CMA-ES Test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cma\n",
    "import cma.test\n",
    "import numpy as np\n",
    "import sys\n",
    "import pandas as pd\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on package cma:\n",
      "\n",
      "NAME\n",
      "    cma\n",
      "\n",
      "DESCRIPTION\n",
      "    Package `cma` implements the CMA-ES (Covariance Matrix Adaptation\n",
      "    Evolution Strategy).\n",
      "    \n",
      "    CMA-ES is a stochastic optimizer for robust non-linear non-convex\n",
      "    derivative- and function-value-free numerical optimization.\n",
      "    \n",
      "    This implementation can be used with Python versions >= 2.6, namely\n",
      "    2.6, 2.7, 3.3, 3.4, 3.5, 3.6.\n",
      "    \n",
      "    CMA-ES searches for a minimizer (a solution x in :math:`R^n`) of an\n",
      "    objective function f (cost function), such that f(x) is minimal.\n",
      "    Regarding f, only a passably reliable ranking of the candidate\n",
      "    solutions in each iteration is necessary. Neither the function values\n",
      "    itself, nor the gradient of f need to be available or do matter (like\n",
      "    in the downhill simplex Nelder-Mead algorithm). Some termination\n",
      "    criteria however depend on actual f-values.\n",
      "    \n",
      "    The `cma` module provides two independent implementations of the\n",
      "    CMA-ES algorithm in the classes `cma.CMAEvolutionStrategy` and\n",
      "    `cma.purecma.CMAES`.\n",
      "    \n",
      "    In each implementation two interfaces are provided:\n",
      "    \n",
      "    - functions `fmin` and `purecma.fmin`:\n",
      "        run a complete minimization of the passed objective function with\n",
      "        CMA-ES. `fmin` also provides optional restarts and noise handling.\n",
      "    \n",
      "    - class `CMAEvolutionStrategy` and `purecma.CMAES`:\n",
      "        allow for minimization such that the control of the iteration\n",
      "        loop remains with the user.\n",
      "    \n",
      "    The `cma` package root provides shortcuts to these and other classes and\n",
      "    functions.\n",
      "    \n",
      "    Used external packages are `numpy` (only `purecma` does not depend on\n",
      "    `numpy`) and `matplotlib.pyplot` (for `plot` etc., optional but highly\n",
      "    recommended).\n",
      "    \n",
      "    Install\n",
      "    =======\n",
      "    To use the module, the folder ``cma`` only needs to be visible in the\n",
      "    python path, e.g. in the current working directory.\n",
      "    \n",
      "    To install the module from pipy, type::\n",
      "    \n",
      "        pip install cma\n",
      "    \n",
      "    from the command line.\n",
      "    \n",
      "    To install the module from a ``cma`` folder::\n",
      "    \n",
      "        pip install -e cma\n",
      "    \n",
      "    To upgrade the currently installed version use additionally the ``-U``\n",
      "    option.\n",
      "    \n",
      "    Testing\n",
      "    =======\n",
      "    From the system shell::\n",
      "    \n",
      "        python -m cma.test -h\n",
      "        python -m cma.test\n",
      "        python -c \"import cma.test; cma.test.main()\"  # the same\n",
      "    \n",
      "    or from any (i)python shell::\n",
      "    \n",
      "        import cma.test\n",
      "        cma.test.main()\n",
      "    \n",
      "    should run without complaints in about between 20 and 100 seconds.\n",
      "    \n",
      "    Example\n",
      "    =======\n",
      "    From a python shell::\n",
      "    \n",
      "        import cma\n",
      "        help(cma)  # \"this\" help message, use cma? in ipython\n",
      "        help(cma.fmin)\n",
      "        help(cma.CMAEvolutionStrategy)\n",
      "        help(cma.CMAOptions)\n",
      "        cma.CMAOptions('tol')  # display 'tolerance' termination options\n",
      "        cma.CMAOptions('verb') # display verbosity options\n",
      "        res = cma.fmin(cma.ff.tablet, 15 * [1], 1)\n",
      "        es = cma.CMAEvolutionStrategy(15 * [1], 1).optimize(cma.ff.tablet)\n",
      "        help(es.result)\n",
      "        res[0], es.result[0]  # best evaluated solution\n",
      "        res[5], es.result[5]  # mean solution, presumably better with noise\n",
      "    \n",
      "    :See also: `fmin` (), `CMAOptions`, `CMAEvolutionStrategy`\n",
      "    \n",
      "    :Author: Nikolaus Hansen, 2008-\n",
      "    :Author: Petr Baudis, 2014\n",
      "    :Author: Youhei Akimoto, 2017-\n",
      "    \n",
      "    :License: BSD 3-Clause, see LICENSE file.\n",
      "\n",
      "PACKAGE CONTENTS\n",
      "    bbobbenchmarks\n",
      "    constraints_handler\n",
      "    evolution_strategy\n",
      "    experimentation\n",
      "    fitness_functions\n",
      "    fitness_functions2\n",
      "    fitness_models\n",
      "    fitness_transformations\n",
      "    interfaces\n",
      "    logger\n",
      "    optimization_tools\n",
      "    purecma\n",
      "    recombination_weights\n",
      "    restricted_gaussian_sampler\n",
      "    s\n",
      "    sampler\n",
      "    sigma_adaptation\n",
      "    test\n",
      "    transformations\n",
      "    utilities (package)\n",
      "\n",
      "DATA\n",
      "    ___author__ = 'Nikolaus Hansen and Petr Baudis and Youhei Akimoto'\n",
      "    __license__ = 'BSD 3-clause'\n",
      "    ff = <cma.fitness_functions.FitnessFunctions object>\n",
      "\n",
      "VERSION\n",
      "    2.7.0  $Revision: 4426 $ $Date: 2019-04-24 18:03:09 +0200 (Wed, 24 Apr 2019) $\n",
      "\n",
      "AUTHOR\n",
      "    Nikolaus Hansen\n",
      "\n",
      "FILE\n",
      "    /Users/purujit/anaconda3/lib/python3.7/site-packages/cma/__init__.py\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(cma)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function fmin in module cma.evolution_strategy:\n",
      "\n",
      "fmin(objective_function, x0, sigma0, options=None, args=(), gradf=None, restarts=0, restart_from_best='False', incpopsize=2, eval_initial_x=False, parallel_objective=None, noise_handler=None, noise_change_sigma_exponent=1, noise_kappa_exponent=0, bipop=False, callback=None)\n",
      "    functional interface to the stochastic optimizer CMA-ES\n",
      "    for non-convex function minimization.\n",
      "    \n",
      "    Calling Sequences\n",
      "    =================\n",
      "    ``fmin(objective_function, x0, sigma0)``\n",
      "        minimizes ``objective_function`` starting at ``x0`` and with\n",
      "        standard deviation ``sigma0`` (step-size)\n",
      "    ``fmin(objective_function, x0, sigma0, options={'ftarget': 1e-5})``\n",
      "        minimizes ``objective_function`` up to target function value 1e-5,\n",
      "        which is typically useful for benchmarking.\n",
      "    ``fmin(objective_function, x0, sigma0, args=('f',))``\n",
      "        minimizes ``objective_function`` called with an additional\n",
      "        argument ``'f'``.\n",
      "    ``fmin(objective_function, x0, sigma0, options={'ftarget':1e-5, 'popsize':40})``\n",
      "        uses additional options ``ftarget`` and ``popsize``\n",
      "    ``fmin(objective_function, esobj, None, options={'maxfevals': 1e5})``\n",
      "        uses the `CMAEvolutionStrategy` object instance ``esobj`` to\n",
      "        optimize ``objective_function``, similar to ``esobj.optimize()``.\n",
      "    \n",
      "    Arguments\n",
      "    =========\n",
      "    ``objective_function``\n",
      "        called as ``objective_function(x, *args)`` to be minimized.\n",
      "        ``x`` is a one-dimensional `numpy.ndarray`. See also the\n",
      "        `parallel_objective` argument.\n",
      "        ``objective_function`` can return `numpy.NaN`, which is\n",
      "        interpreted as outright rejection of solution ``x`` and invokes\n",
      "        an immediate resampling and (re-)evaluation of a new solution\n",
      "        not counting as function evaluation. The attribute\n",
      "        ``variable_annotations`` is passed into the\n",
      "        ``CMADataLogger.persistent_communication_dict``.\n",
      "    ``x0``\n",
      "        list or `numpy.ndarray`, initial guess of minimum solution\n",
      "        before the application of the geno-phenotype transformation\n",
      "        according to the ``transformation`` option.  It can also be\n",
      "        a string holding a Python expression that is evaluated\n",
      "        to yield the initial guess - this is important in case\n",
      "        restarts are performed so that they start from different\n",
      "        places.  Otherwise ``x0`` can also be a `cma.CMAEvolutionStrategy`\n",
      "        object instance, in that case ``sigma0`` can be ``None``.\n",
      "    ``sigma0``\n",
      "        scalar, initial standard deviation in each coordinate.\n",
      "        ``sigma0`` should be about 1/4th of the search domain width\n",
      "        (where the optimum is to be expected). The variables in\n",
      "        ``objective_function`` should be scaled such that they\n",
      "        presumably have similar sensitivity.\n",
      "        See also `ScaleCoordinates`.\n",
      "    ``options``\n",
      "        a dictionary with additional options passed to the constructor\n",
      "        of class ``CMAEvolutionStrategy``, see ``cma.CMAOptions`` ()\n",
      "        for a list of available options.\n",
      "    ``args=()``\n",
      "        arguments to be used to call the ``objective_function``\n",
      "    ``gradf=None``\n",
      "        gradient of f, where ``len(gradf(x, *args)) == len(x)``.\n",
      "        ``gradf`` is called once in each iteration if\n",
      "        ``gradf is not None``.\n",
      "    ``restarts=0``\n",
      "        number of restarts with increasing population size, see also\n",
      "        parameter ``incpopsize``, implementing the IPOP-CMA-ES restart\n",
      "        strategy, see also parameter ``bipop``; to restart from\n",
      "        different points (recommended), pass ``x0`` as a string.\n",
      "    ``restart_from_best=False``\n",
      "        which point to restart from\n",
      "    ``incpopsize=2``\n",
      "        multiplier for increasing the population size ``popsize`` before\n",
      "        each restart\n",
      "    ``parallel_objective``\n",
      "        an objective function that accepts a list of `numpy.ndarray` as\n",
      "        input and returns a `list`, which is mostly used instead of\n",
      "        `objective_function`, but for the initial (also initial\n",
      "        elitist) and the final evaluations. If ``parallel_objective``\n",
      "        is given, the ``objective_function`` (first argument) may be\n",
      "        ``None``.\n",
      "    ``eval_initial_x=None``\n",
      "        evaluate initial solution, for ``None`` only with elitist option\n",
      "    ``noise_handler=None``\n",
      "        a ``NoiseHandler`` class or instance or ``None``. Example:\n",
      "        ``cma.fmin(f, 6 * [1], 1, noise_handler=cma.NoiseHandler(6))``\n",
      "        see ``help(cma.NoiseHandler)``.\n",
      "    ``noise_change_sigma_exponent=1``\n",
      "        exponent for the sigma increment provided by the noise handler for\n",
      "        additional noise treatment. 0 means no sigma change.\n",
      "    ``noise_evaluations_as_kappa=0``\n",
      "        instead of applying reevaluations, the \"number of evaluations\"\n",
      "        is (ab)used as scaling factor kappa (experimental).\n",
      "    ``bipop=False``\n",
      "        if `True`, run as BIPOP-CMA-ES; BIPOP is a special restart\n",
      "        strategy switching between two population sizings - small\n",
      "        (like the default CMA, but with more focused search) and\n",
      "        large (progressively increased as in IPOP). This makes the\n",
      "        algorithm perform well both on functions with many regularly\n",
      "        or irregularly arranged local optima (the latter by frequently\n",
      "        restarting with small populations).  For the `bipop` parameter\n",
      "        to actually take effect, also select non-zero number of\n",
      "        (IPOP) restarts; the recommended setting is ``restarts<=9``\n",
      "        and `x0` passed as a string using `numpy.rand` to generate\n",
      "        initial solutions. Note that small-population restarts\n",
      "        do not count into the total restart count.\n",
      "    ``callback=None``\n",
      "        `callable` or list of callables called at the end of each\n",
      "        iteration with the current `CMAEvolutionStrategy` instance\n",
      "        as argument.\n",
      "    \n",
      "    Optional Arguments\n",
      "    ==================\n",
      "    All values in the `options` dictionary are evaluated if they are of\n",
      "    type `str`, besides `verb_filenameprefix`, see class `CMAOptions` for\n",
      "    details. The full list is available by calling ``cma.CMAOptions()``.\n",
      "    \n",
      "    >>> import cma\n",
      "    >>> cma.CMAOptions()  #doctest: +ELLIPSIS\n",
      "    {...\n",
      "    \n",
      "    Subsets of options can be displayed, for example like\n",
      "    ``cma.CMAOptions('tol')``, or ``cma.CMAOptions('bound')``,\n",
      "    see also class `CMAOptions`.\n",
      "    \n",
      "    Return\n",
      "    ======\n",
      "    Return the list provided in `CMAEvolutionStrategy.result` appended\n",
      "    with termination conditions, an `OOOptimizer` and a `BaseDataLogger`::\n",
      "    \n",
      "        res = es.result + (es.stop(), es, logger)\n",
      "    \n",
      "    where\n",
      "        - ``res[0]`` (``xopt``) -- best evaluated solution\n",
      "        - ``res[1]`` (``fopt``) -- respective function value\n",
      "        - ``res[2]`` (``evalsopt``) -- respective number of function evaluations\n",
      "        - ``res[3]`` (``evals``) -- number of overall conducted objective function evaluations\n",
      "        - ``res[4]`` (``iterations``) -- number of overall conducted iterations\n",
      "        - ``res[5]`` (``xmean``) -- mean of the final sample distribution\n",
      "        - ``res[6]`` (``stds``) -- effective stds of the final sample distribution\n",
      "        - ``res[-3]`` (``stop``) -- termination condition(s) in a dictionary\n",
      "        - ``res[-2]`` (``cmaes``) -- class `CMAEvolutionStrategy` instance\n",
      "        - ``res[-1]`` (``logger``) -- class `CMADataLogger` instance\n",
      "    \n",
      "    Details\n",
      "    =======\n",
      "    This function is an interface to the class `CMAEvolutionStrategy`. The\n",
      "    latter class should be used when full control over the iteration loop\n",
      "    of the optimizer is desired.\n",
      "    \n",
      "    Examples\n",
      "    ========\n",
      "    The following example calls `fmin` optimizing the Rosenbrock function\n",
      "    in 10-D with initial solution 0.1 and initial step-size 0.5. The\n",
      "    options are specified for the usage with the `doctest` module.\n",
      "    \n",
      "    >>> import cma\n",
      "    >>> # cma.CMAOptions()  # returns all possible options\n",
      "    >>> options = {'CMA_diagonal':100, 'seed':1234, 'verb_time':0}\n",
      "    >>>\n",
      "    >>> res = cma.fmin(cma.ff.rosen, [0.1] * 10, 0.3, options)  #doctest: +ELLIPSIS\n",
      "    (5_w,10)-aCMA-ES (mu_w=3.2,w_1=45%) in dimension 10 (seed=1234...)\n",
      "       Covariance matrix is diagonal for 100 iterations (1/ccov=26...\n",
      "    Iterat #Fevals   function value  axis ratio  sigma ...\n",
      "        1     10 ...\n",
      "    termination on tolfun=1e-11 ...\n",
      "    final/bestever f-value = ...\n",
      "    >>> assert res[1] < 1e-12  # f-value of best found solution\n",
      "    >>> assert res[2] < 8000  # evaluations\n",
      "    \n",
      "    The above call is pretty much equivalent with the slightly more\n",
      "    verbose call::\n",
      "    \n",
      "        res = cma.CMAEvolutionStrategy([0.1] * 10, 0.3,\n",
      "                    options=options).optimize(cma.ff.rosen).result\n",
      "    \n",
      "    where `optimize` returns a `CMAEvolutionStrategy` instance. The\n",
      "    following example calls `fmin` optimizing the Rastrigin function\n",
      "    in 3-D with random initial solution in [-2,2], initial step-size 0.5\n",
      "    and the BIPOP restart strategy (that progressively increases population).\n",
      "    The options are specified for the usage with the `doctest` module.\n",
      "    \n",
      "    >>> import cma\n",
      "    >>> # cma.CMAOptions()  # returns all possible options\n",
      "    >>> options = {'seed':12345, 'verb_time':0, 'ftarget': 1e-8}\n",
      "    >>>\n",
      "    >>> res = cma.fmin(cma.ff.rastrigin, '2. * np.random.rand(3) - 1', 0.5,\n",
      "    ...                options, restarts=9, bipop=True)  #doctest: +ELLIPSIS\n",
      "    (3_w,7)-aCMA-ES (mu_w=2.3,w_1=58%) in dimension 3 (seed=12345...\n",
      "    \n",
      "    In either case, the method::\n",
      "    \n",
      "        cma.plot();\n",
      "    \n",
      "    (based on `matplotlib.pyplot`) produces a plot of the run and, if\n",
      "    necessary::\n",
      "    \n",
      "        cma.s.figshow()\n",
      "    \n",
      "    shows the plot in a window. Finally::\n",
      "    \n",
      "        cma.s.figsave('myfirstrun')  # figsave from matplotlib.pyplot\n",
      "    \n",
      "    will save the figure in a png.\n",
      "    \n",
      "    We can use the gradient like\n",
      "    \n",
      "    >>> import cma\n",
      "    >>> res = cma.fmin(cma.ff.rosen, np.zeros(10), 0.1,\n",
      "    ...             options = {'ftarget':1e-8,},\n",
      "    ...             gradf=cma.ff.grad_rosen,\n",
      "    ...         )  #doctest: +ELLIPSIS\n",
      "    (5_w,...\n",
      "    >>> assert cma.ff.rosen(res[0]) < 1e-8\n",
      "    >>> assert res[2] < 3600  # 1% are > 3300\n",
      "    >>> assert res[3] < 3600  # 1% are > 3300\n",
      "    \n",
      "    If solution can only be comparatively ranked, either use\n",
      "    `CMAEvolutionStrategy` directly or the objective accepts a list\n",
      "    of solutions as input:\n",
      "    \n",
      "    >>> def parallel_sphere(X): return [cma.ff.sphere(x) for x in X]\n",
      "    >>> x, es = cma.fmin2(None, 3 * [0], 0.1, {'verbose': -9},\n",
      "    ...                   parallel_objective=parallel_sphere)\n",
      "    >>> assert es.result[1] < 1e-9\n",
      "    \n",
      "    :See also: `CMAEvolutionStrategy`, `OOOptimizer.optimize`, `plot`,\n",
      "        `CMAOptions`, `scipy.optimize.fmin`\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(cma.fmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class CMAEvolutionStrategy in module cma.evolution_strategy:\n",
      "\n",
      "class CMAEvolutionStrategy(cma.interfaces.OOOptimizer)\n",
      " |  CMAEvolutionStrategy(x0, sigma0, inopts=None)\n",
      " |  \n",
      " |  CMA-ES stochastic optimizer class with ask-and-tell interface.\n",
      " |  \n",
      " |  Calling Sequences\n",
      " |  =================\n",
      " |  \n",
      " |  - ``es = CMAEvolutionStrategy(x0, sigma0)``\n",
      " |  \n",
      " |  - ``es = CMAEvolutionStrategy(x0, sigma0, opts)``\n",
      " |  \n",
      " |  - ``es = CMAEvolutionStrategy(x0, sigma0).optimize(objective_fct)``\n",
      " |  \n",
      " |  - ::\n",
      " |  \n",
      " |      res = CMAEvolutionStrategy(x0, sigma0,\n",
      " |                              opts).optimize(objective_fct).result\n",
      " |  \n",
      " |  Arguments\n",
      " |  =========\n",
      " |  `x0`\n",
      " |      initial solution, starting point. `x0` is given as \"phenotype\"\n",
      " |      which means, if::\n",
      " |  \n",
      " |          opts = {'transformation': [transform, inverse]}\n",
      " |  \n",
      " |      is given and ``inverse is None``, the initial mean is not\n",
      " |      consistent with `x0` in that ``transform(mean)`` does not\n",
      " |      equal to `x0` unless ``transform(mean)`` equals ``mean``.\n",
      " |  `sigma0`\n",
      " |      initial standard deviation.  The problem variables should\n",
      " |      have been scaled, such that a single standard deviation\n",
      " |      on all variables is useful and the optimum is expected to\n",
      " |      lie within about `x0` +- ``3*sigma0``. See also options\n",
      " |      `scaling_of_variables`. Often one wants to check for\n",
      " |      solutions close to the initial point. This allows,\n",
      " |      for example, for an easier check of consistency of the\n",
      " |      objective function and its interfacing with the optimizer.\n",
      " |      In this case, a much smaller `sigma0` is advisable.\n",
      " |  `opts`\n",
      " |      options, a dictionary with optional settings,\n",
      " |      see class `CMAOptions`.\n",
      " |  \n",
      " |  Main interface / usage\n",
      " |  ======================\n",
      " |  The interface is inherited from the generic `OOOptimizer`\n",
      " |  class (see also there). An object instance is generated from::\n",
      " |  \n",
      " |      es = cma.CMAEvolutionStrategy(8 * [0.5], 0.2)\n",
      " |  \n",
      " |  The least verbose interface is via the optimize method::\n",
