.

Copy constructors
	tensor (const tensor< data_t, vec_t, vec_size_t > &t)
	Copy using `operator()`

tensor< data_t, vec_t, vec_size_t > &	operator= (const tensor< data_t, vec_t, vec_size_t > &t)
	Copy using `operator=()`

Clear method
void	clear ()
	Clear the tensor of all data and free allocated memory.

Set functions
template<class size_vec_t >
void	set (const size_vec_t &index, data_t val)
	Set the element indexed by `index` to value `val`.

void	set_all (data_t x)
	Set all elements in a tensor to some fixed value.

void	swap_data (vec_t &dat)
	Swap the data vector.

Slice function
template<class size_vec_t >
ubvector_slice	vector_slice (size_t ix, const size_vec_t &index)
	Fix all but one index to create a vector. More...

Resize method
template<class size_vec_t >
void	resize (size_t rank, const size_vec_t &dim)
	Resize the tensor to rank `rank` with sizes given in `dim`. More...

Size functions
size_t	get_rank () const
	Return the rank of the tensor.

size_t	get_size (size_t i) const
	Returns the size of the ith index.

const vec_size_t &	get_size_arr () const
	Return the full vector of sizes.

const vec_t &	get_data () const
	Return the full data vector.

size_t	total_size () const
	Returns the size of the tensor (the product of the sizes over every index)

Index manipulation
template<class size_vec_t >
size_t	pack_indices (const size_vec_t &index)
	Pack the indices into a single vector index.

template<class size_vec_t >
void	unpack_index (size_t ix, size_vec_t &index)
	Unpack the single vector index into indices.

Minimum, maximum, and sum
data_t	min_value ()
	Compute the minimum value in the tensor.

void	min_index (vec_size_t &index)
	Compute the index of the minimum value in the tensor.

void	min (vec_size_t &index, data_t &val)
	Compute the index of the minimum value in the tensor and return the minimum.

data_t	max_value ()
	Compute the maximum value in the tensor.

void	max_index (vec_size_t &index)
	Compute the index of the maximum value in the tensor.

void	max (vec_size_t &index, data_t &val)
	Compute the index and value of the maximum value in the tensor and return the maximum.

void	minmax_value (data_t &min, data_t &max)
	Compute the minimum and maximum values in the tensor.

void	minmax_index (vec_size_t &index_min, vec_size_t &index_max)
	Compute the indices of the minimum and maximum values in the tensor.

void	minmax (vec_size_t &index, size_t &index_min, data_t &min, size_t &index_max, data_t &max)
	Compute the indices and values of the maximum and minimum in the tensor.

double	total_sum () const
	Return the sum over every element in the tensor.

Slicing and converting to table3d objects
void	convert_table3d_sum (size_t ix_x, size_t ix_y, table3d &tab, std::string x_name="x", std::string y_name="y", std::string slice_name="z")
	Convert to a o2scl::table3d object by summing over all but two indices.

tensor< data_t >	rearrange_and_copy (std::vector< index_spec > spec, int verbose=0, bool err_on_fail=true)
	Rearrange, sum and copy current tensor to a new tensor. More...

Protected Attributes
vec_t	data
	The data.

vec_size_t	size
	A rank-sized vector of the sizes of each dimension.

size_t	rk
	Rank.

Get functions
typedef boost::numeric::ublas::vector_slice< boost::numeric::ublas::vector< data_t > >	ubvector_slice

typedef boost::numeric::ublas::slice	slice

template<class size_vec_t >
data_t &	get (const size_vec_t &index)
	Get the element indexed by `index`.

template<class size_vec_t >
const data_t &	get (const size_vec_t &index) const
	Get a const reference to the element indexed by `index`.

Constructor & Destructor Documentation

◆ tensor()

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>

template<class size_vec_t >

o2scl::tensor< data_t, vec_t, vec_size_t >::tensor	(	size_t	rank,
		const size_vec_t &	dim
	)

inline

The parameter dim must be a pointer to a vector of sizes with length rank. If the user requests any of the sizes to be zero, this constructor will call the error handler, create an empty tensor, and will allocate no memory.

Definition at line 260 of file tensor.h.

Member Function Documentation

◆ rearrange_and_copy()

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>

tensor<data_t> o2scl::tensor< data_t, vec_t, vec_size_t >::rearrange_and_copy	(	std::vector< index_spec >	spec,
		int	verbose = `0`,
		bool	err_on_fail = `true`
	)

inline

Idea for Future:: Return a scalar if possible as a rank 1 tensor with 1 element.

Definition at line 816 of file tensor.h.

◆ resize()

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>

template<class size_vec_t >

void o2scl::tensor< data_t, vec_t, vec_size_t >::resize	(	size_t	rank,
		const size_vec_t &	dim
	)

inline

The parameter dim must be a vector of sizes with a length equal to rank. This resize method is always destructive.

If the user requests any of the sizes to be zero, this function will call the error handler.

Definition at line 549 of file tensor.h.

◆ vector_slice()

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>

template<class size_vec_t >

ubvector_slice o2scl::tensor< data_t, vec_t, vec_size_t >::vector_slice	(	size_t	ix,
		const size_vec_t &	index
	)

inline

This fixes all of the indices to the values given in index except for the index number ix, and returns the corresponding vector, whose length is equal to the size of the tensor in that index. The value index[ix] is ignored.

For example, for a rank 3 tensor allocated with

tensor t;
size_t dim[3]={3,4,5};
t.resize(3,dim);

the following code

size_t index[3]={1,0,3};

ubvector_view v=t.vector_slice(1,index);

Gives a vector v of length 4 which refers to the values t(1,0,3), t(1,1,3), t(1,2,3), and t(1,3,3).

Definition at line 504 of file tensor.h.

The documentation for this class was generated from the following file:

/home/awsteiner/wcs/o2scl/src/base/tensor.h

Detailed Description

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>> class o2scl::tensor< data_t, vec_t, vec_size_t >

Public Member Functions

Protected Attributes

Get functions

Constructor & Destructor Documentation

◆ tensor()

Member Function Documentation

◆ rearrange_and_copy()

◆ resize()

◆ vector_slice()

template<class data_t = double, class vec_t = std::vector<data_t>, class vec_size_t = std::vector<size_t>>
class o2scl::tensor< data_t, vec_t, vec_size_t >