Skip to content
GitLab
Commit 4e4c3363 authored by Danny Dunlavy's avatar Danny Dunlavy
Browse files

Merge branch 'release-candidate-3.2.1' into 'master' 

Release Candidate 3.2.1

See merge request !60
parents 1c098db0 7d0f1db3
Pipeline #281710894 passed with stage
in 46 seconds
Hide whitespace changes
Inline Side-by-side
Contents.m
View file @ 4e4c3363
% Tensor Toolbox (Sandia National Labs)
% Version 3.2 (R2021a) 10-Feb-2021
% Version 3.2.1 (R2021a) 05-Apr-2021
% Tensor Toolbox for dense, sparse, and decomposed n-way arrays.
%
% Tensor Toolbox Classes:
......
README.md
View file @ 4e4c3363
# Tensor Toolbox for MATLAB, Version 3.2
February 10, 2021
# Tensor Toolbox for MATLAB, Version 3.2.1
April 5, 2021
by Brett W. Bader, Tamara G. Kolda, Daniel M. Dunlavy, et al.
Sandia National Laboratories
......@@ -12,6 +12,12 @@ For all other information, including download and usage instructions, see [www.t
Release notes follow below.
## Changes from Version 3.2 (February 10, 2021)
- Changes to [`export_data`](export_data.m): support formatting of data and lambda values, improved write times for ['sptensor'](@sptensor/) when exporting (fixes #51)
- New help pages:
* [Importing and Exporting Tensor Data](https://www.tensortoolbox.com/import_export_doc.html)
## Changes from Version 3.1 (June 4, 2019)
- Added new functionality for implicit symmetric CP tensor computation per the paper: S. Sherman, T. G. Kolda. Estimating Higher-Order Moments Using Symmetric Tensor Decomposition, SIMAX, 2020 (see !43)
......
doc/html/helptoc.xml
View file @ 4e4c3363
<?xml version='1.0' encoding='ISO-8859-1' ?>
<!-- $Revision: 1.1 $ $Date: 2006/08/30 23:22:37 $ -->
<toc version="3.2">
<toc version="3.2.1">
<tocitem target="index.html" >Tensor Toolbox
......@@ -45,6 +45,7 @@
</tocitem>
<tocitem target="test_problems_doc.html">Creating Test Problems</tocitem>
<tocitem target="identities_doc.html">Identities</tocitem>
<tocitem target="import_export_doc.html">Importing and Exporting Tensor Data</tocitem>
</tocitem>
<tocitem target="opt_options_doc.html">Optimization Methods
<tocitem target="tt_opt_doc.html">Developer Information for Optimization Methods</tocitem>
......
doc/html/import_export_doc.html 0 → 100644
View file @ 4e4c3363
<!DOCTYPE html
PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<!--
This HTML was auto-generated from MATLAB code.
To make changes, update the MATLAB code and republish this document.
--><title>Importing and Exporting Tensor Data</title><meta name="generator" content="MATLAB 9.8"><link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"><meta name="DC.date" content="2021-03-30"><meta name="DC.source" content="import_export_doc.m"><style type="text/css">
html,body,div,span,applet,object,iframe,h1,h2,h3,h4,h5,h6,p,blockquote,pre,a,abbr,acronym,address,big,cite,code,del,dfn,em,font,img,ins,kbd,q,s,samp,small,strike,strong,sub,sup,tt,var,b,u,i,center,dl,dt,dd,ol,ul,li,fieldset,form,label,legend,table,caption,tbody,tfoot,thead,tr,th,td{margin:0;padding:0;border:0;outline:0;font-size:100%;vertical-align:baseline;background:transparent}body{line-height:1}ol,ul{list-style:none}blockquote,q{quotes:none}blockquote:before,blockquote:after,q:before,q:after{content:'';content:none}:focus{outine:0}ins{text-decoration:none}del{text-decoration:line-through}table{border-collapse:collapse;border-spacing:0}
html { min-height:90%; margin-bottom:1px; }
html body { height:100%; margin:0px; font-family:Arial, Helvetica, sans-serif; font-size:12px; color:#000; line-height:140%; background:#fff none; }
html body td { vertical-align:top; text-align:left; }
h1 { padding:0px; margin:0px 0px 25px; font-family:Arial, Helvetica, sans-serif; font-size:2.5em; color:#d55000; line-height:100%; font-weight:normal; }
