R6 Tenmat Class

A class representing a matricized tensor, equivalent to MATLAB Tensor Toolbox's tenmat.

Usage

tenmat(T, rdims = NULL, cdims = NULL, tsize = NULL)

Arguments

T

A Tensor object, matrix, or Tenmat object

rdims

Dimensions to map to rows

cdims

Dimensions to map to columns (optional). Also accepts 't', 'fc', 'bc'.

tsize

Tensor size (used internally)

Value

A new Tenmat object

Public fields

data

The underlying 2D matrix data

rdims

The modes mapped to the rows

cdims

The modes mapped to the columns

tsize

The size of the original tensor Initialize a new tenmat

Methods

Public methods

• Tenmat$new()

• Tenmat$dim()

• Tenmat$length()

• Tenmat$print()

• Tenmat$show()

• Tenmat$clone_tenmat()

• Tenmat$add()

• Tenmat$subtract()

• Tenmat$negate()

• Tenmat$transpose()

• Tenmat$clone()

---

Tenmat$new()

Usage

Tenmat$new(data = NULL, rdims = NULL, cdims = NULL, tsize = NULL)

Arguments

data

The matrix representation

rdims

The row indices

cdims

The column indices

tsize

The size of the original tensor

Returns

A new Tenmat object Get dimension of the underlying matrix

---

Tenmat$dim()

Usage

Tenmat$dim()

Returns

Integer vector of dimensions Get number of elements

---

Tenmat$length()

Usage

Tenmat$length()

Returns

Integer number of elements Print tenmat information

---

Tenmat$print()

Usage

Tenmat$print()

Returns

Invisible self Show tenmat (alias for print)

---

Tenmat$show()

Usage

Tenmat$show()

Returns

Invisible self Clone the tenmat

---

Tenmat$clone_tenmat()

Usage

Tenmat$clone_tenmat()

Returns

A new Tenmat object Add another Tenmat

---

Tenmat$add()

Usage

Tenmat$add(other)

Arguments

other

Another Tenmat object

Returns

A new Tenmat object Subtract another Tenmat

---

Tenmat$subtract()

Usage

Tenmat$subtract(other)

Arguments

other

Another Tenmat object

Returns

A new Tenmat object Negate Tenmat

---

Tenmat$negate()

Usage

Tenmat$negate()

Returns

A new Tenmat object Transpose Tenmat

---

Tenmat$transpose()

Usage

Tenmat$transpose()

Returns

A new Tenmat object

---

Tenmat$clone()

The objects of this class are cloneable with this method.

Usage

Tenmat$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

t <- tensor(array(1:24, dim = c(3, 4, 2)))
# Unfold mode 1 to rows, modes 2 and 3 to columns
m <- tenmat(t, rdims = 1)
print(m)
#> <Tenmat object>
#> A matrix corresponding to a tensor of size 3 x 4 x 2 
#> rindices = [ 1 ] (modes of tensor corresponding to rows)
#> cindices = [ 2 3 ] (modes of tensor corresponding to columns)
#> data =
#>      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
#> [1,]    1    4    7   10   13   16   19   22
#> [2,]    2    5    8   11   14   17   20   23
#> [3,]    3    6    9   12   15   18   21   24

# Unfold mode 2 to rows, and mode 3 and 1 to columns (forward cyclic)
m2 <- tenmat(t, rdims = 2, cdims = "fc")
print(m2)
#> <Tenmat object>
#> A matrix corresponding to a tensor of size 3 x 4 x 2 
#> rindices = [ 2 ] (modes of tensor corresponding to rows)
#> cindices = [ 3 1 ] (modes of tensor corresponding to columns)
#> data =
#>      [,1] [,2] [,3] [,4] [,5] [,6]
#> [1,]    1   13    2   14    3   15
#> [2,]    4   16    5   17    6   18
#> [3,]    7   19    8   20    9   21
#> [4,]   10   22   11   23   12   24