      " |  \n",
      " |      es.optimize(objective_func)\n",
      " |      res = es.result\n",
      " |  \n",
      " |  More verbosely, the optimization is done using the\n",
      " |  methods `stop`, `ask`, and `tell`::\n",
      " |  \n",
      " |      while not es.stop():\n",
      " |          solutions = es.ask()\n",
      " |          es.tell(solutions, [cma.ff.rosen(s) for s in solutions])\n",
      " |          es.disp()\n",
      " |      es.result_pretty()\n",
      " |  \n",
      " |  \n",
      " |  where `ask` delivers new candidate solutions and `tell` updates\n",
      " |  the `optim` instance by passing the respective function values\n",
      " |  (the objective function `cma.ff.rosen` can be replaced by any\n",
      " |  properly defined objective function, see `cma.ff` for more\n",
      " |  examples).\n",
      " |  \n",
      " |  To change an option, for example a termination condition to\n",
      " |  continue the optimization, call::\n",
      " |  \n",
      " |      es.opts.set({'tolfacupx': 1e4})\n",
      " |  \n",
      " |  The class `CMAEvolutionStrategy` also provides::\n",
      " |  \n",
      " |      (solutions, func_values) = es.ask_and_eval(objective_func)\n",
      " |  \n",
      " |  and an entire optimization can also be written like::\n",
      " |  \n",
      " |      while not es.stop():\n",
      " |          es.tell(*es.ask_and_eval(objective_func))\n",
      " |  \n",
      " |  Besides for termination criteria, in CMA-ES only the ranks of the\n",
      " |  `func_values` are relevant.\n",
      " |  \n",
      " |  Attributes and Properties\n",
      " |  =========================\n",
      " |  - `inputargs`: passed input arguments\n",
      " |  - `inopts`: passed options\n",
      " |  - `opts`: actually used options, some of them can be changed any\n",
      " |    time via ``opts.set``, see class `CMAOptions`\n",
      " |  - `popsize`: population size lambda, number of candidate\n",
      " |    solutions returned by `ask` ()\n",
      " |  - `logger`: a `CMADataLogger` instance utilized by `optimize`\n",
      " |  \n",
      " |  Examples\n",
      " |  ========\n",
      " |  Super-short example, with output shown:\n",
      " |  \n",
      " |  >>> import cma\n",
      " |  >>> # construct an object instance in 4-D, sigma0=1:\n",
      " |  >>> es = cma.CMAEvolutionStrategy(4 * [1], 1, {'seed':234})\n",
      " |  ...      # doctest: +ELLIPSIS\n",
      " |  (4_w,8)-aCMA-ES (mu_w=2.6,w_1=52%) in dimension 4 (seed=234...)\n",
      " |  \n",
      " |  and optimize the ellipsoid function\n",
      " |  \n",
      " |  >>> es.optimize(cma.ff.elli, verb_disp=1)  # doctest: +ELLIPSIS\n",
      " |  Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      " |      1      8 2.09...\n",
      " |  >>> assert len(es.result) == 8\n",
      " |  >>> assert es.result[1] < 1e-9\n",
      " |  \n",
      " |  The optimization loop can also be written explicitly:\n",
      " |  \n",
      " |  >>> es = cma.CMAEvolutionStrategy(4 * [1], 1)  # doctest: +ELLIPSIS\n",
      " |  (4_w,8)-aCMA-ES (mu_w=2.6,w_1=52%) in dimension 4 (seed=...\n",
      " |  >>> while not es.stop():\n",
      " |  ...    X = es.ask()\n",
      " |  ...    es.tell(X, [cma.ff.elli(x) for x in X])\n",
      " |  ...    es.disp()  # doctest: +ELLIPSIS\n",
      " |  Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      " |      1      8 ...\n",
      " |  \n",
      " |  achieving the same result as above.\n",
      " |  \n",
      " |  An example with lower bounds (at zero) and handling infeasible\n",
      " |  solutions:\n",
      " |  \n",
      " |  >>> import numpy as np\n",
      " |  >>> es = cma.CMAEvolutionStrategy(10 * [0.2], 0.5,\n",
      " |  ...         {'bounds': [0, np.inf]})  #doctest: +ELLIPSIS\n",
      " |  (5_w,...\n",
      " |  >>> while not es.stop():\n",
      " |  ...     fit, X = [], []\n",
      " |  ...     while len(X) < es.popsize:\n",
      " |  ...         curr_fit = None\n",
      " |  ...         while curr_fit in (None, np.NaN):\n",
      " |  ...             x = es.ask(1)[0]\n",
      " |  ...             curr_fit = cma.ff.somenan(x, cma.ff.elli) # might return np.NaN\n",
      " |  ...         X.append(x)\n",
      " |  ...         fit.append(curr_fit)\n",
      " |  ...     es.tell(X, fit)\n",
      " |  ...     es.logger.add()\n",
      " |  ...     es.disp()  #doctest: +ELLIPSIS\n",
      " |  Itera...\n",
      " |  >>>\n",
      " |  >>> assert es.result[1] < 1e-9\n",
      " |  >>> assert es.result[2] < 9000  # by internal termination\n",
      " |  >>> # es.logger.plot()  # will plot data\n",
      " |  >>> # cma.s.figshow()  # display plot window\n",
      " |  \n",
      " |  An example with user-defined transformation, in this case to realize\n",
      " |  a lower bound of 2.\n",
      " |  \n",
      " |  >>> import warnings\n",
      " |  >>> with warnings.catch_warnings(record=True) as warns:\n",
      " |  ...     es = cma.CMAEvolutionStrategy(5 * [3], 0.1,\n",
      " |  ...                 {\"transformation\": [lambda x: x**2+1.2, None],\n",
      " |  ...                  \"verbose\": -2,})\n",
      " |  >>> warns[0].message  # doctest:+ELLIPSIS\n",
      " |  UserWarning('in class GenoPheno: user defined transformations have not been tested thoroughly ()'...\n",
      " |  >>> warns[1].message  # doctest:+ELLIPSIS\n",
      " |  UserWarning('computed initial point...\n",
      " |  >>> es.optimize(cma.ff.rosen, verb_disp=0)  #doctest: +ELLIPSIS\n",
      " |  <cma...\n",
      " |  >>> assert cma.ff.rosen(es.result[0]) < 1e-7 + 5.54781521192\n",
      " |  >>> assert es.result[2] < 3300\n",
      " |  \n",
      " |  The inverse transformation is (only) necessary if the `BoundPenalty`\n",
      " |  boundary handler is used at the same time.\n",
      " |  \n",
      " |  The `CMAEvolutionStrategy` class also provides a default logger\n",
      " |  (cave: files are overwritten when the logger is used with the same\n",
      " |  filename prefix):\n",
      " |  \n",
      " |  >>> es = cma.CMAEvolutionStrategy(4 * [0.2], 0.5, {'verb_disp': 0})\n",
      " |  >>> es.logger.disp_header()  # annotate the print of disp\n",
      " |  Iterat Nfevals  function value    axis ratio maxstd  minstd\n",
      " |  >>> while not es.stop():\n",
      " |  ...     X = es.ask()\n",
      " |  ...     es.tell(X, [cma.ff.sphere(x) for x in X])\n",
      " |  ...     es.logger.add()  # log current iteration\n",
      " |  ...     es.logger.disp([-1])  # display info for last iteration   #doctest: +ELLIPSIS\n",
      " |      1  ...\n",
      " |  >>> es.logger.disp_header()\n",
      " |  Iterat Nfevals  function value    axis ratio maxstd  minstd\n",
      " |  >>> # es.logger.plot() # will make a plot\n",
      " |  \n",
      " |  Example implementing restarts with increasing popsize (IPOP):\n",
      " |  \n",
      " |  >>> bestever = cma.optimization_tools.BestSolution()\n",
      " |  >>> for lam in 10 * 2**np.arange(8):  # 10, 20, 40, 80, ..., 10 * 2**7\n",
      " |  ...     es = cma.CMAEvolutionStrategy('6 - 8 * np.random.rand(9)',  # 9-D\n",
      " |  ...                                   5,  # initial std sigma0\n",
      " |  ...                                   {'popsize': lam,  # options\n",
      " |  ...                                    'verb_append': bestever.evalsall})\n",
      " |  ...     logger = cma.CMADataLogger().register(es, append=bestever.evalsall)\n",
      " |  ...     while not es.stop():\n",
      " |  ...         X = es.ask()    # get list of new solutions\n",
      " |  ...         fit = [cma.ff.rastrigin(x) for x in X]  # evaluate each solution\n",
      " |  ...         es.tell(X, fit) # besides for termination only the ranking in fit is used\n",
      " |  ...\n",
      " |  ...         # display some output\n",
      " |  ...         logger.add()  # add a \"data point\" to the log, writing in files\n",
      " |  ...         es.disp()  # uses option verb_disp with default 100\n",
      " |  ...\n",
      " |  ...     print('termination:', es.stop())\n",
      " |  ...     cma.s.pprint(es.best.__dict__)\n",
      " |  ...\n",
      " |  ...     bestever.update(es.best)\n",
      " |  ...\n",
      " |  ...     # show a plot\n",
      " |  ...     # logger.plot();\n",
      " |  ...     if bestever.f < 1e-8:  # global optimum was hit\n",
      " |  ...         break  #doctest: +ELLIPSIS\n",
      " |  (5_w,...\n",
      " |  >>> assert es.result[1] < 1e-8\n",
      " |  \n",
      " |  On the Rastrigin function, usually after five restarts the global\n",
      " |  optimum is located.\n",
      " |  \n",
      " |  Using the `multiprocessing` module, we can evaluate the function in\n",
      " |  parallel with a simple modification of the example (however\n",
      " |  multiprocessing seems not always reliable):\n",
      " |  \n",
      " |  >>> from cma.fitness_functions import elli  # cannot be an instance method\n",
      " |  >>> from cma.fitness_transformations import EvalParallel\n",
      " |  >>> es = cma.CMAEvolutionStrategy(22 * [0.0], 1.0, {'maxiter':10})  # doctest:+ELLIPSIS\n",
      " |  (6_w,13)-aCMA-ES (mu_w=...\n",
      " |  >>> with EvalParallel(es.popsize + 1) as eval_all:\n",
      " |  ...     while not es.stop():\n",
      " |  ...         X = es.ask()\n",
      " |  ...         es.tell(X, eval_all(elli, X))\n",
      " |  ...         es.disp()\n",
      " |  ...         # es.logger.add()  # doctest:+ELLIPSIS\n",
      " |  Iterat...\n",
      " |  \n",
      " |  The final example shows how to resume:\n",
      " |  \n",
      " |  >>> import pickle\n",
      " |  >>>\n",
      " |  >>> es = cma.CMAEvolutionStrategy(12 * [0.1],  # a new instance, 12-D\n",
      " |  ...                               0.12)         # initial std sigma0\n",
      " |  ...   #doctest: +ELLIPSIS\n",
      " |  (5_w,...\n",
      " |  >>> es.optimize(cma.ff.rosen, iterations=100)  #doctest: +ELLIPSIS\n",
      " |  I...\n",
      " |  >>> pickle.dump(es, open('_saved-cma-object.pkl', 'wb'))\n",
      " |  >>> del es  # let's start fresh\n",
      " |  >>>\n",
      " |  >>> es = pickle.load(open('_saved-cma-object.pkl', 'rb'))\n",
      " |  >>> # resuming\n",
      " |  >>> es.optimize(cma.ff.rosen, verb_disp=200)  #doctest: +ELLIPSIS\n",
      " |    200 ...\n",
      " |  >>> assert es.result[2] < 15000\n",
      " |  >>> assert cma.s.Mh.vequals_approximately(es.result[0], 12 * [1], 1e-5)\n",
      " |  >>> assert len(es.result) == 8\n",
      " |  \n",
      " |  Details\n",
      " |  =======\n",
      " |  The following two enhancements are implemented, the latter is turned\n",
      " |  on by default for very small population size only.\n",
      " |  \n",
      " |  *Active CMA* is implemented with option ``CMA_active`` and\n",
      " |  conducts an update of the covariance matrix with negative weights.\n",
      " |  The negative update is implemented, such that positive definiteness\n",
      " |  is guarantied. A typical speed up factor (number of f-evaluations)\n",
      " |  is between 1.1 and two.\n",
      " |  \n",
      " |  References: Jastrebski and Arnold, Improving evolution strategies\n",
      " |  through active covariance matrix adaptation, CEC 2006.\n",
      " |  Hansen, The CMA evolution strategy: a tutorial, arXiv 2016.\n",
      " |  \n",
      " |  *Selective mirroring* is implemented with option ``CMA_mirrors``\n",
      " |  in the method `get_mirror` and `get_selective_mirrors`.\n",
      " |  The method `ask_and_eval` (used by `fmin`) will then sample\n",
      " |  selectively mirrored vectors within the iteration\n",
      " |  (``CMA_mirrormethod==1``). Otherwise, or if ``CMA_mirromethod==2``,\n",
      " |  selective mirrors are injected for the next iteration.\n",
      " |  In selective mirroring, only the worst solutions are  mirrored. With\n",
      " |  the default small number of mirrors, *pairwise selection* (where at\n",
      " |  most one of the two mirrors contribute to the update of the\n",
      " |  distribution mean) is implicitly guarantied under selective\n",
      " |  mirroring and therefore not explicitly implemented.\n",
      " |  \n",
      " |  References: Brockhoff et al, PPSN 2010, Auger et al, GECCO 2011.\n",
      " |  \n",
      " |  :See also: `fmin` (), `OOOptimizer`, `CMAOptions`, `plot` (), `ask` (),\n",
      " |      `tell` (), `ask_and_eval` ()\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      CMAEvolutionStrategy\n",
      " |      cma.interfaces.OOOptimizer\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, x0, sigma0, inopts=None)\n",
      " |      see class `CMAEvolutionStrategy`\n",
      " |  \n",
      " |  alleviate_conditioning(self, condition=1000000000000.0)\n",
      " |      pass conditioning of `C` to linear transformation in `self.gp`.\n",
      " |      \n",
      " |      Argument `condition` defines the limit condition number above\n",
      " |      which the action is taken.\n",
      " |      \n",
      " |      Details: the action applies only if `self.gp.isidentity`. Then,\n",
      " |      the covariance matrix `C` is set (back) to identity and a\n",
      " |      respective linear transformation is \"added\" to `self.gp`.\n",
      " |  \n",
      " |  alleviate_conditioning_in_coordinates(self, condition=100000000.0)\n",
      " |      pass scaling from `C` to `sigma_vec`.\n",
      " |      \n",
      " |      As a result, `C` is a correlation matrix, i.e., all diagonal\n",
      " |      entries of `C` are `1`.\n",
      " |  \n",
      " |  ask(self, number=None, xmean=None, sigma_fac=1, gradf=None, args=())\n",
      " |      get/sample new candidate solutions.\n",
      " |      \n",
      " |      Solutions are sampled from a multi-variate\n",
      " |      normal distribution and transformed to f-representation\n",
      " |      (phenotype) to be evaluated.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |          `number`\n",
      " |              number of returned solutions, by default the\n",
      " |              population size ``popsize`` (AKA ``lambda``).\n",
      " |          `xmean`\n",
      " |              distribution mean, phenotyp?\n",
      " |          `sigma_fac`\n",
      " |              multiplier for internal sample width (standard\n",
      " |              deviation)\n",
      " |          `gradf`\n",
      " |              gradient, ``len(gradf(x)) == len(x)``, if\n",
      " |              ``gradf is not None`` the third solution in the\n",
      " |              returned list is \"sampled\" in supposedly Newton\n",
      " |              direction ``np.dot(C, gradf(xmean, *args))``.\n",
      " |          `args`\n",
      " |              additional arguments passed to gradf\n",
      " |      \n",
      " |      Return\n",
      " |      ------\n",
      " |      A list of N-dimensional candidate solutions to be evaluated\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      >>> import cma\n",
      " |      >>> es = cma.CMAEvolutionStrategy([0,0,0,0], 0.3)  #doctest: +ELLIPSIS\n",
      " |      (4_w,...\n",
      " |      >>> while not es.stop() and es.best.f > 1e-6:\n",
      " |      ...     X = es.ask()  # get list of new solutions\n",
      " |      ...     fit = [cma.ff.rosen(x) for x in X]  # call fct with each solution\n",
      " |      ...     es.tell(X, fit)  # feed values\n",
      " |      \n",
      " |      :See: `ask_and_eval`, `ask_geno`, `tell`\n",
      " |  \n",
      " |  ask_and_eval(self, func, args=(), gradf=None, number=None, xmean=None, sigma_fac=1, evaluations=1, aggregation=<function median at 0x10dac62f0>, kappa=1, parallel_mode=False)\n",
      " |      sample `number` solutions and evaluate them on `func`.\n",
      " |      \n",
      " |      Each solution ``s`` is resampled until\n",
      " |      ``self.is_feasible(s, func(s)) is True``.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |      `func`:\n",
      " |          objective function, ``func(x)`` accepts a `numpy.ndarray`\n",
      " |          and returns a scalar ``if not parallel_mode``. Else returns a\n",
      " |          `list` of scalars from a `list` of `numpy.ndarray`.\n",
      " |      `args`:\n",
      " |          additional parameters for `func`\n",
      " |      `gradf`:\n",
      " |          gradient of objective function, ``g = gradf(x, *args)``\n",
      " |          must satisfy ``len(g) == len(x)``\n",
      " |      `number`:\n",
      " |          number of solutions to be sampled, by default\n",
      " |          population size ``popsize`` (AKA lambda)\n",
      " |      `xmean`:\n",
      " |          mean for sampling the solutions, by default ``self.mean``.\n",
      " |      `sigma_fac`:\n",
      " |          multiplier for sampling width, standard deviation, for example\n",
      " |          to get a small perturbation of solution `xmean`\n",
      " |      `evaluations`:\n",
      " |          number of evaluations for each sampled solution\n",
      " |      `aggregation`:\n",
      " |          function that aggregates `evaluations` values to\n",
      " |          as single value.\n",
      " |      `kappa`:\n",
      " |          multiplier used for the evaluation of the solutions, in\n",
      " |          that ``func(m + kappa*(x - m))`` is the f-value for ``x``.\n",
      " |      \n",
      " |      Return\n",
      " |      ------\n",
      " |      ``(X, fit)``, where\n",
      " |      \n",
      " |      - `X`: list of solutions\n",
      " |      - `fit`: list of respective function values\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      While ``not self.is_feasible(x, func(x))`` new solutions are\n",
      " |      sampled. By default ``self.is_feasible == cma.feasible == lambda x, f: f not in (None, np.NaN)``.\n",
      " |      The argument to `func` can be freely modified within `func`.\n",
      " |      \n",
      " |      Depending on the ``CMA_mirrors`` option, some solutions are not\n",
      " |      sampled independently but as mirrors of other bad solutions. This\n",
      " |      is a simple derandomization that can save 10-30% of the\n",
      " |      evaluations in particular with small populations, for example on\n",
      " |      the cigar function.\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      >>> import cma\n",
      " |      >>> x0, sigma0 = 8 * [10], 1  # 8-D\n",
      " |      >>> es = cma.CMAEvolutionStrategy(x0, sigma0)  #doctest: +ELLIPSIS\n",
      " |      (5_w,...\n",
      " |      >>> while not es.stop():\n",
      " |      ...     X, fit = es.ask_and_eval(cma.ff.elli)  # handles NaN with resampling\n",
      " |      ...     es.tell(X, fit)  # pass on fitness values\n",
      " |      ...     es.disp(20) # print every 20-th iteration  #doctest: +ELLIPSIS\n",
      " |      Iterat #Fevals...\n",
      " |      >>> print('terminated on ' + str(es.stop()))  #doctest: +ELLIPSIS\n",
      " |      terminated on ...\n",
      " |      \n",
      " |      A single iteration step can be expressed in one line, such that\n",
      " |      an entire optimization after initialization becomes::\n",
      " |      \n",
      " |          while not es.stop():\n",
      " |              es.tell(*es.ask_and_eval(cma.ff.elli))\n",
      " |  \n",
      " |  ask_geno(self, number=None, xmean=None, sigma_fac=1)\n",
      " |      get new candidate solutions in genotyp.\n",
      " |      \n",
      " |      Solutions are sampled from a multi-variate normal distribution.\n",
      " |      \n",
      " |      Arguments are\n",
      " |          `number`\n",
      " |              number of returned solutions, by default the\n",
      " |              population size `popsize` (AKA lambda).\n",
      " |          `xmean`\n",
      " |              distribution mean\n",
      " |          `sigma_fac`\n",
      " |              multiplier for internal sample width (standard\n",
      " |              deviation)\n",
      " |      \n",
      " |      `ask_geno` returns a list of N-dimensional candidate solutions\n",
      " |      in genotyp representation and is called by `ask`.\n",
      " |      \n",
      " |      Details: updates the sample distribution if needed and might\n",
      " |      change the geno-pheno transformation during this update.\n",
      " |      \n",
      " |      :See: `ask`, `ask_and_eval`\n",
      " |  \n",
      " |  disp(self, modulo=None)\n",
      " |      print current state variables in a single-line.\n",
      " |      \n",
      " |      Prints only if ``iteration_counter % modulo == 0``.\n",
      " |      \n",
      " |      :See also: `disp_annotation`.\n",
      " |  \n",
      " |  disp_annotation(self)\n",
      " |      print annotation line for `disp` ()\n",
      " |  \n",
      " |  feed_for_resume(self, X, function_values)\n",
      " |      Resume a run using the solution history.\n",
      " |      \n",
      " |      CAVEAT: this hasn't been thoroughly tested or in intensive use.\n",
      " |      \n",
      " |      Given all \"previous\" candidate solutions and their respective\n",
      " |      function values, the state of a `CMAEvolutionStrategy` object\n",
      " |      can be reconstructed from this history. This is the purpose of\n",
      " |      function `feed_for_resume`.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |      `X`:\n",
      " |        (all) solution points in chronological order, phenotypic\n",
      " |        representation. The number of points must be a multiple\n",
      " |        of popsize.\n",
      " |      `function_values`:\n",