h2 { padding:0px; margin:0px 0px 8px; font-family:Arial, Helvetica, sans-serif; font-size:1.5em; color:#000; font-weight:bold; line-height:140%; border-bottom:1px solid #d6d4d4; display:block; }
h3 { padding:0px; margin:0px 0px 5px; font-family:Arial, Helvetica, sans-serif; font-size:1.2em; color:#000; font-weight:bold; line-height:140%; }
a { color:#005fce; text-decoration:none; }
a:hover { color:#005fce; text-decoration:underline; }
a:visited { color:#004aa0; text-decoration:none; }
p { padding:0px; margin:0px 0px 20px; }
img { padding:0px; margin:0px 0px 20px; border:none; }
p img, pre img, tt img, li img, h1 img, h2 img { margin-bottom:0px; }
ul { padding:0px; margin:0px 0px 20px 23px; list-style:square; }
ul li { padding:0px; margin:0px 0px 7px 0px; }
ul li ul { padding:5px 0px 0px; margin:0px 0px 7px 23px; }
ul li ol li { list-style:decimal; }
ol { padding:0px; margin:0px 0px 20px 0px; list-style:decimal; }
ol li { padding:0px; margin:0px 0px 7px 23px; list-style-type:decimal; }
ol li ol { padding:5px 0px 0px; margin:0px 0px 7px 0px; }
ol li ol li { list-style-type:lower-alpha; }
ol li ul { padding-top:7px; }
ol li ul li { list-style:square; }
.content { font-size:1.2em; line-height:140%; padding: 20px; }
.banner{ background-color:#15243c; text-align:center;}
.navigate {font-size:0.8em; padding:0px; line-height:100%; }
pre, code { font-size:14px; }
tt { font-size: 1.0em; font-weight:bold; background:#f7f7f7; padding-right:5px; padding-left:5px }
pre { margin:0px 0px 20px; }
pre.codeinput { padding:10px; border:1px solid #d3d3d3; background:#f7f7f7; }
pre.codeoutput { padding:10px 11px; margin:0px 0px 20px; color:#4c4c4c; }
pre.error { color:red; }
@media print { pre.codeinput, pre.codeoutput { word-wrap:break-word; width:100%; } }
span.keyword { color:#0000FF }
span.comment { color:#228B22 }
span.string { color:#A020F0 }
span.untermstring { color:#B20000 }
span.syscmd { color:#B28C00 }
.footer { width:auto; padding:10px 0px; margin:20px 0px 0px; border-top:1px dotted #878787; font-size:0.9em; line-height:140%; font-style:italic; color:#878787; text-align:left; float:none; }
.footer p { margin:0px; padding:0px 20px; }
.footer a { color:#878787; }
.footer a:hover { color:#878787; text-decoration:underline; }
.footer a:visited { color:#878787; }
table th { padding:7px 5px; text-align:left; vertical-align:middle; border: 1px solid #d6d4d4; font-weight:bold; }
table td { padding:7px 5px; text-align:left; vertical-align:top; border:1px solid #d6d4d4; }
</style></head><body><div class="banner"><a href="index.html"><img src="Tensor-Toolbox-for-MATLAB-Banner.png"></a></div><div class="content"><h1>Importing and Exporting Tensor Data</h1><!--introduction--><p>
<p class="navigate">
&#62;&#62; <a href="index.html">Tensor Toolbox</a>
&#62;&#62; <a href="working.html">Working with Tensors</a>
&#62;&#62; <a href="import_export_doc.html">Importing and Exporting Tensor Data</a>
</p>
</p><p>You can import and export tensor data in ASCII format for the following data types:</p><p>
<ul>
<li><tt>tensor</tt></li>
<li><tt>matrix</tt></li>
<li><tt>sptensor</tt></li>
<li><tt>ktensor</tt></li>
</ul>
</p><!--/introduction--><h2>Contents</h2><div><ul><li><a href="#1">Tensor/Matrix Data</a></li><li><a href="#2">Sptensor Data</a></li><li><a href="#3">Ktensor Data</a></li><li><a href="#4">Formatting Output - Data Values</a></li><li><a href="#5">Formatting Output - Ktensor Lambda Values</a></li></ul></div><h2 id="1">Tensor/Matrix Data</h2><p>In the case of a tensor, the first three lines give details about the tensor. The format for a 4 x 3 x 2 tensor is as follows...</p><pre> tensor
3
4 3 2
&lt;value of A(1,1,1)&gt;
&lt;value of A(2,1,1)&gt;
&lt;value of A(3,1,1)&gt;
&lt;value of A(4,1,1)&gt;
&lt;value of A(1,2,1)&gt;
&lt;value of A(2,2,1)&gt;
&lt;value of A(3,2,1)&gt;
&lt;value of A(4,2,1)&gt;
...
