notebooks/02_spindles_detection_multi.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multi-channel spindles detection\n",
    "\n",
    "This notebook demonstrates how to use YASA to perform **multi-channel sleep spindles detection** from a NumPy array (example 1) or a MNE Raw object (example 2). If you haven't done so already, check out the notebook on [single-channel spindles detection](02_spindles_detection_multi.ipynb).\n",
    "\n",
    "Please make sure to install the latest version of YASA first by typing the following line in your terminal or command prompt:\n",
    "\n",
    "`pip install --upgrade yasa`\n",
    "\n",
    "**Important**\n",
    "- The data must be a numpy array of shape *(n_channels, n_samples)*.\n",
    "- The sampling frequency `sf` must be the same for all channels.\n",
    "- A list of the channel names (`ch_names`) must be provided as well.\n",
    "- The unit of the data must be $\\mu V$.\n",
    "\n",
    "## Example 1: Using a NumPy array\n",
    "\n",
    "To illustrate the multi-channel spindles detection, we load a full-night 3-channels dataset (Cz, Fz, Pz) sampled at 100 Hz. The data is in compressed NumPy format (*.npz*)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 2161058) ['Cz' 'Fz' 'Pz']\n",
      "[[15.797 22.307 39.922 25.657 27.094]\n",
      " [16.896 26.385 40.966 21.833 24.456]\n",
      " [ 5.899 14.297 36.592 26.094 23.395]]\n"
     ]
    }
   ],
   "source": [
    "import yasa\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Load data\n",
    "f = np.load('data_full_6hrs_100Hz_Cz+Fz+Pz.npz')\n",
    "data, chan = f['data'], f['chan']\n",
    "sf = 100.\n",
    "times = np.arange(data.size) / sf\n",
    "\n",
    "print(data.shape, chan)\n",
    "print(np.round(data[:, 0:5], 3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To apply the multi-channel detection, we use the [spindles_detect](https://raphaelvallat.com/yasa/build/html/generated/yasa.spindles_detect.html#yasa.spindles_detect) function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Start</th>\n",
       "      <th>Peak</th>\n",
       "      <th>End</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "      <th>Channel</th>\n",
       "      <th>IdxChannel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>522.54</td>\n",
       "      <td>522.78</td>\n",
       "      <td>523.79</td>\n",
       "      <td>1.25</td>\n",
       "      <td>45.509445</td>\n",
       "      <td>10.850299</td>\n",
       "      <td>2.131626</td>\n",
       "      <td>0.470228</td>\n",
       "      <td>13.082397</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.190476</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>585.51</td>\n",
       "      <td>586.09</td>\n",
       "      <td>586.38</td>\n",
       "      <td>0.87</td>\n",
       "      <td>59.435380</td>\n",
       "      <td>13.185339</td>\n",
       "      <td>2.255297</td>\n",
       "      <td>0.473192</td>\n",
       "      <td>12.945464</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.659091</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>598.06</td>\n",
       "      <td>598.62</td>\n",
       "      <td>599.54</td>\n",
       "      <td>1.48</td>\n",
       "      <td>79.922226</td>\n",
       "      <td>15.976502</td>\n",
       "      <td>2.373750</td>\n",
       "      <td>0.421111</td>\n",
       "      <td>12.889143</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.375839</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>604.36</td>\n",
       "      <td>604.73</td>\n",
       "      <td>605.12</td>\n",
       "      <td>0.76</td>\n",
       "      <td>60.648920</td>\n",