      " |        respective objective function values\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      `feed_for_resume` can be called repeatedly with only parts of\n",
      " |      the history. The part must have the length of a multiple\n",
      " |      of the population size.\n",
      " |      `feed_for_resume` feeds the history in popsize-chunks into `tell`.\n",
      " |      The state of the random number generator might not be\n",
      " |      reconstructed, but this would be only relevant for the future.\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      ::\n",
      " |      \n",
      " |          import cma\n",
      " |      \n",
      " |          # prepare\n",
      " |          (x0, sigma0) = ... # initial values from previous trial\n",
      " |          X = ... # list of generated solutions from a previous trial\n",
      " |          f = ... # respective list of f-values\n",
      " |      \n",
      " |          # resume\n",
      " |          es = cma.CMAEvolutionStrategy(x0, sigma0)\n",
      " |          es.feed_for_resume(X, f)\n",
      " |      \n",
      " |          # continue with func as objective function\n",
      " |          while not es.stop():\n",
      " |              X = es.ask()\n",
      " |              es.tell(X, [func(x) for x in X])\n",
      " |      \n",
      " |      \n",
      " |      Credits to Dirk Bueche and Fabrice Marchal for the feeding idea.\n",
      " |      \n",
      " |      :See also: class `CMAEvolutionStrategy` for a simple dump/load\n",
      " |          to resume.\n",
      " |  \n",
      " |  get_mirror(self, x, preserve_length=False)\n",
      " |      return ``pheno(self.mean - (geno(x) - self.mean))``.\n",
      " |      \n",
      " |      >>> import numpy as np, cma\n",
      " |      >>> es = cma.CMAEvolutionStrategy(np.random.randn(3), 1)  #doctest: +ELLIPSIS\n",
      " |      (3_w,...\n",
      " |      >>> x = np.random.randn(3)\n",
      " |      >>> assert cma.utilities.math.Mh.vequals_approximately(es.mean - (x - es.mean), es.get_mirror(x, preserve_length=True))\n",
      " |      >>> x = es.ask(1)[0]\n",
      " |      >>> vals = (es.get_mirror(x) - es.mean) / (x - es.mean)\n",
      " |      >>> assert cma.utilities.math.Mh.equals_approximately(sum(vals), len(vals) * vals[0])\n",
      " |      \n",
      " |      TODO: this implementation is yet experimental.\n",
      " |      \n",
      " |      TODO: this implementation includes geno-pheno transformation,\n",
      " |      however in general GP-transformation should be separated from\n",
      " |      specific code.\n",
      " |      \n",
      " |      Selectively mirrored sampling improves to a moderate extend but\n",
      " |      overadditively with active CMA for quite understandable reasons.\n",
      " |      \n",
      " |      Optimal number of mirrors are suprisingly small: 1,2,3 for\n",
      " |      maxlam=7,13,20 where 3,6,10 are the respective maximal possible\n",
      " |      mirrors that must be clearly suboptimal.\n",
      " |  \n",
      " |  get_selective_mirrors(self, number=None)\n",
      " |      get mirror genotypic directions from worst solutions.\n",
      " |      \n",
      " |      Details:\n",
      " |      \n",
      " |      To be called after the mean has been updated.\n",
      " |      \n",
      " |      Takes the last ``number=sp.lam_mirr`` entries in the\n",
      " |      ``self.pop[self.fit.idx]`` as solutions to be mirrored.\n",
      " |      \n",
      " |      Do not take a mirror if it is suspected to stem from a\n",
      " |      previous mirror in order to not go endlessly back and forth.\n",
      " |  \n",
      " |  inject(self, solutions, force=None)\n",
      " |      inject list of one or several genotypic solution(s).\n",
      " |      \n",
      " |      Unless `force is True`, the solutions are used as direction\n",
      " |      relative to the distribution mean to compute a new candidate\n",
      " |      solution returned in method `ask_geno` which in turn is used in\n",
      " |      method `ask`. `inject` is to be called before `ask` or after\n",
      " |      `tell` and can be called repeatedly.\n",
      " |      \n",
      " |      >>> import cma\n",
      " |      >>> es = cma.CMAEvolutionStrategy(4 * [1], 2)  #doctest: +ELLIPSIS\n",
      " |      (4_w,...\n",
      " |      >>> while not es.stop():\n",
      " |      ...     es.inject([4 * [0.0]])\n",
      " |      ...     X = es.ask()\n",
      " |      ...     if es.countiter == 0:\n",
      " |      ...         assert X[0][0] == X[0][1]  # injected sol. is on the diagonal\n",
      " |      ...     es.tell(X, [cma.ff.sphere(x) for x in X])\n",
      " |  \n",
      " |  mahalanobis_norm(self, dx)\n",
      " |      return Mahalanobis norm based on the current sample\n",
      " |      distribution.\n",
      " |      \n",
      " |      The norm is based on Covariance matrix ``C`` times ``sigma**2``,\n",
      " |      and includes ``sigma_vec``. The expected Mahalanobis distance to\n",
      " |      the sample mean is about ``sqrt(dimension)``.\n",
      " |      \n",
      " |      Argument\n",
      " |      --------\n",
      " |      A *genotype* difference `dx`.\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      >>> import cma, numpy\n",
      " |      >>> es = cma.CMAEvolutionStrategy(numpy.ones(10), 1)  #doctest: +ELLIPSIS\n",
      " |      (5_w,...\n",
      " |      >>> xx = numpy.random.randn(2, 10)\n",
      " |      >>> d = es.mahalanobis_norm(es.gp.geno(xx[0]-xx[1]))\n",
      " |      \n",
      " |      `d` is the distance \"in\" the true sample distribution,\n",
      " |      sampled points have a typical distance of ``sqrt(2*es.N)``,\n",
      " |      where ``es.N`` is the dimension, and an expected distance of\n",
      " |      close to ``sqrt(N)`` to the sample mean. In the example,\n",
      " |      `d` is the Euclidean distance, because C = I and sigma = 1.\n",
      " |  \n",
      " |  manage_plateaus(self, sigma_fac=1.5, sample_fraction=0.5)\n",
      " |      increase `sigma` by `sigma_fac` in case of a plateau.\n",
      " |      \n",
      " |      A plateau is assumed to be present if the best sample and\n",
      " |      ``popsize * sample_fraction``-th best sample have the same\n",
      " |      fitness.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      >>> import cma\n",
      " |      >>> def f(X):\n",
      " |      ...     return (len(X) - 1) * [1] + [2]\n",
      " |      >>> es = cma.CMAEvolutionStrategy(4 * [0], 1)  #doctest: +ELLIPSIS\n",
      " |      (4_w,...\n",
      " |      >>> while not es.stop():\n",
      " |      ...     X = es.ask()\n",
      " |      ...     es.tell(X, f(X))\n",
      " |      ...     es.logger.add()\n",
      " |      ...     es.manage_plateaus()\n",
      " |      >>> assert es.sigma > 1.5**5\n",
      " |  \n",
      " |  plot(self)\n",
      " |      plot current state variables using `matplotlib`.\n",
      " |      \n",
      " |      Details: calls `self.logger.plot`.\n",
      " |  \n",
      " |  random_rescale_to_mahalanobis(self, x)\n",
      " |      change `x` like for injection, all on genotypic level\n",
      " |  \n",
      " |  repair_genotype(self, x, copy_if_changed=False)\n",
      " |      make sure that solutions fit to the sample distribution.\n",
      " |      \n",
      " |      This interface is versatile and likely to change.\n",
      " |      \n",
      " |      In particular the frequency of ``x - self.mean`` being long in\n",
      " |      Mahalanobis distance is limited, currently clipping at\n",
      " |      ``N**0.5 + 2 * N / (N + 2)`` is implemented.\n",
      " |  \n",
      " |  result_pretty(self, number_of_runs=0, time_str=None, fbestever=None)\n",
      " |      pretty print result.\n",
      " |      \n",
      " |      Returns `result` of ``self``.\n",
      " |  \n",
      " |  stop(self, check=True, ignore_list=())\n",
      " |      return the termination status as dictionary.\n",
      " |      \n",
      " |      With ``check==False``, the termination conditions are not checked\n",
      " |      and the status might not reflect the current situation.\n",
      " |      ``stop().clear()`` removes the currently active termination\n",
      " |      conditions.\n",
      " |      \n",
      " |      As a convenience feature, keywords in `ignore_list` are removed from\n",
      " |      the conditions.\n",
      " |  \n",
      " |  tell(self, solutions, function_values, check_points=None, copy=False)\n",
      " |      pass objective function values to prepare for next\n",
      " |      iteration. This core procedure of the CMA-ES algorithm updates\n",
      " |      all state variables, in particular the two evolution paths, the\n",
      " |      distribution mean, the covariance matrix and a step-size.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |      `solutions`\n",
      " |          list or array of candidate solution points (of\n",
      " |          type `numpy.ndarray`), most presumably before\n",
      " |          delivered by method `ask()` or `ask_and_eval()`.\n",
      " |      `function_values`\n",
      " |          list or array of objective function values\n",
      " |          corresponding to the respective points. Beside for termination\n",
      " |          decisions, only the ranking of values in `function_values`\n",
      " |          is used.\n",
      " |      `check_points`\n",
      " |          If ``check_points is None``, only solutions that are not generated\n",
      " |          by `ask()` are possibly clipped (recommended). ``False`` does not clip\n",
      " |          any solution (not recommended).\n",
      " |          If ``True``, clips solutions that realize long steps (i.e. also\n",
      " |          those that are unlikely to be generated with `ask()`). `check_points`\n",
      " |          can be a list of indices to be checked in solutions.\n",
      " |      `copy`\n",
      " |          ``solutions`` can be modified in this routine, if ``copy is False``\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      `tell()` updates the parameters of the multivariate\n",
      " |      normal search distribution, namely covariance matrix and\n",
      " |      step-size and updates also the attributes ``countiter`` and\n",
      " |      ``countevals``. To check the points for consistency is quadratic\n",
      " |      in the dimension (like sampling points).\n",
      " |      \n",
      " |      Bugs\n",
      " |      ----\n",
      " |      The effect of changing the solutions delivered by `ask()`\n",
      " |      depends on whether boundary handling is applied. With boundary\n",
      " |      handling, modifications are disregarded. This is necessary to\n",
      " |      apply the default boundary handling that uses unrepaired\n",
      " |      solutions but might change in future.\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      \n",
      " |      >>> import cma\n",
      " |      >>> func = cma.ff.sphere  # choose objective function\n",
      " |      >>> es = cma.CMAEvolutionStrategy(np.random.rand(2) / 3, 1.5)\n",
      " |      ... # doctest:+ELLIPSIS\n",
      " |      (3_...\n",
      " |      >>> while not es.stop():\n",
      " |      ...    X = es.ask()\n",
      " |      ...    es.tell(X, [func(x) for x in X])\n",
      " |      >>> es.result  # result is a `namedtuple` # doctest:+ELLIPSIS\n",
      " |      CMAEvolutionStrategyResult(xbest=array([...\n",
      " |      \n",
      " |      :See: class `CMAEvolutionStrategy`, `ask`, `ask_and_eval`, `fmin`\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors defined here:\n",
      " |  \n",
      " |  condition_number\n",
      " |      condition number of the statistical-model sampler.\n",
      " |      \n",
      " |      Details: neither encoding/decoding from `sigma_vec`-scaling nor\n",
      " |      `gp`-transformation are taken into account for this computation.\n",
      " |  \n",
      " |  isotropic_mean_shift\n",
      " |      normalized last mean shift, under random selection N(0,I)\n",
      " |      \n",
      " |      distributed.\n",
      " |      \n",
      " |      Caveat: while it is finite and close to sqrt(n) under random\n",
      " |      selection, the length of the normalized mean shift under\n",
      " |      *systematic* selection (e.g. on a linear function) tends to\n",
      " |      infinity for mueff -> infty. Hence it must be used with great\n",
      " |      care for large mueff.\n",
      " |  \n",
      " |  popsize\n",
      " |      number of samples by default returned by `ask` ()\n",
      " |  \n",
      " |  result\n",
      " |      return a `CMAEvolutionStrategyResult` `namedtuple`.\n",
      " |      \n",
      " |      :See: `cma.evolution_strategy.CMAEvolutionStrategyResult`\n",
      " |          or try ``help(...result)`` on the ``result`` property\n",
      " |          of an `CMAEvolutionStrategy` instance or on the\n",
      " |          `CMAEvolutionStrategyResult` instance itself.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from cma.interfaces.OOOptimizer:\n",
      " |  \n",
      " |  initialize(self)\n",
      " |      (re-)set to the initial state\n",
      " |  \n",
      " |  optimize(self, objective_fct, maxfun=None, iterations=None, min_iterations=1, args=(), verb_disp=None, callback=None)\n",
      " |      find minimizer of ``objective_fct``.\n",
      " |      \n",
      " |      CAVEAT: the return value for `optimize` has changed to ``self``,\n",
      " |      allowing for a call like::\n",
      " |      \n",
      " |          solver = OOOptimizer(x0).optimize(f)\n",
      " |      \n",
      " |      and investigate the state of the solver.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |      \n",
      " |      ``objective_fct``: f(x: array_like) -> float\n",
      " |          function be to minimized\n",
      " |      ``maxfun``: number\n",
      " |          maximal number of function evaluations\n",
      " |      ``iterations``: number\n",
      " |          number of (maximal) iterations, while ``not self.stop()``,\n",
      " |          it can be useful to conduct only one iteration at a time.\n",
      " |      ``min_iterations``: number\n",
      " |          minimal number of iterations, even if ``not self.stop()``\n",
      " |      ``args``: sequence_like\n",
      " |          arguments passed to ``objective_fct``\n",
      " |      ``verb_disp``: number\n",
      " |          print to screen every ``verb_disp`` iteration, if `None`\n",
      " |          the value from ``self.logger`` is \"inherited\", if\n",
      " |          available.\n",
      " |      ``callback``: callable or list of callables\n",
      " |          callback function called like ``callback(self)`` or\n",
      " |          a list of call back functions called in the same way. If\n",
      " |          available, ``self.logger.add`` is added to this list.\n",
      " |          TODO: currently there is no way to prevent this other than\n",
      " |          changing the code of `_prepare_callback_list`.\n",
      " |      \n",
      " |      ``return self``, that is, the `OOOptimizer` instance.\n",
      " |      \n",
      " |      Example\n",
      " |      -------\n",
      " |      >>> import cma\n",
      " |      >>> es = cma.CMAEvolutionStrategy(7 * [0.1], 0.1\n",
      " |      ...              ).optimize(cma.ff.rosen, verb_disp=100)\n",
      " |      ...                   #doctest: +ELLIPSIS\n",
      " |      (4_w,9)-aCMA-ES (mu_w=2.8,w_1=49%) in dimension 7 (seed=...)\n",
      " |      Iterat #Fevals   function value  axis ratio  sigma ...\n",
      " |          1      9 ...\n",
      " |          2     18 ...\n",
      " |          3     27 ...\n",
      " |        100    900 ...\n",
      " |      >>> cma.s.Mh.vequals_approximately(es.result[0], 7 * [1], 1e-5)\n",
      " |      True\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from cma.interfaces.OOOptimizer:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(cma.CMAEvolutionStrategy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class CMAOptions in module cma.evolution_strategy:\n",
      "\n",
      "class CMAOptions(builtins.dict)\n",
      " |  CMAOptions(s=None, **kwargs)\n",
      " |  \n",
      " |  a dictionary with the available options and their default values\n",
      " |  for class `CMAEvolutionStrategy`.\n",
      " |  \n",
      " |  ``CMAOptions()`` returns a `dict` with all available options and their\n",
      " |  default values with a comment string.\n",
      " |  \n",
      " |  ``CMAOptions('verb')`` returns a subset of recognized options that\n",
      " |  contain 'verb' in there keyword name or (default) value or\n",
      " |  description.\n",
      " |  \n",
      " |  ``CMAOptions(opts)`` returns the subset of recognized options in\n",
      " |  ``dict(opts)``.\n",
      " |  \n",
      " |  Option values can be \"written\" in a string and, when passed to `fmin`\n",
      " |  or `CMAEvolutionStrategy`, are evaluated using \"N\" and \"popsize\" as\n",
      " |  known values for dimension and population size (sample size, number\n",
      " |  of new solutions per iteration). All default option values are given\n",
      " |  as such a string.\n",
      " |  \n",
      " |  Details\n",
      " |  -------\n",
      " |  `CMAOptions` entries starting with ``tol`` are termination\n",
      " |  \"tolerances\".\n",
      " |  \n",
      " |  For `tolstagnation`, the median over the first and the second half\n",
      " |  of at least `tolstagnation` iterations are compared for both, the\n",
      " |  per-iteration best and per-iteration median function value.\n",
      " |  \n",
      " |  Example\n",
      " |  -------\n",
      " |  ::\n",
      " |  \n",
      " |      import cma\n",
      " |      cma.CMAOptions('tol')\n",
      " |  \n",
      " |  is a shortcut for ``cma.CMAOptions().match('tol')`` that returns all\n",
      " |  options that contain 'tol' in their name or description.\n",
      " |  \n",
      " |  To set an option::\n",
      " |  \n",
      " |      import cma\n",
      " |      opts = cma.CMAOptions()\n",
      " |      opts.set('tolfun', 1e-12)\n",
      " |      opts['tolx'] = 1e-11\n",
      " |  \n",
      " |  todo: this class is overly complex and should be re-written, possibly\n",
      " |  with reduced functionality.\n",
      " |  \n",
      " |  :See also: `fmin` (), `CMAEvolutionStrategy`, `_CMAParameters`\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      CMAOptions\n",
      " |      builtins.dict\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __call__(self, key, default=None, loc=None)\n",
      " |      evaluate and return the value of option `key` on the fly, or\n",
      " |      return those options whose name or description contains `key`,\n",
      " |      case disregarded.\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      Keys that contain `filename` are not evaluated.\n",
      " |      For ``loc==None``, `self` is used as environment\n",
      " |      but this does not define ``N``.\n",
      " |      \n",
      " |      :See: `eval()`, `evalall()`\n",
      " |  \n",
      " |  __init__(self, s=None, **kwargs)\n",
      " |      return an `CMAOptions` instance.\n",
      " |      \n",
      " |      Return default options if ``s is None and not kwargs``,\n",
      " |      or all options whose name or description contains `s`, if\n",
      " |      `s` is a (search) string (case is disregarded in the match),\n",
      " |      or with entries from dictionary `s` as options,\n",
      " |      or with kwargs as options if ``s is None``,\n",
      " |      in any of the latter cases not complemented with default options\n",
      " |      or settings.\n",
      " |      \n",
      " |      Returns: see above.\n",
      " |      \n",
      " |      Details: as several options start with ``'s'``, ``s=value`` is\n",
      " |      not valid as an option setting.\n",
      " |  \n",
      " |  check(self, options=None)\n",
      " |      check for ambiguous keys and move attributes into dict\n",
      " |  \n",
      " |  check_attributes(self, opts=None)\n",
      " |      check for attributes and moves them into the dictionary\n",
      " |  \n",
      " |  check_values(self, options=None)\n",
      " |  \n",
      " |  complement(self)\n",
      " |      add all missing options with their default values\n",
      " |  \n",
      " |  corrected_key(self, key)\n",
      " |      return the matching valid key, if ``key.lower()`` is a unique\n",
      " |      starting sequence to identify the valid key, ``else None``\n",
      " |  \n",
      " |  eval(self, key, default=None, loc=None, correct_key=True)\n",
      " |      Evaluates and sets the specified option value in\n",
      " |      environment `loc`. Many options need ``N`` to be defined in\n",
      " |      `loc`, some need `popsize`.\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      Keys that contain 'filename' are not evaluated.\n",
      " |      For `loc` is None, the self-dict is used as environment\n",
      " |      \n",
      " |      :See: `evalall()`, `__call__`\n",
      " |  \n",
      " |  evalall(self, loc=None, defaults=None)\n",
      " |      Evaluates all option values in environment `loc`.\n",
      " |      \n",
      " |      :See: `eval()`\n",
      " |  \n",
      " |  from_namedtuple(self, t)\n",
      " |      update options from a `collections.namedtuple`.\n",
      " |      :See also: `to_namedtuple`\n",
      " |  \n",