&lt;value of A(2,3,2)&gt;
&lt;value of A(3,3,2)&gt;
&lt;value of A(4,3,2)&gt;</pre><p>A matrix is formatted the same as a 2-way tensor except that the first line says <tt>matrix</tt> rather than <tt>tensor</tt>.</p><pre class="codeinput">Y1 = tenrand([4 3 2]);
export_data(Y1,<span class="string">'Y.tensor'</span>);
Y2 = import_data(<span class="string">'Y.tensor'</span>);
assert(isequal(Y1,Y2)) <span class="comment">%&lt;-- test if equal when reading tensor from file</span>
type <span class="string">'Y.tensor'</span>
</pre><pre class="codeoutput">
tensor
3
4 3 2
7.8410984944456918e-01
2.6761755162945355e-01
9.7039153429968650e-01
9.4510452140896994e-01
9.1857177911762622e-01
7.6670088559754168e-01
1.5661279895469682e-01
7.5410547015482998e-01
1.4114456638613404e-01
7.1495769667190401e-02
5.9315889926690779e-01
5.5243694756677753e-01
7.7619043346318617e-01
4.7640496891668416e-01
7.2436234142735267e-01
5.7227121419711569e-01
3.5316198205832472e-01
5.8119606757561004e-02
6.4105967325759805e-01
3.1387368807787064e-01
2.5976459387313477e-01
3.4807513253975986e-01
1.2389652883493230e-01
9.2594519156509847e-02
</pre><h2 id="2">Sptensor Data</h2><p>In the case of an sptensor, the first four lines give details about the sptensor. The format for a 4 x 3 x 2 sptensor with 10 nonzeros is as follows...</p><pre> sptensor
3
4 3 2
3
i1 j1 k1
i2 j2 k2 &lt;value of A(i2,j2,k2)&gt;
i3 j3 k3 &lt;value of A(i3,j3,k3)&gt;</pre><pre class="codeinput">X1 = sptensor([1 1 1;2 2 2; 4 3 2],rand(3,1),[4 3 2]);
export_data(X1,<span class="string">'X.sptensor'</span>);
X2 = import_data(<span class="string">'X.sptensor'</span>);
assert(isequal(X1,X2)) <span class="comment">%&lt;-- test if equal when reading tensor from file</span>
type <span class="string">'X.sptensor'</span>
</pre><pre class="codeoutput">
sptensor
3
4 3 2
3
1 1 1 8.9733153495338303e-01
2 2 2 4.9896060922967300e-01
4 3 2 6.5262946333672744e-01
</pre><h2 id="3">Ktensor Data</h2><p>In the case of an ktensor, the first four lines give details about the ktensor. The format for a 4 x 3 x 2 ktensor with rank=2 is as follows, which contains the lambda values and factor matrices, U, (one per tensor dimension), ...</p><pre> ktensor
3
4 3 2
2
&lt;value of lambda1&gt; &lt;value of lambda2&gt;
matrix
2
4 2
&lt;value of U{1}(1,1)&gt; &lt;value of U{1}(1,2)
...
&lt;value of U{1}(4,1)&gt; &lt;value of U{1}(4,2)
matrix
2
3 2
&lt;value of U{2}(1,1)&gt; &lt;value of U{2}(1,2)
...
&lt;value of U{2}(3,1)&gt; &lt;value of U{2}(3,2)
matrix
2
2 2
&lt;value of U{3}(1,1)&gt; &lt;value of U{3}(1,2)
&lt;value of U{3}(2,1)&gt; &lt;value of U{3}(2,2)</pre><pre class="codeinput">K1 = ktensor(rand(2,1), rand(4,2), rand(3,2), rand(2,2));
export_data(K1,<span class="string">'K.ktensor'</span>);
K2 = import_data(<span class="string">'K.ktensor'</span>);
assert(isequal(K1,K2)) <span class="comment">%&lt;-- test if equal when reading tensor from file</span>
type <span class="string">'K.ktensor'</span>
</pre><pre class="codeoutput">
ktensor
3
4 3 2
2
5.5321794848598371e-01 2.8539029379907055e-01
matrix
2
4 2
9.7373498999922170e-01 5.0705186097791988e-01
7.2298481642071333e-01 9.2451864889949120e-01
5.3540633990533504e-01 9.6085221326105419e-03
9.1921869350005314e-01 2.6354160551053118e-01
matrix
2
3 2
4.1401269139957864e-01 7.8129482939465800e-01
8.9449474691739461e-01 6.2150938203639761e-01
4.3032399708548463e-01 3.4315258530567716e-01
matrix
2
2 2
2.0310353424746941e-01 6.4395364377355735e-01
9.2082402018958531e-01 5.7706691115717912e-02
</pre><h2 id="4">Formatting Output - Data Values</h2><p>When exporting tensor data your can specify the format of the data values using character vector that follows the convention used in <tt>fprintf</tt>. For example, this can be useful when exporting tensors that contain only integers, resulting in a file that is smaller in size. The optional <tt>fmt_data</tt> parameter to <tt>export_data</tt> specifies the format string, with <tt>'%.16</tt>' being the default value.</p><pre class="codeinput">X3 = sptensor([1 1 1;2 2 2; 4 3 2],[1;2;3],[4 3 2]);
export_data(X3,<span class="string">'X_fmt_data.sptensor'</span>,<span class="string">'fmt_data'</span>,<span class="string">'%d'</span>);
X4 = import_data(<span class="string">'X_fmt_data.sptensor'</span>);
assert(isequal(X3,X4)) <span class="comment">%&lt;-- test if equal when reading tensor from file</span>
type <span class="string">'X_fmt_data.sptensor'</span>
</pre><pre class="codeoutput">
sptensor
3
4 3 2
3
1 1 1 1
2 2 2 2
4 3 2 3
</pre><h2 id="5">Formatting Output - Ktensor Lambda Values</h2><p>When exporting <tt>ktensor</tt> data your can specify the format of the <tt>lambda</tt> values using character vector that follows the convention used in <tt>fprint</tt>. For example, this can be useful when exporting tensors that contain only integers, resulting in a file that is smaller in size. The optional <tt>fmt_lambda</tt> parameter to <tt>export_data</tt> specifies the format string, with <tt>'%.16'</tt> being the default value.</p><pre class="codeinput">K3 = ktensor([1; 2], 4*ones(4,2), 3*ones(3,2), 2*ones(2,2));
export_data(K3,<span class="string">'K_fmt_data_lambda.ktensor'</span>,<span class="string">'fmt_data'</span>,<span class="string">'%d'</span>,<span class="string">'fmt_lambda'</span>,<span class="string">'%d'</span>);
K4 = import_data(<span class="string">'K_fmt_data_lambda.ktensor'</span>);
assert(isequal(K3,K4)) <span class="comment">%&lt;-- test if equal when reading tensor from file</span>
type <span class="string">'K_fmt_data_lambda.ktensor'</span>
</pre><pre class="codeoutput">
ktensor
3
4 3 2
2
1 2
matrix
2
4 2
4 4
4 4
4 4
4 4
matrix
2
3 2
3 3
3 3
3 3
matrix
2
2 2
2 2
2 2
</pre><p class="footer">Tensor Toolbox for MATLAB: <a href="index.html">www.tensortoolbox.org</a>.</p></div><!--
##### SOURCE BEGIN #####
%% Importing and Exporting Tensor Data
%
% <html>
% <p class="navigate">
% &#62;&#62; <a href="index.html">Tensor Toolbox</a>
% &#62;&#62; <a href="working.html">Working with Tensors</a>
% &#62;&#62; <a href="import_export_doc.html">Importing and Exporting Tensor Data</a>
% </p>
% </html>
%
% You can import and export tensor data in ASCII format for the following
% data types:
%
% <html>
% <ul>
% <li><tt>tensor</tt></li>
% <li><tt>matrix</tt></li>
% <li><tt>sptensor</tt></li>
% <li><tt>ktensor</tt></li>
% </ul>
% </html>
%% Tensor/Matrix Data
% In the case of a tensor, the first three lines give details about the
% tensor. The format for a 4 x 3 x 2 tensor is as follows...
