📂Julia

Julia's Permutation Dimension Function and Its Application permutedims

Overview

The built-in function permutedims in Julia can be seen as a generalization of transpose matrix and is useful for handling dimensions of multi-dimensional arrays.

Code

Tensor Dimension Transformation

julia> A = reshape(Vector(1:8), (2,2,2))
2×2×2 Array{Int64, 3}:
[:, :, 1] =
 1  3
 2  4

[:, :, 2] =
 5  7
 6  8

julia> B = permutedims(A, (3, 1, 2))
2×2×2 Array{Int64, 3}:
[:, :, 1] =
 1  2
 5  6

[:, :, 2] =
 3  4
 7  8

Tensors are programmatically useful but equally confusing. Particularly in deep learning tasks related to images, this function can be very handy.

Transpose of String Matrix

julia> A = ["a" "b" "c"
            "d" "e" "f"]
2×3 Matrix{String}:
 "a"  "b"  "c"
 "d"  "e"  "f"

julia> A'
ERROR: MethodError: no method matching adjoint(::String)

julia> permutedims(A)
3×2 Matrix{String}:
 "a"  "d"
 "b"  "e"
 "c"  "f"

As you can see, a matrix of strings cannot be transposed using ‘', but the permutedims function can be utilized to achieve functionality equivalent to matrix transpose¹.

Environment

• OS: Windows
• julia: v1.11.1