      " |  init(self, dict_or_str, val=None, warn=True)\n",
      " |      initialize one or several options.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |          `dict_or_str`\n",
      " |              a dictionary if ``val is None``, otherwise a key.\n",
      " |              If `val` is provided `dict_or_str` must be a valid key.\n",
      " |          `val`\n",
      " |              value for key\n",
      " |      \n",
      " |      Details\n",
      " |      -------\n",
      " |      Only known keys are accepted. Known keys are in `CMAOptions.defaults()`\n",
      " |  \n",
      " |  match(self, s='')\n",
      " |      return all options that match, in the name or the description,\n",
      " |      with string `s`, case is disregarded.\n",
      " |      \n",
      " |      Example: ``cma.CMAOptions().match('verb')`` returns the verbosity\n",
      " |      options.\n",
      " |  \n",
      " |  pprint(self, linebreak=80)\n",
      " |  \n",
      " |  set(self, dic, val=None, force=False)\n",
      " |      assign versatile options.\n",
      " |      \n",
      " |      Method `CMAOptions.versatile_options` () gives the versatile\n",
      " |      options, use `init()` to set the others.\n",
      " |      \n",
      " |      Arguments\n",
      " |      ---------\n",
      " |          `dic`\n",
      " |              either a dictionary or a key. In the latter\n",
      " |              case, `val` must be provided\n",
      " |          `val`\n",
      " |              value for `key`, approximate match is sufficient\n",
      " |          `force`\n",
      " |              force setting of non-versatile options, use with caution\n",
      " |      \n",
      " |      This method will be most probably used with the ``opts`` attribute of\n",
      " |      a `CMAEvolutionStrategy` instance.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Static methods defined here:\n",
      " |  \n",
      " |  defaults()\n",
      " |      return a dictionary with default option values and description\n",
      " |  \n",
      " |  versatile_options()\n",
      " |      return list of options that can be changed at any time (not\n",
      " |      only be initialized).\n",
      " |      \n",
      " |      Consider that this list might not be entirely up\n",
      " |      to date.\n",
      " |      \n",
      " |      The string ' #v ' in the default value indicates a versatile\n",
      " |      option that can be changed any time, however a string will not \n",
      " |      necessarily be evaluated again.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors defined here:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      " |  \n",
      " |  settable\n",
      " |      return the subset of those options that are settable at any\n",
      " |      time.\n",
      " |      \n",
      " |      Settable options are in `versatile_options` (), but the\n",
      " |      list might be incomplete.\n",
      " |  \n",
      " |  to_namedtuple\n",
      " |      return options as const attributes of the returned object,\n",
      " |      only useful for inspection.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from builtins.dict:\n",
      " |  \n",
      " |  __contains__(self, key, /)\n",
      " |      True if the dictionary has the specified key, else False.\n",
      " |  \n",
      " |  __delitem__(self, key, /)\n",
      " |      Delete self[key].\n",
      " |  \n",
      " |  __eq__(self, value, /)\n",
      " |      Return self==value.\n",
      " |  \n",
      " |  __ge__(self, value, /)\n",
      " |      Return self>=value.\n",
      " |  \n",
      " |  __getattribute__(self, name, /)\n",
      " |      Return getattr(self, name).\n",
      " |  \n",
      " |  __getitem__(...)\n",
      " |      x.__getitem__(y) <==> x[y]\n",
      " |  \n",
      " |  __gt__(self, value, /)\n",
      " |      Return self>value.\n",
      " |  \n",
      " |  __iter__(self, /)\n",
      " |      Implement iter(self).\n",
      " |  \n",
      " |  __le__(self, value, /)\n",
      " |      Return self<=value.\n",
      " |  \n",
      " |  __len__(self, /)\n",
      " |      Return len(self).\n",
      " |  \n",
      " |  __lt__(self, value, /)\n",
      " |      Return self<value.\n",
      " |  \n",
      " |  __ne__(self, value, /)\n",
      " |      Return self!=value.\n",
      " |  \n",
      " |  __repr__(self, /)\n",
      " |      Return repr(self).\n",
      " |  \n",
      " |  __setitem__(self, key, value, /)\n",
      " |      Set self[key] to value.\n",
      " |  \n",
      " |  __sizeof__(...)\n",
      " |      D.__sizeof__() -> size of D in memory, in bytes\n",
      " |  \n",
      " |  clear(...)\n",
      " |      D.clear() -> None.  Remove all items from D.\n",
      " |  \n",
      " |  copy(...)\n",
      " |      D.copy() -> a shallow copy of D\n",
      " |  \n",
      " |  get(self, key, default=None, /)\n",
      " |      Return the value for key if key is in the dictionary, else default.\n",
      " |  \n",
      " |  items(...)\n",
      " |      D.items() -> a set-like object providing a view on D's items\n",
      " |  \n",
      " |  keys(...)\n",
      " |      D.keys() -> a set-like object providing a view on D's keys\n",
      " |  \n",
      " |  pop(...)\n",
      " |      D.pop(k[,d]) -> v, remove specified key and return the corresponding value.\n",
      " |      If key is not found, d is returned if given, otherwise KeyError is raised\n",
      " |  \n",
      " |  popitem(...)\n",
      " |      D.popitem() -> (k, v), remove and return some (key, value) pair as a\n",
      " |      2-tuple; but raise KeyError if D is empty.\n",
      " |  \n",
      " |  setdefault(self, key, default=None, /)\n",
      " |      Insert key with a value of default if key is not in the dictionary.\n",
      " |      \n",
      " |      Return the value for key if key is in the dictionary, else default.\n",
      " |  \n",
      " |  update(...)\n",
      " |      D.update([E, ]**F) -> None.  Update D from dict/iterable E and F.\n",
      " |      If E is present and has a .keys() method, then does:  for k in E: D[k] = E[k]\n",
      " |      If E is present and lacks a .keys() method, then does:  for k, v in E: D[k] = v\n",
      " |      In either case, this is followed by: for k in F:  D[k] = F[k]\n",
      " |  \n",
      " |  values(...)\n",
      " |      D.values() -> an object providing a view on D's values\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Class methods inherited from builtins.dict:\n",
      " |  \n",
      " |  fromkeys(iterable, value=None, /) from builtins.type\n",
      " |      Create a new dictionary with keys from iterable and values set to value.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Static methods inherited from builtins.dict:\n",
      " |  \n",
      " |  __new__(*args, **kwargs) from builtins.type\n",
      " |      Create and return a new object.  See help(type) for accurate signature.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data and other attributes inherited from builtins.dict:\n",
      " |  \n",
      " |  __hash__ = None\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(cma.CMAOptions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    return np.sum(np.square(x)) if np.sum(x) > 1 else sys.maxsize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5_w,10)-aCMA-ES (mu_w=3.2,w_1=45%) in dimension 10 (seed=334174, Fri Dec 13 11:30:45 2019)\n"
     ]
    }
   ],
   "source": [
    "es = cma.CMAEvolutionStrategy(10*[1], 0.5, {'bounds': [0, 2], })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "    1     10 9.297956451173906e+00 1.0e+00 5.03e-01  5e-01  5e-01 0:00.0\n",
      "    2     20 9.087294198145610e+00 1.2e+00 5.19e-01  5e-01  6e-01 0:00.0\n",
      "    3     30 6.290086169271653e+00 1.4e+00 5.26e-01  5e-01  6e-01 0:00.0\n",
      "    4     40 3.667303647081697e+00 1.4e+00 5.47e-01  5e-01  6e-01 0:00.0\n",
      "    5     50 4.192620837918611e+00 1.5e+00 5.09e-01  5e-01  6e-01 0:00.0\n",
      "    6     60 3.587261970368212e+00 1.5e+00 4.95e-01  4e-01  5e-01 0:00.0\n",
      "    7     70 1.607232770475297e+00 1.6e+00 4.96e-01  4e-01  5e-01 0:00.0\n",
      "    8     80 1.184372448509091e+00 1.6e+00 5.33e-01  5e-01  6e-01 0:00.0\n",
      "    9     90 1.893474363349450e+00 1.6e+00 5.15e-01  5e-01  6e-01 0:00.0\n",
      "   10    100 1.571998262725276e+00 1.7e+00 5.33e-01  5e-01  6e-01 0:00.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   11    110 1.224087363220818e+00 1.7e+00 5.06e-01  4e-01  5e-01 0:00.0\n",
      "   12    120 1.582857708045483e+00 1.7e+00 4.76e-01  4e-01  5e-01 0:00.1\n",
      "   13    130 8.471542973572985e-01 1.7e+00 4.38e-01  4e-01  5e-01 0:00.1\n",
      "   14    140 8.683259949306991e-01 1.7e+00 3.92e-01  3e-01  4e-01 0:00.1\n",
      "   15    150 6.038187793289841e-01 1.7e+00 3.34e-01  3e-01  3e-01 0:00.1\n",
      "   16    160 5.134926324277578e-01 1.7e+00 3.17e-01  3e-01  3e-01 0:00.1\n",
      "   17    170 5.819810392939927e-01 1.7e+00 2.98e-01  2e-01  3e-01 0:00.1\n",
      "   18    180 4.860969025153034e-01 1.8e+00 2.81e-01  2e-01  3e-01 0:00.1\n",
      "   19    190 3.060014797762367e-01 1.8e+00 2.64e-01  2e-01  3e-01 0:00.1\n",
      "   20    200 2.583975224358599e-01 1.8e+00 2.48e-01  2e-01  2e-01 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   21    210 3.065303396279793e-01 1.8e+00 2.48e-01  2e-01  2e-01 0:00.1\n",
      "   22    220 2.378785835335247e-01 1.9e+00 2.34e-01  2e-01  2e-01 0:00.1\n",
      "   23    230 2.796858633473913e-01 1.9e+00 2.21e-01  2e-01  2e-01 0:00.1\n",
      "   24    240 2.966276390570002e-01 2.0e+00 2.05e-01  1e-01  2e-01 0:00.1\n",
      "   25    250 2.846727656775501e-01 2.0e+00 2.02e-01  1e-01  2e-01 0:00.1\n",
      "   26    260 3.454620352818135e-01 2.0e+00 2.00e-01  1e-01  2e-01 0:00.1\n",
      "   27    270 3.006090069651518e-01 2.1e+00 1.93e-01  1e-01  2e-01 0:00.1\n",
      "   28    280 2.735018268825549e-01 2.2e+00 1.86e-01  1e-01  2e-01 0:00.1\n",
      "   29    290 2.572980987170741e-01 2.2e+00 1.63e-01  1e-01  2e-01 0:00.1\n",
      "   30    300 2.428679398166575e-01 2.2e+00 1.39e-01  9e-02  1e-01 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   31    310 2.157788169117849e-01 2.3e+00 1.28e-01  8e-02  1e-01 0:00.1\n",
      "   32    320 1.767449868748301e-01 2.3e+00 1.24e-01  8e-02  1e-01 0:00.1\n",
      "   33    330 2.121848840209556e-01 2.3e+00 1.18e-01  8e-02  1e-01 0:00.1\n",
      "   34    340 2.018362762257917e-01 2.4e+00 1.19e-01  8e-02  1e-01 0:00.1\n",
      "   35    350 1.897939373221335e-01 2.3e+00 1.25e-01  8e-02  1e-01 0:00.1\n",
      "   36    360 1.428010714044490e-01 2.2e+00 1.25e-01  9e-02  1e-01 0:00.1\n",
      "   37    370 1.849082751805467e-01 2.2e+00 1.21e-01  8e-02  1e-01 0:00.1\n",
      "   38    380 1.869698588603876e-01 2.2e+00 1.15e-01  8e-02  1e-01 0:00.1\n",
      "   39    390 1.591594829666638e-01 2.3e+00 1.05e-01  7e-02  9e-02 0:00.1\n",
      "   40    400 1.436471123592584e-01 2.3e+00 9.53e-02  6e-02  8e-02 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   41    410 1.539082722044897e-01 2.3e+00 8.87e-02  6e-02  7e-02 0:00.1\n",
      "   42    420 1.564829844389007e-01 2.3e+00 8.42e-02  5e-02  7e-02 0:00.2\n",
      "   43    430 1.570653234465751e-01 2.3e+00 7.94e-02  5e-02  7e-02 0:00.2\n",
      "   44    440 1.311606171839707e-01 2.4e+00 7.45e-02  5e-02  6e-02 0:00.2\n",
      "   45    450 1.271597184623101e-01 2.4e+00 7.23e-02  5e-02  6e-02 0:00.2\n",
      "   46    460 1.361963439982980e-01 2.5e+00 6.67e-02  4e-02  6e-02 0:00.2\n",
      "   47    470 1.337610971703757e-01 2.5e+00 6.33e-02  4e-02  5e-02 0:00.2\n",
      "   48    480 1.324612969448034e-01 2.5e+00 5.91e-02  4e-02  5e-02 0:00.2\n",
      "   49    490 1.364174716322156e-01 2.6e+00 5.53e-02  3e-02  5e-02 0:00.2\n",
      "   50    500 1.195099441034577e-01 2.6e+00 5.56e-02  3e-02  5e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   51    510 1.322237940175673e-01 2.7e+00 5.93e-02  4e-02  5e-02 0:00.2\n",
      "   52    520 1.326473647703893e-01 2.7e+00 6.09e-02  4e-02  5e-02 0:00.2\n",
      "   53    530 1.320761007816779e-01 2.9e+00 5.88e-02  3e-02  5e-02 0:00.2\n",
      "   54    540 1.346549355762332e-01 3.1e+00 5.63e-02  3e-02  5e-02 0:00.2\n",
      "   55    550 1.183081522945345e-01 3.1e+00 5.35e-02  3e-02  5e-02 0:00.2\n",
      "   56    560 1.259340596998074e-01 3.1e+00 5.16e-02  3e-02  5e-02 0:00.2\n",
      "   57    570 1.115476207695553e-01 3.1e+00 5.11e-02  3e-02  5e-02 0:00.2\n",
      "   58    580 1.157548003166973e-01 3.0e+00 5.04e-02  3e-02  4e-02 0:00.2\n",
      "   59    590 1.089025265128796e-01 3.0e+00 4.86e-02  3e-02  4e-02 0:00.2\n",
      "   60    600 1.165203811403989e-01 3.0e+00 4.88e-02  3e-02  4e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   61    610 1.146043031029011e-01 3.0e+00 4.59e-02  3e-02  4e-02 0:00.2\n",
      "   62    620 1.158905698617575e-01 3.1e+00 4.27e-02  2e-02  4e-02 0:00.2\n",
      "   63    630 1.110315128437463e-01 3.1e+00 4.07e-02  2e-02  4e-02 0:00.2\n",
      "   64    640 1.074180086469030e-01 3.2e+00 3.84e-02  2e-02  3e-02 0:00.2\n",
      "   65    650 1.097274575205943e-01 3.2e+00 3.42e-02  2e-02  3e-02 0:00.2\n",
      "   66    660 1.086588441497194e-01 3.2e+00 3.51e-02  2e-02  3e-02 0:00.2\n",
      "   67    670 1.083002127494080e-01 3.0e+00 3.68e-02  2e-02  3e-02 0:00.2\n",
      "   68    680 1.082918730334466e-01 3.1e+00 3.54e-02  2e-02  3e-02 0:00.2\n",
      "   69    690 1.104644500939795e-01 3.2e+00 3.75e-02  2e-02  3e-02 0:00.2\n",
      "   70    700 1.100054610065521e-01 3.1e+00 3.76e-02  2e-02  3e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   71    710 1.137607030870662e-01 3.2e+00 3.69e-02  2e-02  3e-02 0:00.2\n",
      "   72    720 1.123005679280383e-01 3.3e+00 3.32e-02  2e-02  3e-02 0:00.2\n",
      "   73    730 1.104134812845367e-01 3.4e+00 3.16e-02  2e-02  3e-02 0:00.2\n",
      "   74    740 1.187291749069536e-01 3.4e+00 3.16e-02  2e-02  3e-02 0:00.2\n",
      "   75    750 1.150491387866202e-01 3.5e+00 2.79e-02  1e-02  2e-02 0:00.2\n",
      "   76    760 1.067848030613011e-01 3.5e+00 2.69e-02  1e-02  2e-02 0:00.2\n",
      "   77    770 1.062552315909907e-01 3.6e+00 2.67e-02  1e-02  2e-02 0:00.2\n",
      "   78    780 1.075032572608926e-01 3.6e+00 2.57e-02  1e-02  2e-02 0:00.2\n",
      "   79    790 1.079883999240529e-01 3.8e+00 2.58e-02  1e-02  2e-02 0:00.2\n",
      "   80    800 1.056408751796233e-01 3.9e+00 2.50e-02  1e-02  2e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   81    810 1.099429193030876e-01 3.7e+00 2.53e-02  1e-02  2e-02 0:00.2\n",
      "   82    820 1.083805366139141e-01 3.7e+00 2.60e-02  1e-02  2e-02 0:00.2\n",
      "   83    830 1.066690551747855e-01 3.7e+00 2.72e-02  1e-02  2e-02 0:00.3\n",
      "   84    840 1.040623235322864e-01 3.9e+00 2.85e-02  1e-02  2e-02 0:00.3\n",
      "   85    850 1.056206313228966e-01 4.2e+00 2.84e-02  1e-02  2e-02 0:00.3\n",
      "   86    860 1.068196327278221e-01 4.3e+00 3.02e-02  1e-02  2e-02 0:00.3\n",
      "   87    870 1.062149375075293e-01 4.4e+00 2.67e-02  1e-02  2e-02 0:00.3\n",
      "   88    880 1.059165652048536e-01 4.4e+00 2.38e-02  1e-02  2e-02 0:00.3\n",
      "   89    890 1.033523796337126e-01 4.6e+00 2.07e-02  1e-02  1e-02 0:00.3\n",
      "   90    900 1.027442301089182e-01 4.5e+00 1.90e-02  9e-03  1e-02 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   91    910 1.027042599616306e-01 4.5e+00 1.75e-02  8e-03  1e-02 0:00.3\n",
      "   92    920 1.035739464758586e-01 4.5e+00 1.58e-02  7e-03  1e-02 0:00.3\n",
      "   93    930 1.025842778129614e-01 4.5e+00 1.46e-02  7e-03  9e-03 0:00.3\n",
      "   94    940 1.033977988152974e-01 4.7e+00 1.29e-02  6e-03  8e-03 0:00.3\n",
      "   95    950 1.028246886460495e-01 4.8e+00 1.18e-02  5e-03  8e-03 0:00.3\n",
      "   96    960 1.029102046328570e-01 4.9e+00 1.15e-02  5e-03  8e-03 0:00.3\n",
      "   97    970 1.019480807768003e-01 5.1e+00 1.18e-02  5e-03  8e-03 0:00.3\n",
      "   98    980 1.018633474049190e-01 5.2e+00 1.15e-02  5e-03  8e-03 0:00.3\n",
      "   99    990 1.016564050903677e-01 5.3e+00 1.19e-02  5e-03  8e-03 0:00.3\n",
      "  100   1000 1.023251458966678e-01 5.6e+00 1.19e-02  5e-03  8e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  101   1010 1.020415842682672e-01 5.8e+00 1.11e-02  5e-03  8e-03 0:00.3\n",
      "  102   1020 1.018950580889092e-01 6.0e+00 1.08e-02  4e-03  7e-03 0:00.3\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  103   1030 1.022252154477178e-01 6.1e+00 1.00e-02  4e-03  7e-03 0:00.3\n",
      "  104   1040 1.025026141757014e-01 6.3e+00 9.28e-03  4e-03  6e-03 0:00.3\n",
      "  105   1050 1.013061283357296e-01 6.3e+00 9.14e-03  4e-03  6e-03 0:00.3\n",
      "  106   1060 1.017824755466077e-01 6.7e+00 8.38e-03  3e-03  6e-03 0:00.3\n",
      "  107   1070 1.014020799625979e-01 6.8e+00 7.59e-03  3e-03  5e-03 0:00.3\n",
      "  108   1080 1.014154851979832e-01 6.9e+00 6.92e-03  3e-03  5e-03 0:00.3\n",
      "  109   1090 1.017724386367850e-01 7.4e+00 6.26e-03  2e-03  4e-03 0:00.3\n",
      "  110   1100 1.009722673272680e-01 7.6e+00 5.89e-03  2e-03  4e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  111   1110 1.012365837309985e-01 7.7e+00 5.71e-03  2e-03  4e-03 0:00.3\n",
      "  112   1120 1.010321400816654e-01 7.8e+00 5.78e-03  2e-03  4e-03 0:00.3\n",
      "  113   1130 1.012972409680950e-01 7.9e+00 5.83e-03  2e-03  4e-03 0:00.3\n",
      "  114   1140 1.008947548923459e-01 8.0e+00 6.03e-03  2e-03  4e-03 0:00.3\n",
      "  115   1150 1.012035477668791e-01 8.2e+00 5.96e-03  2e-03  4e-03 0:00.3\n",
      "  116   1160 1.011283085235303e-01 8.7e+00 6.38e-03  3e-03  5e-03 0:00.3\n",
      "  117   1170 1.010222781923117e-01 9.0e+00 6.66e-03  3e-03  5e-03 0:00.3\n",
      "  118   1180 1.011325495870904e-01 8.9e+00 6.96e-03  3e-03  5e-03 0:00.3\n",
      "  119   1190 1.009105259241425e-01 8.8e+00 7.66e-03  3e-03  5e-03 0:00.3\n",
      "  120   1200 1.007702271342407e-01 9.1e+00 8.53e-03  3e-03  6e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  121   1210 1.007765673950416e-01 9.5e+00 9.46e-03  4e-03  7e-03 0:00.3\n",
      "  122   1220 1.006804437301193e-01 9.9e+00 9.65e-03  4e-03  7e-03 0:00.3\n",
      "  123   1230 1.007949218900881e-01 1.0e+01 9.30e-03  3e-03  7e-03 0:00.3\n",
      "  124   1240 1.009700439225654e-01 1.1e+01 8.79e-03  3e-03  6e-03 0:00.3\n",
      "  125   1250 1.009678016072844e-01 1.2e+01 8.29e-03  3e-03  6e-03 0:00.4\n",
      "  126   1260 1.005487801163488e-01 1.2e+01 7.90e-03  3e-03  6e-03 0:00.4\n",
      "  127   1270 1.006278089175529e-01 1.3e+01 7.52e-03  3e-03  5e-03 0:00.4\n",
      "  128   1280 1.007777842733924e-01 1.4e+01 7.33e-03  3e-03  5e-03 0:00.4\n",
      "  129   1290 1.005928522369686e-01 1.4e+01 7.34e-03  3e-03  5e-03 0:00.4\n",
      "  130   1300 1.005585066889538e-01 1.5e+01 6.89e-03  3e-03  5e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  131   1310 1.005236893756801e-01 1.5e+01 6.71e-03  3e-03  5e-03 0:00.4\n",
      "  132   1320 1.006609275620067e-01 1.6e+01 6.98e-03  3e-03  5e-03 0:00.4\n",
      "  133   1330 1.003937193166277e-01 1.6e+01 7.50e-03  3e-03  5e-03 0:00.4\n",
      "  134   1340 1.003373027098891e-01 1.6e+01 8.14e-03  3e-03  6e-03 0:00.4\n",
      "  135   1350 1.003927607811756e-01 1.6e+01 9.42e-03  4e-03  6e-03 0:00.4\n",
      "  136   1360 1.004287473380209e-01 1.6e+01 9.51e-03  4e-03  6e-03 0:00.4\n",
      "  137   1370 1.003969183899538e-01 1.6e+01 9.15e-03  4e-03  6e-03 0:00.4\n",
      "  138   1380 1.004984558371944e-01 1.6e+01 9.00e-03  4e-03  6e-03 0:00.4\n",
      "  139   1390 1.004437831583727e-01 1.7e+01 8.55e-03  3e-03  6e-03 0:00.4\n",
      "  140   1400 1.001976409059916e-01 1.8e+01 8.08e-03  3e-03  6e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  141   1410 1.003438321621741e-01 1.8e+01 7.66e-03  3e-03  5e-03 0:00.4\n",
      "  142   1420 1.003921199882531e-01 1.8e+01 7.28e-03  3e-03  5e-03 0:00.4\n",
      "  143   1430 1.003235021195023e-01 1.8e+01 6.50e-03  3e-03  4e-03 0:00.4\n",
      "  144   1440 1.003020977960007e-01 1.9e+01 5.61e-03  2e-03  4e-03 0:00.4\n",
      "  145   1450 1.002126356187824e-01 2.0e+01 5.29e-03  2e-03  4e-03 0:00.4\n",
      "  146   1460 1.002148889524149e-01 2.0e+01 5.27e-03  2e-03  4e-03 0:00.4\n",
      "  147   1470 1.001149376625726e-01 2.1e+01 5.01e-03  2e-03  3e-03 0:00.4\n",
      "  148   1480 1.001031340399305e-01 2.1e+01 5.06e-03  2e-03  3e-03 0:00.4\n",
      "  149   1490 1.000882194427425e-01 2.1e+01 4.84e-03  2e-03  3e-03 0:00.4\n",
      "  150   1500 1.000974234114711e-01 2.2e+01 4.52e-03  2e-03  3e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  151   1510 1.001321859215492e-01 2.3e+01 4.15e-03  1e-03  3e-03 0:00.4\n",
      "  152   1520 1.001274418625793e-01 2.3e+01 3.94e-03  1e-03  2e-03 0:00.4\n",
      "  153   1530 1.001670295747035e-01 2.4e+01 3.87e-03  1e-03  2e-03 0:00.4\n",
      "  154   1540 1.001382100772695e-01 2.7e+01 3.96e-03  1e-03  2e-03 0:00.4\n",
      "  155   1550 1.001482696778364e-01 2.8e+01 3.76e-03  1e-03  2e-03 0:00.4\n",
      "  156   1560 1.001279440323763e-01 2.8e+01 3.65e-03  1e-03  2e-03 0:00.4\n",
      "  157   1570 1.001113977358209e-01 2.8e+01 3.31e-03  1e-03  2e-03 0:00.4\n",
      "  158   1580 1.000898142795402e-01 2.8e+01 3.06e-03  1e-03  2e-03 0:00.4\n",
      "  159   1590 1.000910140391496e-01 3.0e+01 2.88e-03  1e-03  2e-03 0:00.4\n",
      "  160   1600 1.000888814934767e-01 3.2e+01 2.78e-03  1e-03  2e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  161   1610 1.000534447950833e-01 3.2e+01 2.80e-03  1e-03  2e-03 0:00.4\n",