       "      <td>12.520378</td>\n",
       "      <td>2.210118</td>\n",
       "      <td>0.350348</td>\n",
       "      <td>12.645687</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.480519</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>607.53</td>\n",
       "      <td>607.98</td>\n",
       "      <td>608.05</td>\n",
       "      <td>0.52</td>\n",
       "      <td>50.211401</td>\n",
       "      <td>13.846638</td>\n",
       "      <td>2.357938</td>\n",
       "      <td>0.258203</td>\n",
       "      <td>13.449947</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.849057</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1912</th>\n",
       "      <td>20498.20</td>\n",
       "      <td>20498.74</td>\n",
       "      <td>20499.22</td>\n",
       "      <td>1.02</td>\n",
       "      <td>83.897639</td>\n",
       "      <td>19.548310</td>\n",
       "      <td>2.609572</td>\n",
       "      <td>0.515605</td>\n",
       "      <td>12.320190</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.524272</td>\n",
       "      <td>Pz</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1913</th>\n",
       "      <td>20512.68</td>\n",
       "      <td>20512.95</td>\n",
       "      <td>20513.42</td>\n",
       "      <td>0.74</td>\n",
       "      <td>55.426790</td>\n",
       "      <td>11.739013</td>\n",
       "      <td>2.243399</td>\n",
       "      <td>0.355705</td>\n",
       "      <td>13.187207</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.360000</td>\n",
       "      <td>Pz</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1914</th>\n",
       "      <td>20524.99</td>\n",
       "      <td>20525.30</td>\n",
       "      <td>20525.61</td>\n",
       "      <td>0.62</td>\n",
       "      <td>66.723038</td>\n",
       "      <td>17.171995</td>\n",
       "      <td>2.614124</td>\n",
       "      <td>0.302493</td>\n",
       "      <td>12.776808</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.492063</td>\n",
       "      <td>Pz</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1915</th>\n",
       "      <td>20542.93</td>\n",
       "      <td>20543.36</td>\n",
       "      <td>20544.65</td>\n",
       "      <td>1.72</td>\n",
       "      <td>78.694386</td>\n",
       "      <td>14.282672</td>\n",
       "      <td>2.103757</td>\n",
       "      <td>0.287887</td>\n",
       "      <td>12.486788</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.248555</td>\n",
       "      <td>Pz</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1916</th>\n",
       "      <td>20556.28</td>\n",
       "      <td>20556.62</td>\n",
       "      <td>20557.17</td>\n",
       "      <td>0.89</td>\n",
       "      <td>69.733264</td>\n",
       "      <td>17.254953</td>\n",
       "      <td>2.617840</td>\n",
       "      <td>0.490823</td>\n",
       "      <td>12.388176</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.377778</td>\n",
       "      <td>Pz</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1917 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Start      Peak       End  Duration  Amplitude        RMS  AbsPower  \\\n",
       "0       522.54    522.78    523.79      1.25  45.509445  10.850299  2.131626   \n",
       "1       585.51    586.09    586.38      0.87  59.435380  13.185339  2.255297   \n",
       "2       598.06    598.62    599.54      1.48  79.922226  15.976502  2.373750   \n",
       "3       604.36    604.73    605.12      0.76  60.648920  12.520378  2.210118   \n",
       "4       607.53    607.98    608.05      0.52  50.211401  13.846638  2.357938   \n",
       "...        ...       ...       ...       ...        ...        ...       ...   \n",
