postmortem_acceptinvest.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import os\n",
    "from categoricalhack import boxplot\n",
    "figwidth=3.3\n",
    "sns.set()\n",
    "sns.set_palette('colorblind')\n",
    "sns.set_context('paper', font_scale=0.75)\n",
    "sns.set_style(\"whitegrid\")\n",
    "figsize=(figwidth, figwidth/1.5)\n",
    "from categoricalhack import boxplot\n",
    "\n",
    "\n",
    "def removeprefix(self: str, prefix: str, /) -> str:\n",
    "    if self.startswith(prefix):\n",
    "        return self[len(prefix):]\n",
    "    else:\n",
    "        return self[:]\n",
    "%config InlineBackend.figure_format = 'retina'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "paperrun2e200000\n"
     ]
    }
   ],
   "source": [
    "nepisodes = 200000\n",
    "path = f\"logs/paperrun2/e{nepisodes}//\"\n",
    "\n",
    "shortname = ''.join(removeprefix(path, 'logs/').split('/'))\n",
    "print(shortname)\n",
    "figdir = \"figures/\" + shortname + \"/\"\n",
    "os.makedirs(figdir, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = pd.read_csv(path+ \"/ppobiglr/evalatend_log.csv.gz\")\n",
    "a[\"Algorithm\"] = \"PPO-MLP\"\n",
    "b = pd.read_csv(path+ \"/cmafixed/evalatend_log.csv.gz\")\n",
    "b[\"Algorithm\"] = \"CMA-ES\"\n",
    "c = pd.read_csv(\"logs/paperrun3/bignetfastredo/evalatend_log.csv.gz\")\n",
    "c[\"Algorithm\"] = \"PPO-DEEP\"\n",
    "c = pd.concat((a, b, c))\n",
    "c.groupby([\"good_site_prob\", \"Algorithm\"])[\"trial_id\"].nunique()\n",
    "c[\"p\"] = c[\"good_site_prob\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>inv</th>\n",
       "      <th>other</th>\n",
       "      <th>accept</th>\n",
       "      <th>run_id</th>\n",
       "      <th>trial_id</th>\n",
       "      <th>good_site_prob</th>\n",
       "      <th>checkpoint_path</th>\n",
       "      <th>Algorithm</th>\n",
       "      <th>p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>5.211583</td>\n",
       "      <td>12.5</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>5.210622</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5.211958</td>\n",
       "      <td>4.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>5.207606</td>\n",
       "      <td>2.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5.212956</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092647</th>\n",
       "      <td>3786</td>\n",
       "      <td>9.332860</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092648</th>\n",
       "      <td>3787</td>\n",
       "      <td>9.333066</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092649</th>\n",
       "      <td>3788</td>\n",
       "      <td>9.333034</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092650</th>\n",
       "      <td>3789</td>\n",
       "      <td>9.332210</td>\n",
       "      <td>13.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092651</th>\n",
       "      <td>3790</td>\n",
       "      <td>9.333282</td>\n",
       "      <td>6.5</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4346828 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Unnamed: 0       inv  other  accept  run_id     trial_id  \\\n",
       "0                 0  5.211583   12.5       0      10  3c19e_00010   \n",
       "1                 1  5.210622    3.5       1      10  3c19e_00010   \n",
       "2                 2  5.211958    4.5       1      10  3c19e_00010   \n",
       "3                 3  5.207606    2.5       1      10  3c19e_00010   \n",
       "4                 4  5.212956   10.0       0      10  3c19e_00010   \n",
       "...             ...       ...    ...     ...     ...          ...   \n",
       "1092647        3786  9.332860   12.0       1       9  c3315_00009   \n",
       "1092648        3787  9.333066    7.0       1       9  c3315_00009   \n",
       "1092649        3788  9.333034    2.0       0       9  c3315_00009   \n",
       "1092650        3789  9.332210   13.0       1       9  c3315_00009   \n",
       "1092651        3790  9.333282    6.5       1       9  c3315_00009   \n",
       "\n",
       "         good_site_prob                                    checkpoint_path  \\\n",
       "0                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "2                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "3                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "4                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "...                 ...                                                ...   \n",
       "1092647             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092648             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092649             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092650             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092651             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "\n",
       "        Algorithm    p  \n",
       "0         PPO-MLP  0.1  \n",
       "1         PPO-MLP  0.1  \n",
       "2         PPO-MLP  0.1  \n",
       "3         PPO-MLP  0.1  \n",
       "4         PPO-MLP  0.1  \n",
       "...           ...  ...  \n",
       "1092647  PPO-DEEP  0.1  \n",
       "1092648  PPO-DEEP  0.1  \n",
       "1092649  PPO-DEEP  0.1  \n",
       "1092650  PPO-DEEP  0.1  \n",
       "1092651  PPO-DEEP  0.1  \n",
       "\n",
       "[4346828 rows x 10 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "accept_other_val = []\n",
    "for n, g in c.groupby([\"Algorithm\", \"good_site_prob\"]):\n",
    "    accepted = pd.DataFrame(g.query(\"accept == 1\")[\"other\"].copy())\n",
    "    accepted[\"p\"] = n[1]\n",
    "    accepted[\"Algorithm\"] = n[0]\n",
    "    accept_other_val.append(accepted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "allppo =  pd.concat(accept_other_val)\n",
    "alg_order = [\"CMA-ES\", \"PPO-DEEP\", \"PPO-MLP\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 166,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "boxplot(data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, dodge=1.8, width=0.4)\n",
    "plt.ylabel(\"Partners' investment value on accept\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.xlabel(\"$p$\")\n",
    "#plt.legend(title=\"Algorithm\", ncol=2)\n",
    "plt.legend(loc=\"lower left\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/accept_boxplot.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 166,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "ax = boxplot(data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, dodge=1.8, width=0.4, labels=None)\n",
    "plt.ylabel(\"Partners' investment value on accept\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.xlabel(\"$p$\")\n",
    "ax.get_legend().remove()\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/accept_boxplot_no_label.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 166,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "sns.violinplot(data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, bw=0.1, cut=0)\n",
    "plt.ylabel(\"Partners' investment value on accept\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.xlabel(\"$p$\")\n",
    "#plt.legend(title=\"Algorithm\", ncol=2)\n",
    "#plt.legend(bbox_to_anchor=(1, -0.15), ncol=2)\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/accept_violin.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 156,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "ax = sns.violinplot(data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, bw=0.1, cut=0)\n",
    "plt.ylabel(\"Accepted partner's investment\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.xlabel(\"$p$\")\n",
    "#plt.legend(title=\"Algorithm\", ncol=2)\n",
    "ax.get_legend().remove()\n",
    "\n",
    "#plt.legend(bbox_to_anchor=(1, -0.3), ncol=2)\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/accept_violin_nolabel.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x237.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 235,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth))\n",
    "\n",
    "boxplot(data=c, x=\"p\", y=\"inv\", hue=\"Algorithm\", dodge=1.8, width=0.4, hue_order=alg_order)\n",
    "plt.ylabel(\"Mean agent's investment\")\n",
    "plt.xlabel(\"$p$\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/invest_boxplot.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>p</th>\n",
       "      <th>Algorithm</th>\n",
       "      <th>run_id</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>inv</th>\n",
       "      <th>other</th>\n",
       "      <th>accept</th>\n",
       "      <th>good_site_prob</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.1</td>\n",
       "      <td>CMA-ES</td>\n",
       "      <td>0</td>\n",
       "      <td>14260.0</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>7.728674</td>\n",
       "      <td>0.497879</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.1</td>\n",
       "      <td>CMA-ES</td>\n",
       "      <td>1</td>\n",
       "      <td>14563.0</td>\n",
       "      <td>8.500000</td>\n",
       "      <td>7.775861</td>\n",
       "      <td>0.468466</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.1</td>\n",
       "      <td>CMA-ES</td>\n",
       "      <td>2</td>\n",
       "      <td>15180.0</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>7.717829</td>\n",
       "      <td>0.361879</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.1</td>\n",
       "      <td>CMA-ES</td>\n",
       "      <td>3</td>\n",
       "      <td>14314.5</td>\n",
       "      <td>9.500000</td>\n",
       "      <td>7.747782</td>\n",
       "      <td>0.400629</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.1</td>\n",
       "      <td>CMA-ES</td>\n",
       "      <td>4</td>\n",
       "      <td>17035.5</td>\n",
       "      <td>7.500000</td>\n",
       "      <td>7.764323</td>\n",
       "      <td>0.529291</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>1.0</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>19</td>\n",
       "      <td>5088.0</td>\n",
       "      <td>9.774185</td>\n",
       "      <td>7.703302</td>\n",
       "      <td>0.420556</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>1.0</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>20</td>\n",
       "      <td>6652.0</td>\n",
       "      <td>9.128277</td>\n",
       "      <td>7.776287</td>\n",
       "      <td>0.475460</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>1.0</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>21</td>\n",
       "      <td>4984.5</td>\n",
       "      <td>9.583926</td>\n",
       "      <td>7.707523</td>\n",
       "      <td>0.461986</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238</th>\n",
       "      <td>1.0</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>22</td>\n",
       "      <td>7848.0</td>\n",
       "      <td>8.696220</td>\n",
       "      <td>7.738676</td>\n",
       "      <td>0.494426</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>1.0</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>23</td>\n",
       "      <td>5281.5</td>\n",
       "      <td>9.508409</td>\n",
       "      <td>7.752414</td>\n",
       "      <td>0.457119</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>240 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       p Algorithm  run_id  Unnamed: 0        inv     other    accept  \\\n",
       "0    0.1    CMA-ES       0     14260.0   8.000000  7.728674  0.497879   \n",
       "1    0.1    CMA-ES       1     14563.0   8.500000  7.775861  0.468466   \n",
       "2    0.1    CMA-ES       2     15180.0  10.000000  7.717829  0.361879   \n",
       "3    0.1    CMA-ES       3     14314.5   9.500000  7.747782  0.400629   \n",
       "4    0.1    CMA-ES       4     17035.5   7.500000  7.764323  0.529291   \n",
       "..   ...       ...     ...         ...        ...       ...       ...   \n",
       "235  1.0   PPO-MLP      19      5088.0   9.774185  7.703302  0.420556   \n",
       "236  1.0   PPO-MLP      20      6652.0   9.128277  7.776287  0.475460   \n",
       "237  1.0   PPO-MLP      21      4984.5   9.583926  7.707523  0.461986   \n",
       "238  1.0   PPO-MLP      22      7848.0   8.696220  7.738676  0.494426   \n",
       "239  1.0   PPO-MLP      23      5281.5   9.508409  7.752414  0.457119   \n",
       "\n",
       "     good_site_prob  \n",
       "0               0.1  \n",
       "1               0.1  \n",
       "2               0.1  \n",
       "3               0.1  \n",
       "4               0.1  \n",
       "..              ...  \n",
       "235             1.0  \n",
       "236             1.0  \n",
       "237             1.0  \n",
       "238             1.0  \n",
       "239             1.0  \n",
       "\n",
       "[240 rows x 8 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cmean = c.groupby([\"p\", \"Algorithm\", \"run_id\"], as_index=False).mean()\n",
    "cmean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 156,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "\n",
    "ax = boxplot(data=cmean, x=\"p\", y=\"inv\", hue=\"Algorithm\", dodge=1.8, width=0.4, hue_order=alg_order)\n",
    "plt.ylabel(\"Mean agent's investment\")\n",
    "plt.xlabel(\"$p$\")\n",
    "plt.ylim(0, 15)\n",
    "ax.get_legend().remove()\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/invest_boxplot_no_label.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 156,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "\n",
    "ax = sns.violinplot(data=cmean, x=\"p\", y=\"inv\", hue=\"Algorithm\", dodge=1.3, hue_order=alg_order, cut=0)\n",
    "plt.ylabel(\"Mean agent's investment\")\n",
    "plt.xlabel(\"$p$\")\n",
    "plt.ylim(0, 16)\n",
    "ax.get_legend().remove()\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/invest_violinplot_no_label.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "figures/paperrun2e200000/\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAE5CAYAAAA3NOGDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAABYlAAAWJQFJUiTwAABfV0lEQVR4nO3dd3xT5RoH8N/JapM2lC5GoS2jlDIre+8pyBBUFAURBLwIuBDZCCJDxoULgqCAiF4R2UspQ4Zw2SCrlFUKlNVJ0zRt5v0j5DRpkibtSZqc9vl+Pnxoc07PebPOc971vIzBYDCAEEIIIW4h8HQBCCGEkNKMAi0hhBDiRhRoCSGEEDeiQEsIIYS4EQVaQgghxI0o0BJCCCFuRIGWEEIIcSMKtIQQQogbUaAlhBBC3IgCLSGEEOJGFGgJIYQQN6JASwghhLiRyNMF8Ebnz5/3dBEIIYR4oSZNmhT5b6hGSwghhLgR1WgLUZw7F0IIIaUPl5ZOqtESQgghbkSBlhBCCHEjCrSEEEKIG1GgJYQQQtyIAi0hhBDiRhRoCSGEEDeiQEsIIYS4EQVaQgghxI0o0BJCCCFuRIGWEEIIcSMKtIQQQogbeW2gjYuLw7vvvosmTZqgdu3advc7ePAgateujUmTJpVg6QghhBDneG2gValUaNmyJUaNGmV3n/T0dMybNw+NGzcuwZIRQgghzvPa1Xv69esHADh9+rTdfWbNmoUhQ4bgxo0bJVUsQgghpEi8NtA6smfPHjx79gxDhw7FlClT3HKO+Ph4txyXEEJI2cHLQPvs2TN888032LBhAwQCr239JoQQQvgZaGfMmIFhw4ahevXqbj1PnTp13Hp8Qggh/MBl4XdeBtpz587h2LFjWLRoEQBAr9cDAHbt2oXr1697smiEEEKIBV4G2t9++w06nY79fenSpQCAjz/+2DMFIoQQQuzw2kCbmZmJx48f4/79+wDyBybVrFkTNWvWtNi3XLlyAIDo6OiSLSQhhBDigNcG2sOHD2Py5Mns7/379wcAHDp0CFWrVvVQqQghhJCiYQwGg8HThfA2pk7vJk2aeLgkhBBCvAGXuEBzYwghhBA34hxoV6xYAZVKZfV4bm4uVqxYwfXwhBBCCK9xDrTffvstcnJyrB5XqVT49ttvuR6eEEII4TXOgdZgMIBhGKvHk5KS2NHAhBBCSFlV7FHHnTt3BsMwYBgGAwcOtEiFqNfrkZKSgp49e7qkkIQQQghfFTvQvv766zAYDPjPf/6DV155BTKZjN0mFotRtWpVdO3a1SWFJIQQQviq2IH2X//6FwCgcuXK6NWrF3x8fFxWKEIIIaS04Jyw4tVXX3VFOQghhJBSiXOgTUtLw/z583Hq1CmkpaWhYP4LWtOVEEJIWcY50E6dOhU3b97EkCFDULFiRZsjkAkhhJCyinOgPXv2LH744Qc0atTIFeUhhBBCShXO82gDAgIgl8tdURZCCCGk1OEcaEePHo2VK1dCo9G4ojyEEEJIqcK56fjPP//ElStX0KFDB9SsWRNisdhi+7p167ieghBCCOEtzoG2UqVKqFSpkivKQgghhJQ6nAPtvHnzXFEOQgghpFRy2Xq0N27cwJ9//skumadWq6HX6111eEIIIYSXONdoMzMzMXbsWJw7dw4MwyAuLg7h4eGYNWsW/P39MXnyZFeUkxBCCOElzjXaBQsWQCgU4vDhw/D19WUf79GjB/7++2+uhyeEEEJ4jXON9sSJE1i5ciXCwsIsHq9evToePXpU7OPGxcXhl19+wdWrV5GdnY2EhAR22+nTp7F27VpcvnwZGo0G9evXx8SJE1GvXr1in48QQghxB8412ufPnyMgIMDqcaVSabFGbVGpVCq0bNkSo0aNstp26dIlxMbGYs2aNdiyZQsiIiIwYsQIZGRkFPt8hBBCiDtwrtHWrVsXx44dw9tvv23x+Pbt2xEbG1vs4/br1w+AsfZa0OjRoy1+//LLL7F3715cuHABXbp0KfY5CSGEEFfjHGjHjh2LDz/8EE+fPoVer8fOnTtx8+ZNHD58GBs2bHBFGR1SqVTIy8tDuXLlXHpcWnmIEEIIV5ybjtu0aYNVq1bh4sWLEAgE+P7775GWloa1a9eiSZMmriijQytWrEBERAQtbEAIIcTrcK7RAkCrVq3QqlUrVxyqyH7//Xds2bIFP//8M0QilzwdVp06dVx6PEIIIfx0/vz5Yv+tSyOTwWCwWvidy4AoR/bt24e5c+di9erViImJcdt5CCGEkOLiHGifPXuG+fPn43//+x8yMzOttrurn/Pw4cOYMmUKli1bhubNm7vlHIQQQghXnAPtF198gcePH2P06NEIDQ0FwzCuKBcyMzPx+PFj3L9/H0B+wK5ZsybOnz+Pjz76CJ9//jnq1q2LlJQUAIBMJoOfn59Lzk8IIYS4AudAe+nSJfz6668ub7o9fPiwRfrG/v37AwAOHTqEnTt3Qq1W4+uvv8bXX3/N7jN27FiMGzfOpeUghBBCuOAcaCMiIqBWq11RFgsDBgzAgAEDbG6bP38+5s+f7/JzEkIIIa7GeaTSlClT8M033+D69evQ6XSuKBMhhBBSanCu0VarVg16vR4DBw60uZ2SPhBCCCnLOAfaCRMmIC0tDRMmTEBISIjLBkMRQgghpQHnQHv58mVs2rSJkjsQQgghNnDuo42IiIBWq3VFWQghhJBSh3OgnTZtGhYuXIiEhASrrFCEEEJIWce56Xj48OHQ6/Xo378/GIaxSrl49epVrqcghBBCeItzoP3qq69oABQhhBBiB+dAay+pBCGEEEJc0Edbp04dpKWlWT2ekZFBI5EJIYSUeZwDrb0BUBqNBkKhkOvhCSGEEF4rdtPxjh07AAAMw+CPP/6Av78/u02n0+H06dOIjIzkXEBCCCGEz4odaKdOncr+PG/ePIttYrEYVatWxaRJk4pfMkIIIaWCUqkEgDK7jGmxA+21a9cAAJ07d8aWLVsQFBTkskIRQgjhr19//RVnz54FAGi1Wpw6dQoA0LJlS4hE+WGHYRh0794dPXr08Eg5SwrnUceHDx92RTnKjLJ+Z0eICX0XPM8d78HevXvxzTff2Nx2/Phxq8eOHDmC0NBQNG7c2GVl8DacA+2uXbsQEBCADh06AACWLl2KTZs2oVatWli0aBEqVqzIuZB8plQqcezYMahUKuTl5WH58uUAgHHjxsHHxwcSiQQdOnSAXC73cEkJKTmnTp3Cp59+CgD49NNP0bVrV5QvX96zhSpj7t+/jzfffBOA8XpUvnx5dOjQATKZrNjHzMrKwsKFC9nf61TSQKUF7qWKAQDVQjSQGn+ETg/cfCqGXq/HV199hc2bN0MsFhf/CXkxxsAxb2Lv3r0xadIktGvXDteuXcOgQYMwfvx4HD9+HKGhoViyZImrylpizp8/DwBo0qQJ52PNnj0b27dvL3Sf7t27Y8GCBZzPRQgf3Lt3D6+++qrFY+3bt8eyZcs8VKKy6fXXX8ft27ctHhs0aBCnsTXfffcdVq9eDQD4rJsC1YJ1uJ8uxMI4Y0Xi8+4KRATlr1t+/r4YP5401qZnz56NPn36FPvc7sYlLnCe3vPo0SNUr14dAHDo0CF06dIFo0aNwuTJk3H69Gmuh+c1rVbrVNP6sWPHkJubWwIlIiVJqVSyTXMk3927d60eK3jBJ+6l1+tx7949q8evX79e7GNmZ2dj06ZNAICGVdSoFqyz2qdgta5xuAZh5Y37rV+/Hjqd9d+UBpwDrUgkglqtBgCcOXMGLVu2BACUK1cO2dnZXA/Pa1evXsXz588BAGn1+yDlpdfZbSkvvY7U2IEAgNzcXHbgAOG35cuXo0OHDmjbtq3Fv3bt2qFdu3YYOnQo0tPTPV1Mj8rKyrJ6TKFQeKAkZVd6errFqms5FWIAAE+ePCnW8QwGAxYvXsxe77rVzWO3Xbyf3xz83TE/HL8lYX9nGKB7HWMlIzExEevXry/W+b0d50DbsGFDrFy5Ejt27MDFixfRrl07AMDDhw8RGhrKuYB8duzYMQCAXiiGOqAqAPOc0Aw08orQiaUAbA8S4IJqUyXvp59+wrp165CVlWXx2iuVSmRnZyM7OxtXrlzBxx9/XKbfG1tBValU0upfJahgQNVJjP2yqamp0Gg0RTqWTqfD4sWL2dwKzarl12ZP3Jbg4A1fdt/sPAE2n5fhxJ38YNsoQoMaIcag/+233+Lnn38udZ8FzoF28uTJuHnzJubMmYMPP/wQVatWBQDExcUhNja22MeNi4vDu+++iyZNmqB27dpW2//55x8MGDAADRo0QK9evXD06NFin8td/v77bwCAunw4ILCRJYsRIC/QmNTj+PHjLvlw7d+/H0uWLEGXLl3QpUsXLFmyBNu2bSu1TTLeQK/X47vvvsO///1vAIBUrEebmvldAR1q5aJfrIq9mFy5cgUffPABUlNTPVJeT7N1k6HX68v0zUdJe/z4scXv+heB1mAw4NmzZ04fJz09HePGjcMvv/wCAAgP1OKNJjns9rjrvjb/Lu66D/uzgAHea61EkEwPAFi8eDFmzJgBlUrldDm8HedAGxUVhV27duHcuXMYM2YM+/iECROsElkUhUqlQsuWLTFq1CirbRkZGRg5ciQaN26M7du3o1+/fhg7dqzNPgdPefLkCW7dugUAyA2ynyErL6gau/+dO3c4nTMhIQGTJk3Cxo0bkZeXh7y8PGzcuBFfffWVy2vMxOj58+f47LPP2AEgoXIdprysQOua+bWC5tU16FonDx93yUa7WsYmtatXr+Ltt9/GhQsXPFJuT7LXpeSuQEutO9asA62f3W32XL16FW+99Rb+97//AQCiK2owtpMSvi9aipV5DNJzbIeYdKUQyrz8Fr7yMgM+7qJg+2v37NmDd955B/fv33f6OXkzztN7TG7cuIF79+6hQ4cOkEqlEIlEFhOTi6pfv34AYHNA1e7du+Hv74+pU6eCYRhERUXh2LFj+O233/DFF18U+5yudOLECfZnU63VlrzAcBjAgIEBJ0+eRFRUVLHP+ejRI7vbkpOTi33c0sZVcwdPnz6NGTNmsDWA6iFajGyrhNzXgCyzsW2mhgqGAV5vrEI5Xz32XpHi2bNnGDlyJIYPH45Ro0aV2qkNBZmS3RS0Y8cOjB49usjHS0lJwZ9//omEhASrbVqtFocOHQJgTK5j6zX28fFh+9LLynvw9OlTi9914vwpPc4E2vPnz2Ps2LHsIM5udXLRu0EuhGZxVWusoIJhGBgMBshkMuTk5LC/m7abBPoZ8GlXBTafk+HMPQnu3r2LYcOGYe3ateyAW77iHGgzMzMxduxYnDt3DgzDIC4uDuHh4Zg1axb8/f0xefJkV5TTwuXLl9GiRQuLdXBbtWqFkydPuvQ88fHxxf7buLg4AIBGFgy9j7/d/QwiH2jKVYIk6zEOHDiAZs2aFfuctkZu+or0yNUKcP/+fU7PpzQ4evQojh49yrY0REVFQSgUokqVKhg6dKjTi2BotVps2rQJ+/btYx/rGJ2HfrEqiF4couAAkF71c9GulhoMA/Ssl4fwQB1+Pi1Ddh7www8/4NChQxg7diwqV67suifspex9Djdt2oT27ds7dQytVovz58/jyJEjuHz5slPdLqbvpC3btm2Dv78/2rRpg86dOyM8PNypcvCV6TtgYhCKoRdKINCpcfXq1UJv+NVqNT7//HPk5uZCIjJgWCslGlTR2t3fYDBg4cKF6Nq1Kw4cOICJEyfa3ddHBAxpmYMaIVpsPi9FRkYGJkyYwPt1zzkH2gULFkAoFOLw4cPo3bs3+3iPHj3cNjc0PT3dagm+wMBAm8v1eYJer2eHyecFOv7C5pUPhyTrMRISEqDRaIp9V52Xl2f1mEgIQIsyP31Ir9dj3bp1FgM9TDWg69evo3Hjxk6NKcjMzMSSJUvYm5pyvnq83SIHdSvnX2jsDQARCIA2NY0j9OuFaTGppwL/PSPD9cdiJCYmYtq0aRgzZoxL5m97K4VCYfNzChhfW4VCUWjyFoPBgBMnTmDz5s1WfdxVy2shMbuiZeYwyFCJrGpTgTItykuNgdkAIEMpQKZKgOzsbOzfvx/79+/HSy+9hCFDhpTaGx9b10qdjz8EOekOr6N37txBRkYGADgMsgAgk8nQtWtXAEC3bt3Y96IwbaLU0OiArRdluHv3LtLS0hASElLo33gzzoH2xIkTWLlyJcLCwiwer169eqFNmVyU1Ii04q6nm5CQwPZDqctXdbh/XvmqkN8/A41GA61Wi4YNGxbrvKZmdrHQAI3OePfnKzYgO8/YPFaW1wdOS0uzGk2ZF1AFPs+NTeoSicTh65OSkoIvvvgCDx48AAA0qKLB4OY58Pex/DwWNgDEFGgBIEBqwAftlTh2S4Idl6RQqVT497//jblz55ba3K+OBtpUq1YNFSpUsLv9xx9/xMqVK9nfQ/x1aFldjWbV1Ajys3wfPt8SYLM2laNmMKtP/shnvQG480yE0/fEuHhfArWOwaVLl3D79m38/PPPpXIVMltTrHQ+/hDnpEOtVhf6XRAI8tuH76WJHAbanJwcHDhwAN26dcOBAwccBlkA0OuBpHQRe76GDRsiICDA4d+5M62nKWFFcXAOtM+fP7f5AiiVSos3xJWCg4Ot5iJmZGQgODjYLecrqn/++QeA8W5ZXc7xHbFGXhEGRgjGoMPFixeLnfPTNGTfz0ePzBxjG6a/RI9UCK36ZMoaW3NXs6q3QfDVnRBo8xzObTUYDJgyZQobZHs3UKFH3TwUbM0yDQCx1S9lGgDiZxaYGQboEG2cDrH6uB8UucCMGTMQHR3N+34pd9i8eTP788i22ahfRQuBjRZFZR6DXC1jtzZl/j4IGKBWRS1qVdRiQKNc7PrHFyfu+CA7Oxt//vlnsfqNvZlWq7X5eTcNiEpJSSn076Ojo9GsWTOcPXsWcdd9ocxjMLCxCuICPS+iF5d/hmEwceJEqz5akZ3wkKNmsPGUDFcfGVv2+vTpU2iQvX37Nh4/fozc3FzMmjULADBz5kz4+ubf8IaHh6NatWqFPi934hwJ69aty84XNbd9+3ZO03sK07BhQ6tBUqdOnXLb+YrqypUrAACtLAgGkY+DvQEIhNDIjXOO7Q0UcYapKTRYlj/KwHSXn5CQUOrmphVFZmamzcf1It9Ct5ucOXMG586dAwC8XF+FnvWsgyyQPwDEVJM6ceIEvvnmG/a1LzgAxCQyWIdxnbIhERqgVquxdu1ah8+pLDKf6rf7shTnk8TQ2XhNTa+zqTYFwKI2Zet9eK5icOC6D87ey5/jGR0d7brCe4n09HTo9dYvgO5FoHU07YxhGMyfPx+1atUCAJy444OFcXIkZ1qGEz8fA4Jkevazb3rtDQYDgvx0FjecJreeijDvDzkbZFu0aFHoANezZ8/ijTfewPjx4zFx4kR2hPnEiRMxfvx49t9rr73GKesVV5wD7dixY7Fw4UIsWbIEer0eO3fuxPjx4/HLL7/gww8/LPZxMzMzER8fzw7vjo+PR3x8PNRqNfr06YPs7Gx8/fXXuHPnDtasWYN//vkHgwYN4vp0XMIU8DT+9pvACuYiM+1b3AFL6enpbICvGpg/Z7bKi5+fPn2KmzdvFuvYpYEpY40FgwF6sTHQ2mpKM3fp0iUAgFRsQI+6tvsYzdmqSTlSOUCP5tXVFucjlqZPn46mTZsCAJ5kCfHTKT98ubsc9l/zgSLX+s7HVJtq06YNJk6caHNAzb00IX76nwwzd5fDwRu+UOsYiEQijBs3Dh07dnT3UypxNvtgDQZ2Lm16errDefdBQUFYv349unfvDgB4/FyIRXFyHLkpsbi0da9re2xI9wLfIZ0e2PWPL5b/5YdMlTEsDR48GMuXL4dUKrVbjh9++MGpCoROp/No1inOgbZNmzZYtWoVLl68CIFAgO+//x5paWlYu3Ytp0Edhw8fRv/+/TFt2jQAQP/+/dG/f388e/YMgYGBWLNmDc6fP49+/fph+/btWLFihUebBky0Wi0SExONP/tZNmX7puaPCg66vgeyR1fY3zUv9n3y5Emx5vxt3bqV/XLUqpjfZ1I9RAuJ0PhBNOUhLYtsZSMKur4HjM4Y2BwFWtOIZJ0BNmtQBdmrSTmifXF9c1e3C98FBQVh9erVmDdvHmrWrAkAyFQJsOeKFDN2lcPGUzI8zBCwzZK2alOAMUfb2XtiLIzzx+IDcpxNkkCnZyAQCNCjRw9s3rwZw4cP5/VIV3tsNRsHXd8DUbaxJqvX623fmBbg5+eH+fPnY+bMmZBKpdDqGWy9IMO6EzLkvRgO0SZKja518oOtv48ebzTJsRirkJXL4D+H/XEg3hcGMAgKCsJ//vMffP7554UODE1KSsKZM2cAANm1KyO1Ywy7LbVDbTzp3wRP+jdBTnVja+Fff/3lsQGzLplH26pVK7Rq1coVh2INGDAAAwYMsLv9pZdewrZt21x6Tld4/Pgxm0NUKwtkH5c+vgr5w/zkBEKNCgF3jsLACKCqXA9aaf6+9+/fL9LApadPn2LDhg0AgHphGgTK8u/wfERAixpqHL/lg127dmHQoEGIiYmxd6hSy9bNi1CjglCjsrvdnKkWpdYyiLvui1caFj6K216/VGEeZQpwLsnYbNm8efNC9+U7e3MrnSEQCNCzZ090794d//vf//Drr7/ixIkT0OoZnLknwZl7EtQP0yBAqsdzlfUNi7+PHosP+iNdmd+pKJfL0b9/fwwaNAhVqlRx2fP0Rra6SYQaFfye5jetPn/+HEFBQQ6PxTAM+vfvj8aNG2Py5Mm4fv06Lj2UICVbiA/aZ6O8zIBG4RocjDe2HH3QXolIs8UGHj8XYNVRf2S8SGzRunVrfPXVV06d2zRdy8AwUNSvCv/4/FwBgafuQtEwHMroSshqGA5ZYgp0Oh0OHTqEN954w+GxXY3zbfPEiRPZzCDEcrK3zrcc+7P/A9sj1vwfnLPa19nMLICxBj19+nTj4DPGgL4NrdOW9aybC1+xAXq9HlOnTnW6dlWaOAqkjrbHxsaidevWAID9131xOMHHaiUSc/ZqUvY8yhRg5VF/aPUMfH19MWzYsEL35zt7fdhFIRAI0KZNG6xYsQI7d+7E4MGD2dGmVx+JbTYlA8bpVqYgW7NmTUybNg379+/Hp59+WuqDLOC49cbZfcxFRERg/fr17PKHyZlCzP2jHP57RooDZukWD8b74LdzUvx2ToqfTsnwzX45G2Q/+OADLF++3KkgC4CNO3kVy8H3QRrk1/NnuQjzNCh/9i5kt59C5+8LdZDxc2GeSKgkcQ60aWlpeP/999GxY0f8+9//5pxGkO/MmyZMgwsYjQqiPNurk4jyFGA0KuglMpguNc7mwDUYDFi0aBG78k+PerkIK2/drllOasBrjY0X/Lt372Lq1KllLvexo8FOjgItwzCYPXs2eyHeflGKn0/LkFsg/7q9kZT2thsMxibMJQfleK4yjlaeNWsWmzO8tCpOH3ZhIiIi8Pnnn+OPP/7Av/71L0ilUugNDAAD/H2M3wmJKD+Y16pVC8uWLcPvv/+OgQMHFtoP6Cy+pHp05vpSnNWUJBIJpk+fjo8//hgAoNIw+N9dH1x6mD+47NJDCf6+7YO/bxsHnWn1xv7wefPmYfTo0U53mWg0GnbgaF7FAMiv2c58J7/6kN0HgNPJTVyNc6Bdu3Ytjh49iiFDhuDIkSN45ZVX8Prrr+OXX35xeHErjUx9GwYwMAiNHzBGX3hQY/Q6gBGwI5Sd/ZB///33+O233wAAdSpr0LOQQTrNq2nQuoZx+5EjR/D111+XqVHI9pIkmJiWeixMcHAwfvjhB3a05Zl7Esz/U46bT/N7YEwjLW0pONJSkctg/UkZfjrlhzwtA4lEgnnz5rEDTEqz4vZhOyKXyzFq1Cj8+uuv8Pf3B8CgRTU1lg3KhFhgfO07duyIX375Be3bt3dZH+ytW7fQpUsXdO7cGatWrbJYgs7bOLN8aXET+jMMg3fffRfz5s1DgwYNEB0djejoaERFRSEqKor93fSvSZMmWLFiBXr27Fmk8zx8+JD9zmrlPhApbX+/Rco8CPI00AQZs/NlZmZ6ZJlKl/TRhoSEYMSIERgxYgTi4+Oxc+dOrFq1CgsWLMDly5ddcQreMH1ADUIxbM7/KIRBKAa0eU59yH/66SesWrUKgDEjznutlSjsZpBhgDeaqpCRI0D8EzG2b98OiUSCL774olQO+CjI0YXP2QtjpUqV8OOPP2LBggXYtWsX0pRCLP/LH22j8tA/VgUfsXGk5aZz1jU085GWF+6LsfmcFEq18U2LjIzE/Pnzy0z/eXH6sIsiMjIS4eHhiI+Px6EEX5y+J2Ff65YtW7o8p/GyZcvYm7k1a9agVq1abI3d2ziTJY7ryjk9e/YscvAsCtN8dgDQyRxModQZoC2XP6f2wYMHJZ5zweVDGxmGYS/ctuZqlXZskyxT9JfW8OJvHF30f/nlF3ZJtorldBjTUQmpE9cNoQAY0VbJLtf222+/YfHixWWiZuuoqbwoTekymQyzZs3CkiVL2LRwf9/2wfz9cjxIF6JNlBpvNMlBeanx819emj/SMk8D/PQ/Gdaf9INSbWwqfvvtt/Hrr7+WmSALFL0PuzjMs9Vl5wlsPu4qBcdVJCUlufwcruLMerNFXZO2pJmS8xgYQCeVONgb0PnlB+OijIFxFZfUaFNSUrB7927s3LkTN2/eRJ06dTB69Gi88sorrjg8r+TXDotx4TBNPSikhrlt2zYsWrQIAFBBbkxyIPe1f66C1y8fEfCvDtlYecQfiWki/PLLL5BKpZzmPJdVnTp1QuPGjbFw4ULs3bsXqdlC/PuQP95rrUS7Wmq0jVLDAONUEoYxNhWvPOKHh5nGr11ERARmzZqFl156yZNPo9T64osv0LBhQ4saXJUqVdCmTRuXn6tgc2RR1nQtac7cVHr7GA5Tpju9rwSFNuW9oJeIoBcKINDpPfLecA60I0aMwKlTpxAcHIw+ffpg8eLFnJZ64zsfH+OdE6O3rpU6mtJg+hvz1GHmjh8/jq+//hoAEOynw9hO2QiQWgdZeyvHmPiKjcF2xRF/3E8X4YcffkDFihXx2muvFeMZl20BAQGYM2cO2rdvj1mzZiEnJwdrT/hhfKds1AjVwXTLpNYCq47mB9l+/frhiy++cMkgHGJbaGgohg4d6vbz6HQ6q1G6jtIYeoPCrkfe3hppCrQ6mePaLACAYaCXSSBQ5HokHS3npuPAwECsXr0aR48exeeff16mgyyAFwMwXgxwKjAIqtApDQYDBFpjMLSVEPvhw4eYMmUK9Ho95L56jO2ktJgva2Jv5ZgTdyw/kFIJMKaDEhXLGcu4YMECNrNUaeRoNCPXBBHdu3fHDz/8ALlcDp2ewc+nZRaJLQ7G++JBhjHIfvDBB+wkf8J/GRkZVoHJ2ZkDnlTY9cjbx22YbmScDrRm+3qiRss50C5atAht27b1+jempJQvX579WaCxHFBQ2JQGRqcBY9BZHQMwfiFmz56N7OxsCAUGjGqnRIi/7TvOwlaOKcjPx4DR7ZSQivXQarWYMWOG1/fNFJejtWZFIu69KHXq1MH06dMBACnZQlx/bDymTg8cu2X8krdv3x6jRo2i70spYivbkLcs2WmL6btQ2PXIFd8Hd2IDrRP9syamfT1xE+SSV/PevXs4c+YMUlNTre7sxo4d64pT8EZoaCj7szAv22LR98KWixKolTaPAQAnT55k58r2qp+LasG2+09MK8fYYmvlGAAIlesxqKkKP/7PD/fu3cOOHTvw+uuvO/ls+UMiKfwL6apRqF26dEFISAhSU1Nx86lxCbHHz4XsiNdBgwZRkC1lTGuzmvPEFBJnmT7rhV2PXD0q29VM0yj1PvkhzFHXnN5HbPG3JYlzoN26dSumT58OX19fBAcHW1xEGIYpc4HWfKFoYZ4CGlRify9sSoMoN8vmMQDjawwYR692rm1/Pqi9lWEcbW8cocGRm1rcSxNh69atpTLQmi4c9r6Mpr51rgQCAaKjo5GamoojN30hEhqn8piUpZHFZYWtQJubmwuVSuWV3QPsOJJCrkf2xol4C9McWoNZS5WttYfNGYTGm11Hc+rdgXOgXbVqFUaMGIGPP/7YYfNcWRAYGAg/Pz8olUqIVJkW2wqb0iB8sa9QKLQItDqdjl0SsFk1NUROvMRFzSPLMEDL6mrcSxMhISEB6enpTqdB4wvThcPel9FRjbcozLM6mXK8AoBUKkVgYKCtPyE8ZivQmh73xkBr/l0AbF+PvLHc5kxN24xZC6q9tYdNTPu68rvuLM59tKmpqXjjjTcoyL7AMAwiIyMBAKIc219AW0Qq475Vq1a1aLZJTU1lPyyRdpqMCypOHtkIs2ObliYsTUwXF3v9Uq68sLz22muoV68eqlSpwv4LDw/H2LFjqdm4FLLXTGwvAHuarcGWxdnHk0zda8Ls/Nqpo2xjwhfZo0xz30sS5xpty5Ytce3aNYSHh7uiPKVCVFQUrl+/DpHS+QER4hf7mpb+MjHP0CIVOzc319GdnS3mx+aaFcYbmQKpvX4pVwbaWrVq4eeff3bZ8Yh3szfwyVv7aZ0JonK5vARKUny1atXCrVu3IEnNT1dbaLYxgwGSFAX7tyWNc6Dt27cvFi5ciCdPnqBu3bpWo9UaN27M9RS8Ex0dDeBFLdVBnmMAgAEQKVMt/tbEfCRgwQT29hQ2yMGeXE1+Tcvbm42Kw3Rxsfdl9PY7eOK97I1i9da5tOXKlXO4T0BAQAmUpPhatGiBffv2QZylgjDbmJDEblO4kIEkRQHhiwtoixYtSry8nAPtp59+CgCYP3++1TaGYRAfH8/1FLxTu3ZtAABj0EOsTGVX8bFHmPscAp3xQ1BwsExQUBBEIhG0Wi1Ss4UAHOfkLU4e2dTs/F6ESpUqFbInP5nmN9v7MnJdPYaUXQUTIOjEQgg1Oq/NDuVMbdXbA22nTp0wb9485ObmQnYvFVo/2wsLaP18oPcRw+9mIgDj83L12unO4BxoDx065IpylCoxMTFscBMrnkEXXL3Q/UU5+XfEBRd8F4lEqF69Om7duoXkTOf6wYuTR9Z0bH9/f1SsWNGp8/CJoxqrKRAT93I0EMUTA1W4Kpg7V+8rglCj80hOXWcUnKdfkFQqddkofHeRy+Xo1asXtm3bBr+7z/C8YTjKX7TOL62oXxVChQrS+8Zr7KuvvuqR58Z5MJT5gA9b/8oif39/VKtWDQAgznZ8VyvJNvbxhISEoEKFClbbTcH3frr7BpwlvTi26SahtHF0F+/tfVKlRfny5e0ORgkLC3MYBLzN8+fPrZadEymNU08ePnzoiSI55Og15svI+HfffRcCgQCMTg9RjhqKuvnxRucjRmazGsiJqohyVx6CMRinNQ0ePNgjZS1WoL1w4QKbdPrChQuF/iur6tWrBwAQKxzn1RRlG/tyGjRoYHO76fEnWcakE66mNwD3UoWFloHvHAVSqtGWnIEDB9p8fMSIESVcEu7Ml2szYfTGFqSbN2+WdHGc4qh1hy+BNiIiAr179wYA+N1+ityK+X3PaR1joIyuBGGWCtJ7xuvra6+9ZpUMqKQUq+l48ODBOHHiBIKDgzF48GC7fYBltY8WMAasPXv2QJSTDkZX+Cgm04jj+vXr29xuvrrL3VQhGlRx7aLSj58LoNIIrM5Vmjjqg3VmgAhxjYEDB2L16tUWj9WpUwcDBgzwUImKr7CpcNnZ2cjJyeFd/z9fAi0AvP/++9i7dy/0Oj1k98wGn71olZPHP2Jrs8OGDfNMIVHMQHvo0CE2oYEn+2izsrIwf/58HD16FDk5OahduzY+++wzNGvWzGNlMjHVDBnk11jtYaC3+JuCatSogfLlyyMzMxM3n4pdHmhvPjXO2xUIBGjUqJFLj80X1HRccmy1HphagPjGUa31+vXraNq0aQmVxjX4lKwmIiIC3bp1w/79+yF9YDlvmVFrIU00Xnv79u3rkfmzJsVqOq5SpQrbj+fJPtp58+bh2rVrWLlyJXbu3IkGDRpg9OjRUCgUjv/YzaKiothOd3G24yTWDMOgbt26NrcJBAJ2SPq1RyKrNWa5uvrIeL9Vr169MhtwqEZbcnx9fa2mAfL1c+eoH9ZW07K341OgBYA33ngDACDQWk6llD5Ig+DFElqmfTyF82AoT7p8+TJef/11xMbGIiIiAh999BGUSiXu3bvn6aJBLBaz03zEOY4DbfXq1QvtO+nQoQMA46owDzJsD4oSOXg3bW3PUjG4/UxkcY6yiK8Xej5iGMaqVsvXPnJHa5t6Yu1TrvgWaBs1amRzEKnPo0wAxiRAnl6+1bvXQnLgpZdewoEDB9C7d2+UK1cOW7duRaVKlVz2onLtX65cuTIuX74MsdJxoK1atWqh56tUqRJ8fX2Rm5uL47ckeLuFdfYmPx8DgmR6myv4BPnprFbuAYCTdyTQGxgwDINatWqV2j51R+nwnjx5UmqXCPRGBadYZGdn8/Kz9+TJk0K337lzx+uel6Pvgkql8royO1KrVi2recumrFE1a9b0+PPhdY12+vTpkMvlaNmyJRo0aIDvv/8eq1ev9prMRtWrG+fPCtWOMzOZ9rXH19cX7du3BwCcTZJYJJgw171urp3HrSdzq9TAkZvGC95LL73ksRF53oDrwu+kaAp+R/k2YMik4NSegryhG6uovD1ZhS1WKYBfTPmxuc0DeF2j3bBhA5KTk/Hjjz8iICAAO3bswJgxY7B9+3aXfFgKJo8oKqFQiO+++86pfTt06ODwfB9//DH++usvaDQabL/oi5HtrAN4myg1UpUCdtUYfx89etXPRZuaaqt99131ZddJ/fTTTzk/X2/mKEtPVFSUzeYn4h6hoaEWXTzR0dG8+/wZDAbk5hpvbO2tmCUSibzueTn6LsTGxvJuOcfExERs2rSJ/V2gzh8wGhsb65L34Pz588X+W7fcxiuVSsc7cZSbm4vly5dj+vTpaNWqFerWrYspU6ZALBZj7969bj+/M6pXr84O+tCLCs9G4kyi68qVK2PIkCEAgMvJEhy64YN0JWP1r2ZIfhPo4GZK1AvTWO1z7p4YR24ag/HLL79sd2oR8W5qtRpPnjzBkydPkJWVxZsaVMHxCHzMNW1aExWwv2KWJ9Y+5YpvSUMA6zEWjNnAKG8Yf8G5Rrtu3TpUqlQJvXr1AgBMnDgRu3fvRuXKlfH9999brUbjKlqtFhqNxmp5Pmfy+pYUsViMatWq4fbt29BIg+CjsJ2SrVKlSk4PBhk5ciQOHTqEpKQk7LgkxY5LhTeTr/m78A9Z+fLl8dlnnzl1buJdzp07h7Fjx1pdzHv37o05c+Z4qFTOKRhY+TgYSu/EWqimxD58wsdulIJ9/owuPwZ4wyL2nF/RX3/9lZ2fdPbsWRw4cACLFy9G/fr1sXDhQs4FtMff3x+NGzfG3LlzcfnyZSQlJWHJkiVITk5G69at3XbeojL1vTIMoKiav5KRTiyFRmYc3VejRg2nj+fr64uvv/7aJfk6GYbBl19+ieDgYM7HIiXvjz/+sFlj2rdvn1MrNnlSaajRmrO3FmppTGfqjQrmyDZfEN4b8mdzrtE+e/YMVatWBQAcOXIEPXr0QK9evVCrVi22mdNdli5digULFuCDDz6ASqVCzZo18e233zocWFSSTIvAC1WZyK3RDvKHxrSU6XVfQeDNOABF76yvV68etm7diqQk6yTaJqa+o8Lu5ipWrOi2FgfifgWTJeRUC4HsXioMBgNu3rzp1Vm+Cg5+4uNgKPO5wPZWzCrY4kbcw2oArE5vf5sHcA60UqmU7Rc6ffo03n77bQBgp6K4U8WKFbFkyRK3noMr002IUKMCozfL6MQYIMw1vm7FGRVXlhdtIMauk1u3blk8pqxVCdJ7qWAArw+0BS9+3nAxLCqRSMQGVHsrZnlDbaosKDj4NfD8PbvbPIFzoG3evDkWLFiAxo0b48aNG2jXrh0A4O7du6VyXdOiqly5Mvuz+TQfgSYXjEFvtQ8hzrh3755Vs7FBKIC2nBTiLO+fB2keWBmG8Yp+tKJiGAY+Pj6FVii88XmVxqUKg4ODbVbugoODvaK1hHOgnTp1KmbOnIkDBw5g9uzZbH/tsWPHvKqv1FPM13ZlNJaB1tY+hDjDXiDVBPpBnKXCjRs3SrhERWMeaH18fHjblymTyQoNtN5wkS+ofPnyqFy5ss31cv39/Xk56lgsFmPRokWIi4szrgMuFkMgEKBHjx6eLhoAFwTaihUr2pwrOn36dK6HLhXME1kLtfk1EPNA68lk14SfTIFU6yuCKDe/S0IT7AckpRpHums0EIvFnipiobyxplccfn5+SE9PL3S7NxoxYoTNkemtWrXyQGlco02bNmjTpo2ni2ET/8Zx84xMJmObYhiz4CrQ5v/Mp2WpClIqlSUyb5pYYgOt3LJvUx1onCaj1Wpx9+7dEi+Xs8xHzfO1Ngs4XozCG+Zw2jJw4ECbSSm8NVDxXbFrtJ07d7b4gnhyuTxvxjAMAgICkJKSAkaXP8Hd9LNUKuVln8j169exZ88ebNu2DQDQr18/tG3blu2jJ+5jGlUMANoAKZCSn6RCE5jfVHnjxg12YQtvYx5ovWXee3E4ukn25qlztWvXtupi4ON8Zj4odqAdN26cK8tRqvn7+yMlJQUCswXgTYvBe+sdryOTJk2yWAJs8+bN2LZtGw4ePOgVo/xKsydPnrA5drX+ljVag48YWpkEohy11ahkb2J+c8nnGm1YWFih203T+7yRrZHe3trUzXfFDrSvvvqqK8tRqrEfXl1+X5op6PLxg52Xl2dznU2tVov79+/bXcCeuMbt27fZn3Vys77OFzVDbXkZRDlq3Llzp6SL5jTzQMvnGq2jmRXePKPAVj851Wjdg3MfbZcuXWwuu5SVlYUuXbpwPXypYLpzZAxm82gNxtRsfBwUUtjSYI6WDSPcJSYmAgAMAoZdCgwAgo/cgN/NJ9AEyCz280bmg7T4XKM1zZM3p3+x8LNAIOBdoOXjfGY+4Bxok5OTLXJ+mqjVal4ueuwO7GAo89fpxc987J+19b4aXlwrKdC63/379wEAOh8R5PGP2MeFeRqUP3sXApWx///p06dQqazXLfYG5lmV+MxWFrrcyuUBGJPKePP321YaVwq07lHsT/vZs2fZny9evGjRL6fT6XDixAlKWPECe1Ex5Ada5kWN1lunXxTGVqDV+UogUtHNVUlITk4GAAjUthPW+z55zv786NEjr0yzaZ6akM9Nx5GRkRCJRNBq81urRArjNL6oqChPFcsptm4C+NjCxgfFDrRDhgwBwzBgGAZjx4612i6TyTBz5kxOhSstTIGWMb+gvPiZj3f2pvUs9SIBBFrjzYPeVwxQoC0RptdYoLNuSQIAYa7GYl9vD7R8bjoWi8WoXr26xcAzkcLYihAdHe2pYjnFVo3WFYuVEGvFvsofPXoUBoMBHTt2xPbt2xEUFMRuE4vFCAwM5PUXyJXyLypmNVoYAy0fl6QyBVqdjxiCF0k4dL7GmnlKSorHylVWFOU1Tk1NdWNJiq+01GgBoG7duhaB1nQD5O2Lp9u6yadA6x7FDrSmtIHenurNG7A3HDZqtHy8GWFrtL5iQGkMtHofkcU24h5arZad2lMYvYCBQG9AZmam+wtVDOafez5+B8zFxMRg586dVo/XrVvXA6VxXsFuK4ZheNnCxgcueVWzsrJw5coVpKamWt2d9u/f3xWn4LXCLiR8rNGaalR63/yPj/5Fjdb0GeD7xdNbObvOrEEsBPK0Xpu1qzR9PurUqWP1WHBwMEJDQz1QGucVDKp8HC/CF5wD7dGjR/HZZ58hOzvbau1FhmEo0MLeRYW/NVo20PrkfzFNgVaj0SAjI8OiK8EblJYVSwomsDct01ZwHVSD0HgDZ2theG/Ax8+9PdHR0RAIBBazL2JiYrz+ORa8XlNt1n04v7ILFixA165dMWHCBEqOb4fpC2fxteNpt5ROp0NaWhqAAoHW7OeUlBSvC7SFrVgSFhbGmxVLCk6lMxgMWLhwIbp27YoDBw5g4sSJxg0vPnM6ne2RycR1pFIpJk6ciN27d8NgMMDf3x8ffPCBp4vlUMHWNFqk3n04B9rk5GSsWrWKgmwh8u9sraMr35qOMzIy2Iu3ziy46goEWm/MsWtvxZIRI0Z4oDTFU/DzIpPJ0LVrVwBAt27d2JqtaQyAt36++D4AqqBBgwZh0KBBni5GkVCgLTmcv4UNGjSwmY6P5LM5GKrgNp4wH+xkai42/px/z+atI48HDhyIWrVqWTwmlUoxYMAAD5Wo6Ao2cefk5ODAgQMAgAMHDrB9uIzeYHN/b1HaAi0fFQy0fLsW8QnnGu2//vUvLFiwALm5uahTp45VOz8tam52p2h2cWG8vMZhj2WgNXuvBQLofEQQ5mm9euRxpUqVLKZi8G0ASMHc2AzDYOLEiVZ9tIxGa3N/b2ErmxwpWQUDKwVa9+EcaE3NbmPHjrV4o0wjT+Pj47meolDXrl3DN998g0uXLkEikaB169ZYtmyZW89ZVLYCrWlOLd+aa0zJEgyMZb8sAOhkEgjztF6dtKJgTYpvF3yxWJzfPIz851Pwd4HO+L+j9VI9hW+vOyFccA60P/30kyvKUSx37tzBu+++i2HDhmHatGkQCAReuWJJfi3fPAWj8ULItxqVKZexTiphB9yY6GQ+QEaOV+c7Nk+VZ+t3PggKCnJ6mo+3jp2gQOt51HxfcjgH2ubNm7uiHMWydOlS9OjRwyIFpDemmzMFU8ZgvqiA1mIbXzx6ZExir/OzziBjeszWyF5vUXAULh8DbYUKFfDw4UOn9/VG5oGWLviewffWHT5xycSpu3fvYtOmTUhKSsKcOXMQGhqKQ4cOoUqVKm5LQ6bT6XD8+HGMHDkSQ4YMwd27d1G7dm1MmjTJZTlGXdXsbcrkw5hf5F8EXaVS6fbmdVe6e/cugMIDbXJyMq5du+aV/c8KhcLid61Wi+vXr/Oqf6oo/a4KhcIrP1+mzxFgvOB7YxlLu4KDWLVaLb0PbsL5Snju3Dn0798fCQkJOHHiBDuh/s6dO/j22285F9Ce9PR0qFQq/PDDD+jduzfWrFmDihUr4r333nMqRV1JMuUPNV+PlnmxCDzfcouaBjpp/a1X+TA9ptFovDb1n615pXyba+rsAEM/Pz+vXcibak+eV/A9oPfEfTjXaJcsWYIPP/wQo0ePRqNGjdjHW7ZsiV9++YXr4e0yfSh69uyJN998EwAwe/ZstG/fHkeOHMErr7zC+Ry2UqsVx8WLFwHkB1cgf5m88PBwl53H3RQKBbKysgAAOrmNQGv2mI+Pj1c+L1vTXaKioni1DmfTpk2xZcsWh/vVqFHDK98DwDJjlUAg8NpylmambiAThmHofSjE+fPni/23nGu0N27cQK9evaweDw4ORnp6OtfD2xUYGAihUGix8LJYLEZ4eLjX9RGamvoYs4QVAq1xcW5vrXHYYlpwHLAMqibmwdd8X29iq0+Wb/20NWrUcOl+nsC317w0KthHy7eWHT7hHGh9fHys+r0A4N69e25NwyeRSFCnTh0kJSWxj2m1WiQnJyMsLMxt5y0OuVxu9Zgp6PIp0N67d4/9WVvOugZoEAmhlUms9vUmti4marXaAyUpvsjISKf6v7154XEKtJ5X8LtAgdZ9OAfaDh06YPXq1Rbt++np6Vi6dCk6d+7M9fCFGjZsGHbv3o1du3YhMTERc+fOhUAgQMeOHd163qIqbC5jQEBACZaEG1Pw1PmIrObQmpgCcGJiYkkVq0hKQ41WIpEgIiLC4X7eHGjNL+o06tgzCvbJUqB1H859tJ9//jmGDh2Kzp07Q61WY8yYMXj48CGqVq2Kjz/+2AVFtK9Pnz5IS0vD4sWLkZWVhYYNG2L9+vVelw2nsIT1fAq0ppGi2gCZ3X20ATLgyXOLUaXexFZQ1Wg0HigJNzExMQ5bDVw1+t4dzC/qfBrxXZoUDKx6vZ6WuHQTzoE2ODgY27Ztw969e3H16lXo9Xq8++676Nu3b4nkWR02bBiGDRvm9vNwUVgTemBgYAmWhBtTMhBNgP2BQ6Ztjx49gkql8rpBRqWhRgsAtWvXxp9//ml3e0hIiNetoGSOak+eZ2uUsU6no+Xy3MAlr6iPjw8GDBjAq+TsJSkgIMBqvUoTb74YmlOr1WySBIc12hcSExNRt25dt5etKOxdXPjG1upIOh8x9D5CiLNyvXL1JHN8vLkpbSjQlhzOr+ju3bttPs4wDHx8fFCtWjWrFVPKGqFQiIiICKumvpCQEN4MhkpKSmIDktZWjfZFP5v5trt373pdoLUVVPl40bcVSNM61kbwkRt2t3sTPt7clDa23gOaS+seLumjNa0YYo5dRYRhEBsbi9WrV/OqP9LVvvnmG8TFxbEJPXx8fNCpUyfe9IeY97lqXtRapfdT2ceCj9yAomE4lNGVoPMVQ5ir8cp+WlsXEj5eXIKCghASEoLU1Pz3gMnTQphnvGnw5v5ZgJ+veWljaxAa3QC5B+dAu3HjRsybNw+ffPIJm7Di4sWLWLp0KT755BP4+Phg+vTpWLx4MWbPns25wHxVq1YtXtfsTbVxvVgIva8YsltPIL+eP+FdmKdB+bN3YRAw0AZIIczVeOUUH1sXF76Oeo2KikJqaioMDAO9RAiBLj94eftnjS7onkc12pLDeXrPnDlzMHHiRLRr1w7+/v7w9/dHu3bt8Pnnn2PhwoVo1qwZJk+ejL///tsV5SUeYgqa2nJSgGEgv5Zscz/51YfQyqUWf+Pt+BpoTQto6HxFeNqvCURK43xgkUjk1PQfT6ILunfi63fB23EOtImJiQgNDbV6PDQ0lJ1LabrzJvxlyvSklUshyNNApMyzuZ9ImQfdi6QVDx8+pJqLG5kyP4lUGoABRFkqAMaEFt4+oIUCrXeiQOsenANtZGQkNmzYYPEGGQwGbNiwAZGRkQCMieiDg4O5nop4EDviWO4L6Ar/MpqyQ2k0GnYRAm/Gl37ygqpVq8b+LMrKtQi03o4u6KQs4XzbO3nyZIwZMwYnTpxAgwYNwDAMrly5grS0NKxcuRKAMR+yK5L8E88wX0xA6+94tSHzJfSSk5NRuXJlt5WtqIRCoVOP8UF4eDj7szA7FyKFcaCdtzcbE1LWcA60rVu3xv79+/Hf//6XTWjQu3dvDB48mF3O66233uJ6GuJB5qt86Gwsj1eQTpYfaAuuEOJptpKoeOO6uc4ICQmBRCKBWq2GSKGCQGXso/WmGxvivWwtt8jX1h1v55KOnIoVK+KTTz5xxaGIF3ry5An7s60F3wsyiITQS0QQqLVet5JS8+bNsXPnTvb3ChUqWNQM+YRhGFSsWBEPHjyAJDUbpktkpUqVPFouZ9SvXx9isRgajQZNmzb1dHHKpNjYWPj5+UGpVAIwTgkrLC87KT6XjZh49uwZHj16ZJU3tlmzZq46BfGQp0+fAgAMDKCTSiDIdZwbWCeTQKDWsn/rLWbOnIl3332X/ZxGRETA19dxLd1bhYSE4MGDBxBnKNnHbA1O9DbR0dGIi4tDWlqaxVKXpOQEBgZi//79SE5OBsMwqFatGtVo3YRzoH369CkmTJiAc+fOAYBVUur4+HiupyAeZhrQpPcVAwLnvog6mQTizByvGwzFMEypurCbUniajwLnS1rP8uXLF7rgBnE/Pz8/r09uUhpwDrTz58+HVqvF1q1b8fbbb2PlypVITU3FypUrMXXqVFeUkXiYKVia9706Yto3JSXFLWUiRraa+qj5z3M0Gg1SUlKg0WhoChPPCAQCiMVihIaGQiy2vQxocXEOtGfPnsW3336LunXrgmEYVK5cGa1atYJUKsWqVavQtm1bV5STeJApWOqkzq/GpJMaP6jeVqMtbQouCSkUCnndFM5nCoUCjx49gl6vh0Qi4e0gu7JKq9UiJycHCoUCYWFhkMvlLjs250CrVCoREhICwLhKTUZGBqpVq4aYmBhcu3aNcwGJ5+UHWufv8kxBOTMzExqNxuV3iMSo4DKEvr6+1M/mIRkZGQCM85hlMvsrXBHvlZOTgwcPHiAjI8OlgdYlCStMWYOioqKwc+dOqNVq7N69m1drrRL7TIFWX6BGa7qgmy4q5hd4830pK5j7FLyBoRsaz9HpdJBIJBRkeUwmk0Eikbg8ox3nQDtgwADcvn0bADB69Gjs3LkTsbGxWL58OUaOHMm5gMSz8vLy2GQVBZuODQYDFi5ciBMnTuCbb76xyPZjvi/107pPwWQb3p56kZCyiPO3cujQoezPTZs2xb59+3DlyhVERkZ6/ZqYxLG0tDT254KBViaToWvXrgCAbt26QSaTIScnx2pf82MQ16J+QEK8n8tvfytXrkyZaUoR89qovkAfbU5ODg4cOIBu3brhwIEDbJAFjFOBDAAYUI3WnQrmDKYcwoR4H2pnIoVKT09nf9b5WvfRTpw4ka3JMgyTf6EXMNC/WADe/BjEtQr2JdFqSd5HoVBApVKVyLmkUmmRB/E8ePAAy5cvx8mTJ5GVlYWwsDB06tQJI0aMwK+//ooVK1Zg0KBBVuuJ9+jRA/fu3cNPP/2EFi1asI/r9Xp07NgROp0OR48eddidMWTIEJw5c8bq8e+//x7t27eHRqPBt99+i927d+PZs2cICAhAvXr1MHHiRHapSG9XKgLthx9+iIMHD1q94YQ700hKAND7Wn5cTEHVVJMtWJvS+1CgdbeCmdi0Wq2HSkJsUSgUeLn3K1AqskrkfH7ycvhj7x6ng+2dO3fw1ltvoXHjxli6dCnCwsLw6NEjbN26FZs2bQJgTLH7559/Ytq0aWyu8H/++cfuzcPJkydRoUIFSCQSHD9+HJ06dXJYjuHDh2P48OEWjwUEBAAA/vOf/2Dv3r2YOXMmatasibS0NBw7dgwKhcKp5+gNeB9od+zYUWJ3i2VRZmYmAEAvFgJF7A/U+xg/Xs+fP3d1scgLBQNrwcBLPEulUkGpyEJKozegF7t3NLJAkwNc3AyVSuV0oJ01axYiIyOxatUqdtZAWFgYmjZtiqysLHa5U5VKhSNHjqB79+4AgF27dqFPnz744YcfrI65Y8cO9O7dGxKJBDt37nQq0MpkMrupQ48ePYphw4ahQ4cOAICqVasiNjbWqefnLXg9kuLp06dYtmwZ5syZ4+milFqmu0a9pOj3ZKa/yc7OdmmZSD61Wm3xOwVa76QXy6D38XfvvyIG8vT0dJw5cwbvvfeezbnX5hnG+vbti127dgEw3tzt27cPffv2tfobpVKJQ4cOoWfPnujRoweOHj3KueYZEhKCkydP8vo6wusa7dSpUzF69GiEhYW55fiUpzl/mTuDqOj3ZKa/SU1NpdfSTQou2qDX63H16lXerrHLZ3l5efD397doYcvNzS3xcuTm5jrVynf79m0YDAZUrlzZ7v4ajQY6nQ5dunTBkiVL8PTpU1y6dAkhISHsusd5eXns3+/evRvR0dFsDuv69etj165dGDBggN1y6HQ6fPfdd1i7dq3F43/88QfKly+PcePG4dNPP0WLFi1Qp04dNGvWDK+88gpq1KjhzMtRZDqdDtnZ2S69ZnEOtDk5OVizZg1OnTqFtLQ0q/yehw4d4noKmzZt2gStVos333zTLccnRqamSUMxppGY/oYG6LiPrT5ZrVZLgZa4VFBQEBo3boy4uDicO3cOvXr1srnfnj170LNnT/b3Hj16YM+ePRgwYAAeP35sEXCnTZuG3r17AwBef/11q3XLTTXqmJgY7Nq1CxcvXsSFCxfw119/YePGjViyZAnatWvn6qfqFpwD7ZdffokjR47g5ZdfRsWKFUsk/dujR4+wYsUKtrPeXerUqePW4/OBK7J7+fr60mvpJv7+/laP1axZ06Xp44hzEhMTAVimxfRE3mlfX1+r1Jy2REVFAQAeP36MRo0a2dxHLBZDKBRCKpXi1Vdfxfr165GYmIgpU6aw5/Dx8YFUKsXjx49x/vx5nD9/HgsXLmSPodfrkZqaioiICIu1oIODgyGVSiEUChEcHOww70L79u3Rvn17fPTRR3j//fexYcMGts/YlYRCIQICAqxW+Tp//nyxj8k50B49ehRLliwp0cUDrl+/jtTUVKsXediwYXj11Vcxd+7cEitLaWcaZcjoir4SielvTMcgrmerRkv9tMQZQUFBaN68OTZs2ICXX37ZqpJUsG+1a9eumDFjBho2bIiKFStaHW/nzp1o1KgRvvzyS4vHv/76a+zatQsffvghIiMjOZfbtNTlhQsXOB+rpHAOtGKxGFWqVHFFWZzWsmVLtmPepE+fPpgzZw6tFuRiptVhBOqiTxsRaIx/Q7Ur97HVLE/LsxFnzZgxA4MHD8bw4cMxcuRIRERE4NmzZ9i6datV4iGpVIpjx47ZnRe7c+dOvPXWW1br27788stYv349PvzwQ7vlyMnJsUpsI5fL4evri/Hjx6Nly5Z46aWXIJfLceHCBWzduhXvvfdeMZ91yeMcaAcPHoz//ve/Jbr2rL+/v83FiqtWrWrzTosUX3BwMABAkKcB9AanF34HAIHKWLOixSXcx1YmKAq03kegyXG8kwfOUatWLWzZsgXLly/HhAkT2CXiunTpgrfeegv//e9/Lfa3d9N8+fJl3L17F507d7ba1rlzZ3z55Ze4ePGi3SbqdevWYd26dRaPzZkzB6+//jpatmyJPXv2YNmyZcjNzUXlypUxcuRIjBo1qsjP11M4B9pHjx5h//79OHPmDOrWrWt1t/PVV19xPQXxINNdLWMAhDl50Pk72edkMECUbRxx6a5R4YR4O6lUCj95OeDi5hI5n5+8nFP9s+YiIyOxaNEim9vGjRtX6N8mJCTY/NlchQoVcOPGDbvH2LhxY6HnGDx4MAYPHlzoPt6Oc6BNSkpCTEwMAODhw4cW20pyXUx7bzLhplq1auzP4swcY6AVOnhfhQwEKjUEGp3VMYhr2fqO0Xq03kMul+OPvXu8OgUjcT/OgdbR3Qjht/DwcDaXsTg9G7lVg6D3EUPr5wORMs9qf62fD/Q+Yvg+yF+xh1Zxch9bq/fQ1B7vIpfLKfiVcbzODEXcTyAQoEGDBgAAn2f5+VoV9WwPgFPUr2qxb/ny5dmJ7cT1bC30Tou/E+JdXJIZ6tSpU9i3bx+Sk5Otphb89NNPrjgF8aBmzZrh9OnTkKQowGh1MIiEyKlVCaLsPMivJwMAdD5iKBqGIyfKOBjN53EmAOMaxdSU6T62pk5RoCXEu3Cu0e7YsQPvv/8+UlNTcfr0afj7+yM1NRXXr1+nmkwp0bp1awAAozewARQAVBHB7M9pHWOgjK4EABBm50L8XGXxt8Q9fHx8nHqMEOI5nAPt2rVrMXnyZKxcuRJisRiTJ0/G3r170bNnT1SqVMkVZSQeFhMTw06bkt63s+SdWa1V+qJ/lmEYtG/f3u3lK8sKZh7y9fWlFgRCvAznQPvgwQP2YiqRSNgFwIcNG4bffvuNcwGJ5zEMgy5dugAAfB+mA9rCcxdLk4yBtlGjRuw8XOIeMpnlii1FndpBCHE/zoFWLpezQ9crVKiAe/fuATBm+uDzskbEkilRuECrg29yht39RFkqSNKyLf6GuE+FChUsfqdWJEK8D+fBUC+99BLOnj2L6OhodO7cGfPmzcPVq1dx6NAhNGnSxBVlJF6gfv36CA8Px4MHDyBLTEFuZIjN/aSJxjRqIpEI3bp1K8kilkndunXDkydP8OTJE4hEIvTr18/TRSKEFMA50H7xxRfIyTGm/hozZgwUCgWOHDmCqKgoTJkyhXMBiXdgGAa9e/fGd999B2lyBipuPweDWTrGwL8TAKGAHQTVrl07dk1K4j6+vr4YOXKkp4tBCqFQKChhRRnHOdBWrVqV/dnHxwczZszgekjipXr37o01a9ZAr9dDlKO22CbOtkxe0adPn5IsGiFeSaFQ4JXeLyNLoSyR85WT+2HP3j+cCrZDhgzBmTNnABj7+mvWrImPPvoI7dq1K3SbyZ07d7B06VKcPn0aubm5qFGjBoYOHVroIu8AcPr0aQwdOhSAcZ6+XC5HzZo18fLLL+Ott95ip6c9fPiQHRtiLiQkBCdOnABgzKOcnJxstc++fftQs2ZNTJo0Cdu3bwdgjE+RkZF4//33S7zlxyXzaNVqNY4dO4akpCS88cYbkMvlePjwIcqVK8cu3kv4r2rVqli0aBFOnz7NPmaaN20+d7N69ero2LFjSRePEK+jUqmQpVBiYncFykndu9hDlkqAb+KM53S2Vjt8+HAMHz4c2dnZWL9+PcaMGYM9e/YUui0yMhIJCQkYPHgwOnbsiPXr16NcuXI4fvw4Zs+ejUePHmHs2LEOz/3333/DYDAgKysLZ8+excqVKxEXF4d169ZZzA///fffLVYSKpj5bMqUKVYL0QcFBbE/v/zyy5g6dSpyc3Oxc+dOfPHFF6hSpQqaNm3q1GvkCi5ZVOC9997D06dPoVar0b17d8jlcmzYsAF5eXmYPXu2K8pJvESnTp3QqVMnTxeDEF4pJ9UjQGq90pJrFT2Qy2QyhIaGIjQ0FDNmzMCuXbtw8uTJQrdFRkZizpw5qFWrFhYtWsROJxs8eDAEAgFmz56Nfv36ITw8vNBzh4aGAjAO6IuKikKbNm3Qu3dvbNmyxWIRgaCgIHZfW+RyeaHbfX192e1jx47F7t27ceTIkRINtJxHHc+bNw916tTBmTNnLCbKd+nSxaLmQwghxHuJRCKIRCKr7H4Ft6Wnp+PMmTMYOnSo1ZztgQMHwtfXFwcPHizy+SMiItC+fXscPny42M/BGT4+PjafoztxDrTnzp3Dhx9+aJUKrmrVqnj69CnXwxNCCHEzjUaDNWvWIDs726qmV3DbgwcPABi7iAoSi8UIDw9HUlJSscpRo0YNqz7XV155BY0aNWL/jR8/3mL7zJkzLba3adPG5rF1Oh127dqFhIQENGvWrFjlKy7OTce5ubk2c6ump6dTKjhCCPFiq1evxrp165CXlwd/f3/MmDEDdevWLXTbpUuXANhfjtFgyG8inzFjBnbv3g3AuC713r17Cy2PwWCwOu6aNWvYzHSAdVKW8ePHo3v37uzvBVe02r17N/bv3w+NRgORSIQxY8aga9euhZbD1TgH2saNG2PPnj1Wnd8//fRTibaBE0IIKZo333wTQ4YMYftjndlmymF/584ddi1yE41Gg4cPH+LVV18FAHz00UcYMWIEAGPzsyN3795FlSqWK4OFhYVZzG4pKDg4GJGRkXa3d+7cGRMmTGD7am0tLelunAPtJ598giFDhiAxMRE6nQ6rV69GQkIC7t69i02bNrmijIQQQtwgICDAbpCyty0oKAjNmjXDxo0b0atXL4sa6NatW5Gbm8vWGIODg51Ow5qUlITjx4+7PP+Cn59foYG4JHAO7fXr18fvv/8OiUSCiIgIXLhwAdWrV8fmzZtRq1YtV5SREEKIF5k2bRoSEhIwYcIExMfH4+HDh/jvf/+L+fPnY8yYMQ5HHANASkoKnj17htu3b2PTpk145513EBsbi4EDB1rsl56ejpSUFIt/5hQKhdV2tdpynr+nuWQebVRUFObNm+eKQxFCSKmTpRKgONNvin6OkhETE4PNmzdj6dKlGDp0KJuwYvr06VaB0p62bdtCIBDA398fNWvWxIgRIzB48GCrgbWvv/661d9eu3aNbYqeO3cu5s6da7F9/fr1XrVEJ2Mw77nmkVWrVuHPP/9EUlISypUrh+7du+OTTz6Bn58f52OfP38eAChXMyHEaYmJiQAsR+N6c2YoYput9xHgFheKXaOtX7++U/tdvXq1uKco1MWLF/H++++jXr16SEtLw/Tp06FUKqlmTQjxGnK5HHv2/kG5jsu4YgdarVaLKlWqYMCAAQgLC3NlmZyyZs0a9ucaNWrgo48+wsyZM0u8HIQQUhi5XE7Br4wrdqBdsWIFNm/ejFWrVqFVq1YYNGgQOnXqZJWHsqRkZGS4/MMcHx/v0uMRQkov03zTkqq9EvfQ6XTIzs526fW/2L3nXbt2xZo1a3DgwAE0aNAAX331FTp06IB///vfbOaQkqJQKLBu3TqnO+EJIYSQkuKywVB6vR5HjhzBDz/8gEuXLuHUqVMlsnKPWq3GqFGjwDAMvv/+e6cmRTtCg6EIIUVlbxAN4RevGgxV0Llz57B3715cvXoVjRo1KpH0i1qtFp988gmUSiV+/PFHlwRZQgghxJU4Rab09HRs27YNv//+OzIzM9GvXz9s374dNWvWdFX57NLr9fjiiy9w//59bNy40SXTegghhBBXK3agHT9+PP766y+89NJLGDt2LHr06GE10didpk+fjtOnT+P777+HRqNhs4UEBQV5bEAWIYQQUlCxA21cXBzCwsIgFouxfft2bN++3eZ+69atK3bhCrNlyxYAQP/+/S0eP3ToUKEJqAkhhJCSVOxA279/f7vLJJWEhIQEj52bEEKcpVAoKGFFGVfsQDt//nxXloMQQkodhUKBl3v3hlKhKJHz+cnl+GPvXqeC7ZAhQ3DmzBkAgEwmQ82aNfHRRx+hXbt2hW4zuXPnDpYuXYrTp0+zuY6HDh2KAQMGFHre06dPY+jQoahWrRr2799vsW3atGn4/fffMXbsWIwbNw4AULt2bfz0009o0aJFkZ6DN6FhuoQQ4iYqlQpKhQLPXm4Ina97x7AIc9Wo8MdlqFQqp2u1w4cPx/Dhw5GdnY3169djzJgx2LNnT6HbIiMjkZCQgMGDB6Njx45Yv349ypUrh+PHj2P27Nl49OiR1frktqhUKly+fBkNGzYEYJyqGRcXZ7HIO5fn4Oml8cyV/Aq4hBBSxuh8JdDL3PuvOIHctKh79erVMWPGDAiFQpw8edLhtjlz5qBWrVpYtGgR6tWrh/DwcAwePBiTJk3CypUrnUpa9Morr2DXrl3s73/99RciIyOLHCALK6e3oEBLCCEEIpEIIpEIGo2m0G3p6ek4c+YMhg4dajVOZ+DAgfD19cXBgwcdnq9v377Yt28ftFotAGDXrl3o27ev256DJ1GgJYSQMk6j0WDNmjXIzs5G06ZNC91mqq3ayoAlFosRHh6OpKQkh+eMiYlBcHAwTpw4gefPn+PkyZPo3bu3W56Dp1EfLSGElFGrV6/GunXr2AURZsyYgbp16xa67dKlSwBgd9aJeVbfGTNmYPfu3QCAsLAw7N2712Lfvn37YteuXXj8+DGaNWuGoKAglz4Hb0GBlhBCyqg333wTQ4YMYfs5ndkWEREBwDjqOCYmxuJvNBoNHj58iFdffRUA8NFHH2HEiBEAYDNFbt++fbFy5UokJiZi+PDhLn8O3oICLSGElFEBAQF2Bx/Z2xYUFIRmzZph48aN6NWrl0XNduvWrcjNzUXXrl0BAMHBwQgODrZ7/ooVK6JBgwa4cuUKunTp4vLn4C0o0BJCiJsJc9Wl4hwm06ZNw1tvvYUJEybg/fffh1wux7Fjx/DNN99gzJgxCA8Pd/pYq1evhlarhVQqtbvP/fv3LVaDEwgEqF27NqfnUJIo0BJCiJtIpVL4yeWo8MflEjmfn1xeaMBylZiYGGzevBlLly7F0KFD2YQV06dPL/K64M6Ud9q0aRa/y+VynDt3rkjn8SSXrUdbmtB6tISQorK3jimlYOQXr16PlhBCiDW5XE7Br4yjebSEEEKIG1GgJYQQQtyIAi0hhBDiRhRoCSHEBYRCIdRqNXJycjxdFFJMOTk5UKvVEAqFLj0uDYYihBAXCAwMhEqlQlJSEiQSCQQCqsfwiV6vh1qthkAgQGBgoEuPTYGWEEJcQC6Xo0aNGkhJSYFGo4Fer/d0kUgRiEQiSKVShIaGQiwWu/bYLj0aIYSUYWKxGGFhYZ4uBvEy1LZBCCGEuBHvA+3q1avRtm1bxMbGYsyYMUhLS/N0kQghhBAWrwPt1q1b8d1332HmzJnYtGkTFAoFPv30U08XixBCCGHxOtD+/PPPeO+999CtWzfUqVMHc+fOxalTp3Dz5k1PF40QQggBwOPBUGq1Gjdu3MDkyZPZx8LDw1GlShX8888/iI6O5nwOUxJpQgghpLh4W6PNyMiAXq+3WlQ4KCgI6enpHioVIYQQYom3NVp3ouXxCCGEuApva7SBgYEQCARWo4zT09MRFBTkoVIRQgghlngbaCUSCWJiYnD69Gn2sQcPHiA5ORmxsbEeLBkhhBCSj9dNx2+//Tbmzp2LOnXqoGrVqpg7dy5atGjhkoFQhBBCiCswBoPB4OlCcLF69Wps3LgRCoUCrVu3xldffYWQkBBPF4sQQggBUAoCLSGEEOLNeNtHSwghhPABBVpCCCHEjSjQEkIIIW5EgZYQQghxIwq0hBBCiBtRoCWEEELciAItIYQQ4kYUaAkhhBA3okBLCCGEuBEFWkIIIcSNKNASQgghbkSBlsdWr16Ntm3bIjY2FmPGjLFam9ckLy8PEydORM+ePRETE4Ply5eXcElLN2ffh/j4eIwfPx5t27ZFo0aN8MYbb+DkyZMlXNrSydn34OHDh6hdu7bFv6ZNm5ZwaUufuLg4vPvuu2jSpAlq167tcP/ExEQMGTIEDRs2ROfOnbFt27YSKKXnUKDlqa1bt+K7777DzJkzsWnTJigUCnz66ac299XpdJDJZBg5ciRiYmJKuKSlW1Heh+vXr6NKlSpYtmwZduzYgbZt2+KDDz7AnTt3SrjUpUtR3gOT33//HX///Tf+/vtv7N+/v4RKWnqpVCq0bNkSo0aNcrivRqPB6NGjERwcjC1btuBf//oXZsyYgTNnzpRAST3EQHipf//+hmXLlrG/379/3xAdHW1ISEgo9O/eeecdw3/+8x93F6/MKO77YNKrVy/Dhg0b3FW8MqEo78GDBw8M0dHRhgcPHpRkEcuMU6dOGaKjowvd5+DBg4aGDRsalEol+9jnn39uGDdunLuL5zFUo+UhtVqNGzduoGXLluxj4eHhqFKlCv755x8Plqxs4fo+GAwGZGZmoly5cu4sZqlW3PfgnXfeQdu2bTFq1CjcvHmzJIpKXrh8+TIaNmwImUzGPtaqVatSfe0SeboApOgyMjKg1+sRHBxs8XhQUBDS09M9VKqyh+v78PPPP0On06FTp07uKmKpV9T3QCaTYdq0aWjUqBHUajU2bNiAd955B/v27UNISEhJFbtMS09Pt/l+2etXLw0o0BLiAceOHcPixYuxYsUKBAQEeLo4ZUZQUBCGDBnC/h4bG4tevXphx44deP/99z1YsrLDYDB4uggljpqOeSgwMBACgcDqDjA9PR1BQUEeKlXZU9z34dy5c/joo4/w9ddfo23btu4uZqnG9bsgFApRu3ZtJCcnu6uIpIDg4GCb71fBWm5pQoGWhyQSCWJiYnD69Gn2sQcPHiA5ORmxsbEeLFnZUpz34fLlyxg9ejQmTZqE3r17l1RRSy2u3wWDwYA7d+4gLCzMncUkZho2bIjLly9DpVKxj506dapUX7so0PLU22+/jfXr1+PgwYO4ceMGpk6dihYtWiA6OhqXL19Gz5498fTpU3b/27dvIz4+HkqlEikpKYiPj8ejR488+AxKh6K8DwkJCXj//fcxaNAgdO7cGSkpKUhJSYFCofDws+C3orwHf/zxB/bs2YPExEQkJCRg2rRpePLkCfr06ePhZ8FvmZmZiI+Px/379wEY54zHx8dDrVZbvQft2rVDhQoVMHXqVNy6dQtbtmzB3r178c4773jyKbgV9dHy1GuvvYa0tDR8+eWXUCgUaN26Nb766isAxjltiYmJ0Gg07P6jRo1im8euXbuG3377Da+++irmz5/vkfKXFkV5H+Li4vD8+XOsXbsWa9euZY9B7wM3RXkPGIbBypUrkZycDF9fX9SrVw8bNmxApUqVPPkUeO/w4cOYPHky+3v//v0BAIcOHbJ6DyQSCVavXo2ZM2diwIABCA0NxezZs9G8eXNPFL1EMIay2DNNCCGElBBqOiaEEELciAItIYQQ4kYUaAkhhBA3okBLCCGEuBEFWkIIIcSNKNASQgghbkSBlhBCCHEjCrSEEEKIG1GgJYQQQtyIAi0hhBDiRhRoCSGEEDeiQEsIIYS4EQVaQoiVvLw81K1bF6tXr8aQIUPQvHlztGvXDhs3bvR00QjhHQq0hBArCQkJ0Ol02LhxI959911s2bIFffr0wddff40bN254uniE8AoFWkKIlWvXrgEAvv32W3Tt2hURERH47LPPIJVKceHCBQ+XjhB+oUBLCLESHx+P5s2bIzY2ln1MKBRCKBR6sFSE8BMFWkKIlWvXrqFu3boWjyUmJkKhUKB+/foeKhUh/ESBlhBiQavV4tatW9Dr9RaPr1u3DjExMWjQoIGHSkYIP4k8XQBCiHe5ffs28vLycOjQIXTo0AERERH4/fffsXPnTvz8889gGMbTRSSEVyjQEkIsxMfHIzQ0FDNmzMDMmTPx7NkzNGzYEJs3b0ZMTIyni0cI71CgJYRYuHbtGurUqYOOHTuiY8eOni4OIbxHfbSEEAvx8fFUcyXEhSjQEkJYBoMBN27cQJ06dTxdFEJKDcZgMBg8XQhCCCGktKIaLSGEEOJGFGgJIYQQN6JASwghhLgRBVpCCCHEjSjQEkIIIW5EgZYQQghxIwq0hBBCiBtRoCWEEELciAItIYQQ4kYUaAkhhBA3okBLCCGEuBEFWkIIIcSNKNASQgghbkSBlhBCCHEjCrSEEEKIG/0fvzz+SJxwo7MAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 156,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "\n",
    "ax = sns.violinplot(data=cmean, x=\"p\", y=\"inv\", hue=\"Algorithm\", dodge=1.3, hue_order=alg_order, cut=0)\n",
    "plt.ylabel(\"Mean agent's investment\")\n",
    "plt.xlabel(\"$p$\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(figdir + \"/invest_violinplot.pdf\")\n",
    "print(figdir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.1_CMA-ES v.s. 0.1_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=5.114e-01 U_stat=1.640e+02\n",
      "0.1_PPO-DEEP v.s. 0.1_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=1.468e-06 U_stat=2.880e+02\n",
      "0.1_CMA-ES v.s. 0.1_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=2.065e-08 U_stat=5.600e+02\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.2_PPO-DEEP v.s. 0.2_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=1.468e-06 U_stat=2.880e+02\n",
      "0.2_CMA-ES v.s. 0.2_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=7.257e-03 U_stat=6.400e+01\n",
      "0.2_CMA-ES v.s. 0.2_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=6.714e-04 U_stat=4.530e+02\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.5_CMA-ES v.s. 0.5_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=1.082e-01 U_stat=9.600e+01\n",
      "0.5_PPO-DEEP v.s. 0.5_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=1.736e-06 U_stat=2.870e+02\n",
      "0.5_CMA-ES v.s. 0.5_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=7.750e-04 U_stat=4.510e+02\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "1.0_CMA-ES v.s. 1.0_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=7.251e-05 U_stat=2.600e+01\n",
      "1.0_PPO-DEEP v.s. 1.0_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=6.376e-06 U_stat=2.790e+02\n",
      "1.0_CMA-ES v.s. 1.0_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=5.149e-01 U_stat=2.560e+02\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x158.4 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 156,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from statannot import add_stat_annotation\n",
    "\n",
    "plt.figure(figsize=(figwidth, figwidth/1.5))\n",
    "\n",
    "ax = boxplot(data=cmean, x=\"p\", y=\"inv\", hue=\"Algorithm\", dodge=1.8, width=0.4, hue_order=alg_order)\n",
    "plt.ylabel(\"Mean agents' investment\")\n",
    "plt.xlabel(\"$p$\")\n",
    "plt.ylim(0, 16)\n",
    "ax.get_legend().remove()\n",
    "for p in [0.1, 0.2, 0.5, 1.0]:\n",
    "    add_stat_annotation(ax, data=cmean, x=\"p\", y=\"inv\", hue=\"Algorithm\", hue_order=[\"CMA-ES\", \"PPO-DEEP\", \"PPO-MLP\"], verbose=2, test=\"Mann-Whitney\",\n",
    "                        box_pairs=[((p, \"PPO-MLP\"), (p, \"CMA-ES\")), ((p, \"PPO-DEEP\"), (p, \"CMA-ES\")), ((p, \"PPO-MLP\"), (p, \"PPO-DEEP\"))], comparisons_correction=None, text_format=\"star\")\n",
    "#plt.text(0, 0, \"must do bonferri?\")\n",
    "plt.tight_layout()\n",
    "#plt.savefig(figdir + \"/invest_boxplot_no_label.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.1_CMA-ES v.s. 0.1_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.418e+10\n",
      "0.1_PPO-DEEP v.s. 0.1_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=2.771e+09\n",
      "0.1_CMA-ES v.s. 0.1_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.643e+10\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.2_PPO-DEEP v.s. 0.2_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=2.938e+09\n",
      "0.2_CMA-ES v.s. 0.2_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.072e+10\n",
      "0.2_CMA-ES v.s. 0.2_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.688e+10\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "0.5_CMA-ES v.s. 0.5_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.063e+10\n",
      "0.5_PPO-DEEP v.s. 0.5_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=2.758e+09\n",
      "0.5_CMA-ES v.s. 0.5_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.471e+10\n",
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "1.0_CMA-ES v.s. 1.0_PPO-DEEP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=2.024e+10\n",
      "1.0_PPO-DEEP v.s. 1.0_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.041e+10\n",
      "1.0_CMA-ES v.s. 1.0_PPO-MLP: Mann-Whitney-Wilcoxon test two-sided, P_val=0.000e+00 U_stat=1.766e+10\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 237.6x237.6 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 235,
       "width": 237
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(figwidth, figwidth))\n",
    "boxplot(data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, dodge=1.3, width=0.5)\n",
    "plt.ylabel(\"Accepted partner's investment\")\n",
    "plt.ylim(0, 15.1)\n",
    "plt.xlabel(\"$p$\")\n",
    "#plt.legend(title=\"Algorithm\", ncol=2)\n",
    "plt.legend(loc=\"lower left\")\n",
    "#add_stat_annotation(ax, data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=alg_order, verbose=2, test=\"Mann-Whitney\",\n",
    "#                    box_pairs=[((p, \"PPO\"), (p, \"CMA-ES\")) for p in [0.1, 0.2, 0.5, 1.0]], comparisons_correction=None, text_format=\"star\")\n",
    "for p in [0.1, 0.2, 0.5, 1.0]:\n",
    "    add_stat_annotation(ax, data=allppo, x=\"p\", y=\"other\", hue=\"Algorithm\", hue_order=[\"CMA-ES\", \"PPO-DEEP\", \"PPO-MLP\"], verbose=2, test=\"Mann-Whitney\",\n",
    "                        box_pairs=[((p, \"PPO-MLP\"), (p, \"CMA-ES\")), ((p, \"PPO-DEEP\"), (p, \"CMA-ES\")), ((p, \"PPO-MLP\"), (p, \"PPO-DEEP\"))], comparisons_correction=None, text_format=\"star\")\n",
    "\n",
    "plt.tight_layout()\n",
    "#plt.savefig(figdir + \"/accept_boxplot.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>inv</th>\n",
       "      <th>other</th>\n",
       "      <th>accept</th>\n",
       "      <th>run_id</th>\n",
       "      <th>trial_id</th>\n",
       "      <th>good_site_prob</th>\n",
       "      <th>checkpoint_path</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Algorithm</th>\n",
       "      <th>p</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">CMA-ES</th>\n",
       "      <th>0.1</th>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "      <td>332431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "      <td>314054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.5</th>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "      <td>278167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "      <td>307332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">PPO-DEEP</th>\n",
       "      <th>0.1</th>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "      <td>65093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "      <td>58761</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.5</th>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "      <td>59599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "      <td>158031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">PPO-MLP</th>\n",
       "      <th>0.1</th>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "      <td>59770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.2</th>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "      <td>72800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.5</th>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "      <td>76662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "      <td>100381</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Unnamed: 0     inv   other  accept  run_id  trial_id  \\\n",
       "Algorithm p                                                           \n",
       "CMA-ES    0.1      332431  332431  332431  332431  332431    332431   \n",
       "          0.2      314054  314054  314054  314054  314054    314054   \n",
       "          0.5      278167  278167  278167  278167  278167    278167   \n",
       "          1.0      307332  307332  307332  307332  307332    307332   \n",
       "PPO-DEEP  0.1       65093   65093   65093   65093   65093     65093   \n",
       "          0.2       58761   58761   58761   58761   58761     58761   \n",
       "          0.5       59599   59599   59599   59599   59599     59599   \n",
       "          1.0      158031  158031  158031  158031  158031    158031   \n",
       "PPO-MLP   0.1       59770   59770   59770   59770   59770     59770   \n",
       "          0.2       72800   72800   72800   72800   72800     72800   \n",
       "          0.5       76662   76662   76662   76662   76662     76662   \n",
       "          1.0      100381  100381  100381  100381  100381    100381   \n",
       "\n",
       "               good_site_prob  checkpoint_path  \n",
       "Algorithm p                                     \n",
       "CMA-ES    0.1          332431           332431  \n",
       "          0.2          314054           314054  \n",
       "          0.5          278167           278167  \n",
       "          1.0          307332           307332  \n",
       "PPO-DEEP  0.1           65093            65093  \n",
       "          0.2           58761            58761  \n",
       "          0.5           59599            59599  \n",
       "          1.0          158031           158031  \n",
       "PPO-MLP   0.1           59770            59770  \n",
       "          0.2           72800            72800  \n",
       "          0.5           76662            76662  \n",
       "          1.0          100381           100381  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.query(\"accept == 1\").groupby([\"Algorithm\", \"p\"]).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>inv</th>\n",
       "      <th>other</th>\n",
       "      <th>accept</th>\n",
       "      <th>run_id</th>\n",
       "      <th>trial_id</th>\n",
       "      <th>good_site_prob</th>\n",
       "      <th>checkpoint_path</th>\n",
       "      <th>Algorithm</th>\n",
       "      <th>p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>5.211583</td>\n",
       "      <td>12.5</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>5.210622</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5.211958</td>\n",
       "      <td>4.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>5.207606</td>\n",
       "      <td>2.5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5.212956</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3c19e_00010</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-MLP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092647</th>\n",
       "      <td>3786</td>\n",
       "      <td>9.332860</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092648</th>\n",
       "      <td>3787</td>\n",
       "      <td>9.333066</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092649</th>\n",
       "      <td>3788</td>\n",
       "      <td>9.333034</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092650</th>\n",
       "      <td>3789</td>\n",
       "      <td>9.332210</td>\n",
       "      <td>13.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1092651</th>\n",
       "      <td>3790</td>\n",
       "      <td>9.333282</td>\n",
       "      <td>6.5</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>c3315_00009</td>\n",
       "      <td>0.1</td>\n",
       "      <td>/Users/paulecoffet/Documents/isir/These/data/R...</td>\n",
       "      <td>PPO-DEEP</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4346828 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Unnamed: 0       inv  other  accept  run_id     trial_id  \\\n",
       "0                 0  5.211583   12.5       0      10  3c19e_00010   \n",
       "1                 1  5.210622    3.5       1      10  3c19e_00010   \n",
       "2                 2  5.211958    4.5       1      10  3c19e_00010   \n",
       "3                 3  5.207606    2.5       1      10  3c19e_00010   \n",
       "4                 4  5.212956   10.0       0      10  3c19e_00010   \n",
       "...             ...       ...    ...     ...     ...          ...   \n",
       "1092647        3786  9.332860   12.0       1       9  c3315_00009   \n",
       "1092648        3787  9.333066    7.0       1       9  c3315_00009   \n",
       "1092649        3788  9.333034    2.0       0       9  c3315_00009   \n",
       "1092650        3789  9.332210   13.0       1       9  c3315_00009   \n",
       "1092651        3790  9.333282    6.5       1       9  c3315_00009   \n",
       "\n",
       "         good_site_prob                                    checkpoint_path  \\\n",
       "0                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "2                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "3                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "4                   0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "...                 ...                                                ...   \n",
       "1092647             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092648             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092649             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092650             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "1092651             0.1  /Users/paulecoffet/Documents/isir/These/data/R...   \n",
       "\n",
       "        Algorithm    p  \n",
       "0         PPO-MLP  0.1  \n",
       "1         PPO-MLP  0.1  \n",
       "2         PPO-MLP  0.1  \n",
       "3         PPO-MLP  0.1  \n",
       "4         PPO-MLP  0.1  \n",
       "...           ...  ...  \n",
       "1092647  PPO-DEEP  0.1  \n",
       "1092648  PPO-DEEP  0.1  \n",
       "1092649  PPO-DEEP  0.1  \n",
       "1092650  PPO-DEEP  0.1  \n",
       "1092651  PPO-DEEP  0.1  \n",
       "\n",
       "[4346828 rows x 10 columns]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}