Permutation of indices and storage sequence of a tensor

This permutes tensor dimensions like aperm. However the interface is more flexible since not all dimensions have to given and names can be used instead of numbers.

# S3 method for class 'tensor'
reorder(x,i=NULL,...,by=NULL)
<!-- %reorder.tidx(x,i,...) -->

Arguments

x

the tensor

i

numeric or character giving dimensions intended to come first

...

further arguments to other instances of the generic function

by

the complement of i, if i is not given

Value

reorder.tensor returns a tensor equal to x but stored with a different sequence of dimensions.

Details

the remaining dimensions keep their relative sequence and follow at the end of the dimension attribute.

See also

to.tensor

Author

K.Gerald v.d. Boogaart

Examples

A <- to.tensor(1:20,c(A=2,B=2,C=5))
A
#> , , 1
#> 
#>       B
#> A      [,1] [,2]
#>   [1,]    1    3
#>   [2,]    2    4
#> 
#> , , 2
#> 
#>       B
#> A      [,1] [,2]
#>   [1,]    5    7
#>   [2,]    6    8
#> 
#> , , 3
#> 
#>       B
#> A      [,1] [,2]
#>   [1,]    9   11
#>   [2,]   10   12
#> 
#> , , 4
#> 
#>       B
#> A      [,1] [,2]
#>   [1,]   13   15
#>   [2,]   14   16
#> 
#> , , 5
#> 
#>       B
#> A      [,1] [,2]
#>   [1,]   17   19
#>   [2,]   18   20
#> 
#> attr(,"class")
#> [1] "tensor"
reorder(A,"C")
#> , , 1
#> 
#>       A
#> C      [,1] [,2]
#>   [1,]    1    2
#>   [2,]    5    6
#>   [3,]    9   10
#>   [4,]   13   14
#>   [5,]   17   18
#> 
#> , , 2
#> 
#>       A
#> C      [,1] [,2]
#>   [1,]    3    4
#>   [2,]    7    8
#>   [3,]   11   12
#>   [4,]   15   16
#>   [5,]   19   20
#> 
#> attr(,"class")
#> [1] "tensor"
reorder(A,"B")
#> , , 1
#> 
#>       A
#> B      [,1] [,2]
#>   [1,]    1    2
#>   [2,]    3    4
#> 
#> , , 2
#> 
#>       A
#> B      [,1] [,2]
#>   [1,]    5    6
#>   [2,]    7    8
#> 
#> , , 3
#> 
#>       A
#> B      [,1] [,2]
#>   [1,]    9   10
#>   [2,]   11   12
#> 
#> , , 4
#> 
#>       A
#> B      [,1] [,2]
#>   [1,]   13   14
#>   [2,]   15   16
#> 
#> , , 5
#> 
#>       A
#> B      [,1] [,2]
#>   [1,]   17   18
#>   [2,]   19   20
#> 
#> attr(,"class")
#> [1] "tensor"