       "1912  20498.20  20498.74  20499.22      1.02  83.897639  19.548310  2.609572   \n",
       "1913  20512.68  20512.95  20513.42      0.74  55.426790  11.739013  2.243399   \n",
       "1914  20524.99  20525.30  20525.61      0.62  66.723038  17.171995  2.614124   \n",
       "1915  20542.93  20543.36  20544.65      1.72  78.694386  14.282672  2.103757   \n",
       "1916  20556.28  20556.62  20557.17      0.89  69.733264  17.254953  2.617840   \n",
       "\n",
       "      RelPower  Frequency  Oscillations  Symmetry Channel  IdxChannel  \n",
       "0     0.470228  13.082397          16.0  0.190476      Cz           0  \n",
       "1     0.473192  12.945464          11.0  0.659091      Cz           0  \n",
       "2     0.421111  12.889143          19.0  0.375839      Cz           0  \n",
       "3     0.350348  12.645687           9.0  0.480519      Cz           0  \n",
       "4     0.258203  13.449947           6.0  0.849057      Cz           0  \n",
       "...        ...        ...           ...       ...     ...         ...  \n",
       "1912  0.515605  12.320190          12.0  0.524272      Pz           2  \n",
       "1913  0.355705  13.187207          10.0  0.360000      Pz           2  \n",
       "1914  0.302493  12.776808           7.0  0.492063      Pz           2  \n",
       "1915  0.287887  12.486788          20.0  0.248555      Pz           2  \n",
       "1916  0.490823  12.388176          11.0  0.377778      Pz           2  \n",
       "\n",
       "[1917 rows x 13 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Apply the detection\n",
    "sp = yasa.spindles_detect(data, sf, ch_names=chan, multi_only=False)\n",
    "\n",
    "# Get the full detection dataframe\n",
    "all_spindles = sp.summary()  \n",
    "all_spindles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the [summary](https://raphaelvallat.com/yasa/build/html/generated/yasa.SpindlesResults.html#yasa.SpindlesResults.summary) method, we can easily extract the number of spindles detected per channel, as well as the mean spindles properties per channel:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Count</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cz</th>\n",
       "      <td>670</td>\n",
       "      <td>1.041463</td>\n",
       "      <td>78.698054</td>\n",
       "      <td>17.165940</td>\n",
       "      <td>2.428411</td>\n",
       "      <td>0.404349</td>\n",
       "      <td>12.655632</td>\n",
       "      <td>12.856716</td>\n",
       "      <td>0.519365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>551</td>\n",
       "      <td>0.986261</td>\n",
       "      <td>70.968251</td>\n",
       "      <td>15.504383</td>\n",
       "      <td>2.328553</td>\n",
       "      <td>0.382090</td>\n",
       "      <td>12.483724</td>\n",
       "      <td>12.018149</td>\n",
       "      <td>0.516992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>696</td>\n",
       "      <td>1.036351</td>\n",
       "      <td>70.985229</td>\n",
       "      <td>15.399090</td>\n",
       "      <td>2.326625</td>\n",
       "      <td>0.396186</td>\n",
       "      <td>12.663587</td>\n",
       "      <td>12.811782</td>\n",
       "      <td>0.513046</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Count  Duration  Amplitude        RMS  AbsPower  RelPower  Frequency  \\\n",
       "Channel                                                                         \n",
       "Cz         670  1.041463  78.698054  17.165940  2.428411  0.404349  12.655632   \n",
       "Fz         551  0.986261  70.968251  15.504383  2.328553  0.382090  12.483724   \n",