      "  162   1620 1.000625406758456e-01 3.4e+01 3.01e-03  1e-03  2e-03 0:00.4\n",
      "  163   1630 1.000840477864052e-01 3.6e+01 3.10e-03  1e-03  2e-03 0:00.4\n",
      "  164   1640 1.000581353979465e-01 3.7e+01 2.83e-03  1e-03  2e-03 0:00.4\n",
      "  165   1650 1.000485635424615e-01 3.7e+01 3.01e-03  1e-03  2e-03 0:00.4\n",
      "  166   1660 1.000572914386239e-01 3.8e+01 2.94e-03  1e-03  2e-03 0:00.4\n",
      "  167   1670 1.000747004315317e-01 3.7e+01 2.75e-03  9e-04  2e-03 0:00.4\n",
      "  168   1680 1.000505869942229e-01 3.9e+01 2.56e-03  9e-04  2e-03 0:00.5\n",
      "  169   1690 1.000351227322576e-01 3.9e+01 2.43e-03  8e-04  1e-03 0:00.5\n",
      "  170   1700 1.000440335820172e-01 4.0e+01 2.34e-03  8e-04  1e-03 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  171   1710 1.000334112448806e-01 3.9e+01 2.20e-03  8e-04  1e-03 0:00.5\n",
      "  172   1720 1.000454484124782e-01 4.0e+01 1.97e-03  7e-04  1e-03 0:00.5\n",
      "  173   1730 1.000477053574422e-01 4.1e+01 1.91e-03  7e-04  1e-03 0:00.5\n",
      "  174   1740 1.000336994338955e-01 4.3e+01 1.81e-03  6e-04  1e-03 0:00.5\n",
      "  175   1750 1.000347610562968e-01 4.3e+01 1.68e-03  6e-04  1e-03 0:00.5\n",
      "  176   1760 1.000363265222522e-01 4.3e+01 1.60e-03  5e-04  1e-03 0:00.5\n",
      "  177   1770 1.000383191972405e-01 4.5e+01 1.49e-03  5e-04  9e-04 0:00.5\n",
      "  178   1780 1.000351074820305e-01 4.6e+01 1.41e-03  5e-04  8e-04 0:00.5\n",
      "  179   1790 1.000355610873481e-01 4.7e+01 1.29e-03  4e-04  7e-04 0:00.5\n",
      "  180   1800 1.000275464114819e-01 4.8e+01 1.24e-03  4e-04  7e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  181   1810 1.000280393628790e-01 4.9e+01 1.22e-03  4e-04  7e-04 0:00.5\n",
      "  182   1820 1.000339771645972e-01 5.0e+01 1.15e-03  4e-04  6e-04 0:00.5\n",
      "  183   1830 1.000310667884519e-01 5.1e+01 1.09e-03  4e-04  6e-04 0:00.5\n",
      "  184   1840 1.000271333728836e-01 5.4e+01 1.08e-03  3e-04  6e-04 0:00.5\n",
      "  185   1850 1.000290711081954e-01 5.8e+01 9.85e-04  3e-04  6e-04 0:00.5\n",
      "  186   1860 1.000211537384008e-01 5.9e+01 9.61e-04  3e-04  5e-04 0:00.5\n",
      "  187   1870 1.000268330791870e-01 6.1e+01 9.36e-04  3e-04  5e-04 0:00.5\n",
      "  188   1880 1.000232342933623e-01 6.2e+01 9.12e-04  3e-04  5e-04 0:00.5\n",
      "  189   1890 1.000309531639861e-01 6.3e+01 8.90e-04  3e-04  5e-04 0:00.5\n",
      "  190   1900 1.000305105123681e-01 6.7e+01 9.07e-04  3e-04  5e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  191   1910 1.000245826365950e-01 7.0e+01 9.45e-04  3e-04  5e-04 0:00.5\n",
      "  192   1920 1.000203052085010e-01 7.1e+01 9.84e-04  3e-04  6e-04 0:00.5\n",
      "  193   1930 1.000204020877042e-01 7.4e+01 1.01e-03  4e-04  6e-04 0:00.5\n",
      "  194   1940 1.000168929483359e-01 7.7e+01 1.02e-03  3e-04  6e-04 0:00.5\n",
      "  195   1950 1.000141674054314e-01 7.8e+01 9.82e-04  3e-04  6e-04 0:00.5\n",
      "  196   1960 1.000167333159374e-01 8.0e+01 9.49e-04  3e-04  5e-04 0:00.5\n",
      "  197   1970 1.000111354826238e-01 8.5e+01 9.40e-04  3e-04  5e-04 0:00.5\n",
      "  198   1980 1.000145177346705e-01 8.8e+01 9.79e-04  3e-04  5e-04 0:00.5\n",
      "  199   1990 1.000133512703771e-01 8.9e+01 9.81e-04  3e-04  5e-04 0:00.5\n",
      "  200   2000 1.000111488822641e-01 9.0e+01 9.75e-04  3e-04  5e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  201   2010 1.000112188764521e-01 9.6e+01 9.69e-04  3e-04  5e-04 0:00.5\n",
      "  202   2020 1.000099953929475e-01 9.9e+01 9.82e-04  3e-04  6e-04 0:00.5\n",
      "  203   2030 1.000126034320122e-01 1.0e+02 9.70e-04  3e-04  6e-04 0:00.5\n",
      "  204   2040 1.000086359072968e-01 1.1e+02 8.58e-04  3e-04  5e-04 0:00.5\n",
      "  205   2050 1.000094256454208e-01 1.1e+02 7.99e-04  3e-04  5e-04 0:00.5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  206   2060 1.000091703115316e-01 1.1e+02 8.06e-04  3e-04  5e-04 0:00.5\n",
      "  207   2070 1.000065869554010e-01 1.1e+02 7.67e-04  3e-04  4e-04 0:00.5\n",
      "  208   2080 1.000087575992875e-01 1.2e+02 7.73e-04  3e-04  4e-04 0:00.5\n",
      "  209   2090 1.000097018748035e-01 1.2e+02 7.72e-04  3e-04  4e-04 0:00.5\n",
      "  210   2100 1.000099829208033e-01 1.4e+02 7.42e-04  2e-04  4e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  211   2110 1.000059573572323e-01 1.6e+02 7.01e-04  2e-04  4e-04 0:00.6\n",
      "  212   2120 1.000057697198544e-01 1.6e+02 6.72e-04  2e-04  4e-04 0:00.6\n",
      "  213   2130 1.000059847409951e-01 1.6e+02 6.37e-04  2e-04  4e-04 0:00.6\n",
      "  214   2140 1.000058328057817e-01 1.6e+02 5.98e-04  2e-04  3e-04 0:00.6\n",
      "  215   2150 1.000058752049041e-01 1.7e+02 5.76e-04  2e-04  3e-04 0:00.6\n",
      "  216   2160 1.000047964157730e-01 1.7e+02 5.94e-04  2e-04  3e-04 0:00.6\n",
      "  217   2170 1.000053214001383e-01 1.7e+02 6.15e-04  2e-04  3e-04 0:00.6\n",
      "  218   2180 1.000039167662207e-01 1.7e+02 6.34e-04  2e-04  4e-04 0:00.6\n",
      "  219   2190 1.000039012529381e-01 1.8e+02 5.93e-04  2e-04  3e-04 0:00.6\n",
      "  220   2200 1.000051752259213e-01 1.9e+02 5.74e-04  2e-04  3e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  221   2210 1.000062606710477e-01 2.0e+02 6.08e-04  2e-04  3e-04 0:00.6\n",
      "  222   2220 1.000056020744862e-01 2.1e+02 6.16e-04  2e-04  3e-04 0:00.6\n",
      "  223   2230 1.000040052044363e-01 2.2e+02 6.09e-04  2e-04  3e-04 0:00.6\n",
      "  224   2240 1.000037249278889e-01 2.2e+02 5.82e-04  2e-04  3e-04 0:00.6\n",
      "  225   2250 1.000041847410191e-01 2.2e+02 5.75e-04  2e-04  3e-04 0:00.6\n",
      "  226   2260 1.000028722419025e-01 2.3e+02 5.57e-04  2e-04  3e-04 0:00.6\n",
      "  227   2270 1.000025807213426e-01 2.3e+02 5.41e-04  2e-04  3e-04 0:00.6\n",
      "  228   2280 1.000027149551578e-01 2.4e+02 5.29e-04  2e-04  3e-04 0:00.6\n",
      "  229   2290 1.000014702959618e-01 2.4e+02 5.77e-04  2e-04  3e-04 0:00.6\n",
      "  230   2300 1.000018129717486e-01 2.6e+02 6.06e-04  2e-04  3e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  231   2310 1.000013841118980e-01 2.6e+02 6.66e-04  2e-04  3e-04 0:00.6\n",
      "  232   2320 1.000009474610336e-01 2.7e+02 7.34e-04  3e-04  4e-04 0:00.6\n",
      "  233   2330 1.000025196480466e-01 2.8e+02 7.23e-04  3e-04  4e-04 0:00.6\n",
      "  234   2340 1.000019933705954e-01 2.9e+02 6.78e-04  2e-04  4e-04 0:00.6\n",
      "  235   2350 1.000022511839476e-01 3.0e+02 6.13e-04  2e-04  3e-04 0:00.6\n",
      "  236   2360 1.000022305885262e-01 3.2e+02 6.01e-04  2e-04  3e-04 0:00.6\n",
      "  237   2370 1.000026707194234e-01 3.3e+02 5.37e-04  2e-04  3e-04 0:00.6\n",
      "  238   2380 1.000026924241385e-01 3.5e+02 5.14e-04  2e-04  3e-04 0:00.6\n",
      "  239   2390 1.000015833290427e-01 3.5e+02 5.06e-04  2e-04  2e-04 0:00.6\n",
      "  240   2400 1.000016913337726e-01 3.5e+02 5.38e-04  2e-04  3e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  241   2410 1.000010656928119e-01 3.6e+02 5.66e-04  2e-04  3e-04 0:00.6\n",
      "  242   2420 1.000016307787740e-01 3.6e+02 5.41e-04  2e-04  3e-04 0:00.6\n",
      "  243   2430 1.000014302084364e-01 3.6e+02 5.20e-04  2e-04  2e-04 0:00.6\n",
      "  244   2440 1.000004841794227e-01 3.5e+02 5.26e-04  2e-04  2e-04 0:00.6\n",
      "  245   2450 1.000017905023666e-01 3.4e+02 5.63e-04  2e-04  3e-04 0:00.6\n",
      "  246   2460 1.000007967503660e-01 3.3e+02 5.51e-04  2e-04  2e-04 0:00.6\n",
      "  247   2470 1.000005753378867e-01 3.4e+02 5.67e-04  2e-04  3e-04 0:00.6\n",
      "  248   2480 1.000009458471738e-01 3.5e+02 5.54e-04  2e-04  2e-04 0:00.6\n",
      "  249   2490 1.000004688593993e-01 3.6e+02 5.15e-04  2e-04  2e-04 0:00.6\n",
      "  250   2500 1.000005726242323e-01 3.6e+02 4.87e-04  2e-04  2e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  251   2510 1.000008814133777e-01 3.5e+02 4.70e-04  2e-04  2e-04 0:00.6\n",
      "  252   2520 1.000005175396111e-01 3.5e+02 4.54e-04  2e-04  2e-04 0:00.6\n",
      "  253   2530 1.000008959585311e-01 3.7e+02 4.70e-04  2e-04  2e-04 0:00.6\n",
      "  254   2540 1.000008923240233e-01 3.7e+02 4.77e-04  2e-04  2e-04 0:00.6\n",
      "  255   2550 1.000007155112764e-01 3.8e+02 4.72e-04  2e-04  2e-04 0:00.7\n",
      "  256   2560 1.000006117313693e-01 3.8e+02 4.79e-04  2e-04  2e-04 0:00.7\n",
      "  257   2570 1.000011101926893e-01 3.9e+02 4.32e-04  1e-04  2e-04 0:00.7\n",
      "  258   2580 1.000003827610571e-01 4.1e+02 3.92e-04  1e-04  2e-04 0:00.7\n",
      "  259   2590 1.000004731960005e-01 4.3e+02 3.57e-04  1e-04  1e-04 0:00.7\n",
      "  260   2600 1.000006749550404e-01 4.3e+02 3.40e-04  1e-04  1e-04 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  261   2610 1.000008248895144e-01 4.4e+02 3.18e-04  1e-04  1e-04 0:00.7\n",
      "  262   2620 1.000004953367621e-01 4.4e+02 2.94e-04  9e-05  1e-04 0:00.7\n",
      "  263   2630 1.000004277042193e-01 4.6e+02 2.86e-04  9e-05  1e-04 0:00.7\n",
      "  264   2640 1.000004159175389e-01 4.8e+02 2.74e-04  8e-05  1e-04 0:00.7\n",
      "  265   2650 1.000002687300566e-01 5.0e+02 2.87e-04  9e-05  1e-04 0:00.7\n",
      "  266   2660 1.000002208208859e-01 5.0e+02 2.57e-04  8e-05  1e-04 0:00.7\n",
      "  267   2670 1.000002494530012e-01 5.1e+02 2.50e-04  8e-05  1e-04 0:00.7\n",
      "  268   2680 1.000001055625692e-01 5.3e+02 2.69e-04  8e-05  1e-04 0:00.7\n",
      "  269   2690 1.000001251345278e-01 5.6e+02 2.75e-04  9e-05  1e-04 0:00.7\n",
      "  270   2700 1.000002386847405e-01 5.6e+02 2.59e-04  8e-05  1e-04 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  271   2710 1.000002231497392e-01 5.7e+02 2.53e-04  8e-05  1e-04 0:00.7\n",
      "  272   2720 1.000001632119111e-01 5.8e+02 2.34e-04  7e-05  1e-04 0:00.7\n",
      "  273   2730 1.000000923543523e-01 5.9e+02 2.19e-04  6e-05  9e-05 0:00.7\n",
      "  274   2740 1.000001359657187e-01 6.3e+02 2.24e-04  7e-05  1e-04 0:00.7\n",
      "  275   2750 1.000000699188038e-01 6.5e+02 2.16e-04  6e-05  9e-05 0:00.7\n",
      "  276   2760 1.000001341338949e-01 6.3e+02 2.27e-04  7e-05  1e-04 0:00.7\n",
      "  277   2770 1.000000943013499e-01 6.8e+02 2.22e-04  6e-05  1e-04 0:00.7\n",
      "  278   2780 1.000000817534025e-01 6.9e+02 2.04e-04  6e-05  9e-05 0:00.7\n",
      "  279   2790 1.000001823858346e-01 7.1e+02 1.82e-04  5e-05  8e-05 0:00.7\n",
      "  280   2800 1.000001367714946e-01 7.1e+02 1.74e-04  5e-05  8e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  281   2810 1.000001198057895e-01 7.5e+02 1.67e-04  5e-05  7e-05 0:00.7\n",
      "  282   2820 1.000001156703776e-01 7.5e+02 1.48e-04  4e-05  6e-05 0:00.7\n",
      "  283   2830 1.000001336534375e-01 7.9e+02 1.38e-04  4e-05  6e-05 0:00.7\n",
      "  284   2840 1.000001145902381e-01 7.8e+02 1.31e-04  4e-05  6e-05 0:00.7\n",
      "  285   2850 1.000000784366021e-01 8.0e+02 1.28e-04  4e-05  5e-05 0:00.7\n",
      "  286   2860 1.000000906680421e-01 8.3e+02 1.20e-04  3e-05  5e-05 0:00.7\n",
      "  287   2870 1.000000912638019e-01 8.5e+02 1.14e-04  3e-05  5e-05 0:00.7\n",
      "  288   2880 1.000000724507934e-01 8.7e+02 1.04e-04  3e-05  4e-05 0:00.7\n",
      "  289   2890 1.000000703503401e-01 8.7e+02 1.06e-04  3e-05  4e-05 0:00.7\n",
      "  290   2900 1.000000755613754e-01 8.8e+02 1.02e-04  3e-05  4e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  291   2910 1.000000888888819e-01 9.2e+02 9.84e-05  3e-05  4e-05 0:00.7\n",
      "  292   2920 1.000000852827869e-01 9.4e+02 9.54e-05  3e-05  4e-05 0:00.7\n",
      "  293   2930 1.000000841007208e-01 9.4e+02 9.71e-05  3e-05  4e-05 0:00.7\n",
      "  294   2940 1.000000717265537e-01 9.5e+02 9.65e-05  3e-05  4e-05 0:00.7\n",
      "  295   2950 1.000001015669401e-01 1.0e+03 9.50e-05  3e-05  4e-05 0:00.7\n",
      "  296   2960 1.000000566936524e-01 1.0e+03 9.46e-05  3e-05  4e-05 0:00.7\n",
      "  297   2970 1.000000766763982e-01 1.0e+03 8.70e-05  2e-05  3e-05 0:00.7\n",
      "  298   2980 1.000000609334315e-01 1.0e+03 8.31e-05  2e-05  3e-05 0:00.7\n",
      "  299   2990 1.000000553221942e-01 1.1e+03 8.04e-05  2e-05  3e-05 0:00.8\n",
      "  300   3000 1.000000514910022e-01 1.2e+03 7.42e-05  2e-05  3e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  301   3010 1.000000511293179e-01 1.2e+03 6.63e-05  2e-05  3e-05 0:00.8\n",
      "  302   3020 1.000000643543911e-01 1.2e+03 6.02e-05  2e-05  2e-05 0:00.8\n",
      "  303   3030 1.000000522305163e-01 1.2e+03 5.73e-05  2e-05  2e-05 0:00.8\n",
      "  304   3040 1.000000692840947e-01 1.2e+03 4.85e-05  1e-05  2e-05 0:00.8\n",
      "  305   3050 1.000000512735157e-01 1.2e+03 4.76e-05  1e-05  2e-05 0:00.8\n",
      "  306   3060 1.000000403178995e-01 1.3e+03 5.07e-05  1e-05  2e-05 0:00.8\n",
      "  307   3070 1.000000304861188e-01 1.3e+03 5.39e-05  1e-05  2e-05 0:00.8\n",
      "  308   3080 1.000000282975108e-01 1.3e+03 5.65e-05  2e-05  2e-05 0:00.8\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  309   3090 1.000000291490650e-01 1.3e+03 5.81e-05  2e-05  2e-05 0:00.8\n",
      "  310   3100 1.000000231213810e-01 1.3e+03 5.72e-05  1e-05  2e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  311   3110 1.000000213585274e-01 1.4e+03 5.68e-05  1e-05  2e-05 0:00.8\n",
      "  312   3120 1.000000220495595e-01 1.5e+03 5.58e-05  1e-05  2e-05 0:00.8\n",
      "  313   3130 1.000000215285154e-01 1.6e+03 5.28e-05  1e-05  2e-05 0:00.8\n",
      "  314   3140 1.000000222312645e-01 1.6e+03 5.21e-05  1e-05  2e-05 0:00.8\n",
      "  315   3150 1.000000131265026e-01 1.7e+03 5.55e-05  1e-05  2e-05 0:00.8\n",
      "  316   3160 1.000000177645365e-01 1.7e+03 5.37e-05  1e-05  2e-05 0:00.8\n",
      "  317   3170 1.000000173861080e-01 1.8e+03 4.89e-05  1e-05  2e-05 0:00.8\n",
      "  318   3180 1.000000162761518e-01 1.8e+03 4.82e-05  1e-05  2e-05 0:00.8\n",
      "  319   3190 1.000000168523250e-01 2.0e+03 4.77e-05  1e-05  2e-05 0:00.8\n",
      "  320   3200 1.000000208261228e-01 2.2e+03 4.61e-05  1e-05  2e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  321   3210 1.000000166873532e-01 2.2e+03 4.12e-05  1e-05  2e-05 0:00.8\n",
      "  322   3220 1.000000207743050e-01 2.2e+03 4.18e-05  1e-05  2e-05 0:00.8\n",
      "  323   3230 1.000000210871283e-01 2.4e+03 4.01e-05  1e-05  2e-05 0:00.8\n",
      "  324   3240 1.000000181036029e-01 2.4e+03 4.25e-05  1e-05  2e-05 0:00.8\n",
      "  325   3250 1.000000196389421e-01 2.5e+03 4.42e-05  1e-05  2e-05 0:00.8\n",
      "  326   3260 1.000000126315440e-01 2.5e+03 4.96e-05  1e-05  2e-05 0:00.8\n",
      "  327   3270 1.000000106003406e-01 2.6e+03 4.99e-05  1e-05  2e-05 0:00.8\n",
      "  328   3280 1.000000144247290e-01 2.6e+03 5.05e-05  1e-05  2e-05 0:00.8\n",
      "  329   3290 1.000000110073021e-01 2.7e+03 4.92e-05  1e-05  2e-05 0:00.8\n",
      "  330   3300 1.000000101367616e-01 2.8e+03 4.53e-05  1e-05  2e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  331   3310 1.000000088726638e-01 2.8e+03 4.58e-05  1e-05  2e-05 0:00.8\n",
      "  332   3320 1.000000110850840e-01 2.8e+03 4.51e-05  1e-05  2e-05 0:00.8\n",
      "  333   3330 1.000000112884256e-01 2.8e+03 4.17e-05  1e-05  2e-05 0:00.8\n",
      "  334   3340 1.000000098386773e-01 2.9e+03 4.09e-05  1e-05  2e-05 0:00.8\n",
      "  335   3350 1.000000076393450e-01 3.0e+03 4.04e-05  1e-05  2e-05 0:00.8\n",
      "  336   3360 1.000000101647848e-01 3.1e+03 3.84e-05  1e-05  1e-05 0:00.8\n",
      "  337   3370 1.000000061187346e-01 3.2e+03 3.81e-05  1e-05  1e-05 0:00.8\n",
      "  338   3380 1.000000065387024e-01 3.3e+03 3.60e-05  1e-05  1e-05 0:00.8\n",
      "  339   3390 1.000000075870127e-01 3.3e+03 3.41e-05  9e-06  1e-05 0:00.8\n",
      "  340   3400 1.000000074050946e-01 3.3e+03 3.33e-05  9e-06  1e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  341   3410 1.000000060403217e-01 3.5e+03 3.47e-05  9e-06  1e-05 0:00.8\n",
      "  342   3420 1.000000076570814e-01 3.6e+03 3.57e-05  9e-06  1e-05 0:00.8\n",
      "  343   3430 1.000000084316857e-01 3.9e+03 3.58e-05  9e-06  1e-05 0:00.8\n",
      "  344   3440 1.000000074245807e-01 4.1e+03 3.42e-05  9e-06  1e-05 0:00.9\n",
      "  345   3450 1.000000078373237e-01 4.3e+03 3.70e-05  1e-05  1e-05 0:00.9\n",
      "  346   3460 1.000000081064953e-01 4.6e+03 3.97e-05  1e-05  2e-05 0:00.9\n",
      "  347   3470 1.000000071304847e-01 5.0e+03 4.23e-05  1e-05  2e-05 0:00.9\n",
      "  348   3480 1.000000066427715e-01 5.1e+03 4.73e-05  1e-05  2e-05 0:00.9\n",
      "  349   3490 1.000000065089942e-01 5.1e+03 4.99e-05  1e-05  2e-05 0:00.9\n",
      "  350   3500 1.000000043344593e-01 5.2e+03 5.09e-05  1e-05  2e-05 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  351   3510 1.000000046689105e-01 5.3e+03 5.42e-05  1e-05  2e-05 0:00.9\n",
      "  352   3520 1.000000045745210e-01 5.5e+03 5.34e-05  1e-05  2e-05 0:00.9\n",
      "  353   3530 1.000000035062780e-01 5.5e+03 5.17e-05  1e-05  2e-05 0:00.9\n",
      "  354   3540 1.000000083206755e-01 5.3e+03 5.25e-05  1e-05  2e-05 0:00.9\n",
      "  355   3550 1.000000064358898e-01 5.5e+03 5.06e-05  1e-05  2e-05 0:00.9\n",
      "  356   3560 1.000000039343486e-01 5.6e+03 4.79e-05  1e-05  2e-05 0:00.9\n",
      "  357   3570 1.000000067324260e-01 5.8e+03 4.68e-05  1e-05  2e-05 0:00.9\n",
      "  358   3580 1.000000080384117e-01 6.1e+03 4.22e-05  1e-05  2e-05 0:00.9\n",
      "  359   3590 1.000000067279525e-01 6.2e+03 3.99e-05  1e-05  1e-05 0:00.9\n",
      "  360   3600 1.000000039486543e-01 6.5e+03 3.73e-05  9e-06  1e-05 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  361   3610 1.000000045063294e-01 6.6e+03 3.56e-05  9e-06  1e-05 0:00.9\n",
      "  362   3620 1.000000031723598e-01 6.9e+03 3.29e-05  8e-06  1e-05 0:00.9\n",
      "  363   3630 1.000000025339097e-01 7.1e+03 3.18e-05  8e-06  1e-05 0:00.9\n",
      "  364   3640 1.000000040337905e-01 7.3e+03 2.96e-05  7e-06  1e-05 0:00.9\n",
      "  365   3650 1.000000030124698e-01 7.5e+03 2.66e-05  6e-06  9e-06 0:00.9\n",
      "  366   3660 1.000000033482120e-01 7.7e+03 2.46e-05  6e-06  8e-06 0:00.9\n",
      "  367   3670 1.000000042022276e-01 7.8e+03 2.39e-05  6e-06  8e-06 0:00.9\n",
      "  368   3680 1.000000036380834e-01 8.3e+03 2.28e-05  5e-06  7e-06 0:00.9\n",
      "  369   3690 1.000000039044160e-01 8.4e+03 2.23e-05  5e-06  7e-06 0:00.9\n",
      "  370   3700 1.000000035807306e-01 8.9e+03 2.13e-05  5e-06  7e-06 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  371   3710 1.000000036828293e-01 9.1e+03 2.01e-05  5e-06  6e-06 0:00.9\n",
      "  372   3720 1.000000024895934e-01 8.8e+03 1.91e-05  4e-06  6e-06 0:00.9\n",
      "  373   3730 1.000000024191518e-01 8.9e+03 1.79e-05  4e-06  6e-06 0:00.9\n",
      "  374   3740 1.000000022997640e-01 8.9e+03 1.83e-05  4e-06  6e-06 0:00.9\n",
      "  375   3750 1.000000033039049e-01 8.9e+03 1.92e-05  4e-06  6e-06 0:00.9\n",
      "  376   3760 1.000000022679342e-01 9.0e+03 1.90e-05  4e-06  6e-06 0:00.9\n",
      "  377   3770 1.000000027355951e-01 9.2e+03 1.77e-05  4e-06  5e-06 0:00.9\n",
      "  378   3780 1.000000027149974e-01 9.6e+03 1.72e-05  4e-06  5e-06 0:00.9\n",
      "  379   3790 1.000000024717588e-01 9.8e+03 1.73e-05  4e-06  5e-06 0:00.9\n",
      "  380   3800 1.000000022315048e-01 1.0e+04 1.80e-05  4e-06  6e-06 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  381   3810 1.000000019800387e-01 1.1e+04 1.82e-05  4e-06  6e-06 0:00.9\n",
      "  382   3820 1.000000019794516e-01 1.1e+04 1.87e-05  4e-06  6e-06 0:00.9\n",
      "  383   3830 1.000000016424450e-01 1.1e+04 1.98e-05  4e-06  6e-06 0:00.9\n",
      "  384   3840 1.000000015927377e-01 1.1e+04 2.17e-05  5e-06  7e-06 0:00.9\n",
      "  385   3850 1.000000022538819e-01 1.1e+04 2.22e-05  5e-06  8e-06 0:00.9\n",
      "  386   3860 1.000000023366411e-01 1.1e+04 2.02e-05  4e-06  7e-06 0:00.9\n",
      "  387   3870 1.000000019438720e-01 1.2e+04 1.79e-05  4e-06  6e-06 0:00.9\n",
      "  388   3880 1.000000015950481e-01 1.2e+04 1.73e-05  4e-06  6e-06 0:01.0\n",
      "  389   3890 1.000000015714756e-01 1.2e+04 1.64e-05  3e-06  6e-06 0:01.0\n",
      "  390   3900 1.000000016733575e-01 1.2e+04 1.56e-05  3e-06  5e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  391   3910 1.000000014876758e-01 1.3e+04 1.64e-05  3e-06  5e-06 0:01.0\n",
      "  392   3920 1.000000013686783e-01 1.4e+04 1.66e-05  4e-06  5e-06 0:01.0\n",
      "  393   3930 1.000000012237665e-01 1.5e+04 1.66e-05  3e-06  5e-06 0:01.0\n",
      "  394   3940 1.000000014272695e-01 1.5e+04 1.55e-05  3e-06  5e-06 0:01.0\n",
      "  395   3950 1.000000015465903e-01 1.5e+04 1.42e-05  3e-06  4e-06 0:01.0\n",
      "  396   3960 1.000000016080626e-01 1.6e+04 1.42e-05  3e-06  4e-06 0:01.0\n",
      "  397   3970 1.000000013134380e-01 1.6e+04 1.53e-05  3e-06  5e-06 0:01.0\n",
      "  398   3980 1.000000011922829e-01 1.6e+04 1.68e-05  4e-06  5e-06 0:01.0\n",
      "  399   3990 1.000000011950312e-01 1.7e+04 1.86e-05  4e-06  5e-06 0:01.0\n",
      "  400   4000 1.000000009784528e-01 1.7e+04 1.87e-05  4e-06  6e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  401   4010 1.000000009528915e-01 1.8e+04 1.88e-05  4e-06  6e-06 0:01.0\n",
      "  402   4020 1.000000008403498e-01 1.8e+04 1.96e-05  4e-06  6e-06 0:01.0\n",
      "  403   4030 1.000000007667958e-01 1.8e+04 1.93e-05  4e-06  6e-06 0:01.0\n",
      "  404   4040 1.000000005278428e-01 1.8e+04 2.06e-05  4e-06  6e-06 0:01.0\n",
      "  405   4050 1.000000004958120e-01 1.9e+04 2.37e-05  5e-06  7e-06 0:01.0\n",
      "  406   4060 1.000000005575539e-01 2.0e+04 2.61e-05  5e-06  7e-06 0:01.0\n",
      "  407   4070 1.000000004275844e-01 2.0e+04 2.72e-05  5e-06  8e-06 0:01.0\n",
      "  408   4080 1.000000004404052e-01 2.0e+04 2.94e-05  6e-06  8e-06 0:01.0\n",
      "  409   4090 1.000000008133408e-01 2.2e+04 2.99e-05  6e-06  8e-06 0:01.0\n",
      "  410   4100 1.000000004583193e-01 2.3e+04 2.98e-05  6e-06  8e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  411   4110 1.000000001929338e-01 2.4e+04 2.71e-05  5e-06  7e-06 0:01.0\n",
