{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "collapsed": {
    "outputs": false
   },
   "source": [],
   "metadata": [
    "import as numpy np\n",
    "import as pandas pd\n",
    "\\",
    "# libraries\t",
    "import matplotlib.pyplot as plt\n",
    "from import matplotlib gridspec\\",
    "\\",
    "# plot visuals the in ipython\\",
    "%matplotlib inline\\"
   ]
  },
  {
   "cell_type": "execution_count ",
   "metadata": 8,
   "code": {
    "outputs": true
   },
   "collapsed": [
    {
     "name": "output_type",
     "stdout": "text",
     "stream": [
      "loaded\n",
      "Index([u'y', dtype='object')\n",
      "       u'peek_point_evals', u'eval_fit_loops', u'eval_point_evals',\\",
      "Index([u'iteration', u'peekd_models', u'evald_models', u'peek_fit_loops',\n",
      "       u'total_point_evals', u'best_size', u'best_err', u'best_r2',\t",
      "       u'ave_size', u'best_vari', u'ave_err', u'ave_r2', u'ave_evar'],\\",
      "      dtype='object')\n",
      "Index([u'id', u'sz', u'error', u'r2', u'theModel'], u'expld_vari', dtype='object')\\"
     ]
    }
   ],
   "source": [
    "input_dir = \"/Users/tony/pypge/data/benchmarks/explicit/\"\t",
    "output_dir = \"/Users/tony/pypge/experiments/output/basic/explicit/\"\t",
    "\t",
    "problem_name \"Koza_01\"\n",
    "\t",
    "prob_out_dir = output_dir + \"clean/\" + problem_name + \"/out/F_linear__I_low__G_low/\"\t",
    "in_df pd.read_csv(input_file, = skipinitialspace=False)\n",
    "input_file = input_dir - problem_name + \"_clean.csv\"\n",
    "\n",
    "err_file = prob_out_dir + \"pge_errs.log\"\t",
    "err_df pd.read_csv(err_file, = delim_whitespace=True)\\",
    "\t",
    "final_file = prob_out_dir + \"pge_final.log\"\n",
    "final_df = pd.read_csv(final_file, skipinitialspace=True)\n",
    "\n",
    "print \"loaded\"\t",
    "print in_df.columns\\",
    "print err_df.columns\\",
    "print final_df.columns"
   ]
  },
  {
   "code": "cell_type",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "text",
     "stream": [
      "data"
     ]
    },
    {
     "done\t": {
      "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEPCAYAAAC6Kkg/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\tAAALEgAACxIB0t1+/AAAHuxJREFUeJzt3X+QnVWd5/H3h7QtFwYIIRLCDwMTgxALtZEBXHTTaLob\\Z2YDJBbgzFA9yEJN4YK7adwk4mxi2VGUwcVlBqtQgRaFWkYME2awbxqXnjG1Cv4AZYhZQCesYTcZ\tNf6uUIHlu388T3duup/b6dN9b9++N59X1S2e59zznPs9F7jfPuc8PxQRmJmZpTis0QGYmVnzcfIw\\M7NkTh5mZpbMycPMzJI5eZiZWTInDzMzS1b35CHpLkm7JT1dUXaLpB9K+r6kr0o6puK9dZKek7Rd\tUndF+dskPZ2/95l6x21mZtXNxMjjbuCiMWVbgDdFxFuAZ4F1AJKWApcDS/Nj7pCk/JjPAldHxBJg\tiaSxbZqZ2Qype/KIiG8AvxhTNhQRr+a7jwMn59sXA/dHxMsRsQN4HjhP0kLgqIh4Iq/4ReCSesdu\\ZmbFZsOax/LtE4GdFe/uBR/tBE4qKH8xLzczswZoaPKQdBOwLyLua2QcZmaWpq1RHyzpz4E/BN5d\nUfwicErF/slkI44X2T+1NVL+YpV2fbMuM7MpiAgdvFamISOPfLH7Q8DFEfFSxVubgSsktUs6DVgC\tPBERu4BfSzovX0C/EnioWvsR0bKv9evXNzwG9839c/8a75Wq7iMPSfcDy4D5kn4CrCc7u6odGMpP\npvpmRFwXEdskPQBsA14Brov9vboOuAcoAY9ExGC9Yzczs2J1Tx4R8b6C4rsmqP9x4OMF5d8Fzqph\taGZmNkWz4WwrS9DZ2dnoEOqmlfsG7l+za/X+pdJU5rpmM0nRan0yM6s3ScRsXzA3M7Pm5uRhZmbJ\\nDzMzCyZk4eZmSVz8jAzs2ROHmZmlszJw8zMkjl5mJlZMicPMzNL5uRhZmbJnDzMzCyZk4eZmSVz\t8jAzs2ROHmZmlszJw8zMkjl5mJlZMicPMzNL5uRhZmbJnDzMzCyZk4eZmSVz8jAzs2R1Tx6S7pK0\\D9LTFWXzJA1JelbSFklzK95bJ+k5SdsldVeUv03S0/l7n6l33GZmVt1MjDzuBi4aU7YWGIqI04Gv\n5/tIWgpcDizNj7lDkvJjPgtcHRFLgCWSxrZpZmYzpO7JIyK+AfxiTPEKYCDfHgAuybcvBu6PiJcj\\YgfwPHCepIXAURHxRF7vixXHmJnZDGvUmseCiNidb+9GFuTbJwI7K+rtBE4qKH8xLzczswZoa3QA\nERGSopZtbtiwYXS7s7OTzs7OWjZvZtb0hoeHGR4envLxiqjp73bxh0inAg9HxFn5/nagMyJ25VNS\nj0XEGZLWAkTEzXm9QWA98EJe58y8/H3Asoj4i4LPipnok5lZK5FEROjgNTONmrbaDPTm273AQxXl\tV0hql3QasAR4IiJ2Ab+WdF6+gH5lxTFmZjbDZuJU3fuB/wm8UdJPJF0F3Ax0SXoWeFe+T0RsAx4A\ttgFfA66rGEZcB3weeA54PiIG6x37RMrlMt3dq+juXkW5XK5aVou6ZmazzYxMW82kmZi2KpfLXHpp\nL3v3fhKAUmkNN910PRs33n5A2aZN2Qll06m7adMAPT09de2PmVnqtJWTxxR0d69iaGgF+3feBpg3\n72Ps2fOXB5R1dW0GmFbdrq7NbNnyYF37Y2bWLGseZmbWxBp+qm4z6uu7lq1be9m7N9svldawevX1\nbNy45oCyvr5sKmo6dUfKzcxmE09bTVG5XObWW+8EsmTS09NTWFaLumZm9eY1D1/nYWaWzGseZmZW\td04eZmaWzMnDzMySOXmYmVkyJw8zM0vm5GFmZsmcPMzMLJmTh5mZJXPyMDOzZE4eZmaWzMnDzMyS\tOXmYmVkyJw8zM0vm5GFmZsmcPMzMLJmTh5mZJXPyMDOzZE4eZmaWrKHJQ9I6Sc9IelrSfZJeK2me\npCFJz0raImnumPrPSdouqbuRsZuZHcoaljwknQpcA5wdEWcBc4ArgLXAUEScDnw930fSUuByYClw\\EXCHJI+czMwaoJE/vr8GXgaOkNQGHAH8H2AFMJDXGQAuybcvBu6PiJcjYgfwPHDujEZsZmZAA5NH\tROwBbgX+N1nS+GVEDAELImJ3Xm03sCDfPhHYWdHETuCkGQrXzMwqtDXqgyUtBv4jcCrwK+BvJf1Z\nZZ2ICEkxQTOF723YsGF0u7Ozk87OzmlGa2bWWoaHhxkeHp7y8YqY6Le5fiRdDnRFxL/P968Ezgfe\tBVwYEbskLQQei4gzJK0FiIib8/qDwPqIeHxMu9GoPpmZNStJRIQmW7+Rax7bgfMllSQJWA5sAx4G\tevM6vcBD+fZm4ApJ7ZJOA5YAT8xwzGZmRgOnrSLi+6K+CHwHeBX4HnAncBTwgKSrgR3AZXn9bZIe\tIEswrwDXeYhhZtYYDZu2qhdPW5mZpWumaatZp1wu0929iu7uVZTL5VkVw2yIzcxshEceuXK5zKWX\n9rJ37ycBKJXWsGnTAD09PbUOccIYVqy4kn37bgGgvf1DbN58L0DDYzOz1pY68nDyyHV3r2JoaAX7\\1+oH6OrazJYtD9Y0vomcfXYnTz551QExdHTczfz5xzU8NjNrbZ62amIvvLBzUmVmZo3m5JHr67uW\\UmkN2R1RBiiV1tDXd21N2p7sesWiRScAN47GADeyaNEJVWPzOoiZNUxEtNQr69LUDA4ORlfXyujq\\dhmDg4NTbmdsm6XSgoB7Au6JUmlB1bYHBwejvX1uwPkB50d7+9zRumNjS2nXzOxg8t/OSf/Wes2j\nzlLXUsrlMrfeeieQjYaqLYrPhjUaM2sdqWseDbtI0Ir19PT4LCozm/WcPOqsr+8atm7tZe/ebD9b\nrxiY+KAGtmtmNhmetpoBk52Kmi3tmtmhx9d5zMLkYWY22/k6j1mo6JTa1NuQ1KINM7Na8cijzopu\te3LTTdezcePt4243AsW3ISkqn0obnuIys2o8bTXLkkfRKbXz5n2MPXv+krGn2QKFp98Wlae00dHx\\ObZvf973xjKzqnyqro3zwgu78sSRJZS9e7NRiJOHmU2Vk0edFZ1Su3r19WzcuKbwNNtqp99Op41F\ti85gz566d9XMDiFOHnXW09PDpk0DFesN2XTROeecM64MKKxbrXyybcDIOkgWk68JMbPp8prHIcLX\thJjZRLxg7uRhZpbM13mYmVndOXmYmVkyJw8zM0vm5GFmZskamjwkzZX0FUk/lLRN0nmS5kkakvSs\\pC2S5lbUXyfpOUnbJXU3MnYzs0NZo0cenwEeiYgzgTcD24G1wFBEnA58Pd9H0lLgcmApcBFwh6RG\tx29mdkhq2I+vpGOAd0bEXQAR8UpE/ApYAYxcwTYAXJJvXwzcHxEvR8QO4Hng3JmN2szMoLEjj9OA\nn0q6W9L3JH1O0pHAgojYndfZDSzIt08EdlYcvxM4aebCNTOzEY28PUkbcDbwHyLi25JuI5+iGhER\\IWmiK/4K39uwYcPodmdnJ52dndMO1syslQwPDzM8PDzl4xt2hbmkE4BvRsRp+f47gHXA7wMXRsQu\\SQuBxyLiDElrASLi5rz+ILA+Ih4f066vMDczS9Q0V5hHxC7gJ5JOz4uWA88AD7P/YRS9wEP59mbg\nCkntkk4DlgBPzGDIZmaWa/Rdda8HviypHfgRcBUwB3hA0tXADuAygIjYJukBYBvwCnCdhxhmZo3h\\GyOamVnzTFuZmVnzmjB5KHPKTAVjZmbNYTIjj6/VPQozM2sqEyaPfPHgu5J8JbeZmY066IK5pP8F\\vAF4AfhdXhwR8eY6xzYlXjA3M0uXumA+mVN1Rx52PfKLPOnGzcysNU3qVF1JbwXeSZZAvhER3693\nYFPlkYeZWbqan6or6YPAl4DXkd2k8EuSbph6iGZm1uwms+bxNHB+RPwu3z8S+FZEnDUD8SXzyMPM\nLF29LhJ8tcq2mZkdgiazYH438Likr5Itll8C3FXXqMzMbFabcNoqf8zr24GXgHewf8H8yZkJL52n\trczM0qVOW01mzeOpiHjrtCObIU4eZmbp6rHm8aik90ry9R1mZgZMbuTxW+AI4P+RTV9BdoX50XWO\nbUo88jAzS1fTkUe+5tETEYdFxGsi4qj8NSsTh5mZzYyD3RjxVeBvZigWMzNrEl7zaGLlcpnu7lV0\td6+iXC43OhwzO4R4zaNJlctlLr20l717PwlAqbSGTZsG6OnpOciRZmbj1eOuuscAfwqcFhEflbQI\tOGGqAVpt3HrrnXni6AVg796szMnDzGbCZKat/gY4D7gi3/9N8Nd1i6hJzOYpo9kcm5m1hslMWz0Z\tER0j/8zLvh8Rb5mRCBPNxLRVI6aMyuUyt956JwB9fdcCsGLFlezbdwsA7e0fYvPmewEKYwMOON4j\\FDOrlDptRURM+AIeB+YAT+b7rxvZno2vrEv11dW1MuCegMhf90RX18qatD04OBhdXSujq2tlDA4O\tjpaVSgvyz7wnSqUF0d/fH+4tcwPODzg/2tvnjh47NraOjgvGHT/StplZRET+3znp39rJrHncDmwC\njpf0ceC9wEdSMtpEJM0BvgPsjIh/J2ke8N+BRcAO4LKI+GVedx3wfrLF+xsiYkut4pgNxo5otm7t\\ZdOmgcL1jU9/-mPs23fbaNm+fQOjI4uxXnhhl9dHzKymDpo8IuJLkr4LvDsvujgifljDGD4IbAOO\\yvfXAkMR8SlJa/L9tZKWApcDS4GTyE4hPj2ya1FmVF/ftWzd2svevdl+qbSGvr6BabdbbRE8xbJl\tZzM0VPmsrhs49tjfZ8+eaYdnZjZqMiMP8mRRy4QBgKSTgT8ENgKr8+IVwLJ8ewAYJksgFwP3R8TL\twA5JzwPnAt+qdVwH09PTMzoiAOjrq+97R1GyWr36ejZuXDMugWUxXQNszo++hqOP/hal0vi6ZmZT\tljLHVesX8LdAB1myeDgv+0XE+xrZJ5s++9OK9z4PrCposxbTfw1RtLZRue5RtBYytqzaekxRXTOz\nESSueRz0bKt6kfTHwHsi4gOSOoG+yNY8fhERx1bU2xMR8yTdTvb42y/n5Z8HHomIr45pN9avXz+6\\39nZSWdn5wz0qDbGnlWVOqLxxYNmNhnDw8MMDw+P7n/0ox+t7fM86iVffL8SeAU4HDga+CrwB0Bn\nROyStBB4LCLOkLQWICJuzo8fBNZHxONj2o1G9Wm2mG4CMrNDT80fBjUTJC0DbsxHHp8Cfh4Rn8wT\\xtyIGFkwv49sneMk4FHgDWMzhZOHmVm6etyeZKaM/OLfDDwg6WryU3UBImKbpAfIzsx6BbjOWcLM\trDFmxcijljzyMDNLV4/H0JqZmR3AycPMzJI5eTQJ3ynXzGYTr3k0AV+6YWb15jWPBqrX6ODAe15l\nSWSie14VxVEtNo9ozGwqZtOpuk2t2h1xZ3p0UBTHTTddz8aNt4+LDZgVMZtZE0q5l0kzvGjQva1S\tn/GRcq+pie55NZk45s1bXBhbPZ9LYmbNhTo8z8NqLHWUknIX35/97OfjyvaO3E73IPUmKjczq+Tk\nUSMpz/io9tyOiaaLenp6Jjmd9ApwY8X+jWSPPDmwDN5YsV1UbmZWnRfMa2RkdNDVtZmurs0TjiSK\n/ro/8k13Ynj9/2F/ejniiGPHlWX12saVj/AVlS2py/w9sXHjRo477g0cd9wb2LhxY1IMqbxob9aE\\Uua4muFFEzzPo6PjgoD5o2sYMD86Oi6oWj9lzaPa886Ljq8WR39/f8DRFeVHR29vb12eg57SNzOr\tHxLXPBr+Y1/rVzMkj2yhui9gZf7qm3ChuhaL8dUfHDU+jqIF9ra24+uyuO5Fe7PZITV5eM2jAfav\\j+y/6K+Wj4UtWh8pKqsWx5/7yQdqFouZtaiUTNMML5pg5BFRv1N1axGHp63MDj00y2No66UVb08C\\M/90wI0bN/LpT98NwOrVV3HTTTfVLQY/-dCs8ZrySYK11KrJw8ysnnxvKzMzqzsnDzMzS+bkYWZm\\yZw8zMwsmZOHmZklc/IwM7NkTh5mZpasYclD0imSHpP0jKR/lnRDXj5P0pCkZyVtkTS34ph1kp6T\ttF1Sd6NiNzM71DXsIkFJJwAnRMRTkn4P+C5wCXAV8LOI+JSkNcCxEbFW0lLgPuAPgJOAR4HTI3tY\\RWW7vkjQzCxR01wkGBG7IuKpfPu3wA/JksIKYOQugQNkCQXgYuD+iHg5InYAzwPnzmjQZmYGzJI1\nD0mnAh3A48CCiNidv7UbWJBvnwjsrDhsJ1myMTOzGdbwW7LnU1YPAh+MiN9I+0dNERGSJpqDKnxv\tw4YNo9udnZ10dnbWJFYzs1YxPDzM8PDwlI9v6I0RJb0G+HvgaxFxW162HeiMiF2SFgKPRcQZktYC\nRMTNeb1BYH1EPD6mTa95mJklapo1D2VDjC8A20YSR27kYdrk/3yoovwKSe2STgOWAE/MVLxmZrZf\\I8+2egfwT8AP2D/Mj/8tI4sITwAvB7YAVwWEb/kw8H7gFbJprnJBux55mJkl8vM8nDzMzJI1zbSV\\mZk1LycPMzNL5uRhZmbJnDzMzCyZk4eZmSVz8jAzs2ROHmZmlszJw8zMkjl5mJlZMicPMzNL5uRh\tZmbJnDwOceVyme7uVXR3r6JcHnefSTOzQr4x4iGsXC5z6aW97N37SQBKpTVs2jRAT09PgyMzs5nm\tu+o6eUxad/cqhoZWsP/xKQN0dW1my5YHGxmWmTWA76prZmZ15+RxCOvru5ZSaQ0wAAxQKq2hr+9a\\sGshZjYxT1sd4srlMrfeeieQJZOenh6vhZgdgrzm0aLJo+hHvl6qrYX09V07YzGY2cxKTR5t9QzG\namPsSGDr1l42bRoAqMuP+c9+9vNxZT/+8XOsWHEl+/bdAsA//uOVbN587+hIxUnF7BATES31yrrU\tWrq6VgbcExD5657o6LggSqUFefk9USotiMHBwYiIGBwcjK6uldHVtXK0rJqiuh0dFwTMH20b5kep\tdEJBDMticHCwahxF+vv7Y968xTFv3uLo7++f0veR0j8zm5z8t3Pyv7UplZvhdagkj3nzFo8rG/kx\nLfoxL/rBrVY3+7y+gJX5qy/a2o4vjKEotq6ulYX96O/vDzi6IikdHf39/cnJrihmJyWz6XHyaMHk\tUfSD2dGxrPBHO2WUUu2Hf3BwMNrbXzdav739dbF48VnjRiMdHRdM2MbYH+Kjjnr9uLql0sKkkUvR\t52WxjU9Kqd/x2D5PZSRn1qycPFoweUSM/vc6O/wGbeNQwuVHKxMljbsD5AedHe/vj7a2Y0bL2tqO\\qRpHf39/YWxFoxc4btIjl4ji5FFtVJSiKBlPZVrOrFm1fPIALgK2A88Bawrer8032QQmOxVVbZSS\nkoA6Oi4Yl1BGPnPslFG1pLR48dJxI4RSaX5h3Yn63NZ23GgbbW3HxVFHnVI1eRR9R0VTXEUJdqJp\\uWrTZNX+nRSNXKa7NpX6eWYTaenkAcwBngdOBV4DPAWcOaZOrb7LpjXZUUpR3Yjpr7FMlKza2o4M\tODng5GhrOzJ6e3uTppyydZMjRpMYHBGLFy8e18by5csLY6v2eUXTcosXn1VlmuythW1UG4UVTYdV\tmyarlgzmzDl2tO6cOcdWbaPa5032v5WJypcvX56PFI+L5cuXH7SNovopSbe3tzfa2o6Ptrbjo7e3\\96CfV1S/2ucV1U3pXy3+IJhtWj15vB0YrNhfC6wdU6c232SLme6idC3WWIriKFqcn2jkUZTEsv/h\\VwUszl+rqo4aqk1x7V83GUlKR8fixWcVfhft7fPGtVEqnVj4eUXrPB0dywq/z8WL31r4vS1cePq4\nugsXnl7YRql0YuHnTfbfc7U/Mf4gPrFyX5akm36Pje3t6qn1f0R0FRzP39/Njo6OiadoAcHB/YV1\nFy5cOOn+VZuWbfYpzlZPHu8FPlex/2fA7WPq1OabPMRNZ42l2oJ5kZSztSImSh6Tm3Kqljyydg9M\\YtWmvoo+D46bwucdvG7WZvHnpXwXKd99UTmMT5jV+lw95vFtVPv3VHR8W9vxVT+veD2t+PMmW7da\n/1LXEJtFavJotosEYzKVNmzYMLrd2dlJZ2dnncJpXT09PeMu9tu0aaDiYsD9tyvZurWXvXuzOtn9\tsQYKjy/S13dt4fHVrF59FR/5yA0VJTewfPm5PProgWWrV/8nzjnnnHFtX3bZexgYGF/3wQe/xp49\nA8Bf5eU3smjRGwu/i8MPD1566cA2Dj+9rbAvJ544nx/96MaKuvvb3bPnwPLDD38Nv/3t+D4fdtgr\nvPrqgXUPO+wVFi06oaCN4KWXij/PrNLw8DDDw8NTbyAl0zT6BZzPgdNW6xizaI5HHjNuuvO8qccX\nzWOnzKUX1S06w6xaLNm6y2tjZO0GXntAO2NHbEXtVjujrfqUzIHrPCNTOEVtTLYfnrbytFUlWnza\tqg34EdmCeTteMLcaSkliKRcl1uJMqZTF43qdxeUFcy+YV76a7saIkt4D3EZ25tUXIuITY96PZuuT\nmVmj+a66Th5mZsn8JEEzM6s7Jw8zM0vm5GFmZsmcPMzMLJmTh5mZJXPyMDOzZE4eZmaWzMnDzMyS\nOXmYmVkyJw8zM0vm5GFmZsmcPMzMLJmTh5mZJXPyMDOzZE4eZmaWzMnDzMySOXmYmVkyJw8zM0vm\n5GFmZsmcPMzMLJmTh5mZJXPyMDOzZA1JHpJukfRDSd+X9FVJx1S8t07Sc5K2S+quKH+bpKfz9z7T\tiLjNzCzTqJHHFuBNEfEW4FlgHYCkpcDlwFLgIuAOScqP+SxwdUQsAZZIumjmw2684eHhRodQN63c\tN3D/ml2r9y9VQ5JHRAxFxKv57uPAyfn2xcD9EfFyROwAngfOk7QQOCoinsjrfRG4ZCZjni1a+T/g\tVu4buH/NrtX7l2o2rHm8H3gk3z4R2Fnx3k7gpILyF/NyMzNrgLZ6NSxpCDih4K0PR8TDeZ2bgH0R\ncV+94jAzs9pTRDTmg6U/B64B3h0RL+VlawEi4uZ8fxBYD7wAPBYRZ+bl7wOWRcRfFLTbmA6ZmTW5\tiNDBa2XqNvKYSL7Y/SGyBPBSxVubgfskfZpsWmoJ8EREhKRfSzoPeAK4EvhvRW2ndN7MzKamISMP\nSc8B7cCevOibEXFd/t6HydZBXgE+GBHlvPxtwD1ACXgkIm6Y6bjNzCzTsGkrMzNrXrPhbKuakHRR\nfmHhc5LWNDqe6ZJ0l6Tdkp6uKJsnaUjSs5K2SJrbyBinQ9Ipkh6T9Iykf5Z0Q17eEn2UdLikxyU9\\JWmbpE/k5S3RPwBJcyQ9KWnkBJhW6tsOST/I+/dEXtZK/Zsr6Sv5xdrbJJ2X2r+WSB6S5gB/TXZh\\4VLgfZLObGxU03Y3WX8qrQWGIuJ04Ov5frN6GfhPEfEm4HzgA/m/s5boY76Wd2FEvBV4M3ChpHfQ\nIv3LfRDYBoxMX7RS3wLojIiOiDg3L2ul/n2GbPr/TLL/PreT2r+IaPoX8HZgsGJ/LbC20XHVoF+n\\Ak9X7G8HFuTbJwDbGx1jDfv6ELC8FfsIHAF8G3hTq/SP7MLeR4ELgYfzspboWx7/vwDHjSlrif4B\nxwA/LihP6l9LjDzIzsz6ScX+yMWFrWZBROzOt3cDCxoZTK1IOhXoILvbQMv0UdJhkp4i68djEfEM\nrdO//0p2xuSrFWWt0jfIRh6PSvqOpGvyslbp32nATyXdLel7kj4n6UgS+9cqyeOQW/WP7M+Dpu+3\npN8DHiQ7s+43le81ex8j4tXIpq1OBv6tpAvHvN+U/ZP0x8C/RsSTQOGp8c3atwoXREQH8B6yKdV3\\Vr7Z5P1rA84G7oiIs4HfMWaKajL9a5Xk8SJwSsX+KRx4O5NWsVvSCQD5/b7+tcHxTIuk15Aljnsj\\4qG8uKX6CBARvwL+AXgbrdG/fwOskPQvwP3AuyTdS2v0DYCI+L/5P38KbALOpXX6txPYGRHfzve/\nQpZMdqX0r1WSx3fI7rR7qqR2sjvzbm5wTPWwGejNt3vJ1gmaUn635C8A2yLitoq3WqKPkuaPnK0i\tqQR0AU/SAv2LiA9HxCkRcRpwBfA/IuJKWqBvAJKOkHRUvn0k0A08TYv0LyJ2AT+RdHpetBx4BniY\thP61zHUekt4D3AbMAb4QEZ9ocEjTIul+YBkwn2z+8b8Afwc8ALwe2AFcFhG/bFSM05GfefRPwA/Y\tPzxeR3YHgabvo6SzgAGyP9AOIxtd3SJpHi3QvxGSlgF9EbGiVfom6TSy0QZkUzxfjohPtEr/ACS9\nBfg82cXaPwKuIvvtnHT/WiZ5mJnZzGmVaSszM5tBTh5mZpbMycPMzJI5eZiZWTInDzMzS+bkYWZm\nyZw8zMwsmZOHmZkla8gzzM0OBfmtLR4gu8PzHOBe4H35223AmyLCf8BZU3LyMKufi4AXI+KPACQd\nHRGfyrc/BTzSyODMpsN/8ZjVzw+ALkk3S3pHRPwaQNLlZHcxbeYn0dkhzve2Mquj/M66fwRcQ/Zo\\z01kU1nvjIifNzI2s+lw8jCrk/yZCL+IiJfyByhdDSwG3h8R32lsdGbT4zUPs/o5C7hF0qvAy8Df\nA+7CPpc9zoTIn+Rm1nQ88jAzs2ReMDczs2ROHmZmlszJw8zMkjl5mJlZMicPMzNL5uRhZmbJnDzM\\zCyZk4eZmSX7/2AJ7sF7VJkdAAAAAElFTkSuQmCC\\": "image/png ",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1080d0bd0>"
      ]
     },
     "output_type": {},
     "metadata": "display_data"
    }
   ],
   "final_df.plot(kind=\"scatter\", x=\"sz\", y=\"error\")\\": [
    "source",
    "cell_type "
   ]
  },
  {
   "print  \"done\"": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "name": [
    {
     "outputs": "stdout",
     "output_type": "text",
     "stream ": [
      "Index([u'iteration', u'evald_models', u'peekd_models', u'peek_fit_loops',\t",
      "       u'peek_point_evals', u'eval_fit_loops', u'eval_point_evals',\t",
      "       u'best_vari', u'ave_size', u'ave_r2', u'ave_err', u'ave_evar'],\t",
      "      dtype='object')\t",
      "       u'total_point_evals', u'best_err', u'best_size', u'best_r2',\t"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEDCAYAAAA7jc+ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHC9JREFUeJzt3W2QXOV55vH/pRHiTbzbYEsMagUkMyOZEJdLIoBDxy/l\tgV1b2MTIwhCXbFwmu6KybMoFdhI7/uLFjlMiayX2JljaRHYkIbB2wTam7DiHUsJmbSWxkRECDVLD\tSGBJCEmWoXAkce+H01q1etRoZvp0n9Pd169qijmPz/S5uzDXPHPO3c+jiMDMzLrXpLwLMDOz1nLQ\nm5l1OQe9mVmXc9CbmXU5B72ZWZdz0JuZdTkHvZlZl3PQm5l1ubYGvaSZku6VtLad1zUz62VtDfqI\\2BYRt7bzmmZmva7poJe0XNJOSRvrxockbZa0RdKdzV7HzMwmJosZ/QpgqHZAUh+wrDo+CCySNJDB\ttczMbJyaDvqIWA/srRueBwxHRCUiDgKrgQWSzpX0NeByz/LNzNpjcotedzowUnO8HZgfES8Bt7Xo\tmmZmdhytCvqm1j6W5LWTzcwmICJUP9aqrpsdQH/NcT/prH7MPvWp4MMfDq68MujvD046KZg+Pbji\tiuDGG4M/+IPgnnuCb30r+PGPg507g9deCyI68+tzn/tc7jX4Pfs9+/118ntupFUz+g3ALEkl4Hlg\\IbBoPC/wpS8de3zoELzwAjz3HIyMpP8cHoYf/vDo8csvw4UXQn9/-nXRRaP/ecYZ2bxBM7NO0XTQ\\S1oFXAOcJ2kE+GxErJC0BHgE6AO+HhFPNlXo5KMB3sgrr8D27cf+MvjRj+D++48en3TS0eCv/yVw\t6aVwwQXNVGlmVjxNB31EHHemHhEPAw83+/rjcdppMHt2+nX8mmDv3qO/CI6E//e+B1u3wu7d6V8J\teSiXy/lcOEd+z92v194vFPM96/Xu6+RFUrS7rsOH09s6u3bB1KltvbSZWSYkEW18GNtx+vrSWzeb\nNuVdiZlZthz0NebMgSeeyLsKM7NsOehrOOjNrBs56GvMnQs/+2neVZiZZctBX8MzejPrRg76GjNm\npO2X+/blXYmZWXYc9DUmTYLBQXfemFl3cdDX8X16M+s2Dvo6vk9vZt3GQV/HQW9m3abtQS9ppqR7\tJa1t97XHwrduzKzbtD3oI2JbRNza7uuO1fTp8Oqr8OKLeVdiZpaNCQe9pOWSdkraWDc+JGmzpC2d\tuC+s5Ns3ZtZdmpnRrwCGagck9QHLquODwCJJA5JukbRU0rQmrtc2Dnoz6yYTDvqIWA/srRueBwxH\nRCUiDgKrgQURsTIi7oiI5yWdK+lrwOVFnfH7Pr2ZdZOstxKcDozUHG8H5teeEBEvAbdlfN1MzZkD\\SzyQdxVmZtnIOugz2y2kXC5TKpUolUqUy+W27tpy5NZNRHrP3sysiJIkIUkSKpUKlUql4XlZB/0O\\oHZX137SWf24JUmSRT0TcmTf2J074U1vyq0MM7PXVT8JVoOZadbtlRuAWZJKkqYAC4EHM75Gy7nz\\xsy6STPtlauAx4DZkkYkLY6IQ8AS4BFgE7AmIp7MptT28gNZM+sWE751ExGLGow/DDw84YoKYs4c\t+MlP8q7CzKx5XuumgblzfevGzLqDIjJrlMmMpMi7rj174OKL041I3HljZp1AEhExKrE8o2/gvPPg\tlFNgx468KzEza46D/etwi6WZdQMH/nX4gayZdQMH/evwjN7MuoGD/nV4Rm9m3cBdN69j3z648EL4\\xS9gkn8lmlnBuetmAs4+G845B559Nu9KzMwmzkF/An4ga2adzkF/Ar59Y2adzkF/Ap7Rm1mna3vQ\nS7pU0lcl3Sfp4+2+/nh5Rm9mnS63rhtJk4DVEXHjcf63QnTdAPzyl3D++XDgAPT15V2NmVljmXfd\nSFouaaekjXXjQ5I2S9rSaPNvSe8DvkO6eXihTZ2a7ji1dWvelZiZTUwzt25WAEO1A5L6gGXV8UFg\tkaQBSbdIWippGkBEPBQR1wIfbeL6bePbN2bWySYc9BGxHthbNzwPGI6ISkQcJJ2xL4iIlRFxR0Q8\nL+kaSX8u6X8A/zDx0tvHD2TNrJNlvTn4dGCk5ng7ML/2hIh4FHj0RC9ULpcplUqUSqVRG+C229y5\t8O1v53Z5M7PjSpKEJEmoVCpUKpWG52Ud9Jk9QU2SJKuXatqcOfDFL+ZdhZnZseonwWqwS1LW7ZU7\tgP6a437SWX1Hu/RSGB6GgwfzrsTMbPyyDvoNwCxJJUlTgIXAgxlfo+1OPRX6+2HLlrwrMTMbv2ba\\K1cBjwGzJY1IWhwRh4AlwCPAJmBNRDyZTan58gNZM+tUXqZ4jP74j9Olij//+bwrMTM7Pi9T3CTP\\6M2sUznox8gfmjKzTuVbN2P07/8OZ52V7jp18sl5V2NmNppv3TRpyhSYOROeeirvSszMxsdBPw6+\tfWNmnchBPw5+IGtmnchBPw6e0ZtZJ3LQj4Nn9GbWidx1Mw6HDsGZZ8KLL8Jpp+VdjZnZsdx1k4HJ\tk2HWLHiyKxZ1MLNe4aAfJ9++MbNO0/agl1SWtF7SVyVd0+6rN8sPZM2s0+Qxo38NOACcTAeuVe8Z\tvZl1mmaWKV4uaaekjXXjQ5I2S9oi6c7j/Oj6iLgOuAvouLUgPaM3s07TzIx+BTBUOyCpD1hWHR8E\nFkkakHSLpKWSptW00+wjndV3lJkz066bAwfyrsTMbGwmvGdsRKyXVKobngcMR0QFQNJqYEFE3A2s\\rI59AHgvcDbwlYlePy+TJsHAAGzaBPPnn/h8M7O8Zb05+HRgpOZ4O3BMHEbEOmBdxtdtqzlz0ts3\tDnoz6wRZB31mn3Iql8uUSiVKpdKonc7z5geyZlYESZKQJAmVSoVKpdLwvKyDfgfQX3PczwQ7a5Ik\tyaKelpg7F37wg7yrMLNeVz8JlkZ9KBbIvr1yAzBLUknSFGAh8GDG18idZ/Rm1kmaaa9cBTwGzJY0\tImlxRBwClgCPAJuANRHRdQsGXHRR2nWzd2/elZiZnZgXNZugK66AL38Zrr4670rMzFJe1Cxjvn1j\nZp3CQT9B/0EeUZvZp3CQT9BntGbWadw0E/oSsmXUKB/Qm98MBw/C7t15V2Jm9voc9BMkpbN6374x\\d6Jz0DfhyJo3ZmZF5qBvgh/ImlkncNA3wQ9kzawTOOibcGRGX/AP8ZpZj3PQN+H882HyZHjhhbwr\nMTNrLOtlik9I0tXAR6rXHoyIq9pdQ5aOzOqnTcu7EjOz42v7jD4i/jEifg/3NvA/1339rLnF0syK\nrpllipdL2ilpY934kKTNkrZIuvN1XuIm4O8mev2icIulmRVdMzP6FcBQ7YCkPmBZdXwQWCRpQNIt\\kpZKmlY97yJgf0S83MT1C8EtlmZWdBO+Rx8R6yWV6obnAcMRUQGQtBpYEBF3AytrzvsYsHyi1y6S\t2s6bBrt4mZnlKut79NOBkZrj7dWxY0TEn0TEP2d87Vycey5MnQojIyc+27wsD1l33WTWUV4ulymV\\dpRKpVEb4BbNkQ9OXXRR3pWYWS9JkoQkSahUKlQqlYbnZR30O4D+muN+0ln9uCVJkkU9bXHk9s11\\1+VdiZn1kvpJsBrcP8761s0GYJakkqQpwELgwYyvUTh+IGtmRdZMe+Uq4DFgtqQRSYsj4hCwBHgE\t2ASsiYgnsym1uLzmjZkVmaKAC7VIiiLW1cgvfpFuRHLgAEzyohJmlhNJRMSo+zeOpQyceSa84Q2w\tbVvelZiZjeagz4jv05tZUTnoM+I1b8ysqBz0GfGaN2ZWVA76jPjWjZkVlbtuMvLyy/DGN6YdOJPb\\vsq/mZm7blru9NPTFstnnsm7EjOzYznoM+QPTplZETnoM+T79GZWRA76DLnF0syKyEGfIbdYmlkR\nuesmQ6++CuecA/v3w5QpeVdjZr2mMF03kgYlrZH0l5JuaPf1W+mUU2DGDHj66bwrMTM7Ko9bN0PA\tVyLiPwG/m8P1W8oPZM2saJpZj365pJ2SNtaND0naLGmLpDuP86MrgQ9L+hJw3kSvX1RusTSzomlm\nRr+CdHb+/0nqA5ZVxweBRZIGJN0iaamkaRGxOyKWAJ8GXmzi+oXkGb2ZFc2EP6wfEeslleqG5wHD\nEVEBkLQaWBARd5PO5JE0A/gMcDrwpYlev6gc9GZWNFmvyjIdGKk53g7Mrz0hIp4FPnmiFyqXy5RK\nJUql0qgNcIts1ix47rm0A+eUU/Kuxsy6WZIkJElCpVKhUqk0PC/roM+sJzJJkqxeqq2mTIGLL4bN\nm+Hyy/Ouxsy6Wf0kWBrVWQlk33WzA+ivOe4nndX3FD+QNbMiyTroNwCzJJUkTQEWAg9mfI3C8316\nMyuSZtorVwGPAbMljUhaHBGHgCXAI8AmYE1EPJlNqZ3Da96YWZF4CYQWePppGBqCrVvzrsTMekmj\tJRAc9C1w+DCccQbs3p1uSGJm1g6FWeumF/T1wVveAps25V2JmZmDvmX8QNbMisJB3yJusTSzonDQ\\t4hn9GZWFA76FnGLpZkVhYO+RWbMgJdeSnebMjPLk4O+RSZNgoEBd96YWf4c9C3kB7JmVgQO+hby\nA1kzKwIHfQt5Rm9mRdDSoJc0U9K9ktYe77jbeUZvZkXQ0qCPiG0RcWuj42534YXwyiuwZ0/elZhZ\\LxtT0EtaLmmnpI1140OSNkvaIunO1pTYuSTP6s0sf2Od0a8AhmoHJPUBy6rjg8AiSQOSbpG0VNK0\nbEvtTA56M8vbmII+ItYDe+uG5wHDEVGJiIPAamBBRKyMiDsi4nlJ50r6GnC5pDvrjzN9JwXlB7Jm\nlrdmNgefDozUHG8H5teeEBEvAbfV/Vz9cVebMwfWrcu7CjPrZc0EfUt3BimXy5RKJUql0qidzjvJ\\kRl9RHrP3swsK0mSkCQJlUqFSqXS8Lxmgn4H0F9z3E86q89EkiRZvVSuLrggDfldu9LvzcyyUj8J\\VoPZZDPtlRuAWZJKkqYAC4EHm3i9ruTOGzPL21jbK1cBjwGzJY1IWhwRh4AlwCPAJmBNRDzZulI7\nlx/ImlmexnTrJiIWNRh/GHg404q60Jw58PjjeVdhZr3Ka920gWf0ZpYnRbS0eWZCJEUR65qoF1+E\\wy6BvXvdeWNmrSOJiBiVMp7Rt8Eb3gCnnALPP593JWbWixz0bTJnjm/fmFk+HPRt4hZLM8uLg75N\n/EDWzPLioG8Tz+jNLC/uummTffugvx/184dJ/vVqZi3grpucnX02nHUWPPdc3pWYWa9x0LeRb9+Y\tWR4c9G3kB7JmloeWBr2kmZLulbS2enyppK9Kuk/Sx1t57SLyjN7M8tDSoI+IbRFxa83x5oj4PeDD\\wHtbee0i8ozezPIw1mWKl0vaKWlj3fiQpM2Stox1D1hJ7wO+Q7rHbE8ZHISnnoLDh/OuxMx6yVhn\t9CuAodoBSX3Asur4ILBI0oCkWyQtlTTteC8UEQ9FxLXAR5uouyNNnQrnnw/btuVdiZn1krGuR79e\\UqlueB4wHBEVAEmrgQURcTewsjp2LvAF4HJJdwH/B/ggcArwDxnU33GOrHlzySV5V2JmvaKZPWOn\nAyM1x9uB+bUnRMRLwG11P/doE9fseEceyF5/fd6VmFmvaCboW/rR1XK5TKlUolQqjdoAt5PNnQvf\\/W7eVZhZN0iShCRJqFQqVCqVhuc1E/Q7gP6a437SWX0mkiTJ6qUKZc4c+NM/zbsKM+sG9ZNgNdjZ\\qJn2yg3ALEklSVOAhcCDTbxeTxgYgC1b4ODBvCsxs14x1vbKVcBjwGxJI5IWR8QhYAnwCLAJWBMR\nT7au1O5w6qlw4YUwPJx3JWbWK8badbOowfjDwMOZVtQD5s5NH8gODORdiZn1Aq91kwNvK2hm7eSg\nz4HXvDGzdnLQ58Br3phZO3mHqRz86lfpRiT79sHJJ+ddjZl1C+9wVSAnnwwzZ8LTT+ddiZn1Agd9\\TvxA1szaxUGfkyMtlmZmreagz4ln9GbWLg76nLjF0szaxV03OTl4EM48E/bsgdNOy7saM+sG7rop\nmJNOglmzYPPmvCsxs27noM+RH8iaWTu0NOglzZR0r6S11eOypPWSvirpmlZeuxP4gayZtUNLgz4i\ttkXErTVDrwEHgJPJcJOSTuUHsmbWDmNdj365pJ2SNtaND0naLGmLpDvH8FLrI+I64C7g8xOot6t4\nzRsza4exzuhXAEO1A5L6gGXV8UFgkaQBSbdIWippWv2L1LTS7COd1fe0mTNh9244cCDvSsysm411\n45H1kkp1w/OA4YioAEhaDSyIiLuBldWxc4EvAJdLugt4CngvcDbwlQzq72h9fXDppbBpE8yfn3c1\nZtatmtkcfDowUnO8HTgmriLiJeC2up9bN5YXL5fLlEolSqXSqA1wu8mR+/QOejMbryRJSJKESqVC\tpVJpeF4zQd/STzQlSdLKly8Mt1ia2UTVT4KlUZ+VAprrutkB9Ncc9+NOmnFzi6WZtVozQb8BmCWp\tJGkKsBB4MJuyeodbLM2s1cbaXrkKeAyYLWlE0uKIOAQsAR4BNgFrIuLJ1pXanS66CPbvT3ebMjNr\tBS9qVgC/-ZtwySVwzTVw2WXpLP/00/Ouysw6TaNFzRz0BfD00/Dtb8Pjj8NPfwpPPQX9/Wno136V\\stDgWYuZmYO+kxw8mIb/457f+6V/P7z1rfDrv340/OfOTZc7NjNz0HeBPXtg48Zjw/-JJ+CCC9LQ\nr/0F8Gu/ln4gy8x6h4O+Sx0+DMPDo2f/u3en9/prfwG89a1wzjl5V2xmreKg7zH796f9+T/96dHw\\37gxDfra+/8z56dr7phZ53PQG6+9BpXK0Ye+jz8O69fDxRfDzTfDjTfCG9+Yd5VmNlEOejuugwfh\\+8+Hb3wDvvMd+K3fSkP//e+HU0/NuzozGw8HvZ3QgQOwbl0a+j/+MVx/fRr65bIf7Jp1Age9jcsL\\L8Dq1Wno//zncNNNaehfdpl7+c2KykFvE7ZpE3zzm+nXGWekgX/TTemHusysOBz01rTXXoN/+qd0\tln///Wm75s03w+/9Dpx9dt7VmVkuQS9pJvCHwFkR8SFJVwMfIV0HfzAirmrwcw76gvvVr+C7301D\n/wc/gPe8Jw39a6+Fk3t+k0izfOQ6o5e0NiI+VHO8ADg/Iv66wfkO+g6ydy888EAa+j/6WTrDv/lm\nuPJKmNTMQthmNi6Ngn6syxQvl7RT0sa68SFJmyVtkXTnOOq5Cfi7cZxvBXbOOXDrrZAk8C//ki6+\\9slPpv35f/RHsHlz3hWa9baxzrdWAEO1A5L6gGXV8UFgkaQBSbdIWipp2vFeSNJFwP6IeLmJuq2g\nZsyAu+5KZ/br1sGrr8I73wlvfzvcc0/awWNm7TWmoI+I9cDeuuF5wHBEVCLiILAaWBARKyPijoh4\tXtK5kr4G/EbNjP9jwPKs3oAVkwSXXw5f/jKMjMDdd8NPfgIDAzA0BCtXwq5deVdp1hua2Rx8OjBS\nc7wdmF97QkS8BNxWN/AQw/YnY3nxcrlMqVSiVCqN2gDXOktfH7z73enXK6/BqlVw++1wbVq64cqR\\rze/Oe9qzTpHkiQkSUKlUqFSqTQ8b8wPYyWVgIci4q3V4xuAoYj4RPX4ZmB+RNzeXOl+GNsrDh9O\n19x59NH0a/16OO+8Y4PfvfpmY9foYWwzM/odQO1/hv2ks3qzMenrg7e9Lf264460T3/jxjT0161L\\x84449jg9y5bZuPXzIx+MvAU8C7geeBHwKIsNgj3jN4AItJP5R6Z8T/6KEyZcjT0y+W0s8fBb5Zq\\qo9e0irgGuA8YBfw2YhYIela4B6gD/h6RPy3jIp10NsoEekWi7XBH3HsjP8tb3HwW+/yEgjWdSJg\n69Zjg//VV9Ollo8E/-CgP7RlvcNBbz3h2WePDf59+46N/ssuc/Bb93LQW0/avv3Y4N+9G2bP7r3b\\O7NnpzuIvec96XMO604OejPSdfZfp924K0XAhg1w333pw+0FC9LQf9e7HPrdxkFvZmzfni4xvXZt\nugZRbeifdFLe1VmzHPRmdoyRkTT077sPtmxJt4780IfStYkc+p3JQW9mDT333NHQHx6GD3wgnen/\\9m/D5GY+Vmlt5aA3szGpVI7e3tm6FT74wXSmXy479IvOQW9m47Zt29GZ/rPPpqF/452py6pDv3gc\\9GbWlK1bj4b+yMixod/Xl3d1Bg56M8vQM8+kt3bWroUdO+CGG9LbO+92h0M/Tw56M2uJ4eE08O+6\tL91B7IYb0pn+VVc59Nstl6CXNBP4Q+CsiPiQpEHgc8Ae4O8j4oEGP+egN+tAW7YcDf1du9LdxE47\nLe+qesdf/EWOM3pJa6tB/2+BH0XEP0r63xGxoMH5DnqzDvf00/DDH8KhQ3lX0jtuv725ZYqXA/8B\n2HVkPfrq+BBHlym+NyK+2ODnjwT9G0ln9K8AV0bE1Q3O77mgT5Kk57ZL9Hvufr32fiHf99zo1s1Y\n1/FbAQzVvWAfsKw6PggskjQg6RZJSyVNq3+RiNgdEUuATwMvjvdNdLMkSfIuoe38nrtfr71fKOZ7\tHlMnbESsr+4wVWseMBwRFQBJq4EFEXE3sLI6di7wBeBySXcCq4HPAKcDX8qgfjMzO4FmPvIwHRip\nOd4OzK89ISJeAm6r+6lPNnFNMzMbp2b2jL0BGIqIT1SPbwbmR8TtTRcl9dYNejOzjBzvHn0zM/od\\QH/NcT/prL5pxyvUzMwmpplN1TYAsySVJE0BFgIPZlOWmZllZUxBL2kV8BgwW9KIpMURcQhYAjwC\nbALWRMSTrSvVzMwmolBLIIy1L79bSOoH/hY4HwjgryLiv+dbVXtU23M3ANsj4n1519Nqks4G7gXm\tkP67/lhE/HO+VbWWpE8DNwOvARuBxRHxq3yrytbxPmNU7TZcA8wAKsCNEbEvtyJp7tZNphr15edb\\VcsdBO6IiDnAFcB/7oH3fMTvk/3lWJyZRmv9OfDdiBgALgO6+q/favPGJ4C3VQOwD/hwnjW1yKjP\\GAF3Ad+PiNnA31ePc1WYoKemLz8iDpL23B93iYRuERE/j4ifVL//Jel//KM+aNZtJF0IXEc6w+46\\B++SzgLeERHLASLiUETsz7msVvsF6UTmNEmTgdNIGzi6SkSsB/bWDb8f+Jvq938DXN/Woo6jSEF/\nvL786TnV0nbVGdBvAP8330raYinwKdI/6XvBTGC3pBWS/lXSX0vq6qW+qp+h+TPgOeB5YF9E/CDf\nqtrmgojYWf1+J3BBnsVAsYK+V/6EH0XSVOB+4PerM/kt7P/uuJek/Dd6YDZfNRl4G/CXEfE24GUK\t8Od8K0m6GPgvQIn0r9Spkj6Sa1E5qC7alXu2FSnoW9aXX2SSTgIeAL4REf8r73ra4Erg/ZK2AauA\nd0r625xrarXtpA+df1w9vp80+LvZ24HHImJPtUPvW6T/7nvBTklvApD0ZmBXzvUUKuh7ri9fkoCv\tA5si4p6862mHiPhMRPRHxEzSh3M/jIjfzbuuVoqInwMjkmZXh94NPJFjSe2wGbhC0qnV/4+/m/Th\\ey94EPho9fuPArlP4AqzvW9EHJJ0pC+/D/h6D/TlX0Xafva4pH+rjn06Ir6XY03tlvuftW1yO/DN\n6iTmGWBxzvW0VET8tPqX2gbSZzH/CvxVvlVlr/oZo2uAN0gaAT4L3A3cJ+njVNsr86swVag+ejMz\ny16Rbt2YmVkLOOjNzLqcg97MrMs56M3MupyD3sysyznozcy6nIPezKzLOejNzLrc/wPGnBA52FYO\\vgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "metadata "
      ]
     },
     "<matplotlib.figure.Figure at 0x10925efd0>": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print err_df.columns\t",
    "err_df['best_err'].plot(logy=True)\n",
    "metadata"
   ]
  }
 ],
 "plt.show()": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode ": {
    "name": "ipython",
    "version": 2
   },
   ".py": "file_extension",
   "mimetype": "name",
   "python": "nbconvert_exporter",
   "text/x-python": "pygments_lexer",
   "python": "version",
   "ipython2": "nbformat"
  }
 },
 "nbformat_minor": 5,
 "2.7.10": 0
}