       "Pz         696  1.036351  70.985229  15.399090  2.326625  0.396186  12.663587   \n",
       "\n",
       "         Oscillations  Symmetry  \n",
       "Channel                          \n",
       "Cz          12.856716  0.519365  \n",
       "Fz          12.018149  0.516992  \n",
       "Pz          12.811782  0.513046  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp.summary(grp_chan=True, aggfunc='mean')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since version 0.4.0 of YASA, it is possible to create an interactive plot of the detected spindles. For this, we need to enable the interactive mode of matplotlib (make sure you have installed and enabled [ipywidgets](https://ipywidgets.readthedocs.io/en/latest/user_install.html))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %matplotlib widget\n",
    "# sp.plot_detection();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we can also plot an average template of all detected spindles with the [plot_average](https://raphaelvallat.com/yasa/build/html/generated/yasa.SpindlesResults.html#yasa.SpindlesResults.plot_average) method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEyCAYAAADQnlYUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAC6WklEQVR4nOy9d5wkZ33n/34qdZ48m6N2lQNCCQQIkAQGDNjAGRsO28Bhc5zP93M8g33BgCM+DA5gggPgABiTjQ0mCSGEkLQrrbQ5p5mdPJ0r1/P8/qiamZ6Z7plB2tWslnq/Xv2anu6qrqd7putT3yyUUqSkpKSkpDxRtNVeQEpKSkrK05tUSFJSUlJSnhSpkKSkpKSkPClSIUlJSUlJeVKkQpKSkpKS8qRIhSQlJSUl5UmRCklKytMIIcTvCCH+5gnu+0IhxFDL76eEEC86f6tL+VHFWO0FpKSsFCHEd4BnAOuUUt4qL2dVUEr94WqvISVlIalFkvK0QAixDbgDUMBPXIDXTy+qUlKeIKmQpDxd+HngB8DHgTcCCCEyQoiKEOK6mY2EEINCCEcIsSb5/RVCiD3Jdt8XQtzQsu0pIcTbhRCPA00hhCGEeIcQ4rgQoi6EOCCEeHXL9roQ4k+FEJNCiJNCiF8WQqgZERJCdAsh/lYIMSKEGBZC/L4QQm/3ZoQQtwkhdgkhakKIMSHE+5LHtyWv+VYhxLnktX6zZb93CiH+ccG2bxRCnEnW9b9ats0JIT4uhCgLIQ4At3b6cIUQWst7nxJCfEYI0ffD/IFSfnRJhSTl6cLPA/+U3F4ihFibuLc+D7y+ZbufBu5VSo0LIZ4J/B3wX4F+4CPAl4UQmZbtXw+8HOhRSoXAcWLLpxt4F/CPQoj1yba/CLwMuBG4CXjVgjV+HAiBncAzgR8DfqHD+/lz4M+VUl3ADuAzC56/E7g8eY23LxPLeB5wJXA38H+FEFcnj/9u8to7gJeQCHAH/kfyfl4AbADKwAeX2D4lZQ6lVHpLbxf1jfhEGQADye+HgF9L7r8ION6y7f3Azyf3PwT83oLXOgy8ILl/Cvgvyxx7D/CTyf1vA/+15bkXEbvaDGAt4AG5ludfD9zT4XW/SyxUAwse35a85lUtj/0J8LfJ/XcC/7hg200t2z4EvC65fwJ4actzbwWGWn4/BbwouX8QuLvlufXJZ26s9t8/vV38t9QiSXk68Ebg60qpyeT3TzJ3dX0PkBdCPCuJo9wIfCF5bivwG4lbqyKEqACbia+4ZzjbeiAhxM+3uMIqwHXAQPL0hgXbt97fCpjASMu+HwHWdHhPbwGuAA4JIR4WQrxiwfOtr316wZoXMtpy3waKHdZ7eonX2Ap8oWXtB4GIWCBTUpYkDTCmXNQIIXLE7ipdCDFzwswAPUKIZyilHhNCfIb46n8M+IpSqp5sdxb4A6XUHyxxiNn210KIrcBfE7uIHlBKRUKIPYBINhkBNrXsu7nl/llii2RAxS6yJVFKHQVeL4TQgNcAnxVC9C947UPJ/S3AueVesw0jyevsb3mdTpwlts7ufwLHSfkRJ7VIUi52XkV8ZXwNsbVxI3A1cB9x3ARiC+VngDck92f4a+BtibUihBAFIcTLhRClDscqEAvLBIAQ4s3EFskMnwF+RQixUQjRA7x95gml1AjwdeBPhRBdSfB6hxDiBe0OJIT4WSHEoFJKApXkYdmyyf8RQuSFENcCbwb+ucOal+IzwG8LIXqFEJuI4yCd+DDwB4mYziQt/OQTOGbKjyCpkKRc7LwR+JhS6oxSanTmBnwAeIMQwlBKPQg0iV05X53ZUSm1izhA/gHi4PEx4E2dDqSUOgD8KfAAsXVzPXHMZYa/JhaLx4FHgX8nDq5HyfM/D1jAgeR4nyWONbTjpcB+IUSDOPD+OqWU0/L8vcl6vwW8Vyn19U7rXoJ3EbuzTibr/ocltv1z4MvA14UQdeIMuWc9gWOm/AgilEoHW6WkPBGEEC8DPqyU2noeX3Mb8YnfXImLLCXlYiC1SFJSVkhSl/HjSb3JRuL02i8st19KyqVOKiQpKStHELuLysSurYPA/13VFaWkXASkrq2UlJSUlCdFapGkpKSkpDwpUiFJSUlJSXlSXBIFiQMDA2rbtm2rvYyUlJSUpxW7d++eVEoNPtnXuSSEZNu2bezatWu1l5GSkpLytEIIsVTbnBWTurZSUlJSUp4UqZCkpKSkpDwpUiFJSUlJSXlSrHqMJJkgtwsYVkq9QgixHfg08SCi3cDPKaX81VxjSkrKjzZBEDA0NITruqu9lCdENptl06ZNmKZ5QV5/1YUE+BXiCuGu5Pf3AO9XSn1aCPFh4rkNH1qtxaWkpKQMDQ1RKpXYtm0bQojld7iIUEoxNTXF0NAQ27dvvyDHWFXXVtLa+uXA3yS/C+Au4q6pAJ9g8TjTlJSUlKcU13Xp7+9/2okIgBCC/v7+C2pNrXaM5M+A32JuDkM/UGnpejoEbFyFdaWkpKTM4+koIjNc6LWvmpAko0XHlVK7n+D+bxVC7BJC7JqYmDjPq0tJSUlZmtHRUV73utexY8cObr75Zn78x3+cj370o7ziFQunJj91vPCFL1yVmrrVtEieC/yEEOIUcXD9LuLhOj1CiJnYzSZguN3OSqmPKqVuUUrdMjj4pAszU1JSUlaMUopXv/rVvPCFL+T48ePs3r2bP/qjP2JsbGy1l7YqrJqQKKV+Wym1SSm1DXgd8G2l1BuAe4CfSjZ7I/ClVVpiSsqTxvZDqnaw2stIOc/cc889mKbJ2972ttnHnvGMZ3DHHXfQaDT4qZ/6Ka666ire8IY3MNNh/d3vfje33nor1113HW9961tnH3/hC1/I29/+dm677TauuOIK7rvvPgA+/vGP85rXvIaXvvSlXH755fzWb/3W7LG+/vWvc/vtt3PTTTfx2te+lkaj8RS++8WsdoykHW8Hfl0IcYw4ZvK3q7yelJQnjO1HTDa91V5Gynlm37593HzzzW2fe/TRR/mzP/szDhw4wIkTJ7j//nha8y//8i/z8MMPs2/fPhzH4Stf+crsPmEY8tBDD/Fnf/ZnvOtd75p9fM+ePfzzP/8ze/fu5Z//+Z85e/Ysk5OT/P7v/z7f/OY3eeSRR7jlllt43/ved2Hf8DJcDOm/KKW+A3wnuX8CuG0115OScr6QUmH70fIbplwy3HbbbWzatAmAG2+8kVOnTvG85z2Pe+65hz/5kz/Btm2mp6e59tpreeUrXwnAa17zGgBuvvlmTp06Nftad999N93d3QBcc801nD59mkqlwoEDB3juc58LgO/73H777U/hO1zMRSEkKSmXKpFUuKmQXHJce+21fPazn237XCaTmb2v6zphGOK6Lr/0S7/Erl272Lx5M+985zvnpePO7DOz/VKvpZTixS9+MZ/61KfO99t6wlyMrq2UlEuGIJK4YSoklxp33XUXnufx0Y9+dPaxxx9/fDa+sZAZ0RgYGKDRaHQUoZXw7Gc/m/vvv59jx44B0Gw2OXLkyBN+vfNBKiQpKReQUCqCSBLJdKT1pYQQgi984Qt885vfZMeOHVx77bX89m//NuvWrWu7fU9PD7/4i7/Iddddx0te8hJuvfXWJ3zswcFBPv7xj/P617+eG264gdtvv51Dhw494dc7H1wSM9tvueUWlc4jSbnYCCPJ2z/3OLdfNsArnrGerKmv9pJSniAHDx7k6quvXu1lPCnavQchxG6l1C1P9rVTiyQl5QJxfKLJ5x4Z5pGzZcLUIkm5hEmFJCXlAtH046BpzQkII7nM1ikpT19SIUlJuUDYXhxkrzkBQZRaJCmXLqmQpKRcIBpeYpG4IV6auZVyCZMKSUrKBcJucW25QSokKZcuqZCkpFwgmkkhYsUJcII0RpJy6ZIKSUrKBaKZuLaqdoDjh8tsnZKyPO1a1692MSKkLVJSUi4YM0ISKcVkPW3cmPLkmGld/8Y3vpFPf/rTADz22GOMjY1xxRVXrOraUoskJeUCMSMkAFNNH5nWkqQ8CTq1rv/GN77BjTfeyI033sjGjRt585vf/JSvLRWSlJQLRKuQVJ2AQKZxkpQnTqfW9e9+97vZs2cP3/nOd+jr6+OXf/mXn/K1pa6tlJQLRNOP0DWIJJRtnzBSZNJv3NOed/3rfg6cq53X17xmQxe/+8prn/D+Sil+9md/ll//9V/vOCflQpJaJCkpF4imF9JXiJWjYgdEl0Bfu5TV49prr2X37t1tn3vnO9/Jpk2bVsWtBalFkpJywWj6EUVLUjcUZTtApZ6tS4InYzk8Ge666y5+53d+h49+9KO89a1vBeLW9V/60pf45je/yT333LMq64JUSFJSLhi2F2JpimImpGL7KFKLJOWJM9O6/ld/9Vd5z3veQzabZdu2bdi2zfDwMLfdFg+W/Ymf+Ane/e53P6VrS4UkJeUCUQ+aNHs+SJ4fp2x3kyZtpTxZNmzYwGc+85nVXsYi0hhJSsoFwpZjeMYZ9Pxxqk7ApTD7JyWlHamQpKRcINyoCYBu1qg5YTolMeWSJRWSlJQLhBc5AAi9RihVOrs95ZIlFZKUlAuAH0r6tDEAcnJ09rGUlEuRVEhSUi4Ath+S0+KiNVfElonjpRZJyqVJKiQpKReAph9h6HGMpCF8AJzUtZVyiZKm/6akXABsL8TQXADqmgRCHD8VkpQnjq7rXH/99bO/f/GLX2Tbtm2rt6AWUiFJSbkANP0ITXNmfxdGI52SmPKkyOVy7NmzZ7WX0ZbUtZWScgGwvRC0uRkkwqjTSGMkKeeZX/iFX5htIT84OMi73vWuVVlHapGkpFwAmn6E0nyEUigh0IwqTT9Y7WWlPI1xHIcbb7wRgO3bt/OFL3yBv/mbvwHg9OnTvPSlL+VNb3rTqqwtFZKUlAtA0wuRus+GMGLYNMgZUzR9f7WXlXI++Oo7YHTv+X3NddfDy/54yU06ubZc1+W1r30tf/mXf8nWrVvP77pWyKq5toQQWSHEQ0KIx4QQ+4UQ70oe3y6EeFAIcUwI8c9CCGu11piS8kRp+iGRFrIlDNCUIm9O4gWpRZJy/nnb297Ga17zGl70ohet2hpW0yLxgLuUUg0hhAl8TwjxVeDXgfcrpT4thPgw8BbgQ6u4zpSUHxrbiwi0kG5f0hdJIqOMF6QWySXBMpbDU8kHP/hB6vU673jHO1Z1HatmkaiYRvKrmdwUcBfw2eTxTwCveupXl5Ly5Gh4IYEmySsYjCI0o4afWiQp55n3vve97N27dzbg/uEPf3hV1rGqMRIhhA7sBnYCHwSOAxWl1Myw6yFgY4d93wq8FWDLli0XfrEpKT8ETS/E0ySWbjEYhYwYTYK0IDHlSdBoNBY9dvLkyVVYyWJWNf1XKRUppW4ENgG3AVf9EPt+VCl1i1LqlsHBwQu1xJSUJ0TD8/E0yIgMA1LhGx5+lPbaSrk0uSiytpRSFSHEPcDtQI8Qwkiskk3A8OquLiXlh6fqxVePeU3HkhaOHqaurZRLltXM2hoUQvQk93PAi4GDwD3ATyWbvRH40qosMCXlSdAIYiHJaCYFM48S4EbeMnulpDw9WU2LZD3wiSROogGfUUp9RQhxAPi0EOL3gUeBv13FNaakPCGaQRNMyGoW6HlgEk+mQpJyabJqQqKUehx4ZpvHTxDHS1JSnrY0/TqYkNMzyFwRnEn8ME3/Tbk0SXttpaRcAMKwAkBOz5LJ9gEgo0Y6tz3lkiQVkpSUC4CmqgAUrAJZMw9AFDVJdSTliaLrOjfeeCPXXXcdr33ta7Fte7WXNEsqJCkpFwBN1gEomCUyRgEAKW1kqiQpT5CZXlv79u3DsqxVKz5sRyokKSkXAhELScnqxrJiIVGhg0x1JOU8cMcdd3Ds2DE+/OEPz1a1b9++nTvvvHNV1pMKSUrKBUAQj9ktZLvIWiUAFA6KVElSnhxhGPLVr36V66+/nre97W3s2bOHhx9+mE2bNvHrv/7rq7Kmi6IgMSXlkkM45KXEyvZimXEDayG9NEZyCfCeh97DoelD5/U1r+q7irff9vYlt2mdR3LHHXfwlre8Zfa5X/mVX+Guu+7ila985Xld10pJhSQl5UIgHIpSoud6yOgieTAVkpQnTqd5JB//+Mc5ffo0H/jAB576RSWkQpKScp5RSqE0j4JUiFwP1kwPUuWnwfZLgOUsh6eS3bt38973vpf77rsPTVu9SEUaI0lJOc8EkUJpPkUp0XK9ZK0iAAo/jZCknFc+8IEPMD09zZ133smNN97IL/zCL6zKOlKLJCXlPBNEEqkFFJRCs0pYmj7zTGqRpDxh2rWR/9jHPrYKK1lMKiQpKecZP5SEWkg2BN3MYs66HII0RpJySZIKSUrKeSaIJEpEGOgIzSSjZwBQIkxbpKRckqRCkpJynvEjSSQkGhq6biF0E0sqEKlFknJpkgbbU1LOM34oiYTCQEPTLTAyZJRCEaYxkqcxT2dr8kKvPRWSlJTzTBApIqHQMNA1A4xsLCQiIpTpuN2nI9lslqmpqaelmCilmJqaIpvNXrBjpK6tlJTzTBBJQqHQ0BGaDnpskaCF2H642stLeQJs2rSJoaEhJiYmVnspT4hsNsumTZsu2OunQpKScp5xg4hAExhCA00HTcNScbA9FZKnJ6Zpsn379tVexkVL6tpKSTnPOGE8UtdAAxF/xSxAiQgnFZKUS5BUSFJSzjNOMCMkOoi4GNFSAqlFNL1gNZeWknJBSIUkJeU8MyskM64twEIghaTppXPbUy49UiFJSTnPzLi2dGG0uLY0IiFxg9QiSbn0SIUkJeU8MyMkpqbPWiSZGSFJYyQplyCpkKSknGdc3wFoadYIptAIhcLxU4sk5dIjFZKUlPOMH8Rjdg0xl12fEQahpnDDaLWWlZJywUiFJCXlPBMENgCm3iIkmkEoFH6YurZSLj1SIUlJWSErbY/hJ0KS0czZxyxhEoi4D1dKyqVGKiQpKSvk+HhjRZXpQRi7tixjTkgymkkgFF6QWiQplx6pkKSkrJCqG1B3lg+Wh6ELQMawZh/L6CZSCLwwDbanXHqkQpKSskKCUDJlL19QGEZx1lbGyMw+lk3u26F9YRaXkrKKrJqQCCE2CyHuEUIcEELsF0L8SvJ4nxDiG0KIo8nP3tVaY0pKK6FUTDf8ZWMlQRRbJJbeKiRxC28vSivbUy49VtMiCYHfUEpdAzwb+O9CiGuAdwDfUkpdDnwr+T0lZVVRShHK+OYGSwfMw0RIMnpu9rEZi8RNihVTUi4lVk1IlFIjSqlHkvt14CCwEfhJ4BPJZp8AXrUqC0xJaUEqECK+3/CWDphHMhaSrDk3SGjGIvETkUlJuZS4KGIkQohtwDOBB4G1SqmR5KlRYO1qrSslZQbHD6k6ARldp7xMnCRM3FeW0WqRxEISyTRGknLpsexgKyHELcAdwAbAAfYB31BKlc/HAoQQReBzwK8qpWpi5rIPUEopIURbh7QQ4q3AWwG2bNlyPpaSktKRv7nvJH/7vZP87RtvodJcRkikDxpYVn72MStxc6loZUJi+yF5K507l/L0oKNFIoR4sxDiEeC3gRxwGBgHngd8UwjxCSHEkzqDCyFMYhH5J6XU55OHx4QQ65Pn1yfHXIRS6qNKqVuUUrcMDg4+mWWkpCzLuapD3Qs5OWkTyKWD7ZGasUgKs49lzTkhkcvsL6XiyGh92e1SUi4WlrrkyQPPVUo57Z4UQtwIXA6ceSIHFrHp8bfAQaXU+1qe+jLwRuCPk59feiKvn5JyPqm7cVzkwEiV9T2ZJbeNpI8lFZo1JySWEVsnUrosJw+hVLihxI8k2ZbGjykpFytLxUge6iQiAEqpPUqpbz2JYz8X+DngLiHEnuT248QC8mIhxFHgRcnvKSmrykyAff+5GlKxpLUQEWCh0M25GEkmuS+Ut2z6cCQVXhDhR2k7lZSnB0tZJB9N4hefBj6llDpwPg+slPoeIDo8fff5PFZKypOl6c1ZJFIqpFJoHf59pQqwlEKbJySxdSKlx3Ieq1BKvFCmfblSnjZ0tEiUUs8EXkFc7/FZIcRjQoh3JBlWKSk/UjS8EE0obF9yZtpeUgykCjCVQmuJkWTMIgACD7WMcyuSCj+SeEHacj7l6cGS6b9KqcNKqXclRYM/D3QD3xJC3P+UrC4l5SLB9iO2dcee3iNjDeQS7qmIkIxSqJasrczsfZ/lmghHUoGCpp8KScrTgxXVkQghNGANcU1HgQ6ZVCkplypNL2Qg59KdjTg5ubyQWEqhm3NBecssAaBYvmnj949PcnxiZZ2GU1IuBpZMVBdC3AG8nri6fC9xvOTXlFLVC7+0lJSLh6YXYYiAnBHR9L3YauhARERGgW60VLZnism9YEkRAvire47jBBG3bOs7H0tPSbngdBQSIcRZ4DSxeLxTKZVaISk/kvhJKq6lR2T0iIa/dMBcighTgdDmvl5mEiNRBMu6tspOQM0JCKUkkgpd65STkpJycbCURfI8pdTpp2wlKSkXKU0vZG12P4b+AzLa62gE4TLpv7GQ6C2jdoWZJSMlECxbR1JzAmw/ouGG+KEkZ6W1JCkXNx2FZEZEhBD3wOL/faXUXRdwXSkpFw0NL+Ty3n/jwdw0W+wxfGf70jESITGkmCckGNk4AC/CJfcNIomdBNnHay5+JMmRCknKxc1Kmvn8Zsv9LPCfiFOCU1J+JGh4IUqPu/bq+d34tW3LxEgkBqCJllwWIzMrJEu5tir2XDB+rOaltSQpTwuWFRKl1O4FD90vhHjoAq0nJeWio+mFRFrcP2vaOowfKULZ+VoqEhJDCcQ8IcliKVBESNlZHFo7C4/Vbdy0liTlacBKuv+2po5owM3E9SQpKT8S1L0QXw8RSlHVG4T6CF7U+QQfCYWpBEK0uKSMDFklkVpIpDoLyXRLZ+HxuouTCknK04CVuLZ2E8dIBLFL6yTwlgu5qJSUi4mmF+Jqkttcn13ZLFppL03vNW23jaQiEgodDa214aIQWEpgi4hwCYtkqhFPUDR1yXjdW3LblJSLhZW4trY/FQtJSblYabghTU2xyQ0ZNjVOWtPU3fYzSYJIEgqFwQLXFvGXTQlJsIQ1M9300awxBks247UdpH0bU54OLDWP5HlL7SiE6BJCXHf+l5SScnHRbDSo6xpdUlKSIUJzqbntYyR+IiS60mCBkFhKIEWEt0TF+lTTJ7P+C0R9X6bqSGx/+Ur4lJTVZimL5D8JIf4E+Bqxe2uCOGtrJ3AnsBX4jQu+wpSUVcZtjBAIQQFBVxQgNI+a08EiCSWhAANtbsh7gomGFBIn7CwO5aaPZlYI9Liv17myC6lPIOUiZ6k6kl9LAu3/CXgtsJ541O5B4CNJG/iUlEsexx4GQLe6KckGpt6k5nrttw0D5KyQzLdIZoTEW8LKmG56aHqDQISgNxkqpzPeUy5+loyRKKWmgb9ObikpP5L4zihooOf6KTRq6JZDw2kvBrYfWyo6GgvH7ZhoRELhLCEkU04NjNj1pVmTjFcr5+U9pKRcSFbU/Tcl5VJCSkXVXnnswffHANAK6ylIBZqPE7Tfv+nHLildtHFtCY1QSJywc7B9yp2cva9ZE9hOc8XrTElZLVIhSfmRww0jDozUVlzsFwXTAOhdGygAoRbS8NrHSOwgdnkZ6G1cWzqRYEnXVtWfnr2vWxN4nr3saN6UlNUmFZKUHzlCqag4PmenVxZ/CGU8NaGYW0tez6EE2H77fZ1ESHR0Frq2LGEQCLXk5MNmWAbAVIruzFmaQbDsaN6UlNVmWSERQuSFEP9HCPHXye+XCyFeceGXlpJyYZBSkTN1hivO7Cz2pQhVA4DuwhoKIg4rNvxGW0vBCROLpI1ryxIGoQAnaH9MKRWeqgBwvedhWmO4QbRkX6+UlIuBlVgkHwM84Pbk92Hg9y/YilJSLjBRJNFCG1TckHHZ7WkilKI7109Jj6ceNqJm2xP8rEWiGYtjJJpBoIHbIb5ScwOEXseSius9H9ts4oZy2UFYKSmrzUqEZIdS6k8gnhGqlLJZaLOnpKwSUw2P4bKz5HyQhUi/Sb52El0TK+quGwqXvBLkskWKRg4AJ7DbupzsGYukTUKkpZnJvu1Th8t2gGWUGYgitgUBoQaOqiw7CCslZbVZiZD4QogcyUwSIcQOYgslJWXVKds+e4crPHq2vGIxOXSugtuoIohbmixHoPnkpcAy8hSMPABO6LS1FLwka8vUFs8QmRESN3A7vhfDrDIQRay3egEw1OkVWyTDZZuK3T4JICXlQrISIfld4ur2zUKIfwK+BfzWBV1VSsoKUdUhBkSDhhvhr7Ax1f/43FF+7xEfFTYJlxEfL4zw9JCs1MhYeQpWPDLXk257IQniILypLbZIMnpikcj212EV20foDfqjiIGBawDoFcdWLCR1N5zXhj4l5aliWSFRSn0DeA3wJuBTwC1Kqe9c2GWlpKwM2ZzGjGxArSgoHUaSihtxuFzguyePECxR0wHQ9CJcTZJVBqaRoZgISaDctq6tYFZIzEXPWUl8xQ3bC8l0M0AaTQaiiK7Shvh1RB0ZrUxI/Egy3Uh7c6U89XSsbBdC3LTgoZHk5xYhxBal1CMXblkpKStDunVME8iwrHUB84PrX9zj8rwrykBPx+2b9RoNTVBQJpaRQc90QwMC5bUVLi+MCwgt3Vr03MxjXichsR0i3WMwiih0b0FMKhQ+Uq6s3iUIFTXXJ4wkhp5m9qc8dSzVIuVPk59Z4BbgMeIg+w3ALuayuFJSVgcpkX4TNMWEO0oQdQGLLYFWak4sJJf31jlaLrHvXIUXXN55e6c2SU3T6I0sDCNDLhvHLiLptp106AdxjCTbRkgyRhYAr0OMpOKWQUB/JDGK68grhdR8wigAcku+L4Bxe5woMrCDPrpSIUl5Cun436aUulMpdSexJXKTUuoWpdTNwDOJU4BTUi4IKw2aq9ClGTocqx1i1D7Z0WXUSs0NeFf2A/ykdT8AlQ5dfGfwa+PUdA1Ty6IbGTJWF4ZSgEPQxlIIolhIMkZm0XMzj4XJNgspe1MADEQRer6PnIRIhImQLE/Fm8aXNs0OLe7b4QYRJyYaK94+JaUdK7lsuVIptXfmF6XUPuDqC7eklB9lyk2fE5MrO7E1bZsvnoayB4ZSHeszWqk3mvzD5lM8VPg+oGh2aHUyg+tMEwiBpWXRdQuRKZKXEoHTdnphGMbWRs5cLCRZPbZIoqh9VXzFj4WkL5IYxbXkFEQiWJGQjFQc/m1PHV81mG6sPKnSC+JJjGkblpQnw0qE5HEhxN8IIV6Y3P4aePx8HFwI8XdCiHEhxL6Wx/qEEN8QQhxNfvaej2OlPD04U7aprLCh4jcOnOPQkQn+4P6d7DmpsFcgJM7UScYMg0czPmb3LhpeuORJtObGJ/eclkXXTbAKFKUC4RDKxVf+YTQjJItdUTlrRkjaWyT1pM9WDwa6VSCLRqRFuP7ywvD1A6N873DAmZNHCU9/HxWtzCoJpKThhbhBOoox5YmzEiF5M7Af+JXkdiB57HzwceClCx57B/AtpdTlxKnG7zhPx0q5yGlWp5lueNj+0if3GUonv8HnM+/k1uwpvn/UwAuXP3lWp48AYClFZu2/UvMml+xl1fAqAGT1HLrQwcxTUBKE2zbjK4jik362nUVixjUoqkP6rx3VAShoFugZskoj1CIaHcb6tjLdjLc5MRESenXCJYZntRJGEscPse0mrNCFlpKykJWk/7pKqfcrpV6d3N6vlGofLfwhUUp9F5he8PBPAp9I7n8CeNX5OFbKRY6MKJ96DF06NIPGimpCspWjANxePIETKJxg+RNuo3YSgFfXGwjdpxKcXTJtuOnXAMgZeYQQYBUpSIXQvLZzRYJEJDJtLJJs8piS7b8+XmLNmEYOdJOcMAmE7DhEq5Xh+hjWwLc4Wcvjq5BohULiBhJdaLijh6E5saJ9UlIWsuRgKwAhxEmSqvZWlFKXXZAVwVql1Eyq8Siw9gIdJ+ViInSplCdhoJspz8UPN5MxFleHz8M5w7v7e/lK8R6M7HGqzh8texjXPgcmbE0aJ7pR+wr1GbwgSefNxPUjWHkKUoLmUfMXC1cYeRhKYSQZWq1kEotEyvaC50sXU4FmxkKSFSa+5nYcotXKseb3yQx+gzNnL8NVRVS0spRhN4goGAp7aggGUi9yyhNjWSEhTv2dIUs8drfvwixnPkopJYRo+y0XQrwVeCvAli1bnorlpFxAIt9FBA0azhAe+eV7YMmIY9o5/qWrREkqtMw4U832sYdWfH8cTNicCIkT2ktaJF4YB8YzM0Ji5ilKiTJ9Gk6bGInyyCiFbhYWPWclfbqUam9h+NIliwKzAEInq5l4qBW5tmrBFOgQGmMcr+W4tU38ph1ffuwcV3ZFDDhNpN9M50qkPCFW4tqaarkNK6X+DHj5BVzTmBBiPUDyc7zDuj6apCTfMjg4eAGXk/JUEHpNQjR2n56m7no4/jInwtAjInY73eEKhG4z1WguG1vxw9iTujlx/QSqfRrv3PaxOBXyydW6VaCgFFILqHuLLYVQ+mSUQmuX/jtjkaig7ToD6ZFTCmEVQNPI6VlcDRpLDMKaoRnF7yubPcPJKQe5AtdWEEk+9J3jfG7PSUIjh2unacApT4yVzCO5qeV2ixDibazMknmifBl4Y3L/jcCXLuCxUi4SpFNlbzPD3+4aZM9JqHrLxAWaEzhaiKYUO/wmQigm7fKybVICFQe0+7UMQsUWRLSEG2imLqSrkAiJGbu2Ij2g1iZ1OEiEBGNxjMScCbaroG1H30C65GUEVhGETkHP4QhBcwUxEj+ZY3J17kF+ZvTjRCuIF03UXRRwZCogMEv4Tp201XDKE2ElgvCnLfdD4CTw0+fj4EKITwEvBAaEEEPEDSL/GPiMEOItwOnzdayUi5vIq/GVM3FVeqUpqbnL5HNMHaOiaxSlYIPfALJU3CqhVCwVWvGFgymzaNl+cgoCXPwlLJJAuqBDX6EnfiBJ/420iJqz+AQfqVhIhJVf9FwmcXcpgsVBR0DhxBZJpgSaQc4sIEOB7dSW+CCS1xajhMAJy6A/miIIlp/1PlqN17/RPsmN3/4T6nf/cZy5ZSyuyk9JWYqVCMlblFInWh8QQmw/HwdXSr2+w1N3n4/XT3maoBTT5Rq7Rg2yGz7NhPssau61S+8zfZKypmEpi76kMLDq1Za2SEIfVwvIygKy2E+XHGEMj2ApiyTJwioapfgBqxAH24GKs9gVFBJiKYVmLQ62WzNCokKkUugLx/oohxwSkSmCpsedhh3wkkLFTiil8PUm3VFEVdcpGy59wfLxotFqLNY/rd+LETmIyhmI/FRIUn5oVhJb++wKH0tJeWKELnt3fZcXFD6H2b2HmnYIL/SXnBUip05Q1nVQRXoSIaj79bbV5rPUz1HVNCyVQZY20B2FHQsLZ5gRklLS9RfdIp9oVcNvIyQqIKsU6IstEs3MYiiFEu1dW0K45KSadW0VM7F4+X6l83sCxpt1Aj3iDicWhpMZ8O3lrZjRuotByI/pu+IHmhOxkKyAlcxxSfnRYanuv1cB1wLdQojXtDzVRZy9lZKyJEqpuPZiOUKXq8f/jf4+F8gQiCpO4OKHErND88Fg8gQTmgWiQG9yUrPDJn4YAh