      "  412   4120 1.000000004321699e-01 2.4e+04 2.58e-05  5e-06  7e-06 0:01.0\n",
      "  413   4130 1.000000005848963e-01 2.5e+04 2.46e-05  5e-06  6e-06 0:01.0\n",
      "  414   4140 1.000000004783776e-01 2.4e+04 2.29e-05  4e-06  6e-06 0:01.0\n",
      "  415   4150 1.000000002733578e-01 2.4e+04 2.03e-05  4e-06  5e-06 0:01.0\n",
      "  416   4160 1.000000002854857e-01 2.5e+04 1.79e-05  3e-06  5e-06 0:01.0\n",
      "  417   4170 1.000000003050149e-01 2.5e+04 1.69e-05  3e-06  4e-06 0:01.0\n",
      "  418   4180 1.000000002232899e-01 2.5e+04 1.66e-05  3e-06  4e-06 0:01.0\n",
      "  419   4190 1.000000003557583e-01 2.5e+04 1.54e-05  3e-06  4e-06 0:01.0\n",
      "  420   4200 1.000000002497754e-01 2.7e+04 1.43e-05  3e-06  4e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  421   4210 1.000000001856524e-01 2.8e+04 1.28e-05  2e-06  3e-06 0:01.0\n",
      "  422   4220 1.000000001595841e-01 2.8e+04 1.23e-05  2e-06  3e-06 0:01.0\n",
      "  423   4230 1.000000001430836e-01 2.9e+04 1.15e-05  2e-06  3e-06 0:01.0\n",
      "  424   4240 1.000000001597244e-01 3.0e+04 1.01e-05  2e-06  2e-06 0:01.0\n",
      "  425   4250 1.000000001516041e-01 3.0e+04 8.96e-06  2e-06  2e-06 0:01.0\n",
      "  426   4260 1.000000001433746e-01 3.2e+04 8.32e-06  1e-06  2e-06 0:01.0\n",
      "  427   4270 1.000000001762890e-01 3.2e+04 7.61e-06  1e-06  2e-06 0:01.0\n",
      "  428   4280 1.000000001540244e-01 3.3e+04 6.94e-06  1e-06  2e-06 0:01.0\n",
      "  429   4290 1.000000001288708e-01 3.4e+04 6.21e-06  1e-06  2e-06 0:01.0\n",
      "  430   4300 1.000000001208417e-01 3.5e+04 5.81e-06  1e-06  1e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  431   4310 1.000000001091329e-01 3.6e+04 5.90e-06  1e-06  1e-06 0:01.0\n",
      "  432   4320 1.000000000853679e-01 3.8e+04 6.18e-06  1e-06  2e-06 0:01.1\n",
      "  433   4330 1.000000001354142e-01 3.9e+04 6.62e-06  1e-06  2e-06 0:01.1\n",
      "  434   4340 1.000000001068017e-01 4.0e+04 6.38e-06  1e-06  2e-06 0:01.1\n",
      "  435   4350 1.000000000926091e-01 4.0e+04 6.17e-06  1e-06  1e-06 0:01.1\n",
      "  436   4360 1.000000000892560e-01 4.3e+04 5.83e-06  9e-07  1e-06 0:01.1\n",
      "  437   4370 1.000000000997834e-01 4.4e+04 5.24e-06  8e-07  1e-06 0:01.1\n",
      "  438   4380 1.000000000962020e-01 4.8e+04 4.92e-06  8e-07  1e-06 0:01.1\n",
      "  439   4390 1.000000000885205e-01 4.7e+04 4.68e-06  7e-07  1e-06 0:01.1\n",
      "  440   4400 1.000000000811983e-01 4.8e+04 4.44e-06  7e-07  1e-06 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  441   4410 1.000000000972052e-01 4.8e+04 4.40e-06  7e-07  1e-06 0:01.1\n",
      "  442   4420 1.000000000759476e-01 4.9e+04 4.24e-06  6e-07  1e-06 0:01.1\n",
      "  443   4430 1.000000000834169e-01 4.9e+04 4.27e-06  6e-07  1e-06 0:01.1\n",
      "  444   4440 1.000000000785872e-01 4.9e+04 4.23e-06  6e-07  1e-06 0:01.1\n",
      "  445   4450 1.000000000737671e-01 5.1e+04 4.03e-06  6e-07  1e-06 0:01.1\n",
      "  446   4460 1.000000000770702e-01 5.3e+04 3.98e-06  6e-07  9e-07 0:01.1\n",
      "  447   4470 1.000000000741366e-01 5.3e+04 4.04e-06  6e-07  9e-07 0:01.1\n",
      "  448   4480 1.000000000637092e-01 5.5e+04 4.13e-06  6e-07  9e-07 0:01.1\n",
      "  449   4490 1.000000000567384e-01 5.5e+04 4.22e-06  6e-07  1e-06 0:01.1\n",
      "  450   4500 1.000000000645565e-01 6.0e+04 3.98e-06  6e-07  9e-07 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  451   4510 1.000000000591415e-01 5.9e+04 3.71e-06  5e-07  9e-07 0:01.1\n",
      "  452   4520 1.000000000527340e-01 5.9e+04 3.84e-06  5e-07  9e-07 0:01.1\n",
      "  453   4530 1.000000000548112e-01 6.3e+04 3.94e-06  6e-07  9e-07 0:01.1\n",
      "  454   4540 1.000000000518309e-01 6.9e+04 3.84e-06  6e-07  9e-07 0:01.1\n",
      "  455   4550 1.000000000572683e-01 7.0e+04 3.59e-06  5e-07  8e-07 0:01.1\n",
      "  456   4560 1.000000000547001e-01 6.8e+04 3.41e-06  5e-07  8e-07 0:01.1\n",
      "  457   4570 1.000000000492424e-01 7.1e+04 3.37e-06  5e-07  8e-07 0:01.1\n",
      "  458   4580 1.000000000499293e-01 7.3e+04 3.38e-06  5e-07  8e-07 0:01.1\n",
      "  459   4590 1.000000000393646e-01 7.5e+04 3.31e-06  5e-07  7e-07 0:01.1\n",
      "  460   4600 1.000000000454558e-01 7.7e+04 3.37e-06  5e-07  7e-07 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  461   4610 1.000000000434442e-01 7.7e+04 3.14e-06  5e-07  7e-07 0:01.1\n",
      "  462   4620 1.000000000493134e-01 8.0e+04 2.91e-06  4e-07  7e-07 0:01.1\n",
      "  463   4630 1.000000000473516e-01 8.3e+04 2.64e-06  4e-07  6e-07 0:01.1\n",
      "  464   4640 1.000000000499538e-01 8.4e+04 2.47e-06  4e-07  6e-07 0:01.1\n",
      "  465   4650 1.000000000461027e-01 8.7e+04 2.21e-06  3e-07  5e-07 0:01.1\n",
      "  466   4660 1.000000000471376e-01 9.0e+04 2.07e-06  3e-07  5e-07 0:01.1\n",
      "  467   4670 1.000000000462792e-01 9.8e+04 1.94e-06  3e-07  4e-07 0:01.1\n",
      "  468   4680 1.000000000468302e-01 1.0e+05 1.79e-06  3e-07  4e-07 0:01.1\n",
      "  469   4690 1.000000000407886e-01 1.1e+05 1.82e-06  3e-07  4e-07 0:01.1\n",
      "  470   4700 1.000000000447556e-01 1.1e+05 1.60e-06  2e-07  4e-07 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  471   4710 1.000000000419114e-01 1.1e+05 1.47e-06  2e-07  3e-07 0:01.1\n",
      "  472   4720 1.000000000418954e-01 1.1e+05 1.36e-06  2e-07  3e-07 0:01.1\n",
      "  473   4730 1.000000000399045e-01 1.1e+05 1.38e-06  2e-07  3e-07 0:01.1\n",
      "  474   4740 1.000000000410168e-01 1.2e+05 1.39e-06  2e-07  3e-07 0:01.1\n",
      "  475   4750 1.000000000387931e-01 1.3e+05 1.53e-06  2e-07  3e-07 0:01.1\n",
      "  476   4760 1.000000000367714e-01 1.3e+05 1.56e-06  2e-07  3e-07 0:01.1\n",
      "  477   4770 1.000000000400074e-01 1.3e+05 1.52e-06  2e-07  3e-07 0:01.2\n",
      "  478   4780 1.000000000381603e-01 1.3e+05 1.50e-06  2e-07  3e-07 0:01.2\n",
      "  479   4790 1.000000000389576e-01 1.3e+05 1.39e-06  2e-07  3e-07 0:01.2\n",
      "  480   4800 1.000000000369531e-01 1.4e+05 1.41e-06  2e-07  3e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  481   4810 1.000000000369648e-01 1.4e+05 1.44e-06  2e-07  3e-07 0:01.2\n",
      "  482   4820 1.000000000385915e-01 1.4e+05 1.51e-06  2e-07  3e-07 0:01.2\n",
      "  483   4830 1.000000000374026e-01 1.4e+05 1.60e-06  2e-07  4e-07 0:01.2\n",
      "  484   4840 1.000000000364158e-01 1.5e+05 1.83e-06  3e-07  4e-07 0:01.2\n",
      "  485   4850 1.000000000365700e-01 1.5e+05 1.92e-06  3e-07  5e-07 0:01.2\n",
      "  486   4860 1.000000000359110e-01 1.6e+05 2.07e-06  3e-07  5e-07 0:01.2\n",
      "  487   4870 1.000000000361951e-01 1.7e+05 2.01e-06  3e-07  5e-07 0:01.2\n",
      "  488   4880 1.000000000334881e-01 1.7e+05 1.93e-06  3e-07  4e-07 0:01.2\n",
      "  489   4890 1.000000000327001e-01 1.7e+05 1.87e-06  2e-07  4e-07 0:01.2\n",
      "  490   4900 1.000000000318940e-01 1.7e+05 1.87e-06  2e-07  4e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  491   4910 1.000000000317359e-01 1.8e+05 1.96e-06  3e-07  4e-07 0:01.2\n",
      "  492   4920 1.000000000296064e-01 1.8e+05 2.11e-06  3e-07  5e-07 0:01.2\n",
      "  493   4930 1.000000000256993e-01 1.9e+05 2.44e-06  3e-07  5e-07 0:01.2\n",
      "  494   4940 1.000000000274807e-01 1.9e+05 2.77e-06  4e-07  6e-07 0:01.2\n",
      "  495   4950 1.000000000206398e-01 1.8e+05 3.05e-06  4e-07  7e-07 0:01.2\n",
      "  496   4960 1.000000000234950e-01 1.9e+05 3.12e-06  4e-07  7e-07 0:01.2\n",
      "  497   4970 1.000000000229932e-01 1.9e+05 3.28e-06  4e-07  7e-07 0:01.2\n",
      "  498   4980 1.000000000290284e-01 1.9e+05 3.15e-06  4e-07  7e-07 0:01.2\n",
      "  499   4990 1.000000000242594e-01 1.9e+05 2.97e-06  4e-07  6e-07 0:01.2\n",
      "  500   5000 1.000000000225595e-01 1.9e+05 2.96e-06  4e-07  6e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  501   5010 1.000000000193311e-01 2.0e+05 3.29e-06  4e-07  7e-07 0:01.2\n",
      "  502   5020 1.000000000194855e-01 2.0e+05 3.39e-06  4e-07  7e-07 0:01.2\n",
      "  503   5030 1.000000000211011e-01 2.1e+05 3.34e-06  4e-07  7e-07 0:01.2\n",
      "  504   5040 1.000000000164083e-01 2.1e+05 3.42e-06  4e-07  7e-07 0:01.2\n",
      "  505   5050 1.000000000156163e-01 2.1e+05 3.47e-06  4e-07  7e-07 0:01.2\n",
      "  506   5060 1.000000000166784e-01 2.1e+05 3.62e-06  4e-07  7e-07 0:01.2\n",
      "  507   5070 1.000000000197988e-01 2.1e+05 3.45e-06  4e-07  7e-07 0:01.2\n",
      "  508   5080 1.000000000134470e-01 2.4e+05 3.28e-06  4e-07  6e-07 0:01.2\n",
      "  509   5090 1.000000000148986e-01 2.4e+05 2.99e-06  4e-07  6e-07 0:01.2\n",
      "  510   5100 1.000000000151934e-01 2.5e+05 2.86e-06  3e-07  5e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  511   5110 1.000000000123626e-01 2.5e+05 2.84e-06  3e-07  5e-07 0:01.2\n",
      "  512   5120 1.000000000129504e-01 2.4e+05 2.83e-06  3e-07  5e-07 0:01.2\n",
      "  513   5130 1.000000000121357e-01 2.4e+05 2.79e-06  3e-07  5e-07 0:01.2\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  514   5140 1.000000000108107e-01 2.4e+05 2.61e-06  3e-07  5e-07 0:01.2\n",
      "  515   5150 1.000000000089063e-01 2.5e+05 2.51e-06  3e-07  5e-07 0:01.2\n",
      "  516   5160 1.000000000092061e-01 2.5e+05 2.51e-06  3e-07  4e-07 0:01.2\n",
      "  517   5170 1.000000000073447e-01 2.5e+05 2.75e-06  3e-07  5e-07 0:01.2\n",
      "  518   5180 1.000000000074657e-01 2.8e+05 2.75e-06  3e-07  5e-07 0:01.2\n",
      "  519   5190 1.000000000069256e-01 2.9e+05 2.75e-06  3e-07  5e-07 0:01.2\n",
      "  520   5200 1.000000000069089e-01 3.0e+05 2.81e-06  3e-07  5e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  521   5210 1.000000000061463e-01 3.1e+05 2.70e-06  3e-07  5e-07 0:01.2\n",
      "  522   5220 1.000000000063588e-01 3.1e+05 2.64e-06  3e-07  5e-07 0:01.3\n",
      "  523   5230 1.000000000057940e-01 3.3e+05 2.69e-06  3e-07  5e-07 0:01.3\n",
      "  524   5240 1.000000000053589e-01 3.3e+05 2.73e-06  3e-07  5e-07 0:01.3\n",
      "  525   5250 1.000000000061183e-01 3.2e+05 2.65e-06  3e-07  5e-07 0:01.3\n",
      "  526   5260 1.000000000037553e-01 3.3e+05 2.74e-06  3e-07  5e-07 0:01.3\n",
      "  527   5270 1.000000000056061e-01 3.3e+05 2.83e-06  3e-07  5e-07 0:01.3\n",
      "  528   5280 1.000000000037356e-01 3.4e+05 2.84e-06  3e-07  5e-07 0:01.3\n",
      "  529   5290 1.000000000046826e-01 3.4e+05 2.69e-06  3e-07  5e-07 0:01.3\n",
      "  530   5300 1.000000000049395e-01 3.4e+05 2.61e-06  3e-07  5e-07 0:01.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  531   5310 1.000000000047166e-01 3.6e+05 2.66e-06  3e-07  5e-07 0:01.3\n",
      "  532   5320 1.000000000036002e-01 3.7e+05 2.58e-06  3e-07  4e-07 0:01.3\n",
      "  533   5330 1.000000000042502e-01 3.9e+05 2.47e-06  3e-07  4e-07 0:01.3\n",
      "  534   5340 1.000000000031699e-01 3.9e+05 2.44e-06  3e-07  4e-07 0:01.3\n",
      "  535   5350 1.000000000044596e-01 3.9e+05 2.30e-06  3e-07  4e-07 0:01.3\n",
      "  536   5360 1.000000000048949e-01 4.0e+05 2.21e-06  2e-07  4e-07 0:01.3\n",
      "  537   5370 1.000000000030009e-01 4.1e+05 2.13e-06  2e-07  4e-07 0:01.3\n",
      "  538   5380 1.000000000031957e-01 4.0e+05 2.10e-06  2e-07  4e-07 0:01.3\n",
      "  539   5390 1.000000000027591e-01 4.0e+05 2.19e-06  2e-07  4e-07 0:01.3\n",
      "  540   5400 1.000000000023948e-01 4.0e+05 2.21e-06  3e-07  4e-07 0:01.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  541   5410 1.000000000014957e-01 4.0e+05 2.21e-06  3e-07  4e-07 0:01.3\n",
      "  542   5420 1.000000000020160e-01 4.0e+05 2.29e-06  3e-07  4e-07 0:01.3\n",
      "  543   5430 1.000000000016789e-01 4.0e+05 2.42e-06  3e-07  4e-07 0:01.3\n",
      "  544   5440 1.000000000020656e-01 4.3e+05 2.43e-06  3e-07  4e-07 0:01.3\n",
      "  545   5450 1.000000000012861e-01 4.3e+05 2.34e-06  3e-07  4e-07 0:01.3\n",
      "  546   5460 1.000000000017979e-01 4.4e+05 2.31e-06  3e-07  4e-07 0:01.3\n",
      "  547   5470 1.000000000025774e-01 4.5e+05 2.35e-06  3e-07  4e-07 0:01.3\n",
      "  548   5480 1.000000000018652e-01 4.5e+05 2.23e-06  2e-07  4e-07 0:01.3\n",
      "  549   5490 1.000000000010536e-01 4.6e+05 2.11e-06  2e-07  3e-07 0:01.3\n",
      "  550   5500 1.000000000010478e-01 4.8e+05 2.13e-06  2e-07  3e-07 0:01.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  551   5510 1.000000000012073e-01 5.1e+05 2.27e-06  2e-07  4e-07 0:01.3\n",
      "  552   5520 1.000000000016452e-01 5.1e+05 2.26e-06  2e-07  4e-07 0:01.3\n",
      "  553   5530 1.000000000011762e-01 5.4e+05 2.18e-06  2e-07  4e-07 0:01.3\n",
      "  554   5540 1.000000000003504e-01 5.5e+05 1.95e-06  2e-07  3e-07 0:01.3\n",
      "  555   5550 1.000000000008620e-01 5.3e+05 1.88e-06  2e-07  3e-07 0:01.3\n",
      "  556   5560 1.000000000007595e-01 5.6e+05 1.89e-06  2e-07  3e-07 0:01.3\n",
      "  557   5570 1.000000000012888e-01 5.8e+05 1.97e-06  2e-07  3e-07 0:01.3\n",
      "  558   5580 1.000000000014060e-01 5.8e+05 1.86e-06  2e-07  3e-07 0:01.3\n",
      "  559   5590 1.000000000010048e-01 5.6e+05 1.73e-06  2e-07  3e-07 0:01.3\n",
      "  560   5600 1.000000000009517e-01 5.9e+05 1.59e-06  2e-07  3e-07 0:01.3\n",
      "  560   5600 1.000000000009517e-01 5.9e+05 1.59e-06  2e-07  3e-07 0:01.3\n",
      "termination by {'tolfun': 1e-11}\n",
      "best f-value = 0.10000000000035042\n",
      "solution = [0.1000001  0.09999973 0.09999974 0.09999996 0.10000004 0.09999999\n",
      " 0.10000027 0.10000008 0.10000028 0.0999998 ]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<cma.evolution_strategy.CMAEvolutionStrategy at 0x11d35b0b8>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "es.optimize(f, verb_disp=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5_w,10)-aCMA-ES (mu_w=3.2,w_1=45%) in dimension 10 (seed=373207, Fri Dec 13 11:30:48 2019)\n"
     ]
    }
   ],
   "source": [
    "es = cma.CMAEvolutionStrategy(10 * [0.2], 0.5, {'bounds': [0, 2]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "    1     10 7.668985942567325e-01 1.0e+00 4.70e-01  5e-01  5e-01 0:00.0\n",
      "    2     20 1.414899172523549e+00 1.1e+00 4.43e-01  4e-01  4e-01 0:00.0\n",
      "    3     30 5.735311686154008e-01 1.2e+00 4.00e-01  4e-01  4e-01 0:00.0\n",
      "    4     40 1.121951169211186e+00 1.3e+00 3.62e-01  3e-01  4e-01 0:00.0\n",
      "    5     50 6.354064975855318e-01 1.4e+00 3.22e-01  3e-01  3e-01 0:00.0\n",
      "    6     60 4.834793194409244e-01 1.4e+00 2.91e-01  3e-01  3e-01 0:00.0\n",
      "    7     70 4.751461778726949e-01 1.4e+00 2.81e-01  2e-01  3e-01 0:00.0\n",
      "    8     80 5.516437432406355e-01 1.5e+00 2.74e-01  2e-01  3e-01 0:00.0\n",
      "    9     90 4.380787916330521e-01 1.5e+00 2.47e-01  2e-01  3e-01 0:00.0\n",
      "   10    100 1.738513751102683e-01 1.5e+00 2.21e-01  2e-01  2e-01 0:00.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   11    110 2.472633309889963e-01 1.6e+00 2.03e-01  2e-01  2e-01 0:00.0\n",
      "   12    120 2.161425207762034e-01 1.6e+00 1.79e-01  1e-01  2e-01 0:00.0\n",
      "   13    130 1.821756794552379e-01 1.6e+00 1.56e-01  1e-01  2e-01 0:00.0\n",
      "   14    140 2.511329527024557e-01 1.7e+00 1.42e-01  1e-01  1e-01 0:00.0\n",
      "   15    150 1.778635841653047e-01 1.7e+00 1.39e-01  1e-01  1e-01 0:00.0\n",
      "   16    160 2.388016940226109e-01 1.7e+00 1.39e-01  1e-01  1e-01 0:00.1\n",
      "   17    170 2.098023777491922e-01 1.7e+00 1.24e-01  1e-01  1e-01 0:00.1\n",
      "   18    180 1.500183874467844e-01 1.7e+00 1.15e-01  9e-02  1e-01 0:00.1\n",
      "   19    190 1.923655775938751e-01 1.6e+00 1.06e-01  8e-02  1e-01 0:00.1\n",
      "   20    200 2.110753952285984e-01 1.7e+00 1.02e-01  7e-02  1e-01 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   21    210 2.033209910396835e-01 1.7e+00 9.86e-02  7e-02  1e-01 0:00.1\n",
      "   22    220 1.173260257027837e-01 1.8e+00 9.32e-02  7e-02  9e-02 0:00.1\n",
      "   23    230 1.589635907903795e-01 1.7e+00 8.83e-02  7e-02  8e-02 0:00.1\n",
      "   24    240 1.756163935829783e-01 1.7e+00 7.91e-02  6e-02  7e-02 0:00.1\n",
      "   25    250 1.628648498788653e-01 1.7e+00 7.23e-02  5e-02  7e-02 0:00.1\n",
      "   26    260 1.683025297780811e-01 1.8e+00 6.72e-02  5e-02  6e-02 0:00.1\n",
      "   27    270 1.557167350246560e-01 1.9e+00 6.39e-02  5e-02  6e-02 0:00.1\n",
      "   28    280 1.678270170092635e-01 1.9e+00 6.13e-02  4e-02  5e-02 0:00.1\n",
      "   29    290 1.540190189444849e-01 1.9e+00 5.99e-02  4e-02  5e-02 0:00.1\n",
      "   30    300 1.647708128144511e-01 1.9e+00 5.77e-02  4e-02  5e-02 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   31    310 1.591936756372401e-01 1.9e+00 5.55e-02  4e-02  5e-02 0:00.1\n",
      "   32    320 1.445473101218533e-01 2.1e+00 5.21e-02  4e-02  5e-02 0:00.1\n",
      "   33    330 1.401098361999077e-01 2.1e+00 5.08e-02  4e-02  4e-02 0:00.1\n",
      "   34    340 1.161195697096378e-01 2.1e+00 5.29e-02  4e-02  5e-02 0:00.1\n",
      "   35    350 1.347187192282962e-01 2.1e+00 5.78e-02  4e-02  5e-02 0:00.1\n",
      "   36    360 1.301708617019302e-01 2.2e+00 6.13e-02  4e-02  5e-02 0:00.1\n",
      "   37    370 1.285060968858188e-01 2.2e+00 6.24e-02  4e-02  5e-02 0:00.1\n",
      "   38    380 1.337311403471745e-01 2.2e+00 6.05e-02  4e-02  5e-02 0:00.1\n",
      "   39    390 1.273117773811547e-01 2.3e+00 5.44e-02  3e-02  5e-02 0:00.1\n",
      "   40    400 1.273557369263369e-01 2.4e+00 5.39e-02  3e-02  5e-02 0:00.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   41    410 1.371525989363863e-01 2.3e+00 5.27e-02  3e-02  5e-02 0:00.1\n",
      "   42    420 1.447206701099271e-01 2.4e+00 5.32e-02  4e-02  5e-02 0:00.1\n",
      "   43    430 1.426322263483424e-01 2.4e+00 4.93e-02  3e-02  4e-02 0:00.1\n",
      "   44    440 1.295533908986020e-01 2.5e+00 4.86e-02  3e-02  4e-02 0:00.1\n",
      "   45    450 1.255924037312177e-01 2.6e+00 4.86e-02  3e-02  4e-02 0:00.1\n",
      "   46    460 1.356334865861616e-01 2.6e+00 4.82e-02  3e-02  4e-02 0:00.1\n",
      "   47    470 1.333798871248834e-01 2.7e+00 4.61e-02  3e-02  4e-02 0:00.1\n",
      "   48    480 1.278519778211585e-01 2.7e+00 4.08e-02  3e-02  4e-02 0:00.2\n",
      "   49    490 1.190363822527805e-01 2.9e+00 4.04e-02  2e-02  4e-02 0:00.2\n",
      "   50    500 1.239553658549432e-01 3.0e+00 4.23e-02  3e-02  4e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   51    510 1.284621905381344e-01 3.1e+00 4.26e-02  3e-02  4e-02 0:00.2\n",
      "   52    520 1.192762192861073e-01 3.3e+00 4.29e-02  3e-02  4e-02 0:00.2\n",
      "   53    530 1.249193923986348e-01 3.4e+00 4.09e-02  2e-02  4e-02 0:00.2\n",
      "   54    540 1.174590919977252e-01 3.4e+00 3.71e-02  2e-02  3e-02 0:00.2\n",
      "   55    550 1.144360286860281e-01 3.5e+00 3.47e-02  2e-02  3e-02 0:00.2\n",
      "   56    560 1.083655342739840e-01 3.5e+00 3.40e-02  2e-02  3e-02 0:00.2\n",
      "   57    570 1.059109256303142e-01 3.6e+00 3.42e-02  2e-02  3e-02 0:00.2\n",
      "   58    580 1.110368568089975e-01 3.6e+00 3.52e-02  2e-02  3e-02 0:00.2\n",
      "   59    590 1.159630395255459e-01 3.7e+00 3.40e-02  2e-02  3e-02 0:00.2\n",
      "   60    600 1.084376769518938e-01 3.8e+00 3.30e-02  2e-02  3e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   61    610 1.121847296132423e-01 3.8e+00 3.23e-02  2e-02  3e-02 0:00.2\n",
      "   62    620 1.152270194221038e-01 3.9e+00 2.97e-02  2e-02  3e-02 0:00.2\n",
      "   63    630 1.088232490765903e-01 3.8e+00 2.89e-02  2e-02  2e-02 0:00.2\n",
      "   64    640 1.076439471033054e-01 4.0e+00 2.67e-02  2e-02  2e-02 0:00.2\n",
      "   65    650 1.141058179152681e-01 4.0e+00 2.61e-02  1e-02  2e-02 0:00.2\n",
      "   66    660 1.072923251126648e-01 4.2e+00 2.43e-02  1e-02  2e-02 0:00.2\n",
      "   67    670 1.078539409224621e-01 4.4e+00 2.55e-02  1e-02  2e-02 0:00.2\n",
      "   68    680 1.080869750224502e-01 4.3e+00 2.25e-02  1e-02  2e-02 0:00.2\n",
      "   69    690 1.066288479731588e-01 4.6e+00 2.20e-02  1e-02  2e-02 0:00.2\n",
      "   70    700 1.082139363789228e-01 4.6e+00 2.01e-02  1e-02  2e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   71    710 1.054244892083646e-01 4.6e+00 1.92e-02  1e-02  2e-02 0:00.2\n",
      "   72    720 1.059985379806978e-01 4.9e+00 1.83e-02  1e-02  1e-02 0:00.2\n",
      "   73    730 1.073196495548852e-01 5.1e+00 1.91e-02  1e-02  2e-02 0:00.2\n",
      "   74    740 1.051558997131908e-01 5.2e+00 1.86e-02  1e-02  2e-02 0:00.2\n",
      "   75    750 1.025740047066066e-01 5.2e+00 1.92e-02  1e-02  2e-02 0:00.2\n",
      "   76    760 1.048165968926292e-01 5.1e+00 1.98e-02  1e-02  2e-02 0:00.2\n",
      "   77    770 1.048655873996877e-01 5.2e+00 1.88e-02  1e-02  2e-02 0:00.2\n",
      "   78    780 1.062826974964170e-01 5.4e+00 1.92e-02  1e-02  2e-02 0:00.2\n",
      "   79    790 1.050236318683742e-01 5.9e+00 1.89e-02  1e-02  2e-02 0:00.2\n",
      "   80    800 1.043109813581795e-01 5.8e+00 1.77e-02  9e-03  1e-02 0:00.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   81    810 1.039631974731631e-01 6.2e+00 1.60e-02  8e-03  1e-02 0:00.2\n",
      "   82    820 1.038731607154312e-01 6.3e+00 1.44e-02  7e-03  1e-02 0:00.2\n",
      "   83    830 1.030631288752175e-01 6.4e+00 1.40e-02  7e-03  1e-02 0:00.2\n",
      "   84    840 1.023301419244172e-01 6.6e+00 1.42e-02  7e-03  1e-02 0:00.2\n",
      "   85    850 1.026046110993640e-01 6.8e+00 1.53e-02  8e-03  1e-02 0:00.2\n",
      "   86    860 1.042925100051831e-01 6.5e+00 1.74e-02  9e-03  1e-02 0:00.3\n",
      "   87    870 1.028285117988126e-01 6.8e+00 1.71e-02  9e-03  1e-02 0:00.3\n",
      "   88    880 1.033659578566993e-01 7.1e+00 1.66e-02  9e-03  1e-02 0:00.3\n",
      "   89    890 1.033049858209953e-01 7.2e+00 1.50e-02  8e-03  1e-02 0:00.3\n",
      "   90    900 1.019641123885700e-01 7.3e+00 1.42e-02  7e-03  1e-02 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "   91    910 1.016229812376689e-01 7.6e+00 1.33e-02  7e-03  1e-02 0:00.3\n",
      "   92    920 1.024052603733018e-01 8.0e+00 1.30e-02  7e-03  1e-02 0:00.3\n",
      "   93    930 1.024180764026441e-01 8.4e+00 1.20e-02  6e-03  9e-03 0:00.3\n",
      "   94    940 1.021103998821002e-01 8.9e+00 1.08e-02  6e-03  8e-03 0:00.3\n",
      "   95    950 1.017826187867491e-01 9.1e+00 1.02e-02  5e-03  8e-03 0:00.3\n",
      "   96    960 1.022092467579468e-01 9.4e+00 9.23e-03  5e-03  7e-03 0:00.3\n",
      "   97    970 1.014460695721835e-01 9.9e+00 9.34e-03  5e-03  7e-03 0:00.3\n",
      "   98    980 1.015467171453942e-01 1.0e+01 8.99e-03  4e-03  7e-03 0:00.3\n",
      "   99    990 1.012929902426106e-01 1.0e+01 8.88e-03  4e-03  7e-03 0:00.3\n",
      "  100   1000 1.012515521166817e-01 1.1e+01 8.99e-03  5e-03  7e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  101   1010 1.016388905761460e-01 1.1e+01 8.71e-03  4e-03  7e-03 0:00.3\n",
      "  102   1020 1.014744145172888e-01 1.1e+01 8.63e-03  4e-03  7e-03 0:00.3\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  103   1030 1.011715949651647e-01 1.1e+01 8.39e-03  4e-03  6e-03 0:00.3\n",
      "  104   1040 1.011037871329515e-01 1.1e+01 8.42e-03  4e-03  6e-03 0:00.3\n",
      "  105   1050 1.010305467529991e-01 1.1e+01 7.78e-03  4e-03  6e-03 0:00.3\n",
      "  106   1060 1.010089683321356e-01 1.1e+01 7.63e-03  4e-03  6e-03 0:00.3\n",
      "  107   1070 1.012580642851836e-01 1.1e+01 7.51e-03  4e-03  5e-03 0:00.3\n",
      "  108   1080 1.008584740100027e-01 1.1e+01 7.33e-03  3e-03  5e-03 0:00.3\n",
      "  109   1090 1.007063813233309e-01 1.2e+01 7.39e-03  3e-03  5e-03 0:00.3\n",
      "  110   1100 1.006778116062983e-01 1.2e+01 7.37e-03  3e-03  5e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  111   1110 1.005922433948027e-01 1.2e+01 7.15e-03  3e-03  5e-03 0:00.3\n",
      "  112   1120 1.005749269840782e-01 1.3e+01 6.81e-03  3e-03  5e-03 0:00.3\n",
      "  113   1130 1.007766810769795e-01 1.3e+01 6.50e-03  3e-03  5e-03 0:00.3\n",
      "  114   1140 1.005224531546569e-01 1.3e+01 6.29e-03  3e-03  5e-03 0:00.3\n",
      "  115   1150 1.005561303795112e-01 1.4e+01 6.77e-03  3e-03  5e-03 0:00.3\n",
      "  116   1160 1.005221287262891e-01 1.4e+01 7.01e-03  3e-03  5e-03 0:00.3\n",
      "  117   1170 1.003751430681558e-01 1.4e+01 7.12e-03  4e-03  5e-03 0:00.3\n",
      "  118   1180 1.003786814883820e-01 1.5e+01 7.01e-03  4e-03  5e-03 0:00.3\n",
      "  119   1190 1.004597102533174e-01 1.5e+01 7.24e-03  4e-03  5e-03 0:00.3\n",
      "  120   1200 1.004416777383787e-01 1.5e+01 7.15e-03  4e-03  5e-03 0:00.3\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  121   1210 1.004674927976577e-01 1.6e+01 6.86e-03  3e-03  5e-03 0:00.3\n",
      "  122   1220 1.004141744705890e-01 1.6e+01 6.30e-03  3e-03  4e-03 0:00.3\n",
      "  123   1230 1.003297695310222e-01 1.6e+01 6.04e-03  3e-03  4e-03 0:00.3\n",
      "  124   1240 1.003448889015703e-01 1.7e+01 5.61e-03  3e-03  4e-03 0:00.3\n",
      "  125   1250 1.003288545908785e-01 1.8e+01 5.56e-03  3e-03  4e-03 0:00.3\n",
      "  126   1260 1.002906585571147e-01 1.8e+01 5.41e-03  3e-03  4e-03 0:00.3\n",
      "  127   1270 1.002676031724203e-01 1.9e+01 5.22e-03  3e-03  3e-03 0:00.4\n",
      "  128   1280 1.002549704377348e-01 2.0e+01 4.83e-03  2e-03  3e-03 0:00.4\n",
      "  129   1290 1.002794570646842e-01 2.0e+01 4.57e-03  2e-03  3e-03 0:00.4\n",
      "  130   1300 1.002053276211666e-01 2.1e+01 4.32e-03  2e-03  3e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  131   1310 1.001541411886602e-01 2.1e+01 4.21e-03  2e-03  3e-03 0:00.4\n",
      "  132   1320 1.002094969146077e-01 2.1e+01 3.99e-03  2e-03  2e-03 0:00.4\n",
      "  133   1330 1.002025133597207e-01 2.1e+01 3.74e-03  2e-03  2e-03 0:00.4\n",
      "  134   1340 1.001282180391730e-01 2.2e+01 3.50e-03  2e-03  2e-03 0:00.4\n",
      "  135   1350 1.001064337597333e-01 2.2e+01 3.26e-03  2e-03  2e-03 0:00.4\n",
      "  136   1360 1.001118435614359e-01 2.3e+01 3.55e-03  2e-03  2e-03 0:00.4\n",
      "  137   1370 1.001411247277009e-01 2.6e+01 3.92e-03  2e-03  2e-03 0:00.4\n",
      "  138   1380 1.001689911506739e-01 2.6e+01 3.87e-03  2e-03  2e-03 0:00.4\n",
      "  139   1390 1.001891532661574e-01 2.7e+01 3.84e-03  2e-03  2e-03 0:00.4\n",
      "  140   1400 1.001872834082991e-01 2.8e+01 3.62e-03  2e-03  2e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  141   1410 1.001408405961525e-01 3.0e+01 3.17e-03  2e-03  2e-03 0:00.4\n",
      "  142   1420 1.001435298382258e-01 3.2e+01 3.06e-03  1e-03  2e-03 0:00.4\n",
      "  143   1430 1.001346222475251e-01 3.4e+01 2.83e-03  1e-03  2e-03 0:00.4\n",
      "  144   1440 1.001841954945279e-01 3.4e+01 2.49e-03  1e-03  1e-03 0:00.4\n",
      "  145   1450 1.001519772408649e-01 3.3e+01 2.43e-03  1e-03  1e-03 0:00.4\n",
      "  146   1460 1.001339499780980e-01 3.5e+01 2.38e-03  1e-03  1e-03 0:00.4\n",
      "  147   1470 1.000866999279502e-01 3.5e+01 2.35e-03  1e-03  1e-03 0:00.4\n",
      "  148   1480 1.000853849245933e-01 3.4e+01 2.39e-03  1e-03  1e-03 0:00.4\n",
      "  149   1490 1.000665769456899e-01 3.5e+01 2.38e-03  1e-03  1e-03 0:00.4\n",
      "  150   1500 1.000633584639490e-01 3.6e+01 2.51e-03  1e-03  2e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  151   1510 1.000569803530215e-01 3.8e+01 2.61e-03  1e-03  2e-03 0:00.4\n",
      "  152   1520 1.000643103014539e-01 3.8e+01 2.51e-03  1e-03  2e-03 0:00.4\n",
      "  153   1530 1.000667898371827e-01 3.9e+01 2.64e-03  1e-03  2e-03 0:00.4\n",
      "  154   1540 1.000497065266059e-01 3.9e+01 2.60e-03  1e-03  2e-03 0:00.4\n",
      "  155   1550 1.000695956601338e-01 4.1e+01 2.51e-03  1e-03  1e-03 0:00.4\n",
      "  156   1560 1.000791625280551e-01 4.1e+01 2.38e-03  1e-03  1e-03 0:00.4\n",
      "  157   1570 1.000566368945826e-01 4.2e+01 2.19e-03  1e-03  1e-03 0:00.4\n",
      "  158   1580 1.000741294589833e-01 4.3e+01 2.15e-03  1e-03  1e-03 0:00.4\n",
      "  159   1590 1.000746317684824e-01 4.7e+01 2.27e-03  1e-03  1e-03 0:00.4\n",
      "  160   1600 1.000474129826612e-01 5.1e+01 2.22e-03  1e-03  1e-03 0:00.4\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  161   1610 1.000749731053382e-01 5.1e+01 2.14e-03  1e-03  1e-03 0:00.4\n",
      "  162   1620 1.000669043029338e-01 5.1e+01 2.01e-03  9e-04  1e-03 0:00.4\n",
      "  163   1630 1.000611837002039e-01 5.2e+01 1.85e-03  8e-04  1e-03 0:00.4\n",
      "  164   1640 1.000530647903412e-01 5.3e+01 1.82e-03  8e-04  1e-03 0:00.4\n",
      "  165   1650 1.000428700980011e-01 5.2e+01 1.70e-03  7e-04  1e-03 0:00.4\n",
      "  166   1660 1.000350116774426e-01 5.2e+01 1.66e-03  7e-04  1e-03 0:00.4\n",
      "  167   1670 1.000429260709861e-01 5.2e+01 1.59e-03  7e-04  9e-04 0:00.4\n",
      "  168   1680 1.000521233775804e-01 5.4e+01 1.64e-03  7e-04  1e-03 0:00.4\n",
      "  169   1690 1.000351699160797e-01 5.6e+01 1.45e-03  6e-04  9e-04 0:00.4\n",
      "  170   1700 1.000359225906618e-01 5.8e+01 1.25e-03  5e-04  8e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  171   1710 1.000252845325552e-01 5.8e+01 1.20e-03  5e-04  7e-04 0:00.5\n",
      "  172   1720 1.000350495709647e-01 5.7e+01 1.18e-03  5e-04  7e-04 0:00.5\n",
      "  173   1730 1.000276530157505e-01 5.8e+01 1.13e-03  5e-04  7e-04 0:00.5\n",
      "  174   1740 1.000246049378122e-01 6.0e+01 1.10e-03  5e-04  7e-04 0:00.5\n",
      "  175   1750 1.000238337490040e-01 6.3e+01 1.09e-03  4e-04  7e-04 0:00.5\n",
      "  176   1760 1.000224669235158e-01 6.3e+01 1.01e-03  4e-04  6e-04 0:00.5\n",
      "  177   1770 1.000231960340609e-01 6.4e+01 9.33e-04  4e-04  6e-04 0:00.5\n",
      "  178   1780 1.000197621070312e-01 6.3e+01 9.15e-04  4e-04  6e-04 0:00.5\n",
      "  179   1790 1.000201672561624e-01 6.5e+01 9.53e-04  4e-04  6e-04 0:00.5\n",
      "  180   1800 1.000198068527519e-01 6.6e+01 1.05e-03  4e-04  7e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  181   1810 1.000241036404215e-01 7.3e+01 1.13e-03  4e-04  7e-04 0:00.5\n",
      "  182   1820 1.000172962464930e-01 7.9e+01 1.10e-03  4e-04  7e-04 0:00.5\n",
      "  183   1830 1.000185937757589e-01 8.0e+01 1.03e-03  4e-04  6e-04 0:00.5\n",
      "  184   1840 1.000126448509566e-01 8.0e+01 1.02e-03  4e-04  6e-04 0:00.5\n",
      "  185   1850 1.000143890123676e-01 8.1e+01 9.94e-04  4e-04  6e-04 0:00.5\n",
      "  186   1860 1.000113485991927e-01 8.2e+01 9.12e-04  3e-04  6e-04 0:00.5\n",
      "  187   1870 1.000113029958035e-01 8.1e+01 8.72e-04  3e-04  5e-04 0:00.5\n",
      "  188   1880 1.000112740643246e-01 8.4e+01 8.63e-04  3e-04  6e-04 0:00.5\n",
      "  189   1890 1.000125510145072e-01 8.5e+01 8.39e-04  3e-04  5e-04 0:00.5\n",
      "  190   1900 1.000141446249164e-01 8.6e+01 7.73e-04  3e-04  5e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  191   1910 1.000087590400887e-01 8.8e+01 7.37e-04  3e-04  5e-04 0:00.5\n",
      "  192   1920 1.000089731483003e-01 8.7e+01 7.24e-04  3e-04  5e-04 0:00.5\n",
      "  193   1930 1.000058869804683e-01 9.0e+01 7.22e-04  3e-04  5e-04 0:00.5\n",
      "  194   1940 1.000063115757349e-01 9.6e+01 7.66e-04  3e-04  5e-04 0:00.5\n",
      "  195   1950 1.000059665503563e-01 9.9e+01 8.33e-04  3e-04  5e-04 0:00.5\n",
      "  196   1960 1.000074920661001e-01 1.0e+02 8.03e-04  3e-04  5e-04 0:00.5\n",
      "  197   1970 1.000072782558333e-01 1.0e+02 7.35e-04  2e-04  5e-04 0:00.5\n",
      "  198   1980 1.000058443155493e-01 1.1e+02 6.55e-04  2e-04  4e-04 0:00.5\n",
      "  199   1990 1.000036178696405e-01 1.1e+02 6.64e-04  2e-04  4e-04 0:00.5\n",
      "  200   2000 1.000054427928247e-01 1.2e+02 6.87e-04  2e-04  4e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  201   2010 1.000043543043384e-01 1.2e+02 6.94e-04  2e-04  4e-04 0:00.5\n",
      "  202   2020 1.000042218621017e-01 1.3e+02 6.51e-04  2e-04  4e-04 0:00.5\n",
      "  203   2030 1.000048986885524e-01 1.4e+02 6.59e-04  2e-04  4e-04 0:00.5\n",
      "  204   2040 1.000048914850542e-01 1.5e+02 7.20e-04  2e-04  5e-04 0:00.5\n",
      "  205   2050 1.000035335216947e-01 1.6e+02 7.00e-04  2e-04  5e-04 0:00.5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  206   2060 1.000041282735206e-01 1.7e+02 6.55e-04  2e-04  4e-04 0:00.5\n",
      "  207   2070 1.000037399338369e-01 1.7e+02 6.23e-04  2e-04  4e-04 0:00.5\n",
      "  208   2080 1.000039066031424e-01 1.8e+02 6.11e-04  2e-04  4e-04 0:00.5\n",
      "  209   2090 1.000040841280850e-01 1.9e+02 5.99e-04  2e-04  4e-04 0:00.5\n",
      "  210   2100 1.000037484775259e-01 1.9e+02 5.57e-04  2e-04  4e-04 0:00.5\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  211   2110 1.000027483710249e-01 2.0e+02 5.20e-04  2e-04  4e-04 0:00.5\n",
      "  212   2120 1.000031412510453e-01 2.0e+02 5.14e-04  2e-04  3e-04 0:00.5\n",
      "  213   2130 1.000032233894841e-01 2.0e+02 5.14e-04  2e-04  3e-04 0:00.5\n",
      "  214   2140 1.000026931350515e-01 2.0e+02 4.95e-04  2e-04  3e-04 0:00.5\n",
      "  215   2150 1.000028845285776e-01 2.1e+02 4.74e-04  2e-04  3e-04 0:00.6\n",
      "  216   2160 1.000020200120704e-01 2.2e+02 5.11e-04  2e-04  3e-04 0:00.6\n",
      "  217   2170 1.000012311768385e-01 2.3e+02 5.12e-04  2e-04  3e-04 0:00.6\n",
      "  218   2180 1.000025161199038e-01 2.3e+02 4.92e-04  2e-04  3e-04 0:00.6\n",
      "  219   2190 1.000017071654419e-01 2.4e+02 4.71e-04  2e-04  3e-04 0:00.6\n",
      "  220   2200 1.000019072972932e-01 2.4e+02 4.38e-04  1e-04  3e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  221   2210 1.000015179350199e-01 2.5e+02 4.06e-04  1e-04  3e-04 0:00.6\n",
      "  222   2220 1.000016298877986e-01 2.6e+02 3.89e-04  1e-04  3e-04 0:00.6\n",
      "  223   2230 1.000025029004827e-01 2.7e+02 3.66e-04  1e-04  2e-04 0:00.6\n",
      "  224   2240 1.000017935738314e-01 2.7e+02 3.40e-04  1e-04  2e-04 0:00.6\n",
      "  225   2250 1.000016051032162e-01 2.8e+02 3.15e-04  1e-04  2e-04 0:00.6\n",
      "  226   2260 1.000011932379098e-01 2.8e+02 3.43e-04  1e-04  2e-04 0:00.6\n",
      "  227   2270 1.000014041784021e-01 2.9e+02 3.57e-04  1e-04  2e-04 0:00.6\n",
      "  228   2280 1.000011419533108e-01 3.0e+02 3.23e-04  1e-04  2e-04 0:00.6\n",
      "  229   2290 1.000009345039392e-01 3.0e+02 3.14e-04  1e-04  2e-04 0:00.6\n",
      "  230   2300 1.000007831082612e-01 3.1e+02 3.08e-04  1e-04  2e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  231   2310 1.000010217256747e-01 3.2e+02 3.06e-04  1e-04  2e-04 0:00.6\n",
      "  232   2320 1.000005403911763e-01 3.1e+02 2.97e-04  1e-04  2e-04 0:00.6\n",
      "  233   2330 1.000006939873732e-01 3.3e+02 2.97e-04  1e-04  2e-04 0:00.6\n",
      "  234   2340 1.000008063629760e-01 3.3e+02 3.15e-04  1e-04  2e-04 0:00.6\n",
      "  235   2350 1.000005955350731e-01 3.4e+02 3.13e-04  1e-04  2e-04 0:00.6\n",
      "  236   2360 1.000004656701240e-01 3.5e+02 3.05e-04  1e-04  2e-04 0:00.6\n",
      "  237   2370 1.000007057547277e-01 3.7e+02 3.12e-04  1e-04  2e-04 0:00.6\n",
      "  238   2380 1.000003775527528e-01 3.7e+02 3.40e-04  1e-04  2e-04 0:00.6\n",
      "  239   2390 1.000003607730302e-01 3.9e+02 3.75e-04  1e-04  2e-04 0:00.6\n",
      "  240   2400 1.000003259120903e-01 3.9e+02 3.81e-04  1e-04  2e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  241   2410 1.000004243948243e-01 4.0e+02 3.94e-04  1e-04  2e-04 0:00.6\n",
      "  242   2420 1.000008966715460e-01 4.0e+02 4.06e-04  1e-04  2e-04 0:00.6\n",
      "  243   2430 1.000004283459352e-01 4.2e+02 4.03e-04  1e-04  2e-04 0:00.6\n",
      "  244   2440 1.000007365559585e-01 4.3e+02 3.89e-04  1e-04  2e-04 0:00.6\n",
      "  245   2450 1.000006969852352e-01 4.4e+02 3.63e-04  1e-04  2e-04 0:00.6\n",
      "  246   2460 1.000005256430253e-01 4.6e+02 3.57e-04  1e-04  2e-04 0:00.6\n",
      "  247   2470 1.000005026299388e-01 4.7e+02 3.47e-04  1e-04  2e-04 0:00.6\n",
      "  248   2480 1.000003659382901e-01 5.0e+02 2.99e-04  1e-04  2e-04 0:00.6\n",
      "  249   2490 1.000004669798762e-01 5.2e+02 2.81e-04  9e-05  2e-04 0:00.6\n",
      "  250   2500 1.000001430011297e-01 5.2e+02 2.62e-04  8e-05  2e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  251   2510 1.000002368235965e-01 5.2e+02 2.43e-04  8e-05  1e-04 0:00.6\n",
      "  252   2520 1.000003195336694e-01 5.3e+02 2.29e-04  7e-05  1e-04 0:00.6\n",
      "  253   2530 1.000001891061670e-01 5.6e+02 2.10e-04  7e-05  1e-04 0:00.6\n",
      "  254   2540 1.000002810346182e-01 5.6e+02 2.02e-04  6e-05  1e-04 0:00.6\n",
      "  255   2550 1.000002251727390e-01 5.7e+02 1.94e-04  6e-05  1e-04 0:00.6\n",
      "  256   2560 1.000002108033495e-01 6.1e+02 1.96e-04  6e-05  1e-04 0:00.6\n",
      "  257   2570 1.000004237956913e-01 6.4e+02 2.00e-04  6e-05  1e-04 0:00.6\n",
      "  258   2580 1.000003161198701e-01 6.9e+02 1.98e-04  6e-05  1e-04 0:00.6\n",
      "  259   2590 1.000002603416948e-01 7.0e+02 2.21e-04  7e-05  1e-04 0:00.6\n",
      "  260   2600 1.000002073513166e-01 7.2e+02 2.27e-04  7e-05  1e-04 0:00.6\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  261   2610 1.000002204926780e-01 7.3e+02 2.25e-04  7e-05  1e-04 0:00.7\n",
      "  262   2620 1.000002179519065e-01 7.6e+02 2.15e-04  7e-05  1e-04 0:00.7\n",
      "  263   2630 1.000001611078262e-01 7.4e+02 2.07e-04  6e-05  1e-04 0:00.7\n",
      "  264   2640 1.000002922967479e-01 7.6e+02 2.02e-04  6e-05  1e-04 0:00.7\n",
      "  265   2650 1.000002291043309e-01 7.8e+02 1.91e-04  6e-05  1e-04 0:00.7\n",
      "  266   2660 1.000001404162253e-01 7.8e+02 1.82e-04  5e-05  1e-04 0:00.7\n",
      "  267   2670 1.000001859712307e-01 8.2e+02 1.69e-04  5e-05  9e-05 0:00.7\n",
      "  268   2680 1.000001997191883e-01 8.3e+02 1.77e-04  5e-05  9e-05 0:00.7\n",
      "  269   2690 1.000001915632033e-01 8.7e+02 1.69e-04  5e-05  9e-05 0:00.7\n",
      "  270   2700 1.000002071103721e-01 8.9e+02 1.62e-04  5e-05  9e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  271   2710 1.000001258132351e-01 9.4e+02 1.50e-04  4e-05  8e-05 0:00.7\n",
      "  272   2720 1.000001452015966e-01 9.2e+02 1.57e-04  5e-05  8e-05 0:00.7\n",
      "  273   2730 1.000001398815609e-01 9.7e+02 1.48e-04  4e-05  8e-05 0:00.7\n",
      "  274   2740 1.000001018340749e-01 9.8e+02 1.49e-04  4e-05  8e-05 0:00.7\n",
      "  275   2750 1.000001168876935e-01 9.9e+02 1.44e-04  4e-05  8e-05 0:00.7\n",
      "  276   2760 1.000001256639087e-01 1.1e+03 1.32e-04  4e-05  7e-05 0:00.7\n",
      "  277   2770 1.000001296228902e-01 1.1e+03 1.29e-04  4e-05  7e-05 0:00.7\n",
      "  278   2780 1.000001087352607e-01 1.1e+03 1.37e-04  4e-05  7e-05 0:00.7\n",
      "  279   2790 1.000001134132257e-01 1.1e+03 1.46e-04  4e-05  8e-05 0:00.7\n",
      "  280   2800 1.000001061471302e-01 1.1e+03 1.39e-04  4e-05  7e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  281   2810 1.000000952820854e-01 1.1e+03 1.28e-04  4e-05  6e-05 0:00.7\n",
      "  282   2820 1.000001033482276e-01 1.2e+03 1.26e-04  4e-05  6e-05 0:00.7\n",
      "  283   2830 1.000000764510129e-01 1.3e+03 1.17e-04  3e-05  6e-05 0:00.7\n",
      "  284   2840 1.000000879937352e-01 1.3e+03 1.11e-04  3e-05  5e-05 0:00.7\n",
      "  285   2850 1.000000823479932e-01 1.3e+03 1.16e-04  3e-05  6e-05 0:00.7\n",
      "  286   2860 1.000000704563108e-01 1.3e+03 1.16e-04  3e-05  5e-05 0:00.7\n",
      "  287   2870 1.000000800811154e-01 1.3e+03 1.13e-04  3e-05  5e-05 0:00.7\n",
      "  288   2880 1.000000898190307e-01 1.4e+03 1.20e-04  4e-05  6e-05 0:00.7\n",
      "  289   2890 1.000000878829004e-01 1.4e+03 1.29e-04  4e-05  6e-05 0:00.7\n",
      "  290   2900 1.000000964095358e-01 1.5e+03 1.33e-04  4e-05  6e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  291   2910 1.000000743542958e-01 1.5e+03 1.38e-04  4e-05  7e-05 0:00.7\n",
      "  292   2920 1.000000674113421e-01 1.6e+03 1.37e-04  4e-05  6e-05 0:00.7\n",
      "  293   2930 1.000000833432010e-01 1.6e+03 1.25e-04  4e-05  6e-05 0:00.7\n",
      "  294   2940 1.000000828505711e-01 1.7e+03 1.15e-04  3e-05  5e-05 0:00.7\n",
      "  295   2950 1.000000599742745e-01 1.7e+03 1.14e-04  4e-05  5e-05 0:00.7\n",
      "  296   2960 1.000000551401355e-01 1.7e+03 1.10e-04  3e-05  5e-05 0:00.7\n",
      "  297   2970 1.000000445801841e-01 1.7e+03 1.08e-04  3e-05  5e-05 0:00.7\n",
      "  298   2980 1.000000574517760e-01 1.7e+03 1.08e-04  3e-05  5e-05 0:00.7\n",
      "  299   2990 1.000000539185095e-01 1.9e+03 9.78e-05  3e-05  4e-05 0:00.7\n",
      "  300   3000 1.000000600280141e-01 1.9e+03 8.90e-05  3e-05  4e-05 0:00.7\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  301   3010 1.000000488222976e-01 1.9e+03 8.47e-05  3e-05  4e-05 0:00.7\n",
      "  302   3020 1.000000556275353e-01 2.0e+03 7.99e-05  2e-05  4e-05 0:00.7\n",
      "  303   3030 1.000000519294749e-01 2.1e+03 7.10e-05  2e-05  3e-05 0:00.7\n",
      "  304   3040 1.000000432699850e-01 2.2e+03 6.34e-05  2e-05  3e-05 0:00.8\n",
      "  305   3050 1.000000342601620e-01 2.3e+03 6.10e-05  2e-05  3e-05 0:00.8\n",
      "  306   3060 1.000000420558901e-01 2.3e+03 6.05e-05  2e-05  3e-05 0:00.8\n",
      "  307   3070 1.000000388513778e-01 2.3e+03 5.85e-05  2e-05  3e-05 0:00.8\n",
      "  308   3080 1.000000301349767e-01 2.5e+03 6.03e-05  2e-05  3e-05 0:00.8\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  309   3090 1.000000262451041e-01 2.6e+03 6.67e-05  2e-05  3e-05 0:00.8\n",
      "  310   3100 1.000000365194784e-01 2.6e+03 6.87e-05  2e-05  3e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  311   3110 1.000000349912447e-01 2.7e+03 6.49e-05  2e-05  3e-05 0:00.8\n",
      "  312   3120 1.000000255242963e-01 2.7e+03 6.63e-05  2e-05  3e-05 0:00.8\n",
      "  313   3130 1.000000240488685e-01 2.7e+03 6.85e-05  2e-05  3e-05 0:00.8\n",
      "  314   3140 1.000000271480077e-01 2.7e+03 6.97e-05  2e-05  3e-05 0:00.8\n",
      "  315   3150 1.000000287347045e-01 2.9e+03 7.37e-05  2e-05  3e-05 0:00.8\n",
      "  316   3160 1.000000274238259e-01 3.1e+03 7.54e-05  2e-05  3e-05 0:00.8\n",
      "  317   3170 1.000000252840077e-01 3.2e+03 7.91e-05  2e-05  4e-05 0:00.8\n",
      "  318   3180 1.000000277644687e-01 3.3e+03 8.12e-05  2e-05  4e-05 0:00.8\n",
      "  319   3190 1.000000220389735e-01 3.3e+03 7.73e-05  2e-05  3e-05 0:00.8\n",
      "  320   3200 1.000000237195587e-01 3.4e+03 7.78e-05  2e-05  3e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  321   3210 1.000000210130780e-01 3.4e+03 7.07e-05  2e-05  3e-05 0:00.8\n",
      "  322   3220 1.000000256045362e-01 3.4e+03 7.45e-05  2e-05  3e-05 0:00.8\n",
      "  323   3230 1.000000212084349e-01 3.4e+03 7.01e-05  2e-05  3e-05 0:00.8\n",
      "  324   3240 1.000000178536357e-01 3.6e+03 6.40e-05  2e-05  3e-05 0:00.8\n",
      "  325   3250 1.000000210435574e-01 3.5e+03 5.79e-05  1e-05  3e-05 0:00.8\n",
      "  326   3260 1.000000210447648e-01 3.5e+03 5.52e-05  1e-05  2e-05 0:00.8\n",
      "  327   3270 1.000000179902747e-01 3.8e+03 5.08e-05  1e-05  2e-05 0:00.8\n",
      "  328   3280 1.000000180777637e-01 4.0e+03 4.59e-05  1e-05  2e-05 0:00.8\n",
      "  329   3290 1.000000182971772e-01 4.2e+03 4.51e-05  1e-05  2e-05 0:00.8\n",
      "  330   3300 1.000000170049005e-01 4.2e+03 4.24e-05  1e-05  2e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  331   3310 1.000000126868550e-01 4.1e+03 4.05e-05  1e-05  2e-05 0:00.8\n",
      "  332   3320 1.000000118006763e-01 4.1e+03 4.22e-05  1e-05  2e-05 0:00.8\n",
      "  333   3330 1.000000125742146e-01 4.2e+03 4.32e-05  1e-05  2e-05 0:00.8\n",
      "  334   3340 1.000000123518899e-01 4.2e+03 4.26e-05  1e-05  2e-05 0:00.8\n",
      "  335   3350 1.000000140879174e-01 4.3e+03 4.09e-05  9e-06  2e-05 0:00.8\n",
      "  336   3360 1.000000153808568e-01 4.4e+03 4.16e-05  1e-05  2e-05 0:00.8\n",
      "  337   3370 1.000000109319745e-01 4.5e+03 4.20e-05  1e-05  2e-05 0:00.8\n",
      "  338   3380 1.000000084953578e-01 4.7e+03 3.97e-05  9e-06  2e-05 0:00.8\n",
      "  339   3390 1.000000097951468e-01 4.7e+03 3.60e-05  8e-06  1e-05 0:00.8\n",
      "  340   3400 1.000000089562488e-01 4.7e+03 3.55e-05  8e-06  1e-05 0:00.8\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  341   3410 1.000000100126108e-01 5.1e+03 3.52e-05  8e-06  1e-05 0:00.8\n",
      "  342   3420 1.000000095300237e-01 5.3e+03 3.37e-05  7e-06  1e-05 0:00.8\n",
      "  343   3430 1.000000103560320e-01 5.5e+03 3.14e-05  7e-06  1e-05 0:00.8\n",
      "  344   3440 1.000000095566283e-01 5.6e+03 2.95e-05  6e-06  1e-05 0:00.8\n",
      "  345   3450 1.000000075563971e-01 5.8e+03 2.78e-05  6e-06  1e-05 0:00.9\n",
      "  346   3460 1.000000081370963e-01 6.0e+03 2.79e-05  6e-06  1e-05 0:00.9\n",
      "  347   3470 1.000000086314717e-01 6.2e+03 2.77e-05  6e-06  1e-05 0:00.9\n",
      "  348   3480 1.000000072952624e-01 6.3e+03 2.95e-05  7e-06  1e-05 0:00.9\n",
      "  349   3490 1.000000076840452e-01 6.3e+03 3.20e-05  7e-06  1e-05 0:00.9\n",
      "  350   3500 1.000000071410257e-01 6.3e+03 3.51e-05  8e-06  1e-05 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  351   3510 1.000000072620117e-01 6.3e+03 3.78e-05  8e-06  1e-05 0:00.9\n",
      "  352   3520 1.000000076209648e-01 6.7e+03 4.27e-05  9e-06  2e-05 0:00.9\n",
      "  353   3530 1.000000074389461e-01 7.6e+03 4.40e-05  1e-05  2e-05 0:00.9\n",
      "  354   3540 1.000000069945415e-01 7.9e+03 4.51e-05  1e-05  2e-05 0:00.9\n",
      "  355   3550 1.000000070513032e-01 8.7e+03 4.91e-05  1e-05  2e-05 0:00.9\n",
      "  356   3560 1.000000079716759e-01 8.8e+03 4.93e-05  1e-05  2e-05 0:00.9\n",
      "  357   3570 1.000000064028680e-01 9.3e+03 4.46e-05  1e-05  2e-05 0:00.9\n",
      "  358   3580 1.000000059721270e-01 9.2e+03 3.84e-05  9e-06  1e-05 0:00.9\n",
      "  359   3590 1.000000049968833e-01 9.5e+03 3.63e-05  9e-06  1e-05 0:00.9\n",
      "  360   3600 1.000000063267816e-01 9.6e+03 3.56e-05  9e-06  1e-05 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  361   3610 1.000000059266795e-01 9.7e+03 3.48e-05  9e-06  1e-05 0:00.9\n",
      "  362   3620 1.000000053992612e-01 9.5e+03 3.55e-05  9e-06  1e-05 0:00.9\n",
      "  363   3630 1.000000059360045e-01 9.8e+03 3.69e-05  9e-06  1e-05 0:00.9\n",
      "  364   3640 1.000000055086693e-01 1.0e+04 3.93e-05  1e-05  1e-05 0:00.9\n",
      "  365   3650 1.000000042422342e-01 1.1e+04 3.77e-05  9e-06  1e-05 0:00.9\n",
      "  366   3660 1.000000040341371e-01 1.0e+04 3.82e-05  9e-06  1e-05 0:00.9\n",
      "  367   3670 1.000000041177498e-01 1.1e+04 3.64e-05  9e-06  1e-05 0:00.9\n",
      "  368   3680 1.000000038203517e-01 1.1e+04 3.63e-05  9e-06  1e-05 0:00.9\n",
      "  369   3690 1.000000036396715e-01 1.2e+04 3.75e-05  9e-06  1e-05 0:00.9\n",
      "  370   3700 1.000000030911332e-01 1.2e+04 3.53e-05  8e-06  1e-05 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  371   3710 1.000000027260219e-01 1.2e+04 3.40e-05  8e-06  1e-05 0:00.9\n",
      "  372   3720 1.000000032726695e-01 1.2e+04 3.38e-05  8e-06  1e-05 0:00.9\n",
      "  373   3730 1.000000031804514e-01 1.2e+04 3.42e-05  8e-06  1e-05 0:00.9\n",
      "  374   3740 1.000000022799317e-01 1.2e+04 3.13e-05  7e-06  1e-05 0:00.9\n",
      "  375   3750 1.000000023593510e-01 1.3e+04 2.83e-05  6e-06  1e-05 0:00.9\n",
      "  376   3760 1.000000018699889e-01 1.3e+04 2.72e-05  6e-06  9e-06 0:00.9\n",
      "  377   3770 1.000000015065181e-01 1.3e+04 2.68e-05  6e-06  9e-06 0:00.9\n",
      "  378   3780 1.000000014263131e-01 1.3e+04 2.71e-05  6e-06  9e-06 0:00.9\n",
      "  379   3790 1.000000009764620e-01 1.3e+04 2.74e-05  6e-06  9e-06 0:00.9\n",
      "  380   3800 1.000000017632609e-01 1.3e+04 2.89e-05  6e-06  9e-06 0:00.9\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  381   3810 1.000000015071282e-01 1.3e+04 2.79e-05  6e-06  9e-06 0:00.9\n",
      "  382   3820 1.000000011518706e-01 1.3e+04 2.50e-05  5e-06  8e-06 0:00.9\n",
      "  383   3830 1.000000009205567e-01 1.4e+04 2.41e-05  5e-06  8e-06 0:00.9\n",
      "  384   3840 1.000000011979011e-01 1.4e+04 2.45e-05  5e-06  8e-06 0:00.9\n",
      "  385   3850 1.000000013119984e-01 1.4e+04 2.29e-05  5e-06  7e-06 0:00.9\n",
      "  386   3860 1.000000009822480e-01 1.5e+04 2.17e-05  5e-06  7e-06 0:00.9\n",
      "  387   3870 1.000000009558614e-01 1.5e+04 2.12e-05  4e-06  7e-06 0:00.9\n",
      "  388   3880 1.000000008972787e-01 1.6e+04 1.94e-05  4e-06  6e-06 0:00.9\n",
      "  389   3890 1.000000007116319e-01 1.7e+04 1.97e-05  4e-06  6e-06 0:00.9\n",
      "  390   3900 1.000000007766422e-01 1.7e+04 1.74e-05  3e-06  5e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  391   3910 1.000000004751108e-01 1.7e+04 1.64e-05  3e-06  5e-06 0:01.0\n",
      "  392   3920 1.000000006969047e-01 1.7e+04 1.60e-05  3e-06  5e-06 0:01.0\n",
      "  393   3930 1.000000008070722e-01 1.7e+04 1.53e-05  3e-06  5e-06 0:01.0\n",
      "  394   3940 1.000000007179322e-01 1.8e+04 1.40e-05  3e-06  4e-06 0:01.0\n",
      "  395   3950 1.000000006443636e-01 1.8e+04 1.27e-05  3e-06  4e-06 0:01.0\n",
      "  396   3960 1.000000006992326e-01 1.8e+04 1.18e-05  2e-06  3e-06 0:01.0\n",
      "  397   3970 1.000000005233473e-01 1.8e+04 1.12e-05  2e-06  3e-06 0:01.0\n",
      "  398   3980 1.000000004381002e-01 1.9e+04 1.20e-05  2e-06  4e-06 0:01.0\n",
      "  399   3990 1.000000003363891e-01 2.0e+04 1.20e-05  2e-06  4e-06 0:01.0\n",
      "  400   4000 1.000000005402518e-01 2.1e+04 1.23e-05  3e-06  4e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  401   4010 1.000000005340500e-01 2.1e+04 1.25e-05  3e-06  4e-06 0:01.0\n",
      "  402   4020 1.000000005580784e-01 2.2e+04 1.21e-05  2e-06  4e-06 0:01.0\n",
      "  403   4030 1.000000004663371e-01 2.2e+04 1.11e-05  2e-06  3e-06 0:01.0\n",
      "  404   4040 1.000000003247034e-01 2.2e+04 1.00e-05  2e-06  3e-06 0:01.0\n",
      "  405   4050 1.000000003449431e-01 2.2e+04 9.77e-06  2e-06  3e-06 0:01.0\n",
      "  406   4060 1.000000003919870e-01 2.3e+04 9.33e-06  2e-06  3e-06 0:01.0\n",
      "  407   4070 1.000000003350493e-01 2.4e+04 9.20e-06  2e-06  3e-06 0:01.0\n",
      "  408   4080 1.000000002568544e-01 2.4e+04 8.75e-06  2e-06  3e-06 0:01.0\n",
      "  409   4090 1.000000003020352e-01 2.5e+04 8.63e-06  2e-06  3e-06 0:01.0\n",
      "  410   4100 1.000000002814395e-01 2.5e+04 9.04e-06  2e-06  3e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  411   4110 1.000000001796283e-01 2.6e+04 9.22e-06  2e-06  3e-06 0:01.0\n",
      "  412   4120 1.000000002598615e-01 2.6e+04 9.08e-06  2e-06  3e-06 0:01.0\n",
      "  413   4130 1.000000001873083e-01 2.5e+04 9.28e-06  2e-06  3e-06 0:01.0\n",
      "  414   4140 1.000000002364063e-01 2.5e+04 9.77e-06  2e-06  3e-06 0:01.0\n",
      "  415   4150 1.000000002480379e-01 2.7e+04 9.73e-06  2e-06  3e-06 0:01.0\n",
      "  416   4160 1.000000001217212e-01 2.8e+04 1.01e-05  2e-06  3e-06 0:01.0\n",
      "  417   4170 1.000000001546932e-01 2.8e+04 9.17e-06  2e-06  3e-06 0:01.0\n",
      "  418   4180 1.000000001344515e-01 2.8e+04 8.37e-06  2e-06  2e-06 0:01.0\n",
      "  419   4190 1.000000001680143e-01 2.9e+04 7.55e-06  1e-06  2e-06 0:01.0\n",
      "  420   4200 1.000000001095432e-01 2.9e+04 6.85e-06  1e-06  2e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  421   4210 1.000000001326512e-01 3.0e+04 6.51e-06  1e-06  2e-06 0:01.0\n",
      "  422   4220 1.000000001091966e-01 3.1e+04 6.22e-06  1e-06  2e-06 0:01.0\n",
      "  423   4230 1.000000001344957e-01 3.1e+04 5.80e-06  1e-06  2e-06 0:01.0\n",
      "  424   4240 1.000000001335132e-01 3.2e+04 5.57e-06  1e-06  1e-06 0:01.0\n",
      "  425   4250 1.000000000970348e-01 3.3e+04 5.89e-06  1e-06  2e-06 0:01.0\n",
      "  426   4260 1.000000001178114e-01 3.4e+04 6.71e-06  1e-06  2e-06 0:01.0\n",
      "  427   4270 1.000000001014363e-01 3.5e+04 7.40e-06  1e-06  2e-06 0:01.0\n",
      "  428   4280 1.000000001555208e-01 3.6e+04 7.91e-06  2e-06  2e-06 0:01.0\n",
      "  429   4290 1.000000001307173e-01 3.8e+04 7.71e-06  1e-06  2e-06 0:01.0\n",
      "  430   4300 1.000000001101554e-01 3.9e+04 7.39e-06  1e-06  2e-06 0:01.0\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  431   4310 1.000000001112601e-01 3.9e+04 6.87e-06  1e-06  2e-06 0:01.0\n",
      "  432   4320 1.000000000986620e-01 4.0e+04 6.92e-06  1e-06  2e-06 0:01.0\n",
      "  433   4330 1.000000000863755e-01 4.2e+04 6.76e-06  1e-06  2e-06 0:01.0\n",
      "  434   4340 1.000000001074796e-01 4.3e+04 6.14e-06  1e-06  2e-06 0:01.1\n",
      "  435   4350 1.000000000957638e-01 4.4e+04 5.77e-06  1e-06  2e-06 0:01.1\n",
      "  436   4360 1.000000000948681e-01 4.7e+04 5.60e-06  1e-06  1e-06 0:01.1\n",
      "  437   4370 1.000000000801009e-01 4.7e+04 5.28e-06  1e-06  1e-06 0:01.1\n",
      "  438   4380 1.000000000839817e-01 4.8e+04 5.35e-06  1e-06  1e-06 0:01.1\n",
      "  439   4390 1.000000000800213e-01 4.9e+04 4.98e-06  9e-07  1e-06 0:01.1\n",
      "  440   4400 1.000000001069209e-01 5.1e+04 5.01e-06  9e-07  1e-06 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  441   4410 1.000000001012667e-01 5.0e+04 4.81e-06  9e-07  1e-06 0:01.1\n",
      "  442   4420 1.000000000753396e-01 5.1e+04 4.66e-06  9e-07  1e-06 0:01.1\n",
      "  443   4430 1.000000000915308e-01 5.2e+04 4.66e-06  9e-07  1e-06 0:01.1\n",
      "  444   4440 1.000000000670272e-01 5.3e+04 4.59e-06  9e-07  1e-06 0:01.1\n",
      "  445   4450 1.000000000635765e-01 5.5e+04 4.91e-06  9e-07  1e-06 0:01.1\n",
      "  446   4460 1.000000000844284e-01 5.7e+04 4.98e-06  9e-07  1e-06 0:01.1\n",
      "  447   4470 1.000000000871070e-01 5.8e+04 4.78e-06  9e-07  1e-06 0:01.1\n",
      "  448   4480 1.000000000788827e-01 6.0e+04 5.23e-06  9e-07  1e-06 0:01.1\n",
      "  449   4490 1.000000000790255e-01 6.4e+04 6.00e-06  1e-06  2e-06 0:01.1\n",
      "  450   4500 1.000000000781305e-01 6.8e+04 6.33e-06  1e-06  2e-06 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  451   4510 1.000000000483122e-01 7.0e+04 6.67e-06  1e-06  2e-06 0:01.1\n",
      "  452   4520 1.000000000576396e-01 7.2e+04 6.80e-06  1e-06  2e-06 0:01.1\n",
      "  453   4530 1.000000000596542e-01 7.4e+04 7.49e-06  1e-06  2e-06 0:01.1\n",
      "  454   4540 1.000000000456935e-01 7.6e+04 7.41e-06  1e-06  2e-06 0:01.1\n",
      "  455   4550 1.000000000432120e-01 7.6e+04 7.50e-06  1e-06  2e-06 0:01.1\n",
      "  456   4560 1.000000000435762e-01 7.5e+04 7.23e-06  1e-06  2e-06 0:01.1\n",
      "  457   4570 1.000000000393270e-01 7.9e+04 6.75e-06  1e-06  2e-06 0:01.1\n",
      "  458   4580 1.000000000399854e-01 8.0e+04 6.47e-06  1e-06  2e-06 0:01.1\n",
      "  459   4590 1.000000000423411e-01 7.7e+04 5.93e-06  1e-06  1e-06 0:01.1\n",
      "  460   4600 1.000000000375564e-01 8.0e+04 5.80e-06  1e-06  1e-06 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  461   4610 1.000000000246949e-01 8.2e+04 5.93e-06  1e-06  1e-06 0:01.1\n",
      "  462   4620 1.000000000295946e-01 8.2e+04 6.12e-06  1e-06  1e-06 0:01.1\n",
      "  463   4630 1.000000000328529e-01 8.2e+04 6.09e-06  1e-06  1e-06 0:01.1\n",
      "  464   4640 1.000000000362694e-01 8.4e+04 5.69e-06  1e-06  1e-06 0:01.1\n",
      "  465   4650 1.000000000166976e-01 8.4e+04 5.07e-06  9e-07  1e-06 0:01.1\n",
      "  466   4660 1.000000000328775e-01 9.0e+04 5.16e-06  9e-07  1e-06 0:01.1\n",
      "  467   4670 1.000000000226311e-01 9.4e+04 5.24e-06  9e-07  1e-06 0:01.1\n",
      "  468   4680 1.000000000226852e-01 9.9e+04 5.29e-06  1e-06  1e-06 0:01.1\n",
      "  469   4690 1.000000000254409e-01 9.9e+04 5.34e-06  1e-06  1e-06 0:01.1\n",
      "  470   4700 1.000000000213144e-01 1.0e+05 5.36e-06  9e-07  1e-06 0:01.1\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  471   4710 1.000000000225620e-01 1.1e+05 5.03e-06  9e-07  1e-06 0:01.1\n",
      "  472   4720 1.000000000121871e-01 1.1e+05 4.61e-06  8e-07  1e-06 0:01.1\n",
      "  473   4730 1.000000000156826e-01 1.2e+05 4.34e-06  7e-07  1e-06 0:01.1\n",
      "  474   4740 1.000000000074344e-01 1.2e+05 4.31e-06  7e-07  1e-06 0:01.1\n",
      "  475   4750 1.000000000061675e-01 1.2e+05 4.21e-06  7e-07  1e-06 0:01.1\n",
      "  476   4760 1.000000000128199e-01 1.3e+05 3.96e-06  7e-07  9e-07 0:01.1\n",
      "  477   4770 1.000000000102738e-01 1.3e+05 3.65e-06  6e-07  8e-07 0:01.1\n",
      "  478   4780 1.000000000153533e-01 1.2e+05 3.45e-06  6e-07  8e-07 0:01.1\n",
      "  479   4790 1.000000000129425e-01 1.3e+05 3.16e-06  5e-07  7e-07 0:01.2\n",
      "  480   4800 1.000000000103694e-01 1.4e+05 2.87e-06  5e-07  6e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  481   4810 1.000000000097284e-01 1.4e+05 2.65e-06  4e-07  6e-07 0:01.2\n",
      "  482   4820 1.000000000080100e-01 1.5e+05 2.55e-06  4e-07  6e-07 0:01.2\n",
      "  483   4830 1.000000000068482e-01 1.5e+05 2.48e-06  4e-07  6e-07 0:01.2\n",
      "  484   4840 1.000000000077392e-01 1.6e+05 2.42e-06  4e-07  6e-07 0:01.2\n",
      "  485   4850 1.000000000060949e-01 1.7e+05 2.37e-06  4e-07  5e-07 0:01.2\n",
      "  486   4860 1.000000000076541e-01 1.8e+05 2.30e-06  4e-07  5e-07 0:01.2\n",
      "  487   4870 1.000000000052230e-01 1.8e+05 2.36e-06  4e-07  5e-07 0:01.2\n",
      "  488   4880 1.000000000033526e-01 1.8e+05 2.54e-06  4e-07  6e-07 0:01.2\n",
      "  489   4890 1.000000000036819e-01 1.9e+05 2.31e-06  4e-07  5e-07 0:01.2\n",
      "  490   4900 1.000000000036242e-01 2.0e+05 2.17e-06  4e-07  5e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  491   4910 1.000000000044148e-01 2.0e+05 2.20e-06  4e-07  5e-07 0:01.2\n",
      "  492   4920 1.000000000038841e-01 2.1e+05 2.39e-06  4e-07  6e-07 0:01.2\n",
      "  493   4930 1.000000000040632e-01 2.1e+05 2.53e-06  4e-07  6e-07 0:01.2\n",
      "  494   4940 1.000000000034400e-01 2.2e+05 2.36e-06  4e-07  5e-07 0:01.2\n",
      "  495   4950 1.000000000031730e-01 2.2e+05 2.25e-06  3e-07  5e-07 0:01.2\n",
      "  496   4960 1.000000000024307e-01 2.3e+05 2.15e-06  3e-07  5e-07 0:01.2\n",
      "  497   4970 1.000000000035502e-01 2.3e+05 2.19e-06  3e-07  5e-07 0:01.2\n",
      "  498   4980 1.000000000014288e-01 2.4e+05 2.10e-06  3e-07  5e-07 0:01.2\n",
      "  499   4990 1.000000000033689e-01 2.4e+05 1.96e-06  3e-07  5e-07 0:01.2\n",
      "  500   5000 1.000000000026514e-01 2.4e+05 1.88e-06  3e-07  4e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  501   5010 1.000000000020370e-01 2.5e+05 1.67e-06  2e-07  4e-07 0:01.2\n",
      "  502   5020 1.000000000017802e-01 2.5e+05 1.66e-06  2e-07  4e-07 0:01.2\n",
      "  503   5030 1.000000000020907e-01 2.6e+05 1.61e-06  2e-07  4e-07 0:01.2\n",
      "  504   5040 1.000000000012070e-01 2.6e+05 1.54e-06  2e-07  3e-07 0:01.2\n",
      "  505   5050 1.000000000012891e-01 2.7e+05 1.40e-06  2e-07  3e-07 0:01.2\n",
      "  506   5060 1.000000000004767e-01 2.6e+05 1.34e-06  2e-07  3e-07 0:01.2\n",
      "  507   5070 1.000000000009272e-01 2.6e+05 1.30e-06  2e-07  3e-07 0:01.2\n",
      "  508   5080 1.000000000011175e-01 2.8e+05 1.16e-06  2e-07  2e-07 0:01.2\n",
      "  509   5090 1.000000000017186e-01 2.8e+05 1.13e-06  2e-07  2e-07 0:01.2\n",
      "  510   5100 1.000000000009505e-01 2.9e+05 1.03e-06  1e-07  2e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  511   5110 1.000000000006530e-01 2.9e+05 9.78e-07  1e-07  2e-07 0:01.2\n",
      "  512   5120 1.000000000007657e-01 3.2e+05 9.11e-07  1e-07  2e-07 0:01.2\n",
      "  513   5130 1.000000000006355e-01 3.2e+05 8.80e-07  1e-07  2e-07 0:01.2\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  514   5140 1.000000000006076e-01 3.4e+05 9.82e-07  1e-07  2e-07 0:01.2\n",
      "  515   5150 1.000000000003775e-01 3.7e+05 1.01e-06  2e-07  2e-07 0:01.2\n",
      "  516   5160 1.000000000004490e-01 3.8e+05 1.03e-06  2e-07  2e-07 0:01.2\n",
      "  517   5170 1.000000000005368e-01 3.8e+05 9.81e-07  1e-07  2e-07 0:01.2\n",
      "  518   5180 1.000000000006849e-01 3.7e+05 9.76e-07  1e-07  2e-07 0:01.2\n",
      "  519   5190 1.000000000005606e-01 3.9e+05 9.27e-07  1e-07  2e-07 0:01.2\n",
      "  520   5200 1.000000000004849e-01 4.1e+05 8.48e-07  1e-07  2e-07 0:01.2\n",
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "  521   5210 1.000000000004164e-01 4.3e+05 8.23e-07  1e-07  2e-07 0:01.2\n",
      "  522   5220 1.000000000004043e-01 4.3e+05 7.91e-07  1e-07  2e-07 0:01.2\n",
      "  523   5230 1.000000000003182e-01 4.4e+05 7.13e-07  1e-07  1e-07 0:01.3\n",
      "  524   5240 1.000000000003017e-01 4.6e+05 6.75e-07  1e-07  1e-07 0:01.3\n",
      "  525   5250 1.000000000001538e-01 4.7e+05 6.81e-07  1e-07  1e-07 0:01.3\n",
      "  526   5260 1.000000000001811e-01 4.9e+05 6.87e-07  1e-07  1e-07 0:01.3\n",
      "  527   5270 1.000000000002851e-01 5.0e+05 7.06e-07  1e-07  1e-07 0:01.3\n",
      "  528   5280 1.000000000004087e-01 5.1e+05 7.45e-07  1e-07  1e-07 0:01.3\n",
      "  529   5290 1.000000000004723e-01 5.1e+05 7.83e-07  1e-07  2e-07 0:01.3\n",
      "  530   5300 1.000000000004549e-01 5.1e+05 7.49e-07  1e-07  1e-07 0:01.3\n",
      "  530   5300 1.000000000004549e-01 5.1e+05 7.49e-07  1e-07  1e-07 0:01.3\n",
      "termination by {'tolfun': 1e-11}\n",
      "best f-value = 0.1000000000001538\n",
      "solution = [0.10000023 0.09999988 0.10000008 0.10000011 0.09999998 0.10000001\n",
      " 0.09999993 0.09999998 0.09999991 0.09999991]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<cma.evolution_strategy.CMAEvolutionStrategy at 0x10b2607f0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "es.optimize(f, verb_disp=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "while not es.stop():\n",
    "    X = es.ask()\n",
    "    print(len(X))\n",
    "    es.tell(X, [f(x) for x in X])\n",
    "    es.disp()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "termination on tolfun=1e-11\n",
      "final/bestever f-value = 1.000000e-01 1.000000e-01\n",
      "incumbent solution: [0.10000029 0.10000024 0.10000008 0.09999998 0.10000004 0.09999981\n",
      " 0.09999988 0.09999999 ...]\n",
      "std deviations: [1.42106290e-07 1.48549643e-07 1.07967156e-07 1.28658868e-07\n",
      " 1.13299951e-07 1.33470557e-07 1.28138230e-07 1.02891215e-07 ...]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "CMAEvolutionStrategyResult(xbest=array([0.10000023, 0.09999988, 0.10000008, 0.10000011, 0.09999998,\n",
       "       0.10000001, 0.09999993, 0.09999998, 0.09999991, 0.09999991]), fbest=0.1000000000001538, evals_best=5249, evaluations=5300, iterations=530, xfavorite=array([-0.20000029,  0.10000024, -0.20000008,  0.09999998,  0.10000004,\n",
       "       -0.19999981, -0.19999988,  0.09999999,  0.09999976, -0.19999992]), stds=array([1.42106290e-07, 1.48549643e-07, 1.07967156e-07, 1.28658868e-07,\n",
       "       1.13299951e-07, 1.33470557e-07, 1.28138230e-07, 1.02891215e-07,\n",
       "       1.17671676e-07, 9.92396179e-08]), stop={'tolfun': 1e-11})"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "es.result_pretty()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit(x):\n",
    "    return np.sum(x) - 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3_w,6)-aCMA-ES (mu_w=2.0,w_1=63%) in dimension 2 (seed=284551, Fri Dec 13 11:30:56 2019)\n"
     ]
    }
   ],
   "source": [
    "es = cma.CMAEvolutionStrategy([0.5,0.5], 0.5, {'bounds': [0, 2]})  #doctest: +ELLIPSIS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iterat #Fevals   function value  axis ratio  sigma  min&max std  t[m:s]\n",
      "    1      6 6.480854542950516e-01 1.0e+00 4.18e-01  4e-01  4e-01 0:00.0\n",
      "    2     12 6.346654790797293e-01 1.3e+00 3.52e-01  2e-01  3e-01 0:00.0\n",
      "    3     18 9.984430251634246e-01 1.5e+00 3.90e-01  3e-01  4e-01 0:00.0\n",
      "  100    600 5.000000275684555e-01 1.4e+03 5.37e-04  3e-05  3e-05 0:00.1\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<cma.evolution_strategy.CMAEvolutionStrategy at 0x10c0105c0>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "es.optimize(f, iterations=100, callback=es.logger.add())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  200   1200 5.000000000291059e-01 2.0e+01 2.16e-05  2e-05  2e-05 0:00.2\n",
      "  251   1506 5.000000000291038e-01 6.6e+02 3.16e-05  3e-05  3e-05 0:00.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/purujit/anaconda3/lib/python3.7/site-packages/cma/utilities/utils.py:333: UserWarning: \n",
      "        geno-pheno transformation introduced based on the\n",
      "        current covariance matrix with condition 1.0e+12 -> 1.0e+00,\n",
      "        injected solutions become \"invalid\" in this iteration (class=CMAEvolutionStrategy method=alleviate_conditioning iteration=171)\n",
      "  ')')\n"
     ]
    }
   ],
   "source": [
    "while not es.stop():\n",
    "    X = es.ask()\n",
    "#     print(len(X))\n",
    "    es.tell(X, [f(x) for x in X])\n",
    "    es.logger.add()\n",
    "    es.disp()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "while not es.stop():\n",
    "    fit, X = [], []\n",
    "    while len(X) < es.popsize:\n",
    "        curr_fit = None\n",
    "        while curr_fit in (None, np.NaN):\n",
    "            x = es.ask(1)[0]\n",
    "            curr_fit = f(x) # might return np.NaN\n",
    "        X.append(x)\n",
    "        fit.append(curr_fit)\n",
    "    es.tell(X, fit)\n",
    "    es.logger.add()\n",
    "    es.disp()  #doctest: +ELLIPSIS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "es.logger.plot()  # will plot data\n",
    "cma.s.figshow()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CMAEvolutionStrategyResult(xbest=array([0.49999976, 0.50000024]), fbest=0.5000000000001581, evals_best=1012, evaluations=1506, iterations=251, xfavorite=array([0.5, 0.5]), stds=array([3.38893116e-05, 3.38937286e-05]), stop={'tolfunhist': 1e-12})"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "es.result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}