%
% tensor
% 3
% 4 3 2
% <value of A(1,1,1)>
% <value of A(2,1,1)>
% <value of A(3,1,1)>
% <value of A(4,1,1)>
% <value of A(1,2,1)>
% <value of A(2,2,1)>
% <value of A(3,2,1)>
% <value of A(4,2,1)>
% ...
% <value of A(2,3,2)>
% <value of A(3,3,2)>
% <value of A(4,3,2)>
%
% A matrix is formatted the same as a 2-way tensor except that the first
% line says |matrix| rather than |tensor|.
%
Y1 = tenrand([4 3 2]);
export_data(Y1,'Y.tensor');
Y2 = import_data('Y.tensor');
assert(isequal(Y1,Y2)) %<REPLACE_WITH_DASH_DASH test if equal when reading tensor from file
type 'Y.tensor'
%% Sptensor Data
% In the case of an sptensor, the first four lines give details about the
% sptensor. The format for a 4 x 3 x 2 sptensor with 10 nonzeros is as
% follows...
%
% sptensor
% 3
% 4 3 2
% 3
% i1 j1 k1
% i2 j2 k2 <value of A(i2,j2,k2)>
% i3 j3 k3 <value of A(i3,j3,k3)>
%
X1 = sptensor([1 1 1;2 2 2; 4 3 2],rand(3,1),[4 3 2]);
export_data(X1,'X.sptensor');
X2 = import_data('X.sptensor');
assert(isequal(X1,X2)) %<REPLACE_WITH_DASH_DASH test if equal when reading tensor from file
type 'X.sptensor'
%% Ktensor Data
% In the case of an ktensor, the first four lines give details about
% the ktensor. The format for a 4 x 3 x 2 ktensor with rank=2 is as
% follows, which contains the lambda values and factor matrices, U,
% (one per tensor dimension), ...
%
% ktensor
% 3
% 4 3 2
% 2
% <value of lambda1> <value of lambda2>
% matrix
% 2
% 4 2
% <value of U{1}(1,1)> <value of U{1}(1,2)
% ...
% <value of U{1}(4,1)> <value of U{1}(4,2)
% matrix
% 2
% 3 2
% <value of U{2}(1,1)> <value of U{2}(1,2)
% ...
% <value of U{2}(3,1)> <value of U{2}(3,2)
% matrix
% 2
% 2 2
% <value of U{3}(1,1)> <value of U{3}(1,2)
% <value of U{3}(2,1)> <value of U{3}(2,2)
%
K1 = ktensor(rand(2,1), rand(4,2), rand(3,2), rand(2,2));
export_data(K1,'K.ktensor');
K2 = import_data('K.ktensor');
assert(isequal(K1,K2)) %<REPLACE_WITH_DASH_DASH test if equal when reading tensor from file
type 'K.ktensor'
%% Formatting Output - Data Values
% When exporting tensor data your can specify the format of the data values
% using character vector that follows the convention used in |fprintf|. For
% example, this can be useful when exporting tensors that contain only
% integers, resulting in a file that is smaller in size. The optional
% |fmt_data| parameter to |export_data| specifies the format string, with
% |'%.16|' being the default value.
%
X3 = sptensor([1 1 1;2 2 2; 4 3 2],[1;2;3],[4 3 2]);
export_data(X3,'X_fmt_data.sptensor','fmt_data','%d');
X4 = import_data('X_fmt_data.sptensor');
assert(isequal(X3,X4)) %<REPLACE_WITH_DASH_DASH test if equal when reading tensor from file
type 'X_fmt_data.sptensor'
%% Formatting Output - Ktensor Lambda Values
% When exporting |ktensor| data your can specify the format of the |lambda|
% values using character vector that follows the convention used in
% |fprint|. For example, this can be useful when exporting tensors that
% contain only integers, resulting in a file that is smaller in size. The
% optional |fmt_lambda| parameter to |export_data| specifies the format
% string, with |'%.16'| being the default value.
%
K3 = ktensor([1; 2], 4*ones(4,2), 3*ones(3,2), 2*ones(2,2));
export_data(K3,'K_fmt_data_lambda.ktensor','fmt_data','%d','fmt_lambda','%d');
K4 = import_data('K_fmt_data_lambda.ktensor');
assert(isequal(K3,K4)) %<REPLACE_WITH_DASH_DASH test if equal when reading tensor from file
type 'K_fmt_data_lambda.ktensor'
##### SOURCE END #####
--></body></html>
\ No newline at end of file
doc/html/index.html
View file @ 4e4c3363
......@@ -11,7 +11,7 @@
<a href="index.html"><img src="Tensor-Toolbox-for-MATLAB-Banner.png"></img></a>
</div>
<div class="content">
<h1 id="home">Tensor Toolbox for MATLAB, Version 3.2</h1>
<h1 id="home">Tensor Toolbox for MATLAB, Version 3.2.1</h1>
<p>The Tensor Toolbox for MATLAB provides a suite of tools for
working with multidimensional or N-way arrays. Tensor
......@@ -30,7 +30,7 @@
for the terms of the license (2-clause BSD).
</p>
<ul>
<li><a href="https://gitlab.com/tensors/tensor_toolbox/-/releases/v3.2">Tensor Toolbox for MATLAB v3.2 (current release) </a></li>
<li><a href="https://gitlab.com/tensors/tensor_toolbox/-/releases/v3.2.1">Tensor Toolbox for MATLAB v3.2.1 (current release) </a></li>
<li><a href="https://gitlab.com/tensors/tensor_toolbox/-/releases/v3.1">Tensor Toolbox for MATLAB v3.1</a></li>
<li><a href="http://gitlab.com/tensors/tensor_toolbox">Current development version from GITLAB repository</a></li>
</ul>
......@@ -89,9 +89,9 @@
<ul class=citations>
<li><b>General software, latest release:</b>
Brett W. Bader, Tamara G. Kolda and others,
Tensor Toolbox for MATLAB, Version 3.2,
Tensor Toolbox for MATLAB, Version 3.2.1,
<a href="https://www.tensortoolbox.org">www.tensortoolbox.org</a>,
February 10, 2021.
April 5, 2021.
<li><b>General software, development version:</b>
Brett W. Bader, Tamara G. Kolda and others,
Tensor Toolbox for MATLAB, development version <span class="untermstring">[hash]</span>,
......
doc/html/working.html
View file @ 4e4c3363
......@@ -39,6 +39,10 @@
<li><a href="identities_doc.html">Identities</a> - Exploration
of identities and connections among tensor and matrix
operations</li>
<li><a href="import_export_doc.html">Importing and Exporting Tensor
Data</a> - Covers <tt>import_data</tt> and <tt>export_data</tt>
for reading and writing <tt>tensor</tt>, <tt>sptensor</tt>, and
<tt>ktensor</tt> objects</li>
</ul>
<p class="footer">Tensor Toolbox for MATLAB: <a href="index.html">www.tensortoolbox.org</a>.</p>
......
doc/import_export_doc.m 0 → 100644
View file @ 4e4c3363
%% Importing and Exporting Tensor Data
%
% <html>
% <p class="navigate">
% &#62;&#62; <a href="index.html">Tensor Toolbox</a>
% &#62;&#62; <a href="working.html">Working with Tensors</a>
% &#62;&#62; <a href="import_export_doc.html">Importing and Exporting Tensor Data</a>
% </p>
% </html>
%
% You can import and export tensor data in ASCII format for the following
% data types:
%
% <html>
% <ul>
% <li><tt>tensor</tt></li>
% <li><tt>matrix</tt></li>
% <li><tt>sptensor</tt></li>
% <li><tt>ktensor</tt></li>
% </ul>
% </html>
%% Tensor/Matrix Data
% In the case of a tensor, the first three lines give details about the
% tensor. The format for a 4 x 3 x 2 tensor is as follows...
%
% tensor
% 3
% 4 3 2
% <value of A(1,1,1)>
% <value of A(2,1,1)>
% <value of A(3,1,1)>
% <value of A(4,1,1)>
% <value of A(1,2,1)>
% <value of A(2,2,1)>
% <value of A(3,2,1)>
% <value of A(4,2,1)>
% ...
% <value of A(2,3,2)>
% <value of A(3,3,2)>
% <value of A(4,3,2)>
%
% A matrix is formatted the same as a 2-way tensor except that the first
% line says |matrix| rather than |tensor|.
%
Y1 = tenrand([4 3 2]);
export_data(Y1,'Y.tensor');
Y2 = import_data('Y.tensor');
assert(isequal(Y1,Y2)) %<-- test if equal when reading tensor from file
type 'Y.tensor'
%% Sptensor Data
% In the case of an sptensor, the first four lines give details about the
% sptensor. The format for a 4 x 3 x 2 sptensor with 10 nonzeros is as
% follows...
%
% sptensor
% 3
% 4 3 2
% 3
% i1 j1 k1
% i2 j2 k2 <value of A(i2,j2,k2)>
% i3 j3 k3 <value of A(i3,j3,k3)>
%
X1 = sptensor([1 1 1;2 2 2; 4 3 2],rand(3,1),[4 3 2]);
export_data(X1,'X.sptensor');
X2 = import_data('X.sptensor');
assert(isequal(X1,X2)) %<-- test if equal when reading tensor from file
type 'X.sptensor'
%% Ktensor Data
% In the case of an ktensor, the first four lines give details about
% the ktensor. The format for a 4 x 3 x 2 ktensor with rank=2 is as
% follows, which contains the lambda values and factor matrices, U,
% (one per tensor dimension), ...
%
% ktensor
% 3
% 4 3 2
% 2
% <value of lambda1> <value of lambda2>
% matrix
% 2
% 4 2
% <value of U{1}(1,1)> <value of U{1}(1,2)
% ...
% <value of U{1}(4,1)> <value of U{1}(4,2)
% matrix
% 2
% 3 2
% <value of U{2}(1,1)> <value of U{2}(1,2)
% ...
% <value of U{2}(3,1)> <value of U{2}(3,2)
% matrix
% 2
% 2 2
% <value of U{3}(1,1)> <value of U{3}(1,2)
% <value of U{3}(2,1)> <value of U{3}(2,2)
%
K1 = ktensor(rand(2,1), rand(4,2), rand(3,2), rand(2,2));
export_data(K1,'K.ktensor');
K2 = import_data('K.ktensor');
assert(isequal(K1,K2)) %<-- test if equal when reading tensor from file
type 'K.ktensor'
%% Formatting Output - Data Values
% When exporting tensor data your can specify the format of the data values
% using character vector that follows the convention used in |fprintf|. For
% example, this can be useful when exporting tensors that contain only
% integers, resulting in a file that is smaller in size. The optional
% |fmt_data| parameter to |export_data| specifies the format string, with
% |'%.16|' being the default value.
%
X3 = sptensor([1 1 1;2 2 2; 4 3 2],[1;2;3],[4 3 2]);
export_data(X3,'X_fmt_data.sptensor','fmt_data','%d');
X4 = import_data('X_fmt_data.sptensor');
assert(isequal(X3,X4)) %<-- test if equal when reading tensor from file
type 'X_fmt_data.sptensor'
%% Formatting Output - Ktensor Lambda Values
% When exporting |ktensor| data your can specify the format of the |lambda|
% values using character vector that follows the convention used in
% |fprint|. For example, this can be useful when exporting tensors that
% contain only integers, resulting in a file that is smaller in size. The
% optional |fmt_lambda| parameter to |export_data| specifies the format
% string, with |'%.16'| being the default value.
%
K3 = ktensor([1; 2], 4*ones(4,2), 3*ones(3,2), 2*ones(2,2));
export_data(K3,'K_fmt_data_lambda.ktensor','fmt_data','%d','fmt_lambda',<