2uqKvDlHUdXWWRXRspViSmaOB2mKEOEKgAQyksM/mXF4KCiF1wzWDxyTqatUgWCwlGFkspJBGqjXNLEQfbyXSBptOV7Y7X4Fc7rg/g5PQ5AG5yA75a0DicsbimMbbkPgDjNZdrur6GLR16QxDNiRUXJR4erbN9oEAhcyHDpSlPF5aySK4EXkHcY/uVLbebgF+84CtLeVqjlGL/uRrVlbQ78W0G5Tm+l49P0MKoMla3l0wBVuVTlHUdU8vTNTPnQ9r4S7QGicpnqGoaGnlE71ZKUqJpDu4SGU4BAaYSmPrcCTOfzBppRovjECFRnLVltenWq1tklEIRLrJIlFIgYmtGz5ZACEq5nvg1o/qS2WjHy0MA9Ec6A7LIIcvCrS+s813MaK3OqQ3f46PdvZxT/QhnGoL2fcBakVIx1fBW9rdN+ZGg4+WEUupLwJeEELcrpR54CteUcgkwUfc4PdWkO2fSnV+6tTsTh9mdM6jrGiVlUjWrjNYcoiVOnnr9LPWuPgZElkxhLZaM8KSNt4QoeFNnKOsaBVlEK66hJBXK9KgvUacRqBBLghBzX5XijHC1OemGIootEq2NVWRkMJVCES0SkiBSKC0gJyVaLs4Q60pSjqOoiVSgdzDuzlRji6QHg5IoMaZP4TeWF5Kz9WGUCY9luzgu81zfnIQ27rqF+JHEjxRjdZcNvcu3t0+59OlokQghZsbp/mchxF8svD1F60u5SPCWmSTYShBJjo436M5Zsz2glkKNHeAbhRxdUcRdYQZhVpmouZ0D54GDFzaRAgpmHopr6JaKEGfJxo1e+Sw1TcPScxjFNRSTNN6K3Tm9NhAhphLo+lwqWCGpWnfbdPGNiDCVQDPbiGdikUgREqn5n6cbRkRaRE6pWSGZmX0i1dLV9+caowD0kCFn5GloGoFdhaXiRUDFPQvAGTNiSPVjuZPgL2+ReKHE1AVVJyBYQUPJlEufpVxbB5Ofu4DdbW4pPyJUbJ89ZyornhPSKE9glI9RyhrUnPbFd634E8c4ZZpsDU122tMIETFpV/GCDuLlViknsZOiVUTkeumNIiJhL+mmqtgTKCHIJ0JSkpJQC6jYnVONQ2QsJC0z2PNGHP8IpL/ovUVCYiiB3mbULkYmFhKiRSLpeCGRiMhKhZaLG0cUkp+qwzTGGSadCbIyHqZVMvI0NIF066CWFn8viIUkEIqDZpF8WEU6S8djIL6omHJHwS9jn31s2e1TLn2Wcm39a/LzE522Sbn0iaTi8GidquPT9ENK2WXcVEDYmCDXOINQO5FK4QQReatzUDacOsGEbrKGHra6h6F7kLJbwQk8YHGrEdwqFS22EHryPWhag75ahBQ2zhIWScWvgQUFI49ZHKAkJQiYXCLDKRARhrLmC4mZB6YTIYHWfIJQSCwl0LU2xSx6BlNBICRywUm+5jkgIKsUWpKtlc/1AxApb8k6wYo/RX+oEGaWnp7tuOOPETSrIEPQ2/+9lFJEjM7+fszSwIOoPoIWhaB3/nvVHJ+KP0Z/pNN0anRvCzoeJ+VHg6Wytv6VNs0aZ1BK/cQFWVHKRcVw2abi2oy6J6nZ3SsTkto4horwGyNE2gC2v4SQKIWonGG6T2Ot6GN90gq+FlRxO6Witlgk/bluhLDpi0KwmjSWKGSsBA2woMsqoFtdFFXs+pmyKx338VGYSpvn2sqZBQghVD5SKbSWepBQSHRE3HJ+IbpBRikaIlrUur7qxi4lSwlI2qvk8wMASOEnY33bV1rWw2kuiyI0K09vIRYfL2zEQtKBmhOCWWZzEDBiZhjJBuABjUmQwZJCcu/REe45ZnP5ljJ1rRAH6PXujtunXPoslbv33qdsFSlPGX7kI5Uk26Y7bTsm6j6RqOBEZc5Wy2zsa5PW2kroEzh1pujhG3v28twbn0XFLjJQXNx7Kt7eQ9lDeAPrEeYg65JYjB1Vsdt01wXAqVBJTux9+QHQfHojiaY3mbI7F+LVkzbtvbkehG5QVPG/f9Xr7M7xhUJHR9NahaTYIiQL3o5QGEpD65C2bCiBRBItiF9UvaTLMPrs1X0m04OmFBKfKAroNI/ekWXWhQEiV2BNIXaHBVF9SSEZr7tIs8q2IERk+xnJxK1jZHMClqj0B/j3vaPsPmXwE2ua9OZLcVwlmwrJjzJLubbunbkvhLCAq4gtlMNKqZVVLaVcdJyonkAguLLvymW3VUpR91zK3ig502SkUSaM1mN0OEkC4DcIIskH9ioeHOnmym0T9OeXaE9ePkmN+KRaEhFFPYOpBL5eZbzewbpwypS1eA29+bUgbHpkBLrHVLNzsLiejNEZLCUxCC0Wt4bf6FjzEmgKM5ovJHqmhNVUhMpPBCFeSyQjIgEmAiHaf0YWcRxFMV9IajNCorRZi0ToOnml4nG7HURBKomvKmyIfIRVZG0xtkgC5RCFYQcbBkarDr7ZYFMjJMr0cCY7QYSIa0mWEKAwkoxU4s/xsekMO/I6kV1G71rfcZ+US5+VNG18OXAc+AvgA8AxIcTLLvTCUs4/dmAzYU8w4UwQyJV0h1XU/DoSSc4oUPWmafrLZG+5NfZMCh4ciS/Vj09Wlh5nO7af6cS66LJyiEwXA1JDmFWmGx2ExK1Q0TVQGn35fsj1zhYljjXKHZdWT65/1hRil1EpOWE3w2b7KYlK4QnQlY7RGgPIFMjNWAotsQ43mdeuK31+4KQFQwmkkEQLKsPrXuLaEjq0uMVyMp7bHoXtr90aQQOEpDeS6JkiA/keAELlLjnc6lR5klAP2RiGbCttBd3hkNETV7erztleXhgxFR3G7LuXw9NFPE0nbC6fapxyabOSFil/CtyplHqhUuoFwJ3A+y/sslIuBEP1IQzdQClFzVu+hYYfSar+BGcnDaaqOm7UoLzMzA9ZH+WbB0bpzoRoQnFyMj65BlEHISmfZiqxcHpz3WjZbtZFCs2sUOlU32FPU9Z0NJkjb2Uh3zfbJmXarXRcW11E6ErQm48tkq4k+8oNnfZZUaGLJ+J4R6tFglkgrySR8ua5qJpBnAproEEHi8REIxRyUfpvI0m7tdCh5VhZJYhERNih0NJOalkKSqJZRbqSIsYIF7lElfrx8hkANoUhm7q3AnDA6FnWIjkz3UDrvo/Mmq/x4qnPkx9/GOk1YYUz4lMuTVYiJHWl1LGW308A9Qu0npQLhB/5jDvjRF4Gx9MZbY4uv08oGTl3ik888i3+6dEH0DUYt5dID5URw6eO8kHnHfzvvi+zsRQwNK2IZNA5fbU5zmQS2O0urEdkulgfRQijQtX12lsyboUp3UBTeXJmNrZIZhs31tunKStFQ4vIRDq5bNw3qytp6+5G7qKYBQCBgyM0dGVgtAbPrSI5qRB4hC1X7w0/FlkDDRY2ZEww0ZBCESxoy9JMmiyawpgnQhmlEWgRUYeTu50M3spLhZHtomh1ARBqASrsnHhwth6n/q6VGuuLsVvqjNaFZk8u2Sbl4OgkWu4MQihuyn6HvtPfjP+2K6iIT7l0WYmQ7BJC/LsQ4k1CiDcC/wo8LIR4zYIeXCkXMU7o4IeS09MO41XJRHMaf5kGfZVGk8/vDTDX/BvT1tdRkWCquYSQRD7Ouf1oQlGMHiJa+wmGqwovcgk6Fcc1xhlJLIO+0mbI9bAx9BFGg5prt3c5uVWmNAODHDnDgkwXPYmrