<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- Created by GNU Texinfo 6.5, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Rearranging Matrices (GNU Octave (version 5.1.0))</title>
<meta name="description" content="Rearranging Matrices (GNU Octave (version 5.1.0))">
<meta name="keywords" content="Rearranging Matrices (GNU Octave (version 5.1.0))">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html#Top" rel="start" title="Top">
<link href="Concept-Index.html#Concept-Index" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Matrix-Manipulation.html#Matrix-Manipulation" rel="up" title="Matrix Manipulation">
<link href="Special-Utility-Matrices.html#Special-Utility-Matrices" rel="next" title="Special Utility Matrices">
<link href="Finding-Elements-and-Checking-Conditions.html#Finding-Elements-and-Checking-Conditions" rel="prev" title="Finding Elements and Checking Conditions">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
blockquote.smallindentedblock {margin-right: 0em; font-size: smaller}
blockquote.smallquotation {font-size: smaller}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
div.smalldisplay {margin-left: 3.2em}
div.smallexample {margin-left: 3.2em}
div.smalllisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
pre.smalldisplay {font-family: inherit; font-size: smaller}
pre.smallexample {font-size: smaller}
pre.smallformat {font-family: inherit; font-size: smaller}
pre.smalllisp {font-size: smaller}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
<link rel="stylesheet" type="text/css" href="octave.css">
</head>
<body lang="en">
<a name="Rearranging-Matrices"></a>
<div class="header">
<p>
Next: <a href="Special-Utility-Matrices.html#Special-Utility-Matrices" accesskey="n" rel="next">Special Utility Matrices</a>, Previous: <a href="Finding-Elements-and-Checking-Conditions.html#Finding-Elements-and-Checking-Conditions" accesskey="p" rel="prev">Finding Elements and Checking Conditions</a>, Up: <a href="Matrix-Manipulation.html#Matrix-Manipulation" accesskey="u" rel="up">Matrix Manipulation</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<a name="Rearranging-Matrices-1"></a>
<h3 class="section">16.2 Rearranging Matrices</h3>
<a name="XREFfliplr"></a><dl>
<dt><a name="index-fliplr"></a><em></em> <strong>fliplr</strong> <em>(<var>x</var>)</em></dt>
<dd><p>Flip array left to right.
</p>
<p>Return a copy of <var>x</var> with the order of the columns reversed. In other
words, <var>x</var> is flipped left-to-right about a vertical axis. For example:
</p>
<div class="example">
<pre class="example">fliplr ([1, 2; 3, 4])
⇒ 2 1
4 3
</pre></div>
<p><strong>See also:</strong> <a href="#XREFflipud">flipud</a>, <a href="#XREFflip">flip</a>, <a href="#XREFrot90">rot90</a>, <a href="#XREFrotdim">rotdim</a>.
</p></dd></dl>
<a name="XREFflipud"></a><dl>
<dt><a name="index-flipud"></a><em></em> <strong>flipud</strong> <em>(<var>x</var>)</em></dt>
<dd><p>Flip array upside down.
</p>
<p>Return a copy of <var>x</var> with the order of the rows reversed. In other
words, <var>x</var> is flipped upside-down about a horizontal axis. For example:
</p>
<div class="example">
<pre class="example">flipud ([1, 2; 3, 4])
⇒ 3 4
1 2
</pre></div>
<p><strong>See also:</strong> <a href="#XREFfliplr">fliplr</a>, <a href="#XREFflip">flip</a>, <a href="#XREFrot90">rot90</a>, <a href="#XREFrotdim">rotdim</a>.
</p></dd></dl>
<a name="XREFflip"></a><dl>
<dt><a name="index-flip"></a><em></em> <strong>flip</strong> <em>(<var>x</var>)</em></dt>
<dt><a name="index-flip-1"></a><em></em> <strong>flip</strong> <em>(<var>x</var>, <var>dim</var>)</em></dt>
<dd><p>Return a copy of array <var>x</var> flipped across dimension <var>dim</var>.
</p>
<p>If <var>dim</var> is unspecified it defaults to the first non-singleton
dimension.
</p>
<p>Examples:
</p>
<div class="example">
<pre class="example">## row vector
flip ([1 2 3 4])
⇒ 4 3 2 1
## column vector
flip ([1; 2; 3; 4])
⇒ 4
3
2
1
## 2-D matrix along dimension 1
flip ([1 2; 3 4])
⇒ 3 4
1 2
## 2-D matrix along dimension 2
flip ([1 2; 3 4], 2)
⇒ 2 1
4 3
</pre></div>
<p><strong>See also:</strong> <a href="#XREFfliplr">fliplr</a>, <a href="#XREFflipud">flipud</a>, <a href="#XREFrot90">rot90</a>, <a href="#XREFrotdim">rotdim</a>, <a href="#XREFpermute">permute</a>, <a href="Arithmetic-Ops.html#XREFtranspose">transpose</a>.
