Updated docs with new default rhobeg

Author: lindonroberts

docs/build/html/_sources/userguide.rst.txt

@@ -75,7 +75,7 @@ These arguments are:
 * :code:`args` - a tuple of extra arguments passed to the objective function. 
 * :code:`bounds` - a tuple :code:`(lower, upper)` with the vectors :math:`a` and :math:`b` of lower and upper bounds on :math:`x` (default is :math:`a_i=-10^{20}` and :math:`b_i=10^{20}`). To set bounds for either :code:`lower` or :code:`upper`, but not both, pass a tuple :code:`(lower, None)` or :code:`(None, upper)`.
 * :code:`npt` - the number of interpolation points to use (default is :code:`len(x0)+1`). If using restarts, this is the number of points to use in the first run of the solver, before any restarts (and may be optionally increased via settings in :code:`user_params`).
-* :code:`rhobeg` - the initial value of the trust region radius (default is :math:`0.1\max(\|x_0\|_{\infty}, 1)`).
+* :code:`rhobeg` - the initial value of the trust region radius (default is :math:`0.1\max(\|x_0\|_{\infty}, 1)`, or 0.1 if :code:`scaling_within_bounds`).
 * :code:`rhoend` - minimum allowed value of trust region radius, which determines when a successful termination occurs (default is :math:`10^{-8}`).
 * :code:`maxfun` - the maximum number of objective evaluations the algorithm may request (default is :math:`\min(100(n+1),1000)`).
 * :code:`nsamples` - a Python function :code:`nsamples(delta, rho, iter, nrestarts)` which returns the number of times to evaluate :code:`objfun` at a given point. This is only applicable for objectives with stochastic noise, when averaging multiple evaluations at the same point produces a more accurate value. The input parameters are the trust region radius (:code:`delta`), the lower bound on the trust region radius (:code:`rho`), how many iterations the algorithm has been running for (:code:`iter`), and how many restarts have been performed (:code:`nrestarts`). Default is no averaging (i.e. :code:`nsamples(delta, rho, iter, nrestarts)=1`).

docs/build/html/userguide.html

@@ -243,7 +243,7 @@ <h2>Optional Arguments<a class="headerlink" href="#optional-arguments" title="Pe
 <li><code class="code docutils literal"><span class="pre">args</span></code> - a tuple of extra arguments passed to the objective function.</li>
 <li><code class="code docutils literal"><span class="pre">bounds</span></code> - a tuple <code class="code docutils literal"><span class="pre">(lower,</span> <span class="pre">upper)</span></code> with the vectors <span class="math">\(a\)</span> and <span class="math">\(b\)</span> of lower and upper bounds on <span class="math">\(x\)</span> (default is <span class="math">\(a_i=-10^{20}\)</span> and <span class="math">\(b_i=10^{20}\)</span>). To set bounds for either <code class="code docutils literal"><span class="pre">lower</span></code> or <code class="code docutils literal"><span class="pre">upper</span></code>, but not both, pass a tuple <code class="code docutils literal"><span class="pre">(lower,</span> <span class="pre">None)</span></code> or <code class="code docutils literal"><span class="pre">(None,</span> <span class="pre">upper)</span></code>.</li>
 <li><code class="code docutils literal"><span class="pre">npt</span></code> - the number of interpolation points to use (default is <code class="code docutils literal"><span class="pre">len(x0)+1</span></code>). If using restarts, this is the number of points to use in the first run of the solver, before any restarts (and may be optionally increased via settings in <code class="code docutils literal"><span class="pre">user_params</span></code>).</li>
-<li><code class="code docutils literal"><span class="pre">rhobeg</span></code> - the initial value of the trust region radius (default is <span class="math">\(0.1\max(\|x_0\|_{\infty}, 1)\)</span>).</li>
+<li><code class="code docutils literal"><span class="pre">rhobeg</span></code> - the initial value of the trust region radius (default is <span class="math">\(0.1\max(\|x_0\|_{\infty}, 1)\)</span>, or 0.1 if <code class="code docutils literal"><span class="pre">scaling_within_bounds</span></code>).</li>
 <li><code class="code docutils literal"><span class="pre">rhoend</span></code> - minimum allowed value of trust region radius, which determines when a successful termination occurs (default is <span class="math">\(10^{-8}\)</span>).</li>
 <li><code class="code docutils literal"><span class="pre">maxfun</span></code> - the maximum number of objective evaluations the algorithm may request (default is <span class="math">\(\min(100(n+1),1000)\)</span>).</li>
 <li><code class="code docutils literal"><span class="pre">nsamples</span></code> - a Python function <code class="code docutils literal"><span class="pre">nsamples(delta,</span> <span class="pre">rho,</span> <span class="pre">iter,</span> <span class="pre">nrestarts)</span></code> which returns the number of times to evaluate <code class="code docutils literal"><span class="pre">objfun</span></code> at a given point. This is only applicable for objectives with stochastic noise, when averaging multiple evaluations at the same point produces a more accurate value. The input parameters are the trust region radius (<code class="code docutils literal"><span class="pre">delta</span></code>), the lower bound on the trust region radius (<code class="code docutils literal"><span class="pre">rho</span></code>), how many iterations the algorithm has been running for (<code class="code docutils literal"><span class="pre">iter</span></code>), and how many restarts have been performed (<code class="code docutils literal"><span class="pre">nrestarts</span></code>). Default is no averaging (i.e. <code class="code docutils literal"><span class="pre">nsamples(delta,</span> <span class="pre">rho,</span> <span class="pre">iter,</span> <span class="pre">nrestarts)=1</span></code>).</li>

docs/userguide.rst

@@ -75,7 +75,7 @@ These arguments are:
 * :code:`args` - a tuple of extra arguments passed to the objective function. 
 * :code:`bounds` - a tuple :code:`(lower, upper)` with the vectors :math:`a` and :math:`b` of lower and upper bounds on :math:`x` (default is :math:`a_i=-10^{20}` and :math:`b_i=10^{20}`). To set bounds for either :code:`lower` or :code:`upper`, but not both, pass a tuple :code:`(lower, None)` or :code:`(None, upper)`.
 * :code:`npt` - the number of interpolation points to use (default is :code:`len(x0)+1`). If using restarts, this is the number of points to use in the first run of the solver, before any restarts (and may be optionally increased via settings in :code:`user_params`).
-* :code:`rhobeg` - the initial value of the trust region radius (default is :math:`0.1\max(\|x_0\|_{\infty}, 1)`).
+* :code:`rhobeg` - the initial value of the trust region radius (default is :math:`0.1\max(\|x_0\|_{\infty}, 1)`, or 0.1 if :code:`scaling_within_bounds`).
 * :code:`rhoend` - minimum allowed value of trust region radius, which determines when a successful termination occurs (default is :math:`10^{-8}`).
 * :code:`maxfun` - the maximum number of objective evaluations the algorithm may request (default is :math:`\min(100(n+1),1000)`).
 * :code:`nsamples` - a Python function :code:`nsamples(delta, rho, iter, nrestarts)` which returns the number of times to evaluate :code:`objfun` at a given point. This is only applicable for objectives with stochastic noise, when averaging multiple evaluations at the same point produces a more accurate value. The input parameters are the trust region radius (:code:`delta`), the lower bound on the trust region radius (:code:`rho`), how many iterations the algorithm has been running for (:code:`iter`), and how many restarts have been performed (:code:`nrestarts`). Default is no averaging (i.e. :code:`nsamples(delta, rho, iter, nrestarts)=1`).