Behavior Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#some imports...\n",
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from scipy import stats\n",
    "\n",
    "import seaborn as sns\n",
    "#sns.set_style(\"whitegrid\")\n",
    "#sns.set(style=\"ticks\") \n",
    "import os\n",
    "import glob\n",
    "from conversions import get_file_without_path\n",
    "from pingouin import ttest\n",
    "\n",
    "#pandas and pandas options \n",
    "import pandas as pd\n",
    "pd.set_option('display.max_rows', 40)\n",
    "pd.set_option('display.max_columns', 40)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data_df = pd.read_csv(\"data/behavior_df.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Normalising between pre and post answers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Plot normalising\n",
    "df_analysis = all_data_df.reset_index()\n",
    "\n",
    "#select questions\n",
    "q1_df = df_analysis.loc[df_analysis[\"question\"] == \"Q1\"]\n",
    "q2_df = df_analysis.loc[df_analysis[\"question\"] == \"Q2\"]\n",
    "\n",
    "#set index\n",
    "q1_df = q1_df.set_index([\"participant_nb\", \"interacting_partner\",\"question_content\", \"participant_condition\", \"other_condition\", \"congruence\", \"sex\", \"day\"])[[\"measure\"]]\n",
    "q2_df = q2_df.set_index([\"participant_nb\", \"interacting_partner\",\"question_content\", \"participant_condition\", \"other_condition\", \"congruence\", \"sex\", \"day\"])[[\"measure\"]]\n",
    "\n",
    "#substract questions\n",
    "norm_df = q2_df - q1_df\n",
    "norm_df = norm_df.reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Only post answers (Q2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Q2 answers\n",
    "df_analysis = q2_df.groupby([\"participant_nb\",\"participant_condition\", \"other_condition\",\"question_content\"]).mean().reset_index()\n",
    "\n",
    "for question in np.unique(norm_df.question_content.values):\n",
    "    df_aux = df_analysis.copy()\n",
    "    df_aux = df_aux.loc[df_aux[\"question_content\"] == question]\n",
    "    g = sns.catplot( x=\"participant_condition\", y=\"measure\", units=\"participant_nb\"\n",
    "                    , height=5.6\n",
    "                    , aspect=0.7\n",
    "                    , data=df_aux\n",
    "                    , kind=\"point\"\n",
    "                    , capsize=.16\n",
    "                    , hue=\"other_condition\"\n",
    "                    , legend=False\n",
    "                    , scale=1.7\n",
    "                    , dodge= 0.15\n",
    "                    , errwidth=2\n",
    "                    , palette= sns.color_palette([\"red\", \"blue\"])\n",
    "                    \n",
    "               )\n",
    "    \n",
    "    # Add individual measure points\n",
    "    sns.stripplot(x=\"participant_condition\", y=\"measure\"\n",
    "                  , data=df_aux, hue=\"other_condition\", dodge=0.15, size=5, alpha=0.3, palette=[\"red\", \"blue\"]\n",
    "                  , legend=False\n",
    "                 )\n",
    "    \n",
    "    g.set(ylim=(0, 7))\n",
    "    plt.savefig(\"plots/q2_\"+question+\"_legend.pdf\", bbox_inches='tight')\n",
    "    g.set(xticklabels=[], yticklabels=[], title='',xlabel=None, ylabel=None)\n",
    "    plt.savefig(\"plots/q2_\"+question+\".pdf\", bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Post - pre (Q2- Q1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAU8AAAH/CAYAAADXB1+HAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAqxElEQVR4nO3dWWwc2X3v8V9vbO5NUpTERa19HWmsGY3GM9aSGHbsXNiOE9gIrv2Qh5kAhoHAL4MAtp6MIA+yH4zk5eYlmCRIgCQIYjvbGLDheBvNYo2k2RftC0WRIkWK3Vy72ct9+E+pu0lKah72ru8HKJBd3U2eqj79q1Onqk75stlsVgCAVfFXuwAAUI8ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcBBsNgXJhIJJRKJe4+z2aySyaR6e3vl8/nKUjgAqFVFtzxPnjypSCRyb+rq6tKGDRs0PT1dzvIBQE3yFXt55tKWZzweVzQaVSwWU2dnZ9kKCAC1qOjd9nA4rHA4XM6yAEDd4IARADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHwWJfmEgklEgk7j2Ox+NlKRAA1IOiW54nT55UJBK5N0Wj0XKWCwBqmi+bzWaLeeFKLc9oNKpYLKbOzs6yFRAAalHRu+3hcFjhcLicZQGAusEBIwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AZTczIw0MSGlUsufi8elyUkpk6l8uUopWO0CAGgcCwvS2bMWjpIUDEr79klbt0qzs9KZMxaektTUJD3+uDQwULXirgnhCaBk3nwzF5yStTzffVfq6JDeecdapJ5kUjp3TopEpLa2ypd1rYoOz0QioUQice9x3Nt8AICkuTnpzp2Vn3v33cLg9GSz0tCQtHdvectWDkX3eZ48eVKRSOTeFI1Gy1kuAHVmcfH+z83P3/+5ZLL0ZakEXzabzRbzwpVantFoVLFYTJ2dnWUrIID6kM1KP/+59XsutWePdOmSlE4vf+7pp6W+vvKXr9SK3m0Ph8MKh8PlLAuAOubzSfv3Wz9mfpOsu1vasUMKhaT33it8z8aNNtUjDhgBKJmBAam9Xbpxw3bHe3ulwUEpEJC2bZO6uqyPM5Wy0BwYsNCtR0Xvti8Vj8cViUTYbQfwSOIkeQBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4CBY7QKURTYrjY1Jw8OSzycNDkrr10uZjDQ6KiWT9ri9vdolLZm5OVvkQEDq65NCoWqX6BE2PS3duSOFw/Zh+B+dNko8Lk1MrLzo2ax0+7Y0NCQFg9KWLVJPT/XKulaNF54LC9KpU9K5c1aJJSkSkfbssSDNZnOv3b5d2r+/OuUsoYsXpfPnc4sWDEqHD9v2ARX2zjvS9eu5x83N0jPPSJ2d1StThbz9tnTjRu5xc7P07LNSR4d9LX/zG+nsWdvQS/a1PHLEVk8gUJ0yr0XjbRLffVe6cCEXnJIUi0m/+IV0+XLha69csU1hHYvFpI8+KtwmpFK27chkqleuR9KtW4XBKVlqvPlmdcpTQcPDhcEpFS76e+9JH3yQC07J6u6770qXLlWunKVUdHgmEgnF4/GCqeak0xaGk5OF8xcWbF9i6XzJKnwdGx5eeX4yKY2PV7Ysj7z71aV4XJqZqWxZKux+ix6LWTtmZGTlr9/ERP1+BYsOz5MnTyoSidybotFoOcu1NvnNsPzHS+ffbx7g4hGuSw9a9IetlnpdbUWH54kTJxSLxe5NQ0ND5SyXm0BA2rBheS90c7PNW6l3emCgMmUrk/7+leeHQlJvb2XL8si734fR0dFQBydX8qBF7+y0g0ddXcuf7+6u369g0QeMwuGwwuFwOctSGgcOSHfv2r7C7KzN6+yUDh2yIyn5Nm+2T7WOdXdLu3bZQSOP3y898UR9dsLXtcHB3FkenlDIPowGt2mTLXr+LnhTU27R9++3bqRz56REwua1t0uPPSbt3Fnx4paEL5t1azTH43FFIhHFYjF11tqRxEzGPsWhIUuSzZuljRutT/TWrdypSittCuvU9LR19wYCtiWvh+1cw7p7N3eq0sDA8o12A5uczJ2qtHTRMxnbrly/bvO3b7evoc9XvfKuRWOGJwCUWeOdqgQAFUB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOAgWO0ClMXEhHTrluTzSQMDUk9P7rnFRenGDWlkRLpyRQoGpa1bpcOHpaamqhW51BYXpZs3pXhcam+XotHCxctmpdu3pbExaXpa8vulzk5pcFDq6qpasetLJiMND0uTk1JLi7R5s9TcvPJrx8akq1ftp89nr9u40d4TiVS23Gs0NSWdO2dfs54em/x+qbdX6u+332Mx6bXXpIsXpY4O6ZlnpJ07pVCo2qUvHV82m826vDEejysSiSgWi6mzs7PU5XL3/vsWivn27JF275ZmZ6VXXpGGhqRf/tISJhi0ZNm4UfrGN6S2tuqUu4Tm520x5+dz88Jh6ehRW7xsVjp71rYf167lvs9bt0rr10v79llFxwOkUtKrr1pKeIJBS4n8jbUkvfmm9N57liQjI7ZF27DB6t2ePdInPiFt21bZ8ju6elX6t3+TZmbs63TrloXjpz9tG9+eHgvQf/gH6fJlW02SveYP/kD63Oek1tZqLkHpNNZuezy+PDgl6cIFaW5O+vBDKZGQzpyx4JTs052YsNbD//5vZctbJufPFwanZIv94Yf2++3b9h2embHglCxQb9yQ0mnpo4+khYXKlrnuXLlSGJyS1aV33y2cNz5uG+tr1yxt4nGbf+eONeHGx6UPPpCSyUqUek1SKekXv7B6k81a3clmbZG8xZ6clP75ny1UveCUbO/mjTesbjWKosMzkUgoHo8XTDXHS4KlvE96bMwq6eRk4fOzs/ZzpeCtQ7dvP3i+t5qmpgqfT6etknurCw9wvxUUjxduecbGbEuWTNoG3JPJ2ONYzH4fHy9veUvAa2NItjhe+0OS7t61xU4k7GuUv6ie0dH71816VHR4njx5UpFI5N4UjUbLWS43wQd04QaDUiBgHTL+JYvtPW6QPs/7rQZvvvczEFj+Gm9eI/VNlcX9VrLPV7hiA4Hc45Xqnffcg+pujfC+QtL9FyUQsLrj8638/jpYzKIVHZ4nTpxQLBa7Nw0NDZWzXG4GBlZOhFBI6uuzznmvjzOf12d78GD5y1gB99uuefM3bbLKvW5dYSUPh61vqqnJuuTwAPdbyX19hVueaNQODnV02OSt8GDQOv96e+359evLX+Y16umxr5Bki+j1Xfp8dqAxFLLFOnRo5WNg27bdf7XVo6LDMxwOq7Ozs2CqOU1Ny4+ah8PS00/bp7pnj/Vmf/KTucra0WEp8uST0pEj1Sl3ie3caQGZr79f2rvXfu/stGMUra322kDAVtPu3fY9fvrplbdByDM4aCsvvwnW02MrNl9bm9WtPXssUfr6bGUPDlqS9PXZCl/alKtBPp/0+c/bYvt8VvTWVgvFfftsEfbulf7kT6QnnrCvlve+7dulz37W6lijaLyj7ZL1IU1M2KfmnUeRb3bWer2np62jZnBQ6u6uTlnLaG7OFrG9feWTCFIp68Py+62fc6XWKB5iYcH6LVtacnswK0mnrU7OzFh4ZrP2np6eulvh2az1X05M2EY5ELC+zu7uwnbL8LBNkYhtzBvgRJYCjRmeAFBmtb+vAAA1iPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBw0ECD4gPwHD5sY24Wa2zMhhwNBFZ/F4G+Prun4qOG8AQa0OioDUS8Wt6t6PFwhCfQgPr6Vvf6kRELTr/fRocv5/9qFIQn0IBWuxu9aZO1OPv7pZs3y1OmRsMBIwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhN4RM3NSS++KH3xi9Lt2zbv9m17/OKL9jzuz5fNZrMub4zH44pEIorFYurs7Cx1uQCUydiY9L3vSX//99LU1P1f19UlPfec9J3vrP52xI8CwhN4hJw9K33pS6u7p3tfn/TSS9KhQ+UrVz1itx14RJw9K33606sLTsle/7u/K507V5Zi1S1uPZzv8OHV16yxMSmdlgKB1e3b9PWt/v6wgKOxMWtxzsy4vX9mxvpC336bXXgP4ZlvdNRuXu0ik3F/L1Bm3/ve6tsFS42OSt//vvSDH5SmTPWOPs98Li3PkRELTr9f6u8v/n20PFEhc3PS4OCDDw4Vq6vL2gitrWv/W/WOlmc+lzDbtMlqU3+/dPNm6csErNG//EtpglOyv/Ov/yo9/3xp/l4944AR0KAyGen996W//uvS/t0f/rC0f69e0fIEGkQiYUfUT52y6ZVXpMnJ0v+foaHS/816RHgCdWpqSnrtNenlly0sT5+2AC031yP2jYbwBOrE0FCuVXnqlPTuu5Lb4d61aW+v/P+sRYQnUIMyGemDD3JB+fLL0o0b1S6ViUarXYLaQHgCNSCRkN54o7C/slRHyEvtq1+tdglqA+EJVMHdu9Krr+bC8o03yttf6fOVZhe/u1v62tfW/ncaAeEJVMCNG7nd71OnpPfeK+//27VLOn5cOnbMpr/5m9KcsvTcc5wg72nsK4wuX5b+8z+la9ds07t7t12HPjIixWK2Gd2yRbpzx050X1yUnnxSOnpU2rFDamt7+P/wTpIfHKzaSfKplHT1qn1Br1yRpqdt/uiolExKHR32ZVq3zhZ9dtZaIeGwTa2tdsFTNCpt3WqX6WMV7t61D+DWLWlmRpmsNHQ9q4/OSxfOZzX6/oS641eVkV8T6lGr5rRZQ0opqFd0RP+tP9Skep3/fSBgIx55QXn0qLRxY+FrxsakgwfXdolmX1/u2vaLF+2akrt37fqQZ5+1bobf/tbqYTgsPfGEtHev9d8ODdnAIhcvWv3r65M+8xnpk5+0q5bqUeOG54cf2gW9s7PS+Lg0P2+fbnu7hWRLi11SOT6eS5CWFgvZY8ek3/kdq4UdHQ/+P1UOz0zG+sdGR60yX79up5LcuGGh2tFhi5dKST099jget22IZIscjdoXYv9++3nkiK0GFOH2bSVfeUNXT49p/LdXNXX5juZHpjSdalaTkmrVtLZoSLNqk08Z9WtEzUpoVH2aU5sy8usVHdH/07c0ooGi/mVbm/SpT1k1PX7cAqiYI+DnztnoSC6nGrW3S7/+tYX02bPST35iG2bPnTtSU5O1Sebnc+XcssXC8cIF6Z137LzTpib7e+vWSV/+sg1YUo+DjTTubvu//7u0sGA1JZOxC3zn5uzT7emxT3h21ubNz9trwmEL1LNnrQV68eLKgxjOzdk1bz/60fIhuL/yFenrX6/Yvs2tW7ZNGBmx1kU2az9jMQvGWMwWS7Lfo1Erfipl81Ipq/idnfa+YNAWpa+vIsWvS5OT1l/58svS7Esf6tZHGe1P31RYae3QXYW0qAFNKqEmbdU1BZRWSgGt06TaNSO/supUTPNqll/SE3pTh3Va/60/lLR8q7VxY+Eu+MGD9jmt1qFDFoBf/KL7eJ6zs3Y+aX5wJpPSpUtSKGSh6G14Jyft9b299lWambH6mUhYnZyYsO6L7dsbPDwTiYQSeT3a8Xi8LAUqiVQqdxlEImHNrMVF+5lMWlCm0xau6bR9oqmU/e73W7pMTS2/PONBQ3CnUrY5/slPpD//84oNwe0VcXo6t8X3theZjBXL77cK7W0vvOL6fIWrx3tucpLw9GSz1prPP7/y/fftuaAW9X80rbASCmlRzVqQT1k1KakmJRVQSmEllFJQIS0qpKQCSisrv8JKKqi0FhVQs5LaohsKK6GEmrVnTy4ojx2zHqRS7QkcOmS73t//vvR3f/fwkeSff1769rdz1XhqKtct5Jmft6/VwoJtsL1gX1y0+uX3W91aXMy9J5WysB0ft8kb1bGeFB2eJ0+e1F/8xV+UsyylEwjY/mksZp/k4mIuQbyfPp895wWmN0k2v6XFJs9qhuCempL+6q+sdVrmIbi9IjY15RY1FLJ5fr/N8xY5ELDXJZOF87yf3vuam8tW3JqXTltrKD8s79cbk1JQiwopoJSy8in18dcpJb/SCmhRQaVliZBWQBkF5A0nYY/98klq6wjoyKc69NlvhHTkePlbYRs22LByf/mXNsjHD38o/exnFmjBoPT5z9vpSF/72vIdqOZmq0P5gkGrP4FA7isk5epfS4vVLb/fNujec5I919pa+L56UXR4njhxQi+88MK9x/F4XNFaPVvW57Pe6H/6J+t48TaJCwv22O+3/YZg0PZZm5vteW/Tt2WLFIlI27bZY28I7tV2FnlDcHudRWUQjdou08aN1lIYH5fWr7fWgN+fq7jZrO2ad3baYnjz2tpsOxMM2pcqFLJu3EfF/LydJuQdBX/1VesTLo5P17RVu3RRd9Sr9RrXnFokZXVXIUk+jWmDejWhhJo1qW6FlFJrcFGt3e16bGPQ1v3uQT37509LByrb9GpttZbl88/nuu43brTt/f2sW2ct4cnJ3KlPra3WE9bebi1NLyDb2qwu9vbaep6bs7rn91sAh8PSvn3Szp312cdedHiGw2GFvc6zevCFL9j+xc9+Zs2JpiY75NzSYp16yaQF5P799rpk0j757dulP/oj6cABaWCg5ofgbm6WnnnGLtVbXLRO+UjEKu3oqFXstjY7sWDbNgvX6Wmr4KGQrY6uLntuYEB6/PHlLYtGMjFhB9i8VuWZM4W7k6t1XnvkV0aZj/sqUwooqLQy8suvtIZbd+vJdde1tXVckb6N6mweUGD+49Md/H6rk3/6p9Jjj5VoCcvvc5+zsHvrLfvKdHRI3/ymBeqbb1o/vN9vX6EnnrDXbtki/eIXduKLtyE/ftxaubt3V3mBHDXu0XbP4qKlSE9Pbv80kch1Bnot0dFR23SGQpZI3n7ECy/YLvhavfBC2YfgXliwRVxctAra1GSnkrS2WgX3uqy9I+3eIgYCFqaNtruezdqXNf8Sxw8/LM//8iutsBLavjuk3zmS0rNHAzp+JK2tOwLyZTO5TuV16+xDGh21pld3t9vRnxJzOWkkkbDWpLfnIlkdnJuzRQuFchtir+55z7e12VRv/Zz5Gj8814IhuOtKOm0tcG8X/NQpawWVSzAoPfVU7kj4kSPW4q9HNXC6ct2p/iavlpV4CO7Dj81pNFV8eLreW056NO7yMTdnp814Qfnqq8uPBJdSR4cFpHcU/JOfXOO2cLW3faFC1BTC80F+9KOS/rnR6wtyuUUc95Yzd+4U9leePbu2/sqH6e/PtSqPH7f+4JLuZrrecJAKURMIzwcp8RhgfXpAK8M7ryNvGplqUSbjW/W95aT6P08zm7UrHvN3wT/6qLz/c9++wpPRt24t81Hg1X5IrjcbdPlfeCjC80FKPGT2GT19/yczkpIfTx/bpCENa5P6dUs3t/1fOzK7a5ed2+H9bJCRaVMpO1Mg//zKkZHy/b9QyPaavaA8csROqamo1e5Gc7PBmkJ4PkitBFMmk0uUpfr7c2GaH647dxY3sEmVeJf5eUfBX3utvLd36Oxc3l+Zfw0EsFqE54Ns3lz+scPWamTEppdfXv7cwEBhS9Wbduyo+FH/sbHC/spz53LX15fD4GDhLviBA/V9WgxqD+H5IF/5il2rXq9u3bLp179e/tzg4PLWqhesa2ySZbM2GmD+Lvj582v6kw+1f3/h9eBbttTnVSuoH4Tng3z96zbIRylOV2pttcswrl+3IWYuXbJzbapleNimX/1q+XPR6MpdATt2rHgmfSplF1DlD/brDTZVDqGQ9PTThf2V69aV7/8BKyE8H6S11UZHKsUVRt/8ZuEVRtmstQovXbIwzZ8uXbJLMaplaMimX/6ycL7PJ0Wjmt3+uF5v+6xOLT6jU7d36bUL6zQ7X76RHSIRC0hvN/zwYforUX2E58N85zt2svxah+D+9rcL5/l8tus8OGiDh+TLZCxYn+yV7sgOXP3eH1mwXr5c8WC9rQ16RUd1KntMp24c07kbh5QuY9XZtKmwv3L/fvorUXsIz4fZsMGGmVnLENwvvbS6K0L8fksQbxyWSJf04x/b794J0ktbql6wrvEuYllJl7RTp3RML+u4TumYLqq8IzccOLC8vxKodYRnMUoxBHeRHj5IvV+t0aj1S37mM4VvTqft/L/8QPWmK1cKh//+2KKCektP6JSO3ZvGtHHZ60qlSQk9rTd0XC/rmE7pU/7T6pnrlK7ukgI7pTt5/axbt+YGGQVqDAODrMbYmPsQ3EX86fsNUr/0TzsNUp9OS0NDmnnnil7/+YxOvRHWqUsb9frELs1my3c+aJfu6qheuRfNh3VGzSqydRwIWICudFbA1q01MRpRRZVx9A4GBlk9wtPF3Nzqh+B+gNUMUu8ptlE7OmrnV3pHwd96KzckXTls1vW8Nuwp7df78qsENwxfKhi0QUhXOitgy5bGDFbCs6YQnmu1xlrnOki9VHhHQ8kO4F+4UHh+5aVLq/+7xfIpowN6T8d0Ssf1so7qFW3WUPn+YbFCIQvWlS4Q2Ly5fo8+raKurXbAprVeNv8oDtjUgJvn+lGKQeo//3npW9/KnWc5Pl7aMuYLh+2yRu/AzqcOp9Udb5UubpMupqRL63N9rNeulbeJ+yCLi7YVuXBh+XOhkN0tYGk3wK5d1o9ca8HqeKdWBmwqP1qea7WGlmepBqkvl+5uu3W9F5ZPPbWK0eYXFy1AVzor4Nq13I1uaklTU2Gw5gdsNFrZu5StsROcoULLj/BcK8fwLOUg9aWyZUtu7Mpjx2yItrLkRTJp480tPSPg0iW7AqsWgzUctiusVuoKGBws7YoqZyc4Sobd9iop5SD1Lnw+G9zXC8qjR61xVRFNTdKePTYtlUhYsC5trV68aOOrum3r1y6RkD74wKalmptzwbo0XAcGVhesNXynVhQiPKukxIPUP1Q4bHfZvNdf+Snb46s54bC0d69NSy0sFAZrfrgODVUvWBcWpPfft2mplpbCYM0P2IGBwtFLavxOrShEeFZJiQepX6anJ9dfefy4NUbq6c7RK2putr6EffuWP7ewYFdYrdQVMFTFMwDm521Yw5WGNmxtLRzY+ty5tV0GLNn7v//9st+pFfR5rp1jn+e2bXbcpFSCQTvw6rUs9+6t7PGNmjY/b8G6UldAIx5a5k6tFUHLs0pKPUj9vn3SP/5jaf9mw2hpsQvoDxxY/tzcXGGw5odrOe9bXE5TU3YRx/PPV7skDY3wrJJSD1JfsYM9jaa11Y6cPf748udmZy1IV+oKKOcNlkrhhz8kPMuM8KySUg9S/9Wvlu5v4WNtbdLBgzYtNTOTC9Wl4VrOkaCLVc1+3kcEfZ5rVQPneXZ327+mi6tGxOP37woYG6tMGbZts5G0UDa0PPOt9rIMKbf7NjJiQVqk1r4+PffcmZJcYfTccwRnTenslJ580qalYrGVuwJOny7t5ay1cufXBkbLM5/XiqyEwUGNnbupgwfXPkg9p/U1gC9+sbT9OF/4gl1xhLKh5Zmvr2/173G9KLivryqD1KNG0Qled2h51oBz5yoySD1qGZ3gdYfTqGvAoUO26/3CCw+/ZLKry1739tsEZ0Px7tRaCnSCVwQtzxpT4kHqUU/GxkQneP0gPGsYt0Z4BJ07t7ZOcEZVqhh224Fa4t2pdbUHL/v6CM4KIzyBWkMneF0gPIFatGGDDSs3PCy9+KKdt+ndETQYtMcvvmjP/+AH9HFWAX2eNYw+TxSgQtQUWp4A4IDwBAAHhCcAOODa9gpa7aBNjgM2SXp076UNVArhWUGjo26DNmUyjXmrHaCeEZ4VtNrznl0HbHL5XwBWh/CsIHajgcbBASMAcEB4AoADwhMAHBCeAOCA8AQAB4QnADjgVCWgWrjkrK4RnkC1cMlZXWus8IzFpPfes1G1b92yy3P6+qSWFrtrWnu7ND8vNTVJN25I4+NSc7PU3y/t2ydls9L0tNTWJu3caeMm1oFEQvrxj6Wf/tQaJT6ftG6dXZW0ebPdiTYetyEgFxYkv1+6c8dWxcaN0pEj0pe/LHV0VHtJ6szcnPTGG3anvitXrL4NDFi9Wb/e6lUmYx+Iz2cr/soV6a23pMVFq2/r11t9LEaNXHJ27Zp09arVn+5uG5v5zTfta3f3ri1ad7cVsbdX2rbNBryvo69UURpnMOTZWek3v5Heeccq2fCwVc5EwlJh0yb7vbU1lyLZrD0XiUiTk9JTT0lbtuT+5hNPSNFo1RapGJmM9Ld/a8E5Pm7f5zt37Lu6Z49V4tFRq8TBoK2aGzdsNUQiNnV32x7kn/1ZbrByPMTiooXm//yPreA7d2xeImFBdfCgPW5ttS1YNiudPi1dvy719Njf8Pmsvv3xH9sHUAcuXZI+/DD3+NYt6fXX7es0O2vBmkrZ16qlxQJ0507p2WdtcZ98cvU9DrWq6ANGiURC8Xi8YKopV65YAM7M2ObPC86JCfs5OmrJMTMj3b5tKSNZS3NszB5fu2YV3nPxYlUWZTVGRqwra2bGFnlmxgI1nbaKfeWKVeahIWt9xmK2OmZn7TXT01IyaS2JDz6o9tLUkaEhW7l37+budLmwYCs3FrPnR0ftQ7l1yzbYw8P2mmTSXp/NWv08fdo+pBqXyUiXLxc+vnrVvnYTE/bTW4y7d+0rlUjYYk9N2fw6+EoVrejwPHnypCKRyL0pWmstspkZ24+Q7BOTcp9kOm1p4bUMFhcLK+vsbO593nu9+W4N84qJxaySLi5aZc5frIWF3Pxk0hbN+956AZvJ5LYzt29XZxnq0syMrfxstrCeeT+np21+JmMfhPchSYUfUjJZWHdrWDKZqz+SLc7cnC3O4uLy59Jpey6ZzC2eyx2Va1XR4XnixAnFYrF709DQUDnLtXrefoIkhcP2Mxi0fYVAwPoxQyF7LhQq3D9ta8u9r7m5cL7PV5nyO+rqyi2a31+4WM3N1p3m99uihcO57jW/31ZLIGCLGA5b/yeK5HX3+Hy5lR4I5H52dNh8v9/qpfchSYUfUlOT9cV7dbeGNTUVds+GQtYrEQzmvlr5zwUC9lxTU27xGqlfvegernA4rHD+2qk127bZrlJHh/Upzc3Zp7lunf3s68v1efb1WWsgkyns89y2rbBi795dveUpUl+f9PTTuQNA7e22aD6fdc53ddneY1+fLdrCgjWY2tpy3/GmJmn7dumxx6q9NHVk0yZpxw7rAEwkbMU2N9uK7+qyvvJUyh7399t7hoetz9NLIJ/POqOffbYuOpv9fuu/9Lp3/H77yoyN5fo8YzFb7J4eC8xwOFcPJWnXrqoVv+Qa54CRZJ/c++/bQaOREfsU84+2t7XZpxwKWR/U2Fju092718LUO9q+Y0fdHBpMJKT/+A87frH0aHs0ar/H43agaGHBQnN83MK2r086elT60pcseLEKc3PS2bO24i9ftvrT328rvbc3d7Q9m8018a9ezR1t37lT+v3ft4NLdeTGDevu9Y62NzVJ587Z125y0hatp8dOJOjtlbZuzR1tHxiodulLp7HCEwAqhMszAcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAg0cjPDMZaWxMWljIzVtclObmqlemMllYKFxMyRZ/dlZKJu1nJlOdsj3Skkmrg4uLuQ8kna52qbAGwWoXoOxef1361a+ssgYC0r590rZtVpGzWamtTXrsMamvr9olXZPpaentt6W7d+1xT4908KA0OipdvChduSJNTEjr10vbt0u7dtlPlFkmI/30p9KZMxagMzP2IRw4IDU12Yewe3e1SwkHjR2eH3wgvfRS7nE6Lf3sZ9KGDdKRIzZvdlY6e1Y6flzq7KxOOdconbZtRH6Lc3JS+q//sm3DrVsWopL9HgjY9zgclgYHq1PmR8avfy29+qr9PjVlG23vQ3jsMen8eSkYZEtWh4rebU8kEorH4wVTzfvtbwsfLy7alv/6dUsPTyZj8+rU6OjyXXXJFunuXfu+5rt9235evVr+sj3SMhnbMHumpnK/X7yY+/3atUqVCCVUdHiePHlSkUjk3hSNRstZrtKYmSl8nErZz3TagjTfSulTJ+bnV56fTEqJRG6xPd6i1/Ei14dUqvDDyf8gEolc5zMfRF0qOjxPnDihWCx2bxoaGipnuUpj06bCx+Gw5Pfbvmxra+FzPT2VK1eJ3a/oHR3WE9HWVji/vf3B70OJNDVZF5GnpSX3e3e31UWJD6JOFR2e4XBYnZ2dBVPN+/SnC0PS75d6e6VPfELy+XLz29ulzZsrXrxS6emR+vuXz3/qKamry7Yh3uL6fPY4FOI4RUX83u9Zn6ZkH5TPZ48PHrR5gYC0Z0/1ygdnvmw2m3V5YzweVyQSUSwWq+0gnZqSfvMbaXjYmmHPPmtNMq/fs7fXOutDoWqXdE2yWenGDTsWIdmBoGjUjodduSKNjNiq6O6WBgZskZc2vlEmw8PSK6/Y6Q4dHfYBhMO20d6+3eah7jR+eAJAGTwaJ8kDQIkRngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDwBwAHhCQAOCE8AcEB4AoADwhMAHBCeAOCA8AQAB4QnADggPAHAAeEJAA4ITwBwQHgCgAPCEwAcEJ4A4IDwBAAHhCcAOCA8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4IDyBUpmclEZHpWTSHs/M2OOZmeqWq4qWrhKpcVZLsNoFAOre7Kx0+nRhGmQykj+vbdLfLx06VDivgc3N2SqZnrbHfr+0a5cUj0sjI7nXDQxITz5Zn6ulDosM1JizZwuD8+ZN6cwZSwrPyIh0/nzly1Yl587lglOybcnPfy599FHh627dki5cqGzZSqXolmcikVAikbj3OBaLSZLi+RUEeNRMTyuY35SS5B8eli+ZVObmTWW3br03P3v+vNKDgxUuYOXNzkrDw8ujZXjYr3g8q1AoWzD//PmsBgbSlSpeUTo6OuTz+R78omyRvvvd72YlMTExMTX8FIvFHpqJvmw2m1URlrY8M5mMJicntW7duocnNO6Jx+OKRqMaGhpSZ2dntYuDBkZdc1dMy7Po3fZwOKxwOFwwr6ury6lgkDo7O6nQqAjqWnlwwAgAHBCeAOCA8KywcDis7373u8u6QIBSo66VV9EHjAAAObQ8AcAB4QkADghPAHBAeAKAA8ITABwQngDggPAEAAeEJwA4+P9XPlMz+LB65QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUMAAAH/CAYAAADNO78JAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAkgElEQVR4nO3dS2xc133H8e/wNaQeHJJ6URJpS7Ily5JjI47jJm7zblIkaZs23iQoiiJpU3TTTVEg8SqrQsmiLdp1kwLdpECbAingFEiCBAkaJ3Uc+SE5ki1blkRRlCiJ4gwpksMhZ7r4ezQkRVJDcp7U9wNckHNnRJ1z58zvnnvOvXcShUKhgCTd51rqXQBJagSGoSRhGEoSYBhKEmAYShJgGEoSYBhKEmAYShIAbeW+MJvNks1m7zwuFArMzs6yc+dOEolEVQonSbVSds/wxIkTpFKpO0tPTw+7d+9mYmKimuWTpJpIlHs53tKeYSaTYXBwkHQ6TXd3d9UKKEm1UPZhcjKZJJlMVrMsklQ3TqBIEoahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAGGoSQBhqEkAYahJAHQVu4Ls9ks2Wz2zuNMJlOVAklSPZTdMzxx4gSpVOrOMjg4WM1ySVJNJQqFQqGcFy7XMxwcHCSdTtPd3V21AkpSLZR9mJxMJkkmk9UsiyTVjRMokoRhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJwP0WhnNzcPMmTEzUuySSGkxbvQtQMxcuwJkzEYgAfX3wvvdBZ2ddiyWpMdwfPcObN+HUqVIQAoyNwcsv169MkhpK2T3DbDZLNpu98ziTyVSlQFVx6dLy62/cgKkp2LKltuWR1HDK7hmeOHGCVCp1ZxkcHKxmuSorl1vfc5LuG4lCoVAo54XL9QwHBwdJp9N0d3dXrYAV8c47cPr03es7O+ETn4CW+2O0QNLKyj5MTiaTJJPJapaleh54AIaH4dat0rpEAo4fNwglAffLbHJrKzzzDFy+HOOEHR0RkI3eo5VUM/dHGEL0AB94IBZJWsJjREnCMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCQA2updAFVXLgcjI1AowJ490NlZ7xJJjckw3MSuXoWTJ2F+Ph4nEnD8OBw8WN9ySY3Iw+RNam4OXn65FIQQvcPTp2Fysn7lkhpV2T3DbDZLNpu98ziTyVSlQKqM0dEIxOVcuQJHjtS2PFKjK7tneOLECVKp1J1lcHCwmuXSBhUK63tOul8lCoXyPhrL9QwHBwdJp9N0d3dXrYBan1wOfvjDxYfJRR/5CPiWSYuVfZicTCZJJpPVLIsqqL0dnngCXnkF8vnS+qNHDUJpOc4mb2L798OOHXFqTT4P/f2wdWu9SyU1JsNwk+vs9FQaqRyeWiNJGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEGIaSBBiGkgQYhpIEQFu9C1AVhQJcvQrXr0N7OwwOwrZti19z7Rr85Cfw+uvx3Ic+BO97HyST9SlzBWQy8NprUbWdO+GJJ2ITnD4NExOwYwc8+CDs2wd9ffUubRMbH4fhYcjnYft2mJmBbDY2+t690LL5+hizs3DpEty+DakUDAxA27vpMTEBly9DLge7d8OePZBI1Le865EoFAqF9fzDTCZDKpUinU7T3d1d6XKtX6EAv/pVJEJRIgHvfS/s3x+PX34Z/uVf4De/gbm5WJdKwe/9HvzZn0Ej1adMFy/C974X4Qel/UFnZ1Tx1q3YDI88Ak89BYcPw9Gj9S1zUzp/PnagEKF47lyE4MGDsW7HDvjABzZVIE5MwAsvRCAWbd0KzzwDN2/Gx2lhivT3RxtrtkDcPO9Y0cjI4iCEeKdOnYo9+cQE/OhH8PbbpSAESKcjRF96qbblrYBcDn7xi1IQQlTzjTfgwoVosBCb4c034/G5c7GX1xpks3DmTPxeKMTGLRRiw09MxPqbN2FoqG5FrIbf/GZxEEK0nbNn42O1tDt19Wp8DJtN2WGYzWbJZDKLloa0NAiLcjkYG4vnr1yJQ5ulxsbgrbdgfr66Zaywmzej6AsVH2cyi6uaz0f1AUZHa1O+TePGjdiAANPTixMinS79vlIbbELFrF/O22/Hx2o5zdi2yg7DEydOkEql7iyDg4PVLNf6ta0yDNreHs+3tS3fh29tjTHDJuvft7bGsnQdxNHa0iO29vb4udqm0jIWbuSlG3Xh4020YROJu9tW0WrD6824CcoOw+eee450On1nGWrUQ4GVQrq7O8YF9+2Dhx66e0IlkYjB78cea7rxnp07Yzh0YYbv2hUNcscO6Ooqre/ogAceiOf27q19WZva7t2lBOjsLLWhRCI2dFGjdhTWaaXqPPpozB+t5d80srI/9clkku7u7kVLQ+rpgfe8Z/HubNu2GNGFSIM//uOYPS6+ky0tcOAA/MEfxBRsk0kk4MMfhuPHS9XeuhU+/elYt29frO/shA9+EHp74f3vb869d121tMDTT5f2Lg89FDvZQ4ciJFtaIiF27apvOSvs0UdjhrgokYgd6qFD8bFa2K9obYXHH49+R7PZfLPJRcUxwvb25c8jyedjCvb8+Xg3jx5tzndwgeL4zvXr8XnctSvmiN5+O55LpWJf0NfXdJ3fxlIoRNuan48e4cRETK709sYG3qQmJ2PipLt78dEGxOaYm4u21aw72c0bhpK0BvYPJAnDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAw1CSAMNQkgDDUJIAaKt3ARrCU0/B1avlv350FObnobUVdu8u/9/198NLL629fJKqzjCECMLh4bX/u3x+ff9OUsMxDCF6bGsxMhJB2NICe/dW7/+RVDOGIaz90HVgIHqEe/fC5cvVKZOkmnICRZLYjD3DbDZ6ej/+MbzyCrS1weOPwx/+IXR1wbVrMDMTrxsZideOj8OOHfCxj8GnPgXbt9e7FmUpFGBoCC5ejI7q/Hys37IlOq8HD8bR/M9/HlXt6Ij5nt5e6O6GQ4eapqr1NT8PP/tZLLdvw0MPwWc/G23m/Hk4eRJeew1u3IArV+I1XV1w+DB8/OOxoQ8ejA3fBMbG4J13YHo65gpv3IC5OThwAAYHo6rj4zHq8+EPw65dcOpUbIbbt+N1v/M70caaSaJQKBTW8w8zmQypVIp0Ok13o9Q6m4X//E/46U/hF78opcP27bF8+tPxDr7xRiznz8e7l0jE8z098NGPwl//NWzduvL/UzxM3r+/rofJr7wSQXjmDKTTcOlSBN7AQFRz61Y4exba26NhX74cw5xPPx2f59ZWeOaZqLZW8e1vw09+ArOzpXW7d8OTT8bOtbi3uXgxdrSJRLwRXV3w4IPwp38abeXpp9d29kEdjIzAr38dO9rXXouQa2mBBx6IkLxwAd773qgeQGcnPPJIvHZhkvT0wF/+ZXPtbDfXYfL583DuXKRDMQghdmO3bsELL0Qi3L4de/DR0eg6FQrRiG/fjpB84YW6VaFcExPRK7x1K4o9Ph5776mpeHztGrz6arymUICbN+Pn/Hxsnlwufn/jjXrXpMEVe34LgxDgzTfjqOL0achkIimmp2Oj5vOxgXO5aGO//GWsO3OmPnVYgzNnop1ks9E2im3m5s3I+uJOtSiTgf/4j8VBCNEef/nLmhZ9w8o+TM5ms2Sz2TuPM5lMVQq0IcPD8S5OTCxePzcX68fGIj0mJyP88vl4p1ta4jUQXayhodqXfY1u3YqfxapOTZWem56GZBKuX49q5XKxrmhqKjZBb29sEq3inXeirSw1OxvdqNnZaEPF9gSlHWwxVW7ejNdlMqXzUxtQNhs7UogwW/BxZ3Ky1MYWfvSnpqJ6y2mCj9EiZfcMT5w4QSqVurMMDg5Ws1zr090dDa29ffH6RCLGDpPJOHTp6IjHxecgAhGi398EffvOzvhZPFxZWOW2tqhOV1dUr7W1VN3i88lk/N7VVZvyNq3u7sUbr6i1FbZtiw1ZbF/FtgTxBhQ3fjIZz3d0lNpZA2pvL1U1mVyc2cUqFH9f+G9WakNN8DFapOx35rnnniOdTt9Zhhox9g8fjtHcBx9cvH7r1phVOHYsTofp7oa+vmjMxXe4szPe/X374AMfqH3Z12jXrqjWzp1R7J6e0mevuzsa4rFjkEqVni8aGIjNATGur1U88USM9y0MOogNf/x4bMCurtjQxbAr7n2Kb8bjj8fvBw7c/XcaSHFsEKK97NlTeq6vL8ahE4n4iBRt2wa/9Vt3/63W1hgibSZlHyYnk0mSC3cJjai3F559NnZdc3NxiNPeHo3wox+Nxjs5GYcq27ZFKpw+HYcwPT0xIP4nfxJTZg0ukYjMfu21OJy5eLHUSenrK01gPvVUaQi0rS2q+fjjsVkefvju/YaW6OiAr3wF/u3fYjZhdjaC8HOfiw145kxpL9TeXhqH7uqKN+KTn4x2dfAgHDlS79rc06OPxs+LF+G3fzsmU6anY8f75JOxbmwshl62bIlZ46eegu99Lybr5uejjX3yk83XtjbXbHJRoRChl83GO7lrV+m4cmYmdoFtbfH7/HwMvCWTcapEORpkNrlodrY0XJVIxO8LD3nm52MMaPv2UrU7Oxv6iK0xpdPRnnbsKI1LFCff5udjwC2ZjEG1QiGOQubm7j7mbALz89GuOjujetPT0ddIJGL9xESE3sJqTU/H0tPTnG1rc4ZhtTVYGErauCbMb0mqPMOwXFNT8K1vxZUH167FumvX4vG3vrX43BZJTcfD5HsZHYVvfAP+9V9j4G0lPT3wpS/B177W8FcZSLqbYbiaX/8afv/313bj1/5+eP75mHqT1DQ8TF7Jr38dp+OsJQghXv+Rj8QlXJKahmG4nNHR6BFOTq7v309Oxlji6GhlyyWpagzD5XzjG2vvES519Sp885uVKY+kqnPMcKmpqTh/cLXJknL19MT5iMVr3yQ1LHuGS33nO5UJQoi/83d/F7fvKN4VR1JDsme41Gc/C9//fuX/bktLzDQPDMSyf3/p9+Lj/ftLlw1KqqnNd9v/jbp0qaJ/7il+xVX6IQ9ceXd5caVX32CU3czTSmsiz+4tk3Hx59IlcXeH3q9kbkB+H3dTMQyXWu8M8gqu0s8wA2v+d/lCK8O31/CdGbcn4fs/K/U4+/oa+nZR9wW/j7upGIZLbdtW0T/Xz9pmpUfoJ08bLcyxdw3/tn/8ahziF3V2Ln84vnDd7t1NdzeVpuL3cTcVxwyXqtaYYZkGGGKYAfZzmctU+b6KbW1xp87VAnPv3tLttFVd3g2pruwZLvX5z9c1DGtqbi7GSFcbJ00k4pbHqwXm/v2ePqSmZxgu9cUvwt/+beXOM3z++fjGnMuXY69/+fLipfgNPI2qUIixr6tX4xLFlfT1rR6YAwNxC3zHMdWgDMOltmyJu8/84z9u/G99+cvxxcQrKRTirsgLw/FvuiEDJDvhyHsiQJvhK+zGxmI5dWrl12zbtvwpRQsf79xpYKouHDNczuhofBHQRi7J6++PLy5e4+28lh02mpoq9SqX610OD8e9Fdf3VjaWjo57B2Z//+aZ+JmaihP9/+u/4Ac/iKGLtjb41KdiyOaLX3QIokYMw5WcPBl3n1nPqTbbtsFPf7qu23itewy9+D2+qwXmlSub40qY1tbFJ7AvF5j79i3+TstG430yG45huJqTJ2N2uYb3M6zqhOL8fHwIVwvMy5eX/9L0ZrRr1+qBuX9/xU+lKov3yWxIhuG9jI7G3We+/e1778G//GX46lc3tAev+9kVhUKM/d0rMDOZOhSuClKpewdm8WvhKqF4n8waH3Ho3gzDck1Nwb//O3z3u3eP7Tz7LHzhCxUZ26l7GJYrk4mCLg3MhY9v3Kh3KSujq+vegbl7972/H7OOY9G6N8NwPaqYWE0ThuWYmbl3YI6MbI6Jn/b2GKdcLTD/4R/gn/5pw//VU3sucbWt/BPyveS5PJ5ao+rp7ISHHoplJblc6RrelQJzeDhe18hyObh4MZYquzrawvA69h9e8rw6w1D11d4Og4OxrCSfh+vXVw/My5fvm69r7S+MQG8PbNla1uu95Lk8hqEaX0tLXBK4Z8/KkweFQkxw3SswK3Xj3jp6iffDBz8Ts8tlKA697N27CYZeqsgw1OaQSMSsb28vPPbYyq+7ffvegdkMX+Q1NFTvEmw6hmEDWHgRwrVrse7atTjF0YsQKmzrVjhyJJaVZLOlE9hXCsyRkZiVqJcK33dThmFdrXYRwtxc3Dzn+9+P+0Z4EUINJZNw4EAsK5mfjz3WaoE5PBzBWg31OFl8kzMM62QtFyGMj8d9I77zHS9CaBitrXEqzb598PTTy7+mUCjdsegrX6nseSqrTThpXQxDWPt3VYyMlH4OrOGW/u+euLXeixCuXo3Lpb0IoUkkEnEXnp074a/+Cv7iLyr3t599tnJ/S4AnXYfidFu17d/P6MnLXoRwP6rk93H39kZvc5WBZG+Gs3b2DGHtJ1Rt4JT+b3xjY0EI8e+/+U34+7/f2N9RDVXyPplf+tKKSeY49PrZM6yhSnYOenqiM+vevYlU+dpkb4azMfYMa+g736ncOb/j4zHu+OijMfm5dOnoWH79WpaOjnvfe0BrsHt3JM9G7pP5/PMrBqHj0Btjz7CG6vzFe+vS3l6ZYK1UUHd0bIJvBajwfTK9GU5l2DOsodW+hK5R5XKxNNL3Vi0M00YI6vb2NQb0k09G8lToPpmOQ1eGPcMaOngQLlyodylUDesO1tYcyQtvkjx/huQ7Z0kWpkm25EgePUTyqfeQ/O2nSHYnV/xb+XwcHlfiXrv3+zi0PcMa8qKBzSubXe/FJu3A8XeXd+WB37y7/FsFClem8fG4f/GXv1y7/7ORODxeQw88UO8SSKv77nfrXYL6sWdYQ5//fGUnUI4di4sbir2Shcvs7OLHUjnu55vhOGZYQzW+COGOQiEmQZYLzdWWpYG63mW5vzM7u/FtoMo7eBDOn693KerDnmEN1egihLskEjGY39EB27dv/P+uhEJhbWFbiWC+199o9G8WqIX7eVzbMKyxr30tTr7e6DlhX/1q5cpUD4lEaUa0UeTz5YVu1YL5rUtk59vIJrrIbuklm41L6Grpfr4ZjmFYY1W8CEEb1NIS32HV2VmhP7jWuyEVRoA8JFqgJ76sJF9IkCVJttBRWt59PFto5z+nP8uJyb+uUIHv75vhOGZYJxW+CEGNqAZ3Q5qii/2JK4wXejb8t9YyDr0Z2TOskwpfhKBGVIO7IW0BvpR7nn8c/ZO1l2+JtYxDb0b2DBvA1FSc7Prd795977lnn4UvfOH+bqRandcmV4Zh2GCKR1b79/u1jirfyZMbG4f2rjVegSJtCk8+GYG21iPz/n6DsMgwlDaJ4jj03/xNjDOvpqcnXvfqqwZhkWEobSK7d8dtuIaH4Vvfgs98JsafIX5+5jOxfng4Xnc/jxEutXnGDIeH4a234uLKdDp2fQMDMDYGP/95jC4nk/Dww9EqXn4ZZmbi4t6HH45bRnd2xlmnR47U7RbP5YwZzs3BG2/E8/Pz3Dk599SpGEyfmIjZ6Ww2qrhtG/T1RdUeeSSuaT569N69h/teJgNnz8I778TGnZqK9nLsGHziE9DVBf/93/A//xPtrniJz/Xr8c2JLS2xwT//+bgff50u71jLOPTt2/Dii/DKK1HdvXtj8u4HP4DTp6O95XLR3mZm4mdnZ9zTcd8++N3fhc99LtpXs9kcp9YMDcW7d+sWnDsX665dg9dfh1/8IhpoIgEXL8b90XO5SId8Hq5ciWQZHo538ty5aBHve19dq7SaF1+Mr+MFePvtqOrZs5HxQ0OR/1NTUc22tqj61q3xups3o0HfugUf+tD9ffnVqqam4IUXIhB/8pPYYMX109Nw40ZcU/jDH0bw5fPx89at0uU1bW3R3iYmYv1XvlLqpjWgXA5+9KOYjMnnY92PfxwfrZmZCLyRkah+IhFZPz8fr02lopq3b0f7+/M/jz5FMym7+5PNZslkMouWhlEMwCtXSutyuUiNubl4lyYnSylw61asz2bj3cvno1c5Nlb6O410a+cFxsZKQZjNxu+3bsX68fGo5uRk6RKvYjVnZuLzOzER+4RcLjo8WsGFC7GRhoYWnwQ6Ph4b9ezZmHkoPjc1FW1mdjY2dvFOFLOzcVRy+nTsuRrY0FC0jWIQFqs/NhZVKi4Lb/xRfO30dOnjdfFi7EeaTdlheOLECVKp1J1lsFEuYiwUSsE1M1Nan8uVzjOYn48GnM/H78XHxd8hGu3EROnfr+cchRpYWKzp6fg5NRVVm54uVTGfL/1eKER1c7lYd/t2aR+hFRQ3dCYTG7ComASZTGzA4t0dcrnSxl64QGzsycnomjewiYlSm4JSrhfbUza7uFoLfy9WPZeL9jg+XvpoNYuyw/C5554jnU7fWYYa5cZniUTpWK+rq7S+vb20vq0tlpaW+NnaWvrZ2hqv6eiAhWOfDXr8uPCuM11dUf0tW6JqXV3xs7W19LO1NV7T1habpKUlDpnb2hrnDjYNqbhxUqnFX3BSvAVQd3e8pr091re3lzb2wgVKG3vPntrWYY22b1/8EeroiPHAYntKJhdXa+Hvxaq3t0f76u0tfbSaRdkDGMlkkmQj3WJkocOHY0Jk797SIXN7O3zwg/C//1tq2LOzMWtQHEwrhmNLS0yi9PbG6/bvj3e0AfX2xhj+jRvROPv6ojo7dkTjm5iIak5NlaqXSESj3rUrNsWDD8bmOXSo3rVpYAcOxPHewAC8+WZpzLCnJzZ0cRbqBz+I7tSWLdFmZmdLgQnxJvX3w2OPwUMPVaRoa73/w8hI6efAwMqvKxQWH+Fv2waf/GTpMLm9PdrR9HRpx1rsNW7ZEpultzc23TPPrLt6dbN5ZpOvXFk8m9zXF9Nbt27FAMaVK/FOHj4cCXHyZLzDu3ZFIz16tDSbfPhwU8wmDw+XDn/n5uC112Iic2IiNsHUVDTsrVsjLAcGoprHjsUkZypV27o1nYmJ0mzy66/H+MKOHXD8OHz849Fenn8+bl9+6VIEX3E2+cqVaENHj5Zmkyt0TWUN7v8ARFX++Z+Xn02en49QzOVKt2IbGICPfQz+6I+ab/IENlMYbhJejqd7WWvPcB33fwCiQ/vSS2svX7Nq3Hl+Scu6nwKqlrwCRZIwDCUJMAwlCTAMJQlwAqXqqnVO2FL328yfVGmGYZVdvbq+c8Ly+dqcSyYpGIZVVoPvBFrX/yNpMcOwyjx0lZqDEyiShGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSYBhKEmAYShJALTVuwA1k8vF0tUFiQTk8zA9Hc+1tMT6JlSsRjIJbUvezWw2ni8Uoorz89DRAbOz0NkJra31KfOmNTcXG33Llmhjm8zMTLSl5T4qxbbWpB8j4H4Iw/l5OHUKhodL71Z3N4yOwtmzMDEBu3bB8ePwxBPQ11fvEpft0iV4441opC0t8MADUY1sFl55Bd56Cy5ehPFxaG+P1ySTUd0HH4QDB+DRRzfl57a28nk4fRqGhuL3zk44ehQGB+tdsoqYmoJXX4UbN+JxTw88/jikUtH2Xn01Pk4QH633vKepPkZ3bP4wfO01uHy59PjKFfjZz+L3QiF+XrsWiTAzAx//eCRGgxsdjUZYlM/DhQtRjevXYzl3Dm7ditdmsxGI27ZFB7mtLfYTra3wyCN1q8bmcPp07HWKZmZib9TVBTt31q1YlVAowC9/Cbdvl9aNj8e6T3wCXnwR0unSc5kM/N//wcc+FvuEZlL2mGE2myWTySxaGt7sbPQIF7p2LXZ1IyOL11+/Hq8fGqpd+TbgnXeWX3/qVDTI69ejIY+Px/qJiaj21FRU89q1WH/hQmmfoHWYm1u5zaz0JjWR0dHFQVg0Owuvv744CIvm5hb3P5pF2WF44sQJUqnUnWWwGQ4Bstm7P+m5XLxbc3OL18/PR/dqZqZ25duAlYo5PR1VmZ2Nx8VqFqtXXFd8fna2tF7rkMutvAGbpC2tZrUqLBeE5fy7RlV2GD733HOk0+k7y1Az9KC2br37kHfbtui/Lx3p7eyMY8cmGexYqZh79pQOh6FUzY6OWIrjhtu3x/pUyomUDensjAmT5TRJW1rNalU4cGDl8eZmrHrZYZhMJunu7l60NLyWlhjIXmjfvhgBPnKktC6RiMHuvj7o769pEdfr4YfvzvlEAp58MiZHdu2Kz+iOHbF+9+543NcXYTkwEJvn0UfrU/5NI5GINrY0FTo74aGH6lOmCtq+PSbmltq1K9YfOnT3c030MVokUSisb8Qok8mQSqVIp9ONH4w3bsTgWDYb6dDfH2OJ585FX3/PHjh8OHZ1TdRNmpmBt9+OccGuLjh4EHp747nLl6PKw8NR7dbWCM9EImb8du6MhpxK1bMGm8jYWIwRzszEm3DoUPPNIKygUIj2VDwhY+/e2OG2vNuVGh6O5+fn46PUZB+jO+6PMJSke/AKFEnCMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCTAMJQkwDCUJMAwlCQA2updAFXG7CyMjUFHB/T11bs0m0yhADdvwvw87NgBbX5sCoVob7lcbJL29nqXaON8VzeBc+fgzTchn4/H3d3w/vfDli31LdemkE7Dr34F09PxuK0NHnsMBgfrW646mpiITXL7djxubYVjx+DAgboWa8M8TG5y16/D2bOlIATIZODkyfqVadMoFBYHIcDcHLz6KkxO1q9cdfbSS6UghOgwnzoF4+N1K1JFlN0zzGazZLPZO4/T6TQAmUym8qVS2c6caWFq6u592tQUjIzMsXVrHQq1SSRu3qT15s1ln8ufPUv+yJEal6j+xsdhdHT52Dh7Ns+xY/lln2sE27dvJ5FIrPyCQpm+/vWvFwAXFxeXplzS6fSqGZcoFAoFyrC0Z5jP5xkbG2PHjh2rp63IZDIMDg4yNDREd3d3vYujTcA2tXb36hmWfZicTCZJJpOL1vX09Ky7YPej7u5uG64qyjZVOU6gSBKGoSQBhmFNJJNJvv71r981zCCtl22q8sqeQJGkzcyeoSRhGEoSYBhKEmAYShJgGEoSYBhKEmAYShJgGEoSAP8PEPD7FkYfwiUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Q2 answers\n",
    "df_analysis = norm_df.groupby([\"participant_nb\",\"participant_condition\", \"other_condition\",\"question_content\"]).mean(numeric_only=True).reset_index()\n",
    "\n",
    "for question in np.unique(norm_df.question_content.values):\n",
    "    df_aux = df_analysis.copy()\n",
    "    df_aux = df_aux.loc[df_aux[\"question_content\"] == question]\n",
    "    g = sns.catplot( x=\"participant_condition\", y=\"measure\", units=\"participant_nb\"\n",
    "                    , height=5.6\n",
    "                    , aspect=0.7\n",
    "                    , data=df_aux\n",
    "                    , kind=\"point\"\n",
    "                    , capsize=.16\n",
    "                    , hue=\"other_condition\"\n",
    "                    , legend=False\n",
    "                    , scale=1.7\n",
    "                    , dodge= 0.15\n",
    "                    , errwidth=2\n",
    "                    , palette= sns.color_palette([\"red\", \"blue\"])\n",
    "               )\n",
    "    # Add individual measure points\n",
    "    sns.stripplot(x=\"participant_condition\", y=\"measure\"\n",
    "                  , data=df_aux, hue=\"other_condition\", dodge=0.15, size=5, alpha=0.3, palette=[\"red\", \"blue\"]\n",
    "                  , legend=False\n",
    "                 )    \n",
    "    \n",
    "    g.set(ylim=(-2, 6))\n",
    "    \n",
    "    plt.savefig(\"plots/norm_\"+question+\"_legend.pdf\", bbox_inches='tight')\n",
    "    g.set(xticklabels=[], yticklabels=[], title='',xlabel=None, ylabel=None)\n",
    "    plt.savefig(\"plots/norm_\"+question+\".pdf\", bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SRM Q2-Q1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_data_df = pd.read_csv(\"data/behavior_df.csv\").groupby([\"dyad\", \"participant_nb\", \"interacting_partner\", \"participant_condition\", \"other_condition\", \"congruence\", \"sex\"]).mean(numeric_only=True).reset_index()\n",
    "\n",
    "all_data_df = all_data_df[[\"dyad\", \"participant_nb\", \"interacting_partner\", \"participant_condition\", \"other_condition\", \"congruence\", \"sex\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/other_seeing_me_again_SRM_results_norm.csv\n",
      "data/good_conversation_SRM_results_norm.csv\n",
      "data/seeing_again_SRM_results_norm.csv\n",
      "data/smile_SRM_results_norm.csv\n"
     ]
    }
   ],
   "source": [
    "srm_df = pd.DataFrame()\n",
    "for file in glob.glob(\"data/*SRM_results_norm.csv\"):\n",
    "    print(file)\n",
    "    aux_df = pd.read_csv(file)\n",
    "    aux_df = aux_df[[\"participant_nb\", \"interacting_partner\", \"relationship\"]]\n",
    "    question_content = get_file_without_path(file).replace(\"_SRM_results_norm\", \"\")\n",
    "    aux_df[\"question_content\"] = [question_content for i in range(len(aux_df))]\n",
    "    aux_df = aux_df.dropna()\n",
    "    aux_df = aux_df.merge(all_data_df, on=[\"participant_nb\", \"interacting_partner\"], how=\"left\")\n",
