flipdim

Arguments

x, y

vectors, matrices, or hypermatrices of any regular data type. y gets the sizes of x.

dim

positive integer : index of the dimension / direction of x along which the order of x components must be inverted.

blockSize

a positive integer, sub-multiple of size(x,dim): number of rows, of columns, of pages etc in each block. Default value = 1

Description

flipdim(x, 1) inverts the order of rows in the matrix or hypermatrix x.

flipdim(x, 2) inverts the order of columns of x.

flipdim(x, 3) inverts the order of pages in the hypermatrix x. Etc.

The optional parameter blockSize allows splitting x in size(x,1)/blockSize blocks of blockSize rows (dim=1), or in size(x,2)/blockSize blocks of blockSize columns (dim=2), etc (dim>2) and to invert their order. In each block, the order of components (rows, columns, pages etc) is unchanged.

Examples

// Example 1: flip x rows (= components along the first dimension)
x = [1 2 3 4; 5 6 7 8]
y = flipdim(x, 1)

// Example 2: flip x columns (= components along the second dimension)
y = flipdim(x, 2)

// Example 3: flip x pages (= components along the third dimension)
x = matrix(1:24, [3 2 4])
y = flipdim(x, 3)

// Example 4: the first example with complex
x = [1+%i 2*%i 3 4; 5 6-%i 7 8*%pi*%i]
y = flipdim(x, 1)

// Integer-encoded numbers:
x = int16(grand(4, 3, 2, "uin", -9, 9))
y = flipdim(x, 1)

// Booleans:
x = (grand(3, 4, "uin", -9, 9) > 0)
y = flipdim(x, 2)

// Texts:
x = matrix(strsplit("a":"x", 1:23), 4, 6);
x = x+x
flipdim(x, 2)

// Polynomials:
x = inv_coeff(grand(3, 9, "uin", 0, 3), 2)
flipdim(x, 1)

// Rationals:
n = inv_coeff(grand(3, 9, "uin", 0, 3), 2);
d = inv_coeff(grand(3, 9, "uin", 0, 3), 2);
r = n./d
flipdim(r, 2)

Examples using blockSize :

X = [0 1 2 3 4 5 6 7 8 9 10 11];
flipdim(X, 2, 2) // => [10 11   8 9   6 7   4 5   2 3   0 1] // Block size = 2.
flipdim(X, 2, 3) // => [9 10 11   6 7 8   3 4 5   0 1 2]
flipdim(X, 2, 4) // => [8 9 10 11   4 5 6 7   0 1 2 3]
flipdim(X, 2, 6) // => [6 7 8 9 10 11   0 1 2 3 4 5]

// Error if blockSize does not divide the targeted dimension of x.
y = flipdim(x, 2, 5); // size(X) = [1 12] and blockSize=5 does not divide 12.

Example of results:

--> x
 x  =
 -5 -2  0  9
  0 -7 -6  9
 -1 -8 -7  8

--> flipdim(x, 1)
 ans  =
 -1 -8 -7  8
  0 -7 -6  9
 -5 -2  0  9

--> flipdim(x, 2)
 ans  =
  9  0 -2 -5
  9 -6 -7  0
  8 -7 -8 -1

--> x
 x  =
(:,:,1)
  9  4 -3
 -4 -8 -3

(:,:,2)
  5  8  9
  4  4  9

--> flipdim(x, 3)
 ans  =
(:,:,1)
  5  8  9
  4  4  9

(:,:,2)
  9  4 -3
 -4 -8 -3

--> x
 x  =
 -2  3 -5  9 -4 -8
  2  8  4 -9  6 -6
 -9  8  3  4 -3  4

--> flipdim(x, 2, 2)
 ans  =
 -4 -8 -5  9 -2  3
  6 -6  4 -9  2  8
 -3  4  3  4 -9  8

--> flipdim(x, 2, 3)
 ans  =
  9 -4 -8 -2  3 -5
 -9  6 -6  2  8  4
  4 -3  4 -9  8  3

See also

• fftshift — rearranges the fft output, moving the zero frequency to the center of the spectrum
• ifftshift — inverse of fftshift
• colon — (:) colon operator

History