</p></dd></dl>
<a name="XREFrot90"></a><dl>
<dt><a name="index-rot90"></a><em></em> <strong>rot90</strong> <em>(<var>A</var>)</em></dt>
<dt><a name="index-rot90-1"></a><em></em> <strong>rot90</strong> <em>(<var>A</var>, <var>k</var>)</em></dt>
<dd><p>Rotate array by 90 degree increments.
</p>
<p>Return a copy of <var>A</var> with the elements rotated counterclockwise in
90-degree increments.
</p>
<p>The second argument is optional, and specifies how many 90-degree rotations
are to be applied (the default value is 1). Negative values of <var>k</var>
rotate the matrix in a clockwise direction.
For example,
</p>
<div class="example">
<pre class="example">rot90 ([1, 2; 3, 4], -1)
⇒ 3 1
4 2
</pre></div>
<p>rotates the given matrix clockwise by 90 degrees. The following are all
equivalent statements:
</p>
<div class="example">
<pre class="example">rot90 ([1, 2; 3, 4], -1)
rot90 ([1, 2; 3, 4], 3)
rot90 ([1, 2; 3, 4], 7)
</pre></div>
<p>The rotation is always performed on the plane of the first two dimensions,
i.e., rows and columns. To perform a rotation on any other plane, use
<code>rotdim</code>.
</p>
<p><strong>See also:</strong> <a href="#XREFrotdim">rotdim</a>, <a href="#XREFfliplr">fliplr</a>, <a href="#XREFflipud">flipud</a>, <a href="#XREFflip">flip</a>.
</p></dd></dl>
<a name="XREFrotdim"></a><dl>
<dt><a name="index-rotdim"></a><em></em> <strong>rotdim</strong> <em>(<var>x</var>)</em></dt>
<dt><a name="index-rotdim-1"></a><em></em> <strong>rotdim</strong> <em>(<var>x</var>, <var>n</var>)</em></dt>
<dt><a name="index-rotdim-2"></a><em></em> <strong>rotdim</strong> <em>(<var>x</var>, <var>n</var>, <var>plane</var>)</em></dt>
<dd><p>Return a copy of <var>x</var> with the elements rotated counterclockwise in
90-degree increments.
</p>
<p>The second argument <var>n</var> is optional, and specifies how many 90-degree
rotations are to be applied (the default value is 1). Negative values of
<var>n</var> rotate the matrix in a clockwise direction.
</p>
<p>The third argument is also optional and defines the plane of the rotation.
If present, <var>plane</var> is a two element vector containing two different
valid dimensions of the matrix. When <var>plane</var> is not given the first two
non-singleton dimensions are used.
</p>
<p>For example,
</p>
<div class="example">
<pre class="example">rotdim ([1, 2; 3, 4], -1, [1, 2])
⇒ 3 1
4 2
</pre></div>
<p>rotates the given matrix clockwise by 90 degrees. The following are all
equivalent statements:
</p>
<div class="example">
<pre class="example">rotdim ([1, 2; 3, 4], -1, [1, 2])
rotdim ([1, 2; 3, 4], 3, [1, 2])
rotdim ([1, 2; 3, 4], 7, [1, 2])
</pre></div>
<p><strong>See also:</strong> <a href="#XREFrot90">rot90</a>, <a href="#XREFfliplr">fliplr</a>, <a href="#XREFflipud">flipud</a>, <a href="#XREFflip">flip</a>.
</p></dd></dl>
<a name="XREFcat"></a><dl>
<dt><a name="index-cat"></a><em></em> <strong>cat</strong> <em>(<var>dim</var>, <var>array1</var>, <var>array2</var>, …, <var>arrayN</var>)</em></dt>
<dd><p>Return the concatenation of N-D array objects, <var>array1</var>,
<var>array2</var>, …, <var>arrayN</var> along dimension <var>dim</var>.
</p>
<div class="example">
<pre class="example">A = ones (2, 2);
B = zeros (2, 2);
cat (2, A, B)
⇒ 1 1 0 0
1 1 0 0
</pre></div>
<p>Alternatively, we can concatenate <var>A</var> and <var>B</var> along the
second dimension in the following way:
</p>
<div class="example">
<pre class="example">[A, B]
</pre></div>
<p><var>dim</var> can be larger than the dimensions of the N-D array objects
and the result will thus have <var>dim</var> dimensions as the
following example shows:
</p>
<div class="example">
<pre class="example">cat (4, ones (2, 2), zeros (2, 2))
⇒ ans(:,:,1,1) =
1 1
1 1
ans(:,:,1,2) =
0 0
0 0
</pre></div>
<p><strong>See also:</strong> <a href="#XREFhorzcat">horzcat</a>, <a href="#XREFvertcat">vertcat</a>.
</p></dd></dl>
<a name="XREFhorzcat"></a><dl>
<dt><a name="index-horzcat"></a><em></em> <strong>horzcat</strong> <em>(<var>array1</var>, <var>array2</var>, …, <var>arrayN</var>)</em></dt>
<dd><p>Return the horizontal concatenation of N-D array objects, <var>array1</var>,
<var>array2</var>, …, <var>arrayN</var> along dimension 2.
</p>
<p>Arrays may also be concatenated horizontally using the syntax for creating
new matrices. For example:
</p>
<div class="example">
<pre class="example"><var>hcat</var> = [ <var>array1</var>, <var>array2</var>, … ]
</pre></div>
<p><strong>See also:</strong> <a href="#XREFcat">cat</a>, <a href="#XREFvertcat">vertcat</a>.
</p></dd></dl>
<a name="XREFvertcat"></a><dl>
<dt><a name="index-vertcat"></a><em></em> <strong>vertcat</strong> <em>(<var>array1</var>, <var>array2</var>, …, <var>arrayN</var>)</em></dt>
<dd><p>Return the vertical concatenation of N-D array objects, <var>array1</var>,
<var>array2</var>, …, <var>arrayN</var> along dimension 1.
</p>
<p>Arrays may also be concatenated vertically using the syntax for creating
new matrices. For example:
</p>
<div class="example">
<pre class="example"><var>vcat</var> = [ <var>array1</var>; <var>array2</var>; … ]
</pre></div>
<p><strong>See also:</strong> <a href="#XREFcat">cat</a>, <a href="#XREFhorzcat">horzcat</a>.
</p></dd></dl>
<a name="XREFpermute"></a><dl>
<dt><a name="index-permute"></a><em></em> <strong>permute</strong> <em>(<var>A</var>, <var>perm</var>)</em></dt>
<dd><p>Return the generalized transpose for an N-D array object <var>A</var>.
</p>
<p>The permutation vector <var>perm</var> must contain the elements
<code>1:ndims (A)</code><!-- /@w --> (in any order, but each element must appear only
once). The <var>N</var>th dimension of <var>A</var> gets remapped to dimension
<code><var>PERM</var>(<var>N</var>)</code>. For example:
</p>
<div class="example">
<pre class="example"><var>x</var> = zeros ([2, 3, 5, 7]);
size (<var>x</var>)
⇒ 2 3 5 7
size (permute (<var>x</var>, [2, 1, 3, 4]))
⇒ 3 2 5 7
size (permute (<var>x</var>, [1, 3, 4, 2]))
⇒ 2 5 7 3
## The identity permutation
size (permute (<var>x</var>, [1, 2, 3, 4]))
⇒ 2 3 5 7
</pre></div>
<p><strong>See also:</strong> <a href="#XREFipermute">ipermute</a>.
</p></dd></dl>
<a name="XREFipermute"></a><dl>
<dt><a name="index-ipermute"></a><em></em> <strong>ipermute</strong> <em>(<var>A</var>, <var>iperm</var>)</em></dt>
<dd><p>The inverse of the <code>permute</code> function.
</p>
<p>The expression
</p>
<div class="example">
<pre class="example">ipermute (permute (A, perm), perm)
</pre></div>
<p>returns the original array <var>A</var>.
</p>
<p><strong>See also:</strong> <a href="#XREFpermute">permute</a>.
</p></dd></dl>
<a name="XREFreshape"></a><dl>
<dt><a name="index-reshape"></a><em></em> <strong>reshape</strong> <em>(<var>A</var>, <var>m</var>, <var>n</var>, …)</em></dt>
<dt><a name="index-reshape-1"></a><em></em> <strong>reshape</strong> <em>(<var>A</var>, [<var>m</var> <var>n</var> …])</em></dt>
<dt><a name="index-reshape-2"></a><em></em> <strong>reshape</strong> <em>(<var>A</var>, …, [], …)</em></dt>
<dt><a name="index-reshape-3"></a><em></em> <strong>reshape</strong> <em>(<var>A</var>, <var>size</var>)</em></dt>
<dd><p>Return a matrix with the specified dimensions (<var>m</var>, <var>n</var>, …)
whose elements are taken from the matrix <var>A</var>.
</p>
<p>The elements of the matrix are accessed in column-major order (like Fortran
arrays are stored).
</p>
<p>The following code demonstrates reshaping a 1x4 row vector into a 2x2 square
matrix.
</p>
<div class="example">
<pre class="example">reshape ([1, 2, 3, 4], 2, 2)
⇒ 1 3
2 4
</pre></div>
<p>Note that the total number of elements in the original matrix
(<code>prod (size (<var>A</var>))</code>) must match the total number of elements
in the new matrix (<code>prod ([<var>m</var> <var>n</var> …])</code>).
</p>
<p>A single dimension of the return matrix may be left unspecified and Octave
will determine its size automatically. An empty matrix ([]) is used to flag
the unspecified dimension.
</p>
<p><strong>See also:</strong> <a href="#XREFresize">resize</a>, <a href="#XREFvec">vec</a>, <a href="#XREFpostpad">postpad</a>, <a href="#XREFcat">cat</a>, <a href="Object-Sizes.html#XREFsqueeze">squeeze</a>.
</p></dd></dl>
<a name="XREFresize"></a><dl>
<dt><a name="index-resize"></a><em></em> <strong>resize</strong> <em>(<var>x</var>, <var>m</var>)</em></dt>
<dt><a name="index-resize-1"></a><em></em> <strong>resize</strong> <em>(<var>x</var>, <var>m</var>, <var>n</var>, …)</em></dt>
<dt><a name="index-resize-2"></a><em></em> <strong>resize</strong> <em>(<var>x</var>, [<var>m</var> <var>n</var> …])</em></dt>
<dd><p>Resize <var>x</var> cutting off elements as necessary.
</p>
<p>In the result, element with certain indices is equal to the corresponding
element of <var>x</var> if the indices are within the bounds of <var>x</var>;
otherwise, the element is set to zero.
</p>
<p>In other words, the statement
</p>
<div class="example">
<pre class="example">y = resize (x, dv)
</pre></div>
<p>is equivalent to the following code:
</p>
<div class="example">
<pre class="example">y = zeros (dv, class (x));
sz = min (dv, size (x));
for i = 1:length (sz)
idx{i} = 1:sz(i);
endfor
y(idx{:}) = x(idx{:});
</pre></div>
<p>but is performed more efficiently.
</p>
<p>If only <var>m</var> is supplied, and it is a scalar, the dimension of the
result is <var>m</var>-by-<var>m</var>.
If <var>m</var>, <var>n</var>, … are all scalars, then the dimensions of
the result are <var>m</var>-by-<var>n</var>-by-….
If given a vector as input, then the
dimensions of the result are given by the elements of that vector.
</p>
<p>An object can be resized to more dimensions than it has;
in such case the missing dimensions are assumed to be 1.
Resizing an object to fewer dimensions is not possible.
</p>
<p><strong>See also:</strong> <a href="#XREFreshape">reshape</a>, <a href="#XREFpostpad">postpad</a>, <a href="#XREFprepad">prepad</a>, <a href="#XREFcat">cat</a>.
</p></dd></dl>
<a name="XREFcircshift"></a><dl>
<dt><a name="index-circshift"></a><em><var>y</var> =</em> <strong>circshift</strong> <em>(<var>x</var>, <var>n</var>)</em></dt>
<dt><a name="index-circshift-1"></a><em><var>y</var> =</em> <strong>circshift</strong> <em>(<var>x</var>, <var>n</var>, <var>dim</var>)</em></dt>
<dd><p>Circularly shift the values of the array <var>x</var>.
</p>
<p><var>n</var> must be a vector of integers no longer than the number of dimensions
in <var>x</var>. The values of <var>n</var> can be either positive or negative, which
determines the direction in which the values of <var>x</var> are shifted. If an
element of <var>n</var> is zero, then the corresponding dimension of <var>x</var> will
not be shifted. If <var>n</var> is a scalar and no <var>dim</var> is specified then
the shift is applied to the first non-singular dimension.
</p>
<p>If a scalar <var>dim</var> is given then operate along the specified dimension.
In this case <var>n</var> must be a scalar as well.
</p>
<p>Examples:
</p>
<div class="example">
<pre class="example">x = [1, 2, 3;
4, 5, 6;
7, 8, 9];
## positive shift on rows (1st non-singular dim)
circshift (x, 1)
⇒
7 8 9
1 2 3
4 5 6
## negative shift on rows (1st non-singular dim)
circshift (x, -2)
⇒
7 8 9
1 2 3
4 5 6
## no shift of rows, shift columns by 1 (2nd dimension)
circshift (x, [0,1])
⇒
3 1 2
6 4 5
9 7 8
## shift columns (2nd dimension)
circshift (x, 1, 2)
⇒
3 1 2
6 4 5
9 7 8
</pre></div>
<p><strong>See also:</strong> <a href="#XREFpermute">permute</a>, <a href="#XREFipermute">ipermute</a>, <a href="#XREFshiftdim">shiftdim</a>.
</p></dd></dl>
<a name="XREFshift"></a><dl>
<dt><a name="index-shift"></a><em></em> <strong>shift</strong> <em>(<var>x</var>, <var>b</var>)</em></dt>
<dt><a name="index-shift-1"></a><em></em> <strong>shift</strong> <em>(<var>x</var>, <var>b</var>, <var>dim</var>)</em></dt>
<dd><p>If <var>x</var> is a vector, perform a circular shift of length <var>b</var> of
the elements of <var>x</var>.
</p>
<p>If <var>x</var> is a matrix, do the same for each column of <var>x</var>.
</p>
<p>If the optional <var>dim</var> argument is given, operate along this dimension.
</p></dd></dl>
<a name="XREFshiftdim"></a><dl>
<dt><a name="index-shiftdim"></a><em><var>y</var> =</em> <strong>shiftdim</strong> <em>(<var>x</var>, <var>n</var>)</em></dt>
<dt><a name="index-shiftdim-1"></a><em>[<var>y</var>, <var>ns</var>] =</em> <strong>shiftdim</strong> <em>(<var>x</var>)</em></dt>
<dd><p>Shift the dimensions of <var>x</var> by <var>n</var>, where <var>n</var> must be
an integer scalar.
</p>
<p>When <var>n</var> is positive, the dimensions of <var>x</var> are shifted to the left,
with the leading dimensions circulated to the end. If <var>n</var> is negative,
then the dimensions of <var>x</var> are shifted to the right, with <var>n</var>
leading singleton dimensions added.
</p>
<p>Called with a single argument, <code>shiftdim</code>, removes the leading
singleton dimensions, returning the number of dimensions removed in the
second output argument <var>ns</var>.
</p>
<p>For example:
</p>
<div class="example">
<pre class="example">x = ones (1, 2, 3);
size (shiftdim (x, -1))
⇒ 1 1 2 3
size (shiftdim (x, 1))
⇒ 2 3
[b, ns] = shiftdim (x)
⇒ b =
1 1 1
1 1 1
⇒ ns = 1
</pre></div>
<p><strong>See also:</strong> <a href="#XREFreshape">reshape</a>, <a href="#XREFpermute">permute</a>, <a href="#XREFipermute">ipermute</a>, <a href="#XREFcircshift">circshift</a>, <a href="Object-Sizes.html#XREFsqueeze">squeeze</a>.
</p></dd></dl>
<a name="XREFsort"></a><dl>
<dt><a name="index-sort"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sort</strong> <em>(<var>x</var>)</em></dt>
<dt><a name="index-sort-1"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sort</strong> <em>(<var>x</var>, <var>dim</var>)</em></dt>
<dt><a name="index-sort-2"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sort</strong> <em>(<var>x</var>, <var>mode</var>)</em></dt>
<dt><a name="index-sort-3"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sort</strong> <em>(<var>x</var>, <var>dim</var>, <var>mode</var>)</em></dt>
<dd><p>Return a copy of <var>x</var> with the elements arranged in increasing order.
</p>
<p>For matrices, <code>sort</code> orders the elements within columns
</p>
<p>For example:
</p>
<div class="example">
<pre class="example">sort ([1, 2; 2, 3; 3, 1])
⇒ 1 1
2 2
3 3
</pre></div>
<p>If the optional argument <var>dim</var> is given, then the matrix is sorted
along the dimension defined by <var>dim</var>. The optional argument <var>mode</var>
defines the order in which the values will be sorted. Valid values of
<var>mode</var> are <code>"ascend"</code> or <code>"descend"</code>.
</p>
<p>The <code>sort</code> function may also be used to produce a matrix
containing the original row indices of the elements in the sorted
matrix. For example:
</p>
<div class="example">
<pre class="example">[s, i] = sort ([1, 2; 2, 3; 3, 1])
⇒ s = 1 1
2 2
3 3
⇒ i = 1 3
2 1
3 2
</pre></div>
<p>For equal elements, the indices are such that equal elements are listed
in the order in which they appeared in the original list.
</p>
<p>Sorting of complex entries is done first by magnitude
(<code>abs (<var>z</var>)</code><!-- /@w -->) and for any ties by phase angle
(<code>angle (z)</code><!-- /@w -->). For example:
</p>
<div class="example">
<pre class="example">sort ([1+i; 1; 1-i])
⇒ 1 + 0i
1 - 1i
1 + 1i
</pre></div>
<p>NaN values are treated as being greater than any other value and are sorted
to the end of the list.
</p>
<p>The <code>sort</code> function may also be used to sort strings and cell arrays
of strings, in which case ASCII dictionary order (uppercase ’A’ precedes
lowercase ’a’) of the strings is used.
</p>
<p>The algorithm used in <code>sort</code> is optimized for the sorting of partially
ordered lists.
</p>
<p><strong>See also:</strong> <a href="#XREFsortrows">sortrows</a>, <a href="#XREFissorted">issorted</a>.
</p></dd></dl>
<a name="XREFsortrows"></a><dl>
<dt><a name="index-sortrows"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sortrows</strong> <em>(<var>A</var>)</em></dt>
<dt><a name="index-sortrows-1"></a><em>[<var>s</var>, <var>i</var>] =</em> <strong>sortrows</strong> <em>(<var>A</var>, <var>c</var>)</em></dt>
<dd><p>Sort the rows of the matrix <var>A</var> according to the order of the columns
specified in <var>c</var>.
</p>
<p>By default (<var>c</var> omitted, or a particular column unspecified in <var>c</var>)
an ascending sort order is used. However, if elements of <var>c</var> are
negative then the corresponding column is sorted in descending order. If
the elements of <var>A</var> are strings then a lexicographical sort is used.
</p>
<p>Example: sort by column 2 in descending order, then 3 in ascending order
</p>
<div class="example">
<pre class="example">x = [ 7, 1, 4;
8, 3, 5;
9, 3, 6 ];
sortrows (x, [-2, 3])
⇒ 8 3 5
9 3 6
7 1 4
</pre></div>
<p><strong>See also:</strong> <a href="#XREFsort">sort</a>.
</p></dd></dl>
<a name="XREFissorted"></a><dl>
<dt><a name="index-issorted"></a><em></em> <strong>issorted</strong> <em>(<var>a</var>)</em></dt>
<dt><a name="index-issorted-1"></a><em></em> <strong>issorted</strong> <em>(<var>a</var>, <var>mode</var>)</em></dt>
<dt><a name="index-issorted-2"></a><em></em> <strong>issorted</strong> <em>(<var>a</var>, "rows", <var>mode</var>)</em></dt>
<dd><p>Return true if the array is sorted according to <var>mode</var>, which
may be either <code>"ascending"</code>, <code>"descending"</code>, or
<code>"either"</code>.
</p>
<p>By default, <var>mode</var> is <code>"ascending"</code>. NaNs are treated in the
same manner as <code>sort</code>.
</p>
<p>If the optional argument <code>"rows"</code> is supplied, check whether
the array is sorted by rows as output by the function <code>sortrows</code>
(with no options).
</p>
<p>This function does not support sparse matrices.
</p>
<p><strong>See also:</strong> <a href="#XREFsort">sort</a>, <a href="#XREFsortrows">sortrows</a>.
</p></dd></dl>
<a name="XREFnth_005felement"></a><dl>
<dt><a name="index-nth_005felement"></a><em></em> <strong>nth_element</strong> <em>(<var>x</var>, <var>n</var>)</em></dt>
<dt><a name="index-nth_005felement-1"></a><em></em> <strong>nth_element</strong> <em>(<var>x</var>, <var>n</var>, <var>dim</var>)</em></dt>
<dd><p>Select the n-th smallest element of a vector, using the ordering defined by
<code>sort</code>.
</p>
<p>The result is equivalent to <code>sort(<var>x</var>)(<var>n</var>)</code>.
</p>
<p><var>n</var> can also be a contiguous range, either ascending <code>l:u</code>
or descending <code>u:-1:l</code>, in which case a range of elements is returned.
</p>
<p>If <var>x</var> is an array, <code>nth_element</code> operates along the dimension
defined by <var>dim</var>, or the first non-singleton dimension if <var>dim</var> is
not given.
</p>
<p>Programming Note: nth_element encapsulates the C++ standard library
algorithms nth_element and partial_sort. On average, the complexity of the
operation is O(M*log(K)), where <code>M = size (<var>x</var>, <var>dim</var>)</code><!-- /@w --> and
<code>K = length (<var>n</var>)</code><!-- /@w -->. This function is intended for cases where
the ratio K/M is small; otherwise, it may be better to use <code>sort</code>.
</p>
<p><strong>See also:</strong> <a href="#XREFsort">sort</a>, <a href="Utility-Functions.html#XREFmin">min</a>, <a href="Utility-Functions.html#XREFmax">max</a>.
</p></dd></dl>
<a name="XREFtril"></a><dl>
<dt><a name="index-tril"></a><em><var>A_LO</var> =</em> <strong>tril</strong> <em>(<var>A</var>)</em></dt>
<dt><a name="index-tril-1"></a><em><var>A_LO</var> =</em> <strong>tril</strong> <em>(<var>A</var>, <var>k</var>)</em></dt>
<dt><a name="index-tril-2"></a><em><var>A_LO</var> =</em> <strong>tril</strong> <em>(<var>A</var>, <var>k</var>, <var>pack</var>)</em></dt>
<dd><p>Return a new matrix formed by extracting the lower triangular part of the
matrix <var>A</var>, and setting all other elements to zero.
</p>
<p>The optional second argument specifies how many diagonals above or below the
main diagonal should also be set to zero. The default value of <var>k</var> is
zero which includes the main diagonal as part of the result. If the value of
<var>k</var> is a nonzero integer then the selection of elements starts at an offset
of <var>k</var> diagonals above the main diagonal for positive <var>k</var> or below the
main diagonal for negative <var>k</var>. The absolute value of <var>k</var> may not be
greater than the number of subdiagonals or superdiagonals.
</p>
<p>Example 1 : exclude main diagonal
</p>
<div class="example">
<pre class="example">tril (ones (3), -1)
⇒ 0 0 0
1 0 0
1 1 0
</pre></div>
<p>Example 2 : include first superdiagonal
</p>
<div class="example">
<pre class="example">tril (ones (3), 1)
⇒ 1 1 0
1 1 1
1 1 1
</pre></div>
<p>If the optional third argument <code>"pack"</code> is given then the extracted
elements are not inserted into a matrix, but instead stacked column-wise one
above another, and returned as a column vector.
</p>
<p><strong>See also:</strong> <a href="#XREFtriu">triu</a>, <a href="Predicates-for-Numeric-Objects.html#XREFistril">istril</a>, <a href="#XREFdiag">diag</a>.
</p></dd></dl>
<a name="XREFtriu"></a><dl>
<dt><a name="index-triu"></a><em><var>A_UP</var> =</em> <strong>triu</strong> <em>(<var>A</var>)</em></dt>
<dt><a name="index-triu-1"></a><em><var>A_UP</var> =</em> <strong>triu</strong> <em>(<var>A</var>, <var>k</var>)</em></dt>
<dt><a name="index-triu-2"></a><em><var>A_UP</var> =</em> <strong>triu</strong> <em>(<var>A</var>, <var>k</var>, <var>pack</var>)</em></dt>
<dd><p>Return a new matrix formed by extracting the upper triangular part of the
matrix <var>A</var>, and setting all other elements to zero.
</p>
<p>The optional second argument specifies how many diagonals above or below the
main diagonal should also be set to zero. The default value of <var>k</var> is
zero which includes the main diagonal as part of the result. If the value of
<var>k</var> is a nonzero integer then the selection of elements starts at an offset
of <var>k</var> diagonals above the main diagonal for positive <var>k</var> or below the
main diagonal for negative <var>k</var>. The absolute value of <var>k</var> may not be
greater than the number of subdiagonals or superdiagonals.
</p>
<p>Example 1 : exclude main diagonal
</p>
<div class="example">
<pre class="example">triu (ones (3), 1)
⇒ 0 1 1
0 0 1
0 0 0
</pre></div>
<p>Example 2 : include first subdiagonal
</p>
<div class="example">
<pre class="example">triu (ones (3), -1)
⇒ 1 1 1
1 1 1
0 1 1
</pre></div>
<p>If the optional third argument <code>"pack"</code> is given then the extracted
elements are not inserted into a matrix, but instead stacked column-wise one
above another, and returned as a column vector.
</p>
<p><strong>See also:</strong> <a href="#XREFtril">tril</a>, <a href="Predicates-for-Numeric-Objects.html#XREFistriu">istriu</a>, <a href="#XREFdiag">diag</a>.
</p></dd></dl>
<a name="XREFvec"></a><dl>
<dt><a name="index-vec"></a><em><var>v</var> =</em> <strong>vec</strong> <em>(<var>x</var>)</em></dt>
<dt><a name="index-vec-1"></a><em><var>v</var> =</em> <strong>vec</strong> <em>(<var>x</var>, <var>dim</var>)</em></dt>
<dd><p>Return the vector obtained by stacking the columns of the matrix <var>x</var>
one above the other.
</p>
<p>Without <var>dim</var> this is equivalent to <code><var>x</var>(:)</code>.
</p>
<p>If <var>dim</var> is supplied, the dimensions of <var>v</var> are set to <var>dim</var>
with all elements along the last dimension. This is equivalent to
<code>shiftdim (<var>x</var>(:), 1-<var>dim</var>)</code>.
</p>
<p><strong>See also:</strong> <a href="#XREFvech">vech</a>, <a href="#XREFresize">resize</a>, <a href="#XREFcat">cat</a>.
</p></dd></dl>
<a name="XREFvech"></a><dl>
<dt><a name="index-vech"></a><em></em> <strong>vech</strong> <em>(<var>x</var>)</em></dt>
<dd><p>Return the vector obtained by eliminating all superdiagonal elements of
the square matrix <var>x</var> and stacking the result one column above the
other.
</p>
<p>This has uses in matrix calculus where the underlying matrix is symmetric
and it would be pointless to keep values above the main diagonal.
</p>
<p><strong>See also:</strong> <a href="#XREFvec">vec</a>.
</p></dd></dl>
<a name="XREFprepad"></a><dl>
<dt><a name="index-prepad"></a><em></em> <strong>prepad</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-prepad-1"></a><em></em> <strong>prepad</strong> <em>(<var>x</var>, <var>l</var>, <var>c</var>)</em></dt>
<dt><a name="index-prepad-2"></a><em></em> <strong>prepad</strong> <em>(<var>x</var>, <var>l</var>, <var>c</var>, <var>dim</var>)</em></dt>
<dd><p>Prepend the scalar value <var>c</var> to the vector <var>x</var> until it is of length
<var>l</var>. If <var>c</var> is not given, a value of 0 is used.
</p>
<p>If <code>length (<var>x</var>) > <var>l</var></code>, elements from the beginning of <var>x</var>
are removed until a vector of length <var>l</var> is obtained.
</p>
<p>If <var>x</var> is a matrix, elements are prepended or removed from each row.
</p>
<p>If the optional argument <var>dim</var> is given, operate along this dimension.