    "    srm_df = pd.concat([srm_df, aux_df])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/g5/mqbcmmn575z0s3c021r942c80000gp/T/ipykernel_77849/4199865681.py:1: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df_analysis = srm_df.groupby([\"participant_nb\",\"participant_condition\", \"other_condition\",\"question_content\"]).mean().reset_index()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUMAAAH/CAYAAADNO78JAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAybUlEQVR4nO3deXDc9X3/8efq8MqyrPUp27JljC+MDwHG5rIN5gqEK4RwBkJJWwi0TTpN0mk77QzNtDPudPpLp5AWNwkhF+UMAQLhCJjDXDaHjcHYGN+3jS1La1nSyqvd3x/vrFcSa2lX2tV3v5/v6zGj8a7Q2h9s6bWf8/0JJZPJJCIiAVfidQNERIqBwlBEBIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREQDKsv3CWCxGLBY79jyZTNLe3s6oUaMIhUIFaZyIyEDJume4ZMkSIpHIsY9hw4ZRU1PD4cOHC9k+EZEBEcr2OF73nmE0GqWuro6mpiaqq6sL1kARkYGQ9TA5HA4TDocL2RYREc9oAUVEBIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBoMzrBoj4xubN8OabEArBuefCxIlet0jySD1DkWx0dMArr0BrK7S0wLJlXrdI8kxhKJKNjg77SInHIZHwrj2SdwpDkWwMGgRz56afz58PJfrxcYnmDEWyNW8ezJhhIVhZ6XVrJM8UhiK5qKryugVSIOrni4igMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBICybL8wFosRi8WOPY9GowVpkIiIF7LuGS5ZsoRIJHLso66urpDtEhEZUKFkMpnM5gsz9Qzr6upoamqiurq6YA0UERkIWQ+Tw+Ew4XC4kG0REfGMFlBERFAYiogACkMREUBhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAEENw+yOY4tIgGR9NtkZa9bAypVQXg4XXggTJnjdIhEpAsHqGR45Au+8A4kExGLw+utet0hEikSwwjCR6Pm5iARWsMJw6FA49VR7XFoK55zjaXNEpHhkXdy1u2g0SiQS8Wdx11jMwrAseFOmIpJZMNOgc5HaI0dg40aoqIDp0yEU8q5dIuKZYIZhSns7PPUUNDfb8wMHYMECb9skIp4I1pxhd4cOpYMQYPt279oiIp4Kbhh+/jls2AANDenP1dR41x4R8VQwh8mHDsHTT0NHBwwaZBuwZ8+G007zumUi4pFghuG+fRaEAFVVMHYszJ/vbZtEisyWLRCNwokngt82jPRFMMOwpgZKStKbrseN87Y9IkXmww9hxQp7vHo1XHstDBniaZMKLphhOGIEXHGFvfUNGwYnn+x1i0SKyrZt6cexmA2mJk/2rj0DIZhhCDY0HjvW61aIFKXRo2HvXntcUmL9B9cFNwxF5LjOPNPOJkSjMG2aDaBcpzAUkS8oKYG5c71uxcAK7j5DEZFOgheG27bBqlVdN1uLSOAFa5i8bh0sX26PV62Ca64JxmSIiPQqWD3DHTvSj+Nx2L3bu7aISFEJVhiOHp1+HAp1fS4igRasYfKpp1pB14YGO2OkMBSRPwpWGIZCMGeO160QkSIUrGGyiMhxBKtnmNLcbNeEtrZCfb1tsReRQAtmz/DVV2HnTjh40B5Ho163SEQ8FswwPHIk/TiZhJYW79oiIkUhmGHYeRFl7FiV+xeRgM4ZzpxpIdjWZr+WBPM9QUTSghmGEIwCbSKSNXWJRERQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoUjfxGKwd69dNytOCO5VoSJ9dfgwPPUUtLRAOAxXXqmrZx2gnqFIrjZssCAE6yGuW+dteyQvFIYiuRo8uOvzykpv2iF5pWGySK5OPhkaGmDnThg7FurrvW6R5IHCUCRXoRAsXOh1KyTPNEwWEUFhKCICuDxMbmmBF1+ExkaYMgUWLfK6RSJSxNztGa5cCfv3Q3u7bX3YssXrFnmmsREOHPC6FSLFzd2e4dGjx3/e1mY9x2HDoMTd9wOANWvgnXfs8ZQpcOGF3rZHpFi5mwSnnmqnAwBqamDyZHu8axf83//B44/DM89AR4dnTRwIq1alH2/aBNGod20RKWbu9gxHj4avf916gEOHpnuAq1ZBPG6P9+6FHTtg0iTPmlloFRV2SALsr2DQIG/b44QNG+DIEZg61b63xAnuhiFAeTlEIl0/1z0NHE+HCy+E11+3WYL58y0cpR9WroTVq+3xxx/DddfpL9URbodhJuecA62tNl6cMQNqa71uUUGNGgXXXON1KxyyfXv6cWurrUxNmOBdeyRvsg7DWCxGLDXeAqJ+nXyqqoKvfMXrVuRdPG6LJbEYzJoF1dVet8hRNTV2FA9s5DF8uLftkbzJOgyXLFnCD37wg0K2ZWA0NcGyZTaXWF8Pc+Z43aK8WLYMtm61x5s2wQ032M+q5NmCBfaGeuSIjSyGDPG6RZ5rbrYZg/Jy+5Hy6/ddKJlMJrP5wkw9w7q6Opqamqj2UzfkmWdg9+7082uvdaIW3a9+ZaO2lK99DUaO9K494pZoFF54wfoSU6fCeefZEe14HB591AIRoK4Ovvxlb9vaV1n3DMPhMOHUVhU/SiQsBPft6/r5TgHvZ+PHw8aN9riq6ovrRiL9sXIlHDpkjzdsgBNOgBNPtA5yKgjBNmj4VTAWUJJJePZZ22N48KAF4IQJ9jFmjNety4vFi203USxmFabKgvEvKwMkkej6PLU9t6rK5qdTSwjjxw9su/IpGD8yq1bBY49ZKE6aZG9rX/2qDY9DIa9blxclJc5Mf0oRmjs3feXL+PHWKwQoLYWrrrITr+XltnjnV8EIw/fft412u3bZWPJLX3IqCEUKbdQouPlmG3l0L+xdWQmnn+5Nu/LJ3eN4nQ0aZP35UMg2yFZVpZdeRSQrpaVu33Dgdhg2NtoM76JFdmxq4kSrUByJpI/kiYjg8jD597+Hl1+2x9dfD3//9/a59na7tyI16SEiQg77DLuLRqNEIpHi3GfY2Gjhl1rzHzoUfvQjW0BpbU0PmUVE/sjNnmF5efpeW7B9AG1tdlpAVUZE8iaRsNMnLS12IGfYMK9b1HduzhkOGQIXXGALJ5WVNmfo8syviEfeeMOKB69ZA08/bX0Ov3KzZwhw6622WBKP2/khDYtFctbSAh99ZPtY6+vT9ZJT9uxJP25rs1Mq48YNbBvzxd0wLCmxEBSRnLW2Wm/v5Zdt80V5uZ1m7V7wqbbWziuD7VrzcxEfd8NQRPrsmWfsGP8nn9g0+2mn2fNksusga8ECC8CWFjjpJH/XufV3GM6bl9vJ8P37bTGltNTq0mVr7Fh4773c2yfiQ7GYDXfLy23jxeHDtlAyYcIXZ5tKSmD2bG/amW/+DsO9e+2IXa4Sib69TiQAwmE7fnfggJ13j8XgrLP8fe44G/4Ow7Fjc/v6PXssCEtKcpvlzfXPEfG5yy+HtWutJzhrlvNXBQGubro+ngkTrEc4fjzs3Ol1a0SkiPi7ZygiebdqlW2nqaqy2xWDUijYzU3XItInn38O775rewYPHLBN1UGhMBSRY9rbe37uMoWhiBwzblz6GujSUqtwHRRuzRk2NMCrr1off+5cOzkuIlkrKbHb7RobbQP14MFet2jguNUzfPVVm+hobobly7te29VdIgHbt3e9NlRECIXsVEmQghBc6xl2LpmRTNpu0aqqL35dMgnPP5/eXlNfb7tKRSSw3OoZzp2bPi904onHv0U9mey6z/CTTwrfNhEpam71DGfMsNnfWOz4QQgWmOXldmMe2MWvItnYt88uCZ4wIXjjSMe5FYZgw+JMQ+POQiG49FIrvlBeDmefPTBtE3/bsMHmpcEKCF9zjQLRIe6FYbbGjYMrr/S6FQNi40ZYuRLKyuC882DMGK9b5FMbNqQfHzliRztTNTNXr7brZ0eNsjfX0lIvWij94NacYU/WrrVvYLBf1671tj0DpK3NOjPNzbZdYtkyr1vkY90v+EidU9u61d5t9u+3+edVqwa6ZZIHbodhMgmPPmrl/2fPtjQA+3X2bPv8o4/a1zkqHrddRCmxmHdt8b1UHasJE+yOndGj7fOHD3f9uu7PxRfcHSbHYnDbbfDww8f/mjfftI/f/hZ+/vMvXvBQxA4ehNdes+NSp58O06Zl/rqqKltXWr/ens+bN3BtdE5ZmZV2Tibt0O7KlVYkeN48Gya3ttrw+KSTvG6p9IGbYZhM9h6EnT38sC2qPPigby6OeuUVO3ADForjxx//AsBzz7UinWVluik1L9avh3Xr7HFzsw2fr73WVpqHD/d9mZd16+DDD21Ukbpdd/583/9v9crNYfJjj2UfhCkPPWSv84nO+8sTid6Hv8OHKwjzpvt9mG1ttqo8aZLvE6Ox0Q5vpeaXX3oJNm+G557zumWF52YY3nNP31537735bUcBdT5AP3myv28l853p09PbtyoqYOZMb9uTR6k31Y4Om4JJbcWNRrvOPbvIvWHy2rU2D9gXb7xhq4E++OaeOdPm8dvbbTeHDKAhQ+C66+zWpEjEV3PNvampse+rnTvt3ELqe2vKFCvi4DL3wvDdd/v/eh+EIejgjKfKy3O7YdEnQiGbHxw1Cq66ynahhUJ2utV17mV9f7c1PP64TYSLBNCmTfDkk7Y4/tJLdjZh8mTfrCv2i3th2N9Vgmeegbo6uPFGW6Z1eA+i9EMiYYnx+uu53d1d5D79NP0t39ISrHvT3AvD+fP7/3scPQqPPAKLF9sm23vuSW/YluCJx63uZVNT+nMrVtg+w/Xr4dlnbYXBAd0Xw4M0FeNeGM6aZRtj82XdOvjrv4baWvizP+v/nKT4SzwOTz9tI4ZHH7VxJNjRu5SOjvSmT58780ybMq+ttXPsma4Mj0btx8KhDjHgYhgCfOc7+f89W1vhZz+DM86wEwc//Wn6rLO4a9cuq54ONn78+GN7XFeX/ppwOH00z+fKyuyU6hVXZD5IE43CE0/YXsSnn7Y9iK5wMwyvu87m/HJ9zU9+YmfbevP++3D77fb2+e1vB6boQyB1P9aTej53rp1Pnj8frr7attsEwM6dXW/MS3WUXeBmGIZCdtb4ppuy+/qLLoJf/Qr+/M+txuG778Kf/mnvteqiUfjRj6zow3nn2SkWVUJwy+jRNu0yfDhMnNh1CmbqVDjtNN+fOslF98I9Lm32DyWTfVsujUajRCIRmpqaqC7WWdZk0o7Y3Xtv5tuwp06F88+33uCtt34x/Bob4Ze/hKVL02dRezN6tAXpt74VjM1ZEjiffWbD45EjrYPsymZst8Ows7VrbTKksdHe3v7u72DECPtvpaUWhuXlmV+bTNoWiqVL4Te/SZ9R6kkoBJdcAnfdBZddZpMxIlK0ghOGYOeMdu2yEi/vvWcB19FhlYknTcru99i3Dx54AP73f62oZ7Z/7h132Gp0bW1fWy8iBRTcMOzvbtKODnjxRbjvPttnls0p9tJSm2y/806bfM/D+CIetypS1dXuDFeKSiIBa9bYX/JJJzmzaixfpDDMh+3bbSX6pz/NfvPVtGk2r3jbbT3f5NeDxkb43e9s18/w4XaW1KGaAcXhzTfTuwXKyuD663u/cEx8ya2+RDRqZ4vvvz/zgkmhTJwI//IvFoqPPQYXXtj7az77DL7/fQvmW2+Ft9/O+ejfmjUWhGAFVDrfVyR50vmcejzuzOZq+SK3wnDFCvtm7eiwUlzbtw/sn19ebhWPX3rJjml997u97z2IxWxbzznn2DaNpUuzLjbRfb3neOs/KR0d9rPd3JzVby9go4mUcFj10hzm1jD5uedgx47084suspIbKYUaJvektdV6i/fdB++8k91rqqrglltsbvGUU477ZW1tlrsHDljndPHi488bpk6VHThgU5cXX2yvkSxs2GBvUFOn9r6ncNeurgtzU6YMTBul39wKw/374fe/ty3ytbW2paVzOngRhp2tXm09v1//OvujfGefbaF43XX9urB8yxb4wx/Sz2tr7ciV5Nkvf5m+FqCkBP7kT3rvsktRcGuYXFMDN98MX/86XH558S2vnnqqheHu3fA//2O3NPXm7bftB2rCBPje92yusQ+652hFRZ9+G+lN5z2oiYR1ycUXiiwt8qC83IaZxVyNsrraNmN/+KGtVt5yS+/LwA0N8MMf2v0bF12U/ebvPxo71o7RVlfb4/HjuxZekTzpXEKuvr5fvXkZWG4Nk3vj9TC5JwcO2HnqpUuzP/0+bpydp7799q5VVHrQ0WFzh59/bs/POst+ZqUfmpvtPHs8botggwfbX7Rffi4E8HsYzpuXW1G1PXts6FJSYkGSrbFj7cTKQEgk4OWXLRSfesp+qHpTUmITgHfeaUcAe5ge2LvXwjClujr3Aj/SzRNPpMt8VVTYNI2OX/qOv//F9u61nl6uEom+vW4glJTYUu/FF1sb778ffvzjntubSFjCPf20FYe44w4rFpHhwqKqKvsjUgdm1HnJg0OH0o/b2mwHgS6p9p1g9Qz377eeVmlpOigSCRvmpD5fXm7v7p3nHAeyZ5hJPG5H/u67D1544Qv/uYXB7GUsI2hgGH8sTV9eDl/7ms1NLlrU5f9n61bbsD14sFWk6l6yT3L02mt2eQjAmDF2FKiY56wlI3+HYT48/7zdd5JajJg7F775zeIt1LZpk/UUf/YzOHCAKEN5kqtpo4ISEnyZ5xjP7q6vmTnThtC33hqo2nsDJpm0vUvxuO1r9fkQef9++OgjW9ObPz84RzwVhr/9LSxblh7qzJsHf/EXMGiQt+3qTSwGv/kNa/7jRd5ZlW7rND7jfF7N/JrKSit4e9dd2VX0Fud1H1wlk3YrXkppaddtWJkGV9nwenCVDYXh5s029Ny61SbQ7rjDV5VJtm6FFx/446mHt9/m9NibnM4HNBJhI1OpJsp0MuxNnDfPQvHGGzVODrDUBotCK8YNHN0pDMHmDFta7NxpsW3UzsKaNRaKoypbOGvTg7T8z895fM002rEe41w+YB7vZ35xJGKbur/1LRtOS6Bk6hm2ttqvyaT1AlOj/rIy2+5a7Bsy+ip4YXjkiG1daWy0Mlpnn+11i/Juy+Ykf/jJVustrlxJTXwXV/NU7y887zybW7zmmuKfJvBCS4tVQ2putntvpk/3ukUFEYvZ1HQ0am+0KdXVVmipWLfq9pe/Z3r7YsWK9FvhRx/Zv6pjFQtGjgpRNu1E4ieeCNdey9jtz8Dzn1olnZ689pp91NTY1pw77tA9Lp29/nq6EtJrr9l0SrEutPVDOGyDhNZWWyRP3XGW6Q5ll/hvTJitxkarEvPhh12rUHe+5zDTcwdUV9se7Jkz4awLh3DG/7vBSpq98ooVJ+1ttXP/fvi3f7OKK5ddZhVks9n87brOxTWSSefvzR48GK68EmbNsuH0okVet6iw3Bwmt7XBo4+mq4dMn271rcB6hc8/byFYU2Op4fOtEDnbu9e25vz4x7BtW3avqatL3+OSy2SRS9avt94h2PzyVVcF7nunmE+09pebYdjbmbNYzOZ/IhFfLpjkTUeHvTEsXWqburP5Vigrs3tc7rrLrll1bXNxbxv5Ewn7KC21/3eX95pk4HIYuvm2Nny49fFTNfG730gXDgdnJ2lPSkut1Nnll1sPMXWPS+dS993F43a1wuOPW4/7zjttNTp17arfuXjEU7LiZs8QbCns008tFGfOdLYHmEhYByVvHbT2dnjySestvvJKdq+pqIAbbrBgPPNMf/cWXSz+kUepnmFtrXvZ724YBsCaNbY4XlpqN49me/Vz1tavt/uhf/5zW5DKxqmnWijefHMwbpFzedyYwdixNnAYNswqp8+b53WL8sfN7lIAtLbaYnkyaSPX1Lx+Xs2YAf/5n/bD/sADcMYZvb9m9WoLw9paO9b40UcFaJh4pfPmiw8+cGtBXWHoU6n+/KFDtmtmw4acCl/nprLS7ndesQLef9+KyfZ2hO/wYauyU19vpXF+/ev06r44w88zIt0pDH2qstKmQteutSNSgwfb4Yho1LYFPvIIrFtXgD947lzbkrN7N/zoR7YJrTdvvQXf+IYNKf/2b2HjxgI0TAZC54NJZ5/t1rF2haGPTZpk35DnnGOHIQ4etOHynj3Q1ATLl2c/1ZezSAT+8i9tGLx8uc0R9naE7+BB+I//sGOQX/qSVYjWhUm+ktpWWVmZ3X1mfqIw9LHRo7tulZw8Ob2bKGXrVlu0LNjcfigECxfaMHjnTvj3f+96V/Xx/OEPVnz2hBPg7rsDsfjgd2vXpucIW1rsuUu0muxzLS0WeFVVdsR6wwY7NptM2lbKtrb0vM4ll1j2FFwiYbfb33efbX7vfBzyeEpK7OzXXXfZlQd+2Qrl+GpyMgmPPQb33GMXOXa3YAF85zt2rbff5w8Vhg6KRq2HuGVL16ojs2bZN++A2rnTNnL/5Cc2z5iNyZOtpNg3v1n8tSUdDsNYzNbNHn6496+98UbbgeXnswzuh2Fzs010OHBreq77gffts5NiJSV2P1E43PtR2oLtBY7HbWXnvvtsiJyNQYPg2mutt7hgQXF2PRwNw2TSLvnLJghTbroJHnywOP+ZspF1GMZiMWKpWj5YGNbV1RV3GL75pk1slJTAuef6vv7cQFQlHpCf6Y0bbTP3Aw/Yoko2Zs+2/Yu33FJc97g4GoaPPmqHinL1yCNWGMmPsg7Df/7nf+YHP/jBFz5flGGYTFrv4+GHbTJtxgz79dZbvW5Zv2TqGba3d91fOHhwerqtLyfFBvSUWFubnXFeujTzhFQmQ4ZYl+XOO22bj9ccDcOFC7P/J+n+uuXL89+egeBmz/Czz+xekxUrLBjHj7djYjfd5HXL8u7pp7sG5EUXpRdzffVzumaN9RZ/9SvbsJ2NM86wULzhBu82vPnqL7ln8Tjs2GHV3f71X/v++6xd688bJLKuWhMOhwn7ZXa0rc3mm6ZMsaXW8vJ0PUPH1Namw7C8vPjXG46rvh7++7+tqOxDD9nc4urVPb9m5Ur7+O53rXLOnXfaKEAyisctt7duzfyxY0d+avi++67jYegr06bZGTWwsvWXX27FOB00b56NHA8ftuwfOtTrFvXT0KFWRPb22y3o7rvPJqJ6OsrX2Aj/9V/2sXixLbhcfXXg7nEZqLDrTTRa+D+jENwMw4oK29Db2JheRs1k3z6bW4zF4LTTimMOqg9OPtnrFhRAKGTlwM48E374Q/jFL2xuccOGnl/36qv2MWaMVeW+/fYClPPxRjxuI/Hjhd3OncVxO0OxzZply/2tNT15/HE72Jty/fVWm8iHdu+2hZS6uvQCikPTWSaZtBqL991nNRezOcoXCtk9LnfdBZdeavXO8imPf8l+CbveOD9n6KTu31l++E7L4O2305WyJkyAL3/5+Hu9Dh2y/+bLzA+FrHDjBRfYUvn991vRiB07jv+aZNKuNHj2WTt+c8cddvNfPq56278/u9M1f3T0aNew27atuMJu1CjrRE+aZN9TfdnGtXChP4MQgh6GZ51ldyjH4zbxPnKk1y3qk87VaXbutH3mmeYOO4fm3Lk+L8w5bhz80z/BP/wD/P73NoR+7rme73HZtg3+8R/tLPQ119iCy+LFfdsl/NJLsHlzl8PgncNu+2cxtuwoY8v20i5hl0N25t3o0emw6/5xwgk295zS132G3/52PlrqjWAPk8G+g+Nx26DnU51H+4MG2Ta8QYO6juA2b7bjUimhkE2p+eUIcFa2bLFjf/ffb722bJx0Uvoel17uQE6F3Y61UUofe5iDB+Hgi++z9+hwlpb+FTuTE0gk4GzeYjYfE6eMl7iIHQzMvdy5hF1vdAIlB74Nw82b05UMFiywHwafa2qyXl88Dqefnt5g3TkMt2+3MExNs4XD9vPvpPZ2Kw+2dKn9W2ejooKj19/Mjqu/zdbqerZuC31hOJvq2YVp42YepBQb0zYyjMewYxfDOMR1PHbst41SzSPc+MU/rw/yGXbZiMXsePhDD/X+tTfdZAeK/LL7LpPgheEDD6SPbJSUWCKUl3vbpgLpPre/Y4fVWU1V3ep+aaCTPvnENnP/4hccbTrCDurYyqSMH7sYT4LMCyw17GMI9vo45ZzAVubzLkcpZzmLaMCmWKpp4gYeOfa6zkHZm5qa44fdxIn5D7tspKrW3HuvFQ/ubuFCGxqrao3fwjCZtO31nWepb7vN2f1ozq0m96K93QI/0+LE1i0Jdu2CRDL3eYFZfMw5vAVAAyN4kqvp6GG6/TQ+YC4f0M4gXuIi9mDvOsUYdrlYu9bCr7HRFuDeeCO7Qud+EawFlFDIykK/8YYF4xlnOBeE7e3O/S8d0znsMn3s2tXT+knfJ0dnsP7Y4xE0UMP+YwHXWQ37jvUzR540mkl/eRk3TS0/Noz1e4n8WbMssBsb7VeXghCCFoZgO5SnTElXP3VEa6vtHmlosB7IZZd53aLc9S/sCidKNSNooJomRnCIS3ieMezvMsieyHYq6VRm/FPg7Zvgr3y8ohAwwQrDtja7X7Olxd7WBqTs88BYsya9orx/v5UOTJVoP3LEhjhev5PHYpnDLjWk9SrsUsZUNDIp9imTklu6BN2JbOYEtlFBe++/SWcPPWTHAv1a0ypgghWGy5fb9guwIxvXX+/fs0MZJJN2k+eyZbBpU/rzjY1WDrDQJdqPF3apj927PQ67MT3P2VVWDoP9J8IDr8L//jj9vdIf996rMPSJYC2gpDbkdXRYGlx5pa0uOKCxEa64IrsadH0t0V7sYTd2bM9hl9NW0kQCXnzRjv797nf9+x/z6/m0DFxelAtWz3D6dKuAsm2bFXs9/3yvW5QXyaQdvc22GOfDD9t7QfcNsrGY7Uc8Xtjt2eNQ2PWmpMTOMl96qRWK+N73+v57+bWmVcAEKwynTLE9AYMHw4gRViLKgU3Xjz2W20kBsOms1lZb4ezcs/PSgIZdLvq7D9WvNa0CJlhhCF3vz3Bkle+ee/r2uiefzGszejVuXM9hV7R3dvW3SKRfppFydPCgjUbicdulNmGC1y3qn2CF4ZAhVh9v5Uq7Jm7RIq9b1G9r1/btropC8G3Y9Wb+fG9fX6T+8Id0p/fFF+Eb3/D3Ya5ghSHAKafAnDnWK3SgZ/juuwP3Zzkbdr1JXTjd1xuSHJ0vTG3dAusdtrcrDP3HoVIthw7l7/eqrc1cAMD5sMvGd77TtzD0c02rXsyZk76mZvLk4j9O2JtghqFDersUvjff/77VO62rC3jY9ea66+C3v829ptV11xWuTR474wy7Yigez0+tXK8pDH2uv9Oe3/ym3Z8lvQiFbHNmKJRbTSsHpmJ64tvbGDNwZ7wYUKeeagW7+8Lh6azCCIdtc+Yjj9hfXiYLF9p/f/BBZ86+x+PeVugeKOoZOuBv/iZ4Jdo9EwrZ8brrr3e/phW2QLdqlS2MXHSR160pLIWhAzSdNUBiMbsz5/PPbUXpvPN8X9Nq3jzYuzfzf0skulzxQihkVdXBTiOl9hWmVpLBOsOZLiAcOxbeey9/7S4EhaEDNJ01QFatSh/I/eyz9P0KPrZ3b99uwUsk+va6YqYwdERqOuvqq4NRot0T7d1KeKWuj/Cx3laB29vT/5vhcLrOSWmp1c1MJq0iXmeVlV/8/vLDarPC0CEBm84aeHPmWJGP1lb7S3VgGT6boWtHh23NPd4b6MqV6f2GZ5xhi3p+5E4Jr44OePVVqzYwbpxVpMk0eREgLpdb8szRo3YxdXW1fX91/kvets3uU96xw25kv/TSwGzePHzYwrKqyuuW9J07W2s+/tgqmra22nWgqdvSRfKpvNzuV870RvvppxaIiYSVG1+1auDb55GhQ/0dhOBSGMZiPT8XKbTOty5mei5FzZ0wPPnk9OHIykrtJpaBd9JJNjwG6ybV13vbHsmJOwsoQ4faykFTk9Us9HP5DPGnQYPgq1+1ci6VlU4VBAkCd8IQLABT78wiXvD7KkKWVq2C9ettHen88/1/JzS4NEwWkYJobrZNGqn9hnv32jG9w4dtIf2tt7xtX7641TMUkbzatQuef97WgiIR+MpXuh7RA7uO3AXqGYrIcX38cXpRvKnJdg7V1aVLd5WW2jpRLGZD5w8/tLPKfqSeoaPa29Pv4K2t9nzQIG/bJP7TeS4wkbAaFaWldkf3oUO2gaOyEp54wi6IAtvgf/nl3rS3P9QzdNSaNekadImEvcOL5CpVzXrYMDuH/MknsGyZXQZVU2Nh2NaWDkKwobWX92v3lcLQUd2LcQahOKfkXzgMF18M117b9dDNzp3pcw0VFV1v4K2p8WchEA2THTVnTnqbW0kJzJ7tbXvE30pKbBtN6mrQysr0tEsoZMPmjz6yr/PrXnOFoY/0VIgzk9TNeYcOwdSp6c93dKTLMHU/YuuHIpzijcsusy01ySScfnrX3t+QIX2/fqJYKAx9pK+FOJNJ9wpxFq2mJnjhBetCzZhx/LtSfKi6Gi680OtWFI7C0EdyLZC5b1+6Ft2oUXataOdinWCf63xvkR+KcBZUrt3vPXvs1927rZxXW1vXAg0VFf6tgx8wCkMfyfVn57XXrKoU2JDmmmvsZ/XZZ9Nfs2CBir52oe73cbW3Z55acYXC0GGpy3vAflajUdsmcfHFsH27bZxVcZ9ucukaJ5PW/U4krPs9dKh1s1PLrKWl9jzT0qrPuuBvvWXbs8rK7Punrs7rFuWfwtBh06enR3xDh6bvLzrxRPuQDHLpfieTdhNXat6htBRuu80+H4ulS8r53KFD6X2q8bgFY/eraT//3C4ObGuD006DU04Z+Hb2l8LQYTNmWFHmw4dtOisgFegHTihkKwqvv24BuHBhegxZ5s6PVveObaaO7vLl6W03K1bACSfYRm0/cedfTDIaM8Y+wI7lrVhhnZb6eiduuvTexIlwyy1et6Kghg2DuXPt7HE4nHmBvPtFgX48n6wwDJBXXklfDLVrl92fPHiwt20Sf5g3zwLxePVq58+376+ODtvT6seyosELw3XrbDtEba2NIwMktQkb7J27uVlhKNnrqXD35Ml2QWB7u81P+1GwwvCzz2xyA2DjRquMPWWKt20qgI8/tkINQ4ZYFeLUTa5Tp1qJJYARI+xD+mDrVnjnHZsXXLQoPQ8RcOFw1z2rfhOsQg0HDvT83AENDbba19xsuz5efz393848067yXbwYrrrK3f1iBdXebsum0aj9Zb/0ktctkjwJVhiecEJ6KSwUcnKzVPcbUrtXIZ440bbcqLZhH8XjXU+YuFLmuZ9iMZuB2rLF65b0XbCGybW11iXat882vdbUeN2ivBszxuZudu2yOZ7TTvO6RY6prLS55vXr7fnpp3vbniIQj8NTT0Fjoz2vr/dn0YZQMtm3MozRaJRIJEJTUxPVqUkpKQqJhI3gKioCcVGbNxoabJ6hcyE/Rxw+bLUmGhttFHHuuT1//f798OST6edVVfD1rxeyhYURrGFyQKQKMygIC2jECCeDEGwvakODvamuX2/rRT0ZOrTrNeV+XZgL1jBZRHrVeUo00/PuBg+2WocffWSjkfnzC9e2QlIYikgXp59u0+ptbTbNPmlS76/pfNLJrzRnKNJfBw/aitWoUZYeDojHbYW4stKf95n0hXqGIv1x4IAtpabGkhdf7ERJoLIyp2pNZEULKCL9sXNn10m1bdu8a4v0i8JQpD9Gjuz63I8VCgTQMFmkf+rq4IIL0qXDdSerbykMRbKVOpf8+ed2rvHcc21T59SpXe9iFV/SMNkhDQ3w5pvwwQe97w2TPli1CnbssD0nGzbYhzhDPUNHtLXB736XLtQQjVp1Gsmj9vauz2Mx+1yq6sX69VYmLhKBs8/ueixDip7C0BGNjV0r1uRy9a9kac4cO5vW2mrFItets7NrI0bYsYtUvbQ9e2z47NAF8kGgMHTE8OG2QbalxZ5PmOBte5w0bBjceKNVMti0yYbNYPMT3W/VS92OJL6hMHREOAxf+YpNYw0eDCef7HWLHFVebj3BVPWCWMyGxmPG2F98VZUd2Zg+3dNmSu4Uhg4ZOlTl9QbM7Nm2nebll+3s2rBhFoJnnmkFJR3db9jRYZ3gQ4fsxoxp07xuUf4oDEX6YtAguPpqW7nqPCSuqXE2CAHefz99j8727fYGPHast23KF22tEemPU05JP66pcbJ6emedb1iEdHVrF6hnKNIfJ59s84UtLTBunPO3bE2Zkj5+XVHh1kKdwlCkvwJ07+rUqTY0bmy0qVGXqqkrDEUkJy4Ucs1Ec4YiIuTQM4zFYsQ6HXGIalOpiDgk657hkiVLiEQixz7qHLyAXUSCK+s7UDL1DOvq6nQHiog4IethcjgcJhwOF7It+ZVM2u3W4bCdDhAZSImEFWsQ33BzNTmZhBdesC3yAAsWwKxZ3rZpgOzfb3UDamtBHXYPtLXBc89ZAdjaWrjkEpXy8gk337oaGtJBCLB6tWdNGUhbt9pFba+/Dk88AU1NXrcogNassSAE2L0bPvnE2/ZI1twMw4qKrkOUykrv2jKANm2yTjFYzdHO7wcdHbZRNh73pGnBkUj0/FyKlpvD5CFDrMzzBx9YMJ57rtctGhDDh2d+3tICTz9t9QSqquCqq9w6OVBU6uvtXaix0Qo2zJzpdYskS1mvJncXjUaJRCJaTS4iiYTl/4EDMGkSzJhhn3/vPft8Sn09nHWWJ00MhmTS6hxWVHjdEsmBmz3DgCopgXnzvvj57vP3qSs7pEBCIQWhD7k5ZyhdzJplPcXycrvhcs4cr1skUnzUMwyAsjL40pe8boUjWlqsTJef9txKVhSGItlascLKPJeUwKJFcNJJXrdI8kjDZJFsHDmSrnefSMDbb3vbHsk7haFINkpLbWEkpUyDKtcoDEWyUVFhl8KXl9uVoIsXe90iyTO9vYlk6+STdSG1w9QzFBFBYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoYgIoAuhRKQHra3w8svQ2AhTp8JZZ3ndosJRz1BEjmvFCti9G1paYM0a2LbN6xYVjsJQRI4rFuv5uUsUhiJyXKecAuXl9njkSJg0ydPmFFQomUwm+/LCaDRKJBKhqamJ6urqfLdLRIpEWxs0N8Pw4VBa6nVrCkcLKCLSo4oK+3CdhskiIigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgLoBIrT2tvhnXfsKNXMmW6fKxXpL/UMHfbGG7B+PezcCS+9ZDXpRCQzhaHDmprSjxMJiEa9a4tIsVMYOmzatPTjoUNhzBjv2iJS7DRn6LDZs60GXXMz1NVBOOx1i0SKl8LQcePGed0CEX/QMFlEBIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACdQAmcRAKWLYMdO6CmBi6+GAYN8rpVIt5TzzBg1q2DzZvh6FHYtQtWr/a6RSLFQWEYMEePdn3e3u5NO0SKjcIwYGbMgGHD7PGQIVBf72lzRIpGKJlMJvvywmg0SiQSoampierq6ny3SwookYDDhy0MyzRrLAJoASWQSkogEvG6FSLFRcNkEREUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICJDDputYLEYsFjv2PBqNFqRBIiJeyLpnuGTJEiKRyLGPurq6QrZLRGRAZX02OVPPsK6uTmeTRcQJWQ+Tw+Ew4XC4kG0REfGMFlBERFAYiogACkMREUBhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBFAYiogACkMREUBhKCICKAxFRACFoUh+JJNw8CAcOeJ1S6SPyrxugIjvJZPw/POwYweUlMD558OUKV63SnKknqFIf+3bZ0EIkEjA++97254COXoU1q2DjRst/12jnqFIf4XDXZ9XVHjTjgJKJOCZZ+Dzz+35zp2weLGnTco79QxF+mv4cDjnHBgyBEaPhnPP9bpFedfcnA5CgM2bvWtLoahn6KhkEj74wEZwEyZAfb3XLXLc7Nn24ajKSuvwtrXZ85EjvW1PISgMHbV2bXrqaudO67RoTl/6qqwMrrgCVq+2x/Pmed2i/FMYOqqxsefnIrkaMQIuuMDrVhSO5gwdNXmy7fIAeyefNMnT5ogUvVAy2bdF8mg0SiQSoampierq6ny3S/KgocEmvceOhUjE69aIFDcNkx02YoR9iEjvNEwWEUFhKCICKAxFRACFoYgIoDAUEQEUhiIigMJQRARQGIqIAApDERFAYSgiAigMRUQAhaGICKAwFBEBFIYiIoDCUEQEUBiKiAAKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIrlpa4N43OtWSAHoqlCRbL35JqxdC2VlcOGFcMIJXrdI8kg9Q5FsHDpkQQjWM3z7bW/bI3mnMBTJRkm3H5XSUm/aIQWjMBTJRiQC8+ZZKA4eDAsXet0iybNQMplM9uWF0WiUSCRCU1MT1dXV+W6XSHFKJiEU8roVUgDqGYrkQkHoLIWhiAgKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBFAYiogAOZTwisVixGKxY8+bmpoAO5YnIlLshg4dSqinE0TJLN19991JQB/60Ic+fPnR1NTUY8ZlXaihe88wkUjQ0NDAyJEje05bIRqNUldXx44dO1TUQvJC31O5661nmPUwORwOEw6Hu3xu2LBhfW5YEFVXV+sbV/JK31P5owUUEREUhiIigMJwQITDYe6+++4vTDOI9JW+p/Kvz5WuRURcop6hiAgKQxERQGEoIgIoDEVEAIWhiAigMBQRARSGIiKAwlBEBID/D7ZjR6WyB5zNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 392x560 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "df_analysis = srm_df.groupby([\"participant_nb\",\"participant_condition\", \"other_condition\",\"question_content\"]).mean().reset_index()\n",
    "\n",
    "for question in np.unique(df_analysis.question_content.values):\n",
    "    df_aux = df_analysis.copy()\n",
    "    df_aux = df_aux.loc[df_aux[\"question_content\"] == question]\n",
    "    g = sns.catplot( x=\"participant_condition\", y=\"relationship\", units=\"participant_nb\"\n",
    "                    , height=5.6\n",
    "                    , aspect=0.7\n",
    "                    , data=df_aux\n",
    "                    , kind=\"point\"\n",
    "                    , capsize=.16\n",
    "                    , hue=\"other_condition\"\n",
    "                    , legend=False\n",
    "                    , scale=1.7\n",
    "                    , dodge= 0.4\n",
    "                    , errwidth=2\n",
    "                    , palette= sns.color_palette([\"red\", \"blue\"])\n",
    "               )\n",
    "    \n",
    "    g.set(ylim=(-3, 3))\n",
    "    \n",
    "    # Add individual measure points\n",
    "    sns.stripplot(x=\"participant_condition\", y=\"relationship\"\n",
    "                  , data=df_aux, hue=\"other_condition\", dodge=0.15, size=3, alpha=0.4, palette=[\"red\", \"blue\"]\n",
    "                  , legend=False\n",
    "                 )\n",
    "\n",
    "    plt.savefig(\"plots/srm_\"+question+\"_legend.pdf\", bbox_inches='tight')\n",
    "    g.set(xticklabels=[], yticklabels=[], title='',xlabel=None, ylabel=None)    \n",
    "    plt.savefig(\"plots/srm_\"+question+\".pdf\", bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Debriefing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Format interacting partner\n",
    "debrief_df = pd.read_csv(\"data/Q3_Speed_dating.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "name_map =  { \"Q1\"    : \"participant_nb\"\n",
    "            , \"Q21\"   : \"age\"\n",
    "            , \"Q11\"   : \"seance\"\n",
    "            , \"Q17\"   : \"date_passtion\"\n",
    "            , \"Q2\"    : \"sexe\"\n",
    "            , \"Q6_1\"  : \"rating_quality\"\n",
    "            , \"Q18\"   : \"comm_quality\"\n",
    "            , 'Q28_1' : \"rating_manip_sourire\"\n",
    "            , 'Q29'   : \"comm_manip_sourire\"\n",
    "            , \"Q25_1\" : \"rating_conversation_fidelity\"\n",
    "            , \"Q26\"   : \"comm_conversation_fidelity\"\n",
    "          }\n",
    "\n",
    "debrief_df = debrief_df.rename(name_map, axis=1)[list(name_map.values())]\n",
    "\n",
    "debrief_df = debrief_df.iloc[1: , :]\n",
    "\n",
    "debrief_df[\"age\"] = [float(x) for x in debrief_df[\"age\"].values]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "rating_df = debrief_df.set_index([\"participant_nb\", \"age\", \"seance\", \"date_passtion\", \"sexe\"])\n",
    "\n",
    "rating_df = rating_df[[\"rating_conversation_fidelity\", \"rating_manip_sourire\", \"rating_quality\"]]\n",
    "\n",
    "rating_df = rating_df.stack().to_frame(\"rating\").reset_index()\n",
    "\n",
    "rating_df = rating_df.rename({\"level_5\" : \"question\"}, axis=1)\n",
    "\n",
    "rating_df[\"rating\"] = [float(x) for x in rating_df[\"rating\"].values]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "T-test results for rating_quality\n",
      "              T  dof alternative  p-val         CI95%  cohen-d       BF10  \\\n",
      "T-test  6.99083   30   two-sided    0.0  [5.07, 5.96]  1.25559  1.639e+05   \n",
      "\n",
      "        power  \n",
      "T-test    1.0  \n",
      "\n",
      "T-test results for rating_conversation_fidelity\n",
      "              T  dof alternative    p-val        CI95%  cohen-d   BF10  \\\n",
      "T-test -2.18816   30   two-sided  0.03658  [3.0, 3.97]    0.393  1.518   \n",
      "\n",
      "          power  \n",
      "T-test  0.56273  \n",
      "\n",
      "T-test results for rating_manip_sourire\n",
      "              T  dof alternative  p-val       CI95%  cohen-d   BF10    power\n",
      "T-test -2.03286   30   two-sided  0.051  [2.9, 4.0]  0.36511  1.163  0.50309\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABFoAAABlCAYAAAB9XYf4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAMZklEQVR4nO3dwUsc9/sH8GeTxT2EKgTqTQ8lPXlqaYyHQiH3JqsG+w/02JMgtZtDb7Gm4B+Q/8BizOCppzS5Wderp0oJ2pPQQEwJmJrO7/DDNMadMdX5Zmb09QKh+ZCEB3l3Zn0yu+9GmqZpAAAAAHBqF8oeAAAAAOCssGgBAAAAKIhFCwAAAEBBLFoAAAAACmLRAgAAAFAQixYAAACAgli0AAAAABTkxIuWNE1jd3c30jQtch4AAACA2jrxouX58+cxMDAQz58/L3IeAAAAgNry1iEAAACAgli0AAAAABTEogUAAACgIBYtAAAAAAWxaAEAAAAoSLPsAQAAoAhpmsbLly/LHoOKejMffX190Wg0Sp6oOnw/oFgWLQAAnAlLS0vxzTffxNOnT+Py5cvx6aefxkcffVT2WFB5CwsL0Wq1yh4DzgxvHQIAoPaSJImpqanY2dmJ/f392NnZiZ9//jl+//33skcD4JzxRAsAALV3586dnuebm5vx9ddfv+dpqKJXr17F48ePIyLiiy++iIsXL5Y8Ubne/H4AxbJoAQCg9jY2Nnqe//HHH9FsesnLYRcvXpQL4H/GW4cAAKi9kZGRnudDQ0PveRIAzjuLFgAAaq/T6RxpTWk0GnHr1q2SJgLgvLJoAQCg9trtdiwuLsbg4GA0m824cuVKzM7OxrVr18oeDYBzxqIFAIAzKU3TskegQtbW1mJpaSnu3bsXs7Ozsbq6WvZIVEiSJDE6OhqXLl2K0dHRSJKk7JGoMYsWAABq7+16583NzZifn/fDNBERsbq6Gnfv3pUPekqSJMbHx6Pb7caLFy+i2+3GxMSEZQsnZtECAEDt9ap3TtM07t+/X8I0VE2vHMgHB7KuH3NzcyVMw1lg0QIAQO1l1Ttvb2+/50mooq2trZ7n8kFE9vUj6xyOY9ECAEDtqXcmz/DwcM9z+SAi+/qRdQ7HsWgBAKD21DuTZ3JyUj7IlHX96HQ6JU1E3Vm0AABQe+qdyTM2NhYzMzPyQU/tdjuWl5cPtQ49ePAgbt68WfZo1FSz7AEA4ECapvHy5cuIiOjr6zvyr0sA/4V6Z/LIB29K0zT++eefSNP09ReclCdaAKiMly9fxvT0dExPT79euAC8C/XO5FHvTJ4kSWJiYiLW19fVO1MIixYAAGpPvTN51DuTR70zRbNoAQCg9tQ7k0e9M3nUO1M0ixYAAGpPvTN51DuTR70zRbNoAQCg9tQ7k0e9M3nUO1M0ixYAAGpPvTN51DuT5+3rx9WrV9U7cyoWLQBAbSRJEqOjo3Hp0qUYHR3VCEEu9azkkQ+yqHfmtCxaAIBaSJIkxsfHo9vtqt/kCPXO5FHvTJ63rx/r6+vuL5yKRQsAUAvqN8mj3pk86p3J4/5C0SxaAIBaUL9JHvXO5FHvTB73F4pm0QIA1IL6TfKodyaPemfyuL9QNIsWAKAW1G+SR70zedQ7k8f9haJZtAAAtdBut2N5eflQ65D6TQ6odyaPemfyqHemaM2yBwAAeFftdjva7XbZY1BRN27ciEePHkVExPXr16PZ9FKXf42OjsZff/0VEfLBUW9ePxYWFqLVapU7ELXmiRYAKmNlZSWWlpbi3r178fnnn6tV5IgkSQ490SIjvOnNa8js7KzqXg5ZW1uTDzJ5DUKRGmmapif5g7u7uzEwMBDPnj2L/v7+oucC4JxJkiTGx8cPnTUajVhcXIwbN26UNBVVsrKyElNTU4fOZIQDWfn49ttvY2xsrKSpqIrV1dX44YcfDp2d93z8/fff8csvv0RExNzc3Ll+giPr+rG8vOwpSk7knRcte3t7sbe39/rXu7u7MTQ0ZNECQCFGR0ej2+0eOR8cHPRhhURExNLSUuzs7Bw5lxEisvPx8ccfx48//ljCRFTJzMxM/Pbbb0fOz3M+9vb24vHjx2WPUQlZ14/R0dH49ddfS5iIunvntw7Nzc3FwMDA6y9VaAAUaWNjo+f506dP3/MkVFVWFmSEiOwcbG9vv+dJqKKtra2e5/JBRPb1I+u1CRznnT8B6rvvvovp6enXvz54ogUAijAyMtLziZZPPvkkFhYWSpiIqllbW4v19fUj5zJCRHY+vF4lImJ4eLjnEy3nOR8XLvz7b+7n/a1DWdePkZGREqbhLHjnRUur1TrX//MB8L/V6XRiYmIi3nxHa6PRiNu3b7v/EBERt2/flhEyZeXD28qIiJicnIz5+Xn5eEOj0Xj93+f9Z72s60en0ylxKupM6xAAldBut2NxcTEGBwej2WzG1atX48GDB3Hz5s2yR6Mi2u12LC8vH2odkhEOvH0NuXLlSszOzsa1a9fKHo0KGBsbi5mZGfmgJ69BKJpFCwCVlKZpnLAYjzPuIBsywnHkgzzyQRb3F05LvTMAlZBV76xakQMyQp6sfJzn+l7+pd75qP39/Xj48GFERCwsLJzrtw65v1A0T7QAUAl37tw5cpamaczNzZUwDVUkI+TJysf9+/dLmIaq6ZUD+eCA+wtFs2gBoBKyKhRVK3JARsiTlQP1vUSodyaf+wtFs2gBoBKyKhRVK3JARsiTlYPzXN/Lv4aHh3ueywcR7i8Uz6IFgErodDqHqiYjVCtymIyQJysf57m+l39NTk7KB5ncXyiaRQsAlaBakeOodyaPemfyqHcmj9cgFK1Z9gAAcODGjRvx6NGjiNCAQDb1zrwr+SCPfJDF/YXT8kQLAFALB/Wb3W43Xrx4Ed1uNyYmJiJJkrJHowKSJImpqanY2dmJ/f392NzcjPn5+VhdXS17NCpgdXU17t69Kx/09Pb1Y3193f2FU7FoAQBqQf0medQ7k0e9M3ncXyiaRQsAUAvqN8mj3pk86p3J4/5C0SxaAIBaUL9JHvXO5FHvTB73F4pm0QJAZfT19cXCwkIsLCxEX19f2eNQMeo3yaPemTzqncnj/kLRLFoAqIxGoxGtVitardaRFzyg3pk86p3Jo96ZPO4vFK2RnrC3and3NwYGBuLZs2fR399f9FwAAPCf7O3txfT0dEREXL9+PZrNZskTUSX7+/vx8OHDiJCPiMPfj4WFhWi1WiVPBGeHJ1oAADgTVlZWYmlpKe7duxezs7OqezlkbW1NPsiUJMmhJ1pUO3MaFi0AANRekiQxNTUVOzs7sb+/H5ubmzE/P++HaSIiYnV1Ne7evSsf9JQkSYyPj0e3240XL15Et9uNiYkJyxZOzKIFAIDau3PnzpGzNE3j/v37JUxD1fTKgXxwIOv6MTc3V8I0nAUWLQAA1N7GxkbP8+3t7fc8CVW0tbXV81w+iMi+fmSdw3EsWgAAqL2RkZGe50NDQ+95EqpoeHi457l8EJF9/cg6h+NYtAAAUHudTudILXyj0Yhbt26VNBFVMjk5KR9kyrp+dDqdkiai7ixaAACovXa7HYuLizE4OBjNZjOuXLkSs7Ozce3atbJHowLGxsZiZmZGPuip3W7H8vLyodahBw8exM2bN8sejZqyaAEA4ExK07TsEagw+eBtaZoe+oKTsmgBAKD21DuTR70zedQ7U7Rm2QMAAMBpZdWzLi0txWeffVbCRFTJ0tLSkbPzno9Xr16VPUJl5NU7t9vt9z8QtWfRAgBA7WXVsD558iQePnz4nqehap48eZJ5Lh+od6Zo3joEAEDtZdWwXr58+T1PQhVl5UA+iFDvTPEa6Qk/5Wd3dzcGBgbi2bNn0d/fX/RcAADwzpIkiYmJiUMfYNloNOKnn36KL7/8ssTJqIKVlZX46quv5CNDX1/fkXrj8yTr+qF5iJOyaAEA4ExIkiTm5uZiY2MjRkZGotPp+CGJ1+SDPPJBkSxaAAAAAAriM1oAAAAACmLRAgAAAFAQixYAAACAgjRP+gcPPtpld3e3sGEAAAAAquqDDz44tqXrxIuWP//8MyIihoaGTvpXAAAAANTGzs5OfPjhh7m/58SLlsuXL0dExNbWVgwMDJz0r+GM2t3djaGhodje3tZKxRHyQR754DgyQh75II98kEc+yHOQj76+vmN/74kXLRcu/P/HuwwMDAghmfr7++WDTPJBHvngODJCHvkgj3yQRz7Ic9zbhiJ8GC4AAABAYSxaAAAAAApy4kVLq9WK77//PlqtVpHzcEbIB3nkgzzywXFkhDzyQR75II98kOe/5KORHvQ0AwAAAHAq3joEAAAAUBCLFgAAAICCWLQAAAAAFMSiBQAAAKAgFi0AAAAABbFoAQAAACiIRQsAAABAQSxaAAAAAAryf+3CPztroosIAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1245x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1245x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABCcAAABlCAYAAAB+6onQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAMfUlEQVR4nO3dPU8cZ9sG4GsfEzZNQLIUdyBk4YqKKLbpErmPjc1H/kBKKksWZLdIZ2wn2i49KRFgr1xFLqKkQSvi1lWQFUGKCCkree0Gy9E8xSp+ITAJDLzcz3qOQ0rh8YcuiZOb2Su7c1ayLMsCAAAAIJH/pB4AAAAAKDfLCQAAACApywkAAAAgKcsJAAAAICnLCQAAACApywkAAAAgKcsJAAAAIKnCy4ksy6LT6USWZac5DwAAAFAyhZcTL1++jMHBwXj58uVpzgMAAACUjI91AAAAAElZTgAAAABJWU4AAAAASVlOAAAAAElZTgAAAABJ9aUeAOhtWZbF69evU48BJLb3LOjv749KpZJ4IlKTAwCOw3ICOJHV1dWYm5uLdrsd58+fj48++iguXryYeiwAEms0GlGtVlOPAUCP8LEOoLBmsxmzs7Oxs7MTb968iZ2dnfj+++/j+fPnqUcDAAB6iHdOAIXdvXv30Oubm5vxxRdfnPE0QEp//vln/PTTTxER8cknn8S5c+cST0QKe3MAAMdhOQEU9uzZs0Ov//bbb9HX53iBsjp37pwzAAA4Fh/rAAobGxs79PrQ0NAZTwIAAPQyywmgsFqtduBJ7JVKJaanpxNNBAAA9CLLCaCwycnJWF5ejgsXLkRfX1+Mjo7GwsJCXL16NfVoAABAD/GBUOBUZVmWegQggY2NjVhdXY12ux1PnjyJ6enpmJiYSD0WZ2xvDjY2NqJer8fk5GTqsQDoAZWs4CuJTqcTg4OD8eLFixgYGDjtuYAe0Gw24+bNm/uuVSqVmJ+f96IESqTVasW9e/f2XXMWlE9eDh4+fGhBAcC/8rEOoLDDqkSzLIu1tbUE0wCpHPY97ywon7wcLC4uJpgGgF5jOQEUllclur29fcaTACltbW0det1ZUC55Ocj7WQEAe1lOAIWpEgUiIoaHhw+97iwol7wc5P2sAIC9LCeAwlSJAhERU1NTzgJyc1Cr1RJNBEAv8UBM4ERWVlZibm4u2u12jIyMxMzMjCpRKKH19fVYWlpyFpTc3hyMj49HvV6PGzdupB4LgB6gSpTCsiyL169fR0REf3//gf9bQjmpEgUinAV0MyAHUF5eK3Bc3jlBYbu7u3H79u2IiGg0GlGtVhNPxFlTJQpEqBKlS5UosJfXChyXZ04AhakSBSJUidKlShSAk7CcAApTJQpEqBKlS5UoACdhOQEUpkoUiFAlSpcqUQBOwnICKEyVKBChSpQuVaIAnITlBFDY5ORkLC8vx4ULF6Kvry9GR0djYWFBfSCUzMTERNy5c8dZUHJ/z8Hly5fj0aNHqkQBOBLLCeBUqY0DIpwFqBIF4HgsJ4DCms1mzM7Oxs7OTrx58yY2Nzfj/v370Wq1Uo8GnKFWqxUPHjxwFpTc33Pw9OnTuHXrVjSbzdSjAdADLCeAwlSJAhGqROlSJQrASVhOAIWpEgUiVInSpUoUgJOwnAAKUyUKRKgSpUuVKAAnYTkBFKZKFIhQJUqXKlEATsJyAihMlSgQoUqULlWiAJxEX+oBgN52/fr1+PHHHyMi4tq1a9HX51iBMrpy5Uq8evUqIpwFZbY3B41GI6rVauKJAOgV7hwo7PHjx7G6uhrtdjs2NjaiXq/H5ORk6rE4Y3tz8OTJk5ieno6JiYnUY3GGWq1WrK2txdbWVgwPD8fU1JQMlNDGxoazgH05cG8A5ea1AsdVybIsK/IXO51ODA4OxosXL2JgYOC05+J/XLPZjJs3b+67VqlUYnl5Oa5fv55oKs7a48ePY3Z2dt+1SqUS8/PzXpSURKvVinv37u27JgPlIwdE5OfAvQGUT9494sOHDy0oyHXkd07s7u7G7u7u2193Op3/l4HoDXfv3j1wLcuymJube/sWf959q6urB65lWRZra2tekJTE2tragWsyUD5yQER+DtwbQPnk3SMuLi5aTpDryA/EXFxcjMHBwbf/qQcrt7zO8na7fcaTkFLe13t7e/uMJyGVra2tQ6/LQLnIARH5OXBvAOWT932f9xoCIo7xzokvv/wybt++/fbXnU7HgqLExsbG4ueffz5wfXx8PBqNRoKJSGFjYyOePn164LqzoTyGh4fjl19+OXBdBspFDojIz4F7AyifvHvEsbGxBNPQK468nKhWq564zFu1Wi1u3boVex9ZUqlUol6vy0mJ1Ov1Q3MwPT2dcCrO0tTUVNy/f18GSk4OiMjPgXsDKJ+8e8RarZZwKv7XeSAmha2srMTc3Fy02+0YHx+Per2uy7yE9uZgZGQkZmZm4urVq6nH4gz91daxvb0dQ0NDMT09LQMltL6+HktLS86CktubA/cGUG5eK3BcqkQ5FVmWRcE9F+8YOSivv772MkCEHODeAMru+vXrbx+G22g0vIOKf+WdExSSVyWqHqhc8nKgPrA8VEgSIQd05eXAvQGU0+7u7ttnFlpOcBRHbuuAvfKqRBcXFxNMQyp5OTisTo530z9VSFIeckBEfg7cGwBwFJYTFJJXA6QeqFzyvt7qA8tDhSQRckBXXg7cGwBwFJYTFJJXA6QeqFzyvt7qA8tjeHj40OsyUC5yQER+DtwbAHAUlhMUUqvVolKp7LumHqh88nKgPrA8pqamZAA5ICLyc+DeAICjsJygkMnJyVheXo4LFy5EX19fXL58OR49eqQeqGT+noPR0dFYWFhQH1giExMTMT8/H5cuXYr3338/Ll26JAMlNDExEXfu3HEWlNzfc+DeAIDjUCVKYeqBOIzauPKZmJjQyMABzgJUiQJwHN45ARTWbDZjdnY2dnZ24s2bN7G5uRn379+PVquVejTgDLVarXjw4IGzoOT+noOnT5/GrVu3otlsph4NgB5gOQEUpkoUiFAlSpcqUQBOwnICKEyVKBChSpQuVaIAnITlBFCYKlEgQpUoXapEATgJywkK6+/vj0ajEY1GI/r7+1OPQwKqRIEIVaJ0qRIF9vJageOqZAUfo9zpdGJwcDBevHgRAwMDpz0X0CNWVlZibm4u2u12jIyMxMzMjPpAKKH19fVYWlpyFpTc3hyMj49HvV5XJQrAkagSBU5kb6XstWvXoq/PsQJldOXKlXj16lVEOAvKbG8O1IwDcBzuHIATefz4cayurka73Y4nT57E9PR0TExMpB4LOGMbGxvOAvblYGNjI+r1ekxOTqYeC4Ae4GMdQGHNZjNu3ry571qlUon5+XkvSqBEWq1W3Lt3b981Z0H55OXg4cOHFhQA/CsPxAQKu3v37oFrWZYd2nUPvLsO+553FpRPXg4WFxcTTANAr7GcAArL667f3t4+40mAlLa2tg697iwol7wc5P2sAIC9LCeAwvK664eGhs54EiCl4eHhQ687C8olLwd5PysAYC/LCaCwWq12aKf99PR0oomAFKamppwF5OagVqslmgiAXuKBmMCJrKysxNzcXLTb7RgZGYmZmZm4evVq6rGAM7a+vh5LS0vOgpLbm4Px8fGo1+tx48aN1GMB0AO8cwI4VQX3ncA7xllAlmVyAMCReecEUNhhVaIREQsLC+oDoURUiRKhShSAk+lLPQDQu+7cuXPo9e+++y4+/vjjM54GSGV1dfXAtSzLYnV11VlQInk5WFxctJwA4F9ZTgCFPX/+/NDrv//+e/zwww9nPA2Qyq+//pp73VlQHnk5UCUKwFF45gRw6nzGGMrl/Pnzx7rOuynv661KFICj8M4JoLCLFy/G5ubmgeujo6PRaDQSTASk8Omnn8bnn3++bzFZqVTi22+/jc8++yzhZJylvByoEgXgKCwngMK+/vrrQx+I+c0330S1Wk0wEZDCzMxMvPfee7G4uBjPnj2LsbGxqNVqKiRLRg4AOAltHcCJNJtNN6IAAMCJWE4AAAAASXkgJgAAAJCU5QQAAACQVOEHYv71aZBOp3NqwwAAAADvng8++CAqlUru7xdeTvzxxx8RETE0NFT0nwAAAABKYGdnJz788MPc3y+8nDh//nxERGxtbcXg4GDRf4Ye1+l0YmhoKLa3tz0YtcTkABkgQg7okgMi5IAuOSDi/3LQ39//j3+u8HLiP//pPq5icHBQ0IiBgQE5QA6QASJCDuiSAyLkgC45ICL+8SMdER6ICQAAACRmOQEAAAAkVXg5Ua1W46uvvopqtXqa89Bj5IAIOUAG6JIDIuSALjkgQg7oOmoOKtlfnaAAAAAACfhYBwAAAJCU5QQAAACQlOUEAAAAkJTlBAAAAJCU5QQAAACQlOUEAAAAkJTlBAAAAJCU5QQAAACQ1H8BwAgJB6L5cAAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1245x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "palette = [(0.691, 0.691, 0.691), (0.691, 0.691, 0.691), (0.691, 0.691, 0.691)]\n",
    "questions = [\"rating_quality\", 'rating_conversation_fidelity', \"rating_manip_sourire\"]\n",
    "\n",
    "with sns.color_palette(palette):\n",
    "    for question in questions:\n",
    "        df_analysis = rating_df.loc[rating_df[\"question\"] ==  question]\n",
    "\n",
    "        g = sns.catplot(x=\"rating\", y=\"question\", data=df_analysis, kind=\"box\", aspect=8.3,  height=1.5)\n",
    "        sns.swarmplot(y=\"question\", x=\"rating\", data=df_analysis, color = \"black\", size =5)\n",
    "\n",
    "        g.set(xticklabels=[], yticklabels=[], title='',xlabel=None, ylabel=None)\n",
    "        \n",
    "        g.set(xlim=(0, 8))\n",
    "        \n",
    "        plt.savefig(\"plots/ind_debriefing_results_\"+ question + \".pdf\", bbox_inches='tight')\n",
    "\n",
    "        #Test\n",
    "        print(\"\")\n",
    "        print(\"T-test results for \" + question)\n",
    "        print(ttest(df_analysis[\"rating\"].values, 4).round(5))\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "rating_quality\n",
      "31\n",
      "0.1935483870967742\n",
      "\n",
      "rating_conversation_fidelity\n",
      "31\n",
      "0.7419354838709677\n",
      "\n",
      "rating_manip_sourire\n",
      "31\n",
      "0.7096774193548387\n"
     ]
    }
   ],
   "source": [
    "palette = [(0.691, 0.691, 0.691), (0.691, 0.691, 0.691), (0.691, 0.691, 0.691)]\n",
    "questions = [\"rating_quality\", 'rating_conversation_fidelity', \"rating_manip_sourire\"]\n",
    "\n",
    "with sns.color_palette(palette):\n",
    "    for question in questions:\n",
    "        df_analysis = rating_df.loc[rating_df[\"question\"] ==  question]\n",
    "        print()\n",
    "        print(question)\n",
    "        print(len(df_analysis))\n",
    "        print(str(len(df_analysis.loc[df_analysis[\"rating\"]<=4]) / len(df_analysis)))\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:stim39]",
   "language": "python",
   "name": "conda-env-stim39-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "320px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}