build based on fa9f6d0

dev/.documenter-siteinfo.json

@@ -1 +1 @@
-{"documenter":{"julia_version":"1.9.4","generation_timestamp":"2023-12-10T07:49:51","documenter_version":"1.2.1"}}
+{"documenter":{"julia_version":"1.10.0","generation_timestamp":"2024-01-23T04:19:51","documenter_version":"1.2.1"}}

dev/algorithms/auglag/index.html

@@ -4,4 +4,4 @@
 
 alg = AugLag()
 options = AugLagOptions()
-result = optimize(model, alg, x0, options = options)</code></pre><p>Percival is an optional dependency of Nonconvex so you need to import it in order to use it.</p><h2 id="Construct-an-instance"><a class="docs-heading-anchor" href="#Construct-an-instance">Construct an instance</a><a id="Construct-an-instance-1"></a><a class="docs-heading-anchor-permalink" href="#Construct-an-instance" title="Permalink"></a></h2><p>To construct an instance of the Ipopt algorithm, use:</p><pre><code class="language-julia hljs">alg = AugAlg()</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The options keyword argument to the <code>optimize</code> function shown above must be an instance of the <code>AugLagOptions</code> struct when the algorihm is an <code>AugLag</code>. To specify options use keyword arguments in the constructor of <code>AugLagOptions</code>, e.g:</p><pre><code class="language-julia hljs">options = AugLagOptions(first_order = false, rtol = 1e-4)</code></pre><p>The most important option is <code>first_order</code> which is <code>true</code> by default. When <code>first_order</code> is <code>true</code>, the first order augmented Lagrangian algorithm will be used. And when it is <code>false</code>, the second order augmented Lagrangian algorithm will be used. Other arguments include:</p><ul><li><code>atol</code>: absolute tolerance in the subproblem optimizer</li><li><code>rtol</code>: relative tolerance in the subproblem optimizer</li><li><code>ctol</code>: absolute feasibility tolerance</li><li><code>max_iter</code>: maximum number of iterations</li><li><code>max_time</code>: maximum time in seconds</li><li><code>max_eval</code>: maximum number of function evaluations</li></ul><p>When using the first order augmented Lagrangian and a block constraint (i.e. a constraint function that returns a vector), the use of reverse-mode AD will only require calling the adjoint operator of the block constraint function in order to compute the gradient of the augmented Lagrangian. This is particularly suitable for constraint functions whose Jacobians are expensive but the adjoint operator is relatively inexpensive.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nlopt/">« Various optimization algorithms from <code>NLopt.jl</code></a><a class="docs-footer-nextpage" href="../minlp/">First and second order mixed integer nonlinear programming algorithms »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.2.1 on <span class="colophon-date" title="Sunday 10 December 2023 07:49">Sunday 10 December 2023</span>. Using Julia version 1.9.4.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+result = optimize(model, alg, x0, options = options)</code></pre><p>Percival is an optional dependency of Nonconvex so you need to import it in order to use it.</p><h2 id="Construct-an-instance"><a class="docs-heading-anchor" href="#Construct-an-instance">Construct an instance</a><a id="Construct-an-instance-1"></a><a class="docs-heading-anchor-permalink" href="#Construct-an-instance" title="Permalink"></a></h2><p>To construct an instance of the Ipopt algorithm, use:</p><pre><code class="language-julia hljs">alg = AugAlg()</code></pre><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The options keyword argument to the <code>optimize</code> function shown above must be an instance of the <code>AugLagOptions</code> struct when the algorihm is an <code>AugLag</code>. To specify options use keyword arguments in the constructor of <code>AugLagOptions</code>, e.g:</p><pre><code class="language-julia hljs">options = AugLagOptions(first_order = false, rtol = 1e-4)</code></pre><p>The most important option is <code>first_order</code> which is <code>true</code> by default. When <code>first_order</code> is <code>true</code>, the first order augmented Lagrangian algorithm will be used. And when it is <code>false</code>, the second order augmented Lagrangian algorithm will be used. Other arguments include:</p><ul><li><code>atol</code>: absolute tolerance in the subproblem optimizer</li><li><code>rtol</code>: relative tolerance in the subproblem optimizer</li><li><code>ctol</code>: absolute feasibility tolerance</li><li><code>max_iter</code>: maximum number of iterations</li><li><code>max_time</code>: maximum time in seconds</li><li><code>max_eval</code>: maximum number of function evaluations</li></ul><p>When using the first order augmented Lagrangian and a block constraint (i.e. a constraint function that returns a vector), the use of reverse-mode AD will only require calling the adjoint operator of the block constraint function in order to compute the gradient of the augmented Lagrangian. This is particularly suitable for constraint functions whose Jacobians are expensive but the adjoint operator is relatively inexpensive.</p></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../nlopt/">« Various optimization algorithms from <code>NLopt.jl</code></a><a class="docs-footer-nextpage" href="../minlp/">First and second order mixed integer nonlinear programming algorithms »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.2.1 on <span class="colophon-date" title="Tuesday 23 January 2024 04:19">Tuesday 23 January 2024</span>. Using Julia version 1.10.0.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>