</p>
<p>If <var>dim</var> is larger than the dimensions of <var>x</var>, the result will have
<var>dim</var> dimensions.
</p>
<p><strong>See also:</strong> <a href="#XREFpostpad">postpad</a>, <a href="#XREFcat">cat</a>, <a href="#XREFresize">resize</a>.
</p></dd></dl>
<a name="XREFpostpad"></a><dl>
<dt><a name="index-postpad"></a><em></em> <strong>postpad</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-postpad-1"></a><em></em> <strong>postpad</strong> <em>(<var>x</var>, <var>l</var>, <var>c</var>)</em></dt>
<dt><a name="index-postpad-2"></a><em></em> <strong>postpad</strong> <em>(<var>x</var>, <var>l</var>, <var>c</var>, <var>dim</var>)</em></dt>
<dd><p>Append the scalar value <var>c</var> to the vector <var>x</var> until it is of length
<var>l</var>. If <var>c</var> is not given, a value of 0 is used.
</p>
<p>If <code>length (<var>x</var>) > <var>l</var></code>, elements from the end of <var>x</var> are
removed until a vector of length <var>l</var> is obtained.
</p>
<p>If <var>x</var> is a matrix, elements are appended or removed from each row.
</p>
<p>If the optional argument <var>dim</var> is given, operate along this dimension.
</p>
<p>If <var>dim</var> is larger than the dimensions of <var>x</var>, the result will have
<var>dim</var> dimensions.
</p>
<p><strong>See also:</strong> <a href="#XREFprepad">prepad</a>, <a href="#XREFcat">cat</a>, <a href="#XREFresize">resize</a>.
</p></dd></dl>
<a name="XREFdiag"></a><dl>
<dt><a name="index-diag"></a><em><var>M</var> =</em> <strong>diag</strong> <em>(<var>v</var>)</em></dt>
<dt><a name="index-diag-1"></a><em><var>M</var> =</em> <strong>diag</strong> <em>(<var>v</var>, <var>k</var>)</em></dt>
<dt><a name="index-diag-2"></a><em><var>M</var> =</em> <strong>diag</strong> <em>(<var>v</var>, <var>m</var>, <var>n</var>)</em></dt>
<dt><a name="index-diag-3"></a><em><var>v</var> =</em> <strong>diag</strong> <em>(<var>M</var>)</em></dt>
<dt><a name="index-diag-4"></a><em><var>v</var> =</em> <strong>diag</strong> <em>(<var>M</var>, <var>k</var>)</em></dt>
<dd><p>Return a diagonal matrix with vector <var>v</var> on diagonal <var>k</var>.
</p>
<p>The second argument is optional. If it is positive, the vector is placed on
the <var>k</var>-th superdiagonal. If it is negative, it is placed on the
<var>-k</var>-th subdiagonal. The default value of <var>k</var> is 0, and the vector
is placed on the main diagonal. For example:
</p>
<div class="example">
<pre class="example">diag ([1, 2, 3], 1)
⇒ 0 1 0 0
0 0 2 0
0 0 0 3
0 0 0 0
</pre></div>
<p>The 3-input form returns a diagonal matrix with vector <var>v</var> on the main
diagonal and the resulting matrix being of size <var>m</var> rows x <var>n</var>
columns.
</p>
<p>Given a matrix argument, instead of a vector, <code>diag</code> extracts the
<var>k</var>-th diagonal of the matrix.
</p></dd></dl>
<a name="XREFblkdiag"></a><dl>
<dt><a name="index-blkdiag"></a><em></em> <strong>blkdiag</strong> <em>(<var>A</var>, <var>B</var>, <var>C</var>, …)</em></dt>
<dd><p>Build a block diagonal matrix from <var>A</var>, <var>B</var>, <var>C</var>, …
</p>
<p>All arguments must be numeric and either two-dimensional matrices or
scalars. If any argument is of type sparse, the output will also be sparse.
</p>
<p><strong>See also:</strong> <a href="#XREFdiag">diag</a>, <a href="#XREFhorzcat">horzcat</a>, <a href="#XREFvertcat">vertcat</a>, <a href="Creating-Sparse-Matrices.html#XREFsparse">sparse</a>.
</p></dd></dl>
<hr>
<div class="header">
<p>
Next: <a href="Special-Utility-Matrices.html#Special-Utility-Matrices" accesskey="n" rel="next">Special Utility Matrices</a>, Previous: <a href="Finding-Elements-and-Checking-Conditions.html#Finding-Elements-and-Checking-Conditions" accesskey="p" rel="prev">Finding Elements and Checking Conditions</a>, Up: <a href="Matrix-Manipulation.html#Matrix-Manipulation" accesskey="u" rel="up">Matrix Manipulation</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>
distrib
>
Mageia
>
7
>
i586
>
media
>
core-updates
>
by-pkgid
>
641ebb3060c35990cc021d8f7aaf9aca
>
files
>
445