dev/algorithms/hyperopt/index.html

@@ -7,4 +7,4 @@
 result = optimize(model, alg, x0, options = options)</code></pre><p>Hyperopt is an optional dependency of Nonconvex so you need to import it in order to use it. <code>HyperoptAlg</code> can wrap any other algorithm in Nonconvex, e.g. <code>IpoptAlg()</code>. When the algorithm is a <code>HyperoptAlg</code>, the <code>options</code> keyword argument must of type <code>HyperoptOptions</code>. For more on the options available see below.</p><h2 id="Construct-an-instance"><a class="docs-heading-anchor" href="#Construct-an-instance">Construct an instance</a><a id="Construct-an-instance-1"></a><a class="docs-heading-anchor-permalink" href="#Construct-an-instance" title="Permalink"></a></h2><p>To construct an instance of the Hyperopt + Ipopt algorithm, use:</p><pre><code class="language-julia hljs">alg = HyperoptAlg(IpoptAlg())</code></pre><p><code>HyperoptAlg</code> can wrap any other algorithm in Nonconvex, e.g. <code>NLoptAlg(:LD_MMA)</code> or <code>AugLag()</code>.</p><h2 id="Options"><a class="docs-heading-anchor" href="#Options">Options</a><a id="Options-1"></a><a class="docs-heading-anchor-permalink" href="#Options" title="Permalink"></a></h2><p>The options keyword argument to the <code>optimize</code> function shown above must be an instance of the <code>HyperoptOptions</code> struct when the algorihm is a <code>HyperoptAlg</code>. To specify options, use keyword arguments in the constructor of <code>HyperoptOptions</code>. The <code>sampler</code> keyword argument determines the sampling algorithm used to propose new starting points in the multi-start procedure. The <code>sub_options</code> keyword argument can be used to pass in the options for the sub-optimizer. There are 2 different ways to pass the sub-options depending on the sampler type.</p><p>The <code>sampler</code> argument can be of type:</p><ol><li><code>RandomSampler</code></li><li><code>LHSampler</code></li><li><code>CLHSampler</code></li><li><code>GPSampler</code></li><li><code>Hyperband</code></li></ol><p>When optimizing the starting point, the upper and lower bounds on the initial solution must be finite, or finite bounds must be passed in to the <code>options</code> constructor. All the options that can be passed to the <code>HyperoptOptions</code> constructor are listed below:</p><article class="docstring"><header><a class="docstring-article-toggle-button fa-solid fa-chevron-down" href="javascript:;" title="Collapse docstring"></a><a class="docstring-binding" id="NonconvexMultistart.HyperoptOptions" href="#NonconvexMultistart.HyperoptOptions"><code>NonconvexMultistart.HyperoptOptions</code></a> — <span class="docstring-category">Type</span></header><section><div><pre><code class="language-julia hljs">HyperoptOptions: options performing starting point optimization using Hyperopt.jl</code></pre><ul><li><code>sub_options</code>: options for the sub-optimizer.</li><li><code>lb</code>: Lower bound of starting point, if don&#39;t specify it, the default value will be <code>nothing</code>,            then will end up be replaced by the lower bound of optimization problem.</li><li><code>ub</code>: Upper bound of starting point, same as above. </li><li><code>searchspace_size::Integer</code>: How many potential starting points we generate.</li><li><code>iters::Integer</code>: Among all generated potential starting points, how many of them will be evaluated. </li><li><code>sampler::Hyperopt.Sampler</code>: An instance of &#39;Hyperopt.Sampler&#39;, which decides search algorithm. </li><li><code>ctol</code>: infeasibility tolerance for accepting a solution as feasible</li><li><code>keep_all</code>: if true, all the solutions of the sub-problems will be saved</li></ul></div><a class="docs-sourcelink" target="_blank" href="https://github.com/JuliaNonconvex/NonconvexMultistart.jl/blob/v0.1.3/src/hyperopt.jl#L5-L17">source</a></section></article><h3 id="Sampler-choice"><a class="docs-heading-anchor" href="#Sampler-choice">Sampler choice</a><a id="Sampler-choice-1"></a><a class="docs-heading-anchor-permalink" href="#Sampler-choice" title="Permalink"></a></h3><h4 id="RandomSampler,-LHSampler,-CLHSampler-and-GPSampler"><a class="docs-heading-anchor" href="#RandomSampler,-LHSampler,-CLHSampler-and-GPSampler">RandomSampler, LHSampler, CLHSampler and GPSampler</a><a id="RandomSampler,-LHSampler,-CLHSampler-and-GPSampler-1"></a><a class="docs-heading-anchor-permalink" href="#RandomSampler,-LHSampler,-CLHSampler-and-GPSampler" title="Permalink"></a></h4><p>All the sampler constructors are functions defined in Nonconvex wrapping the Hyperopt alternatives to define defaults. For <code>GPSampler</code>, <code>Hyperopt.Min</code> is always used by default in Nonconvex so you should not pass this argument. All the other arguments that can be passed to the sampler constructor can be found in the <a href="https://github.com/baggepinnen/Hyperopt.jl#details">Hyperopt documentation</a>. Example:</p><pre><code class="language-julia hljs">options = HyperoptOptions(sub_options = IpoptOptions(), sampler = GPSampler())</code></pre><h4 id="Hyperband"><a class="docs-heading-anchor" href="#Hyperband">Hyperband</a><a id="Hyperband-1"></a><a class="docs-heading-anchor-permalink" href="#Hyperband" title="Permalink"></a></h4><p>The <a href="https://github.com/baggepinnen/Hyperopt.jl#hyperband">Hyperband algorithm</a> in Hyperopt requires a different way to pass in the sub-options. The Hyperband algorithm tries to optimize the allocation of resources. The <code>sub_options</code> argument must be a function with input as the &quot;resources&quot; and output as the sub-solver options. The <code>Hyperband</code> constructor accepts 3 arguments:</p><ol><li>The maximum resources <code>R</code></li><li><code>η</code> which roughly determines the proportion of trials discarded between each round of successive halving</li><li><code>inner</code> which specifies an inner sampler of type <code>RandomSampler</code>, <code>LHSampler</code> or <code>CLHSampler</code>.</li></ol><p>Example:</p><pre><code class="language-julia hljs">options = HyperoptOptions(
     sub_options = max_iter -&gt; IpoptOptions(max_iter = max_iter), 
     sampler = Hyperband(R=100, η=3, inner=RandomSampler()),
-)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../minlp/">« First and second order mixed integer nonlinear programming algorithms</a><a class="docs-footer-nextpage" href="../surrogate/">Surrogate-assisted continuous and discrete, constrained optimization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.2.1 on <span class="colophon-date" title="Sunday 10 December 2023 07:49">Sunday 10 December 2023</span>. Using Julia version 1.9.4.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+)</code></pre></article><nav class="docs-footer"><a class="docs-footer-prevpage" href="../minlp/">« First and second order mixed integer nonlinear programming algorithms</a><a class="docs-footer-nextpage" href="../surrogate/">Surrogate-assisted continuous and discrete, constrained optimization »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="auto">Automatic (OS)</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.2.1 on <span class="colophon-date" title="Tuesday 23 January 2024 04:19">Tuesday 23 January 2024</span>. Using Julia version 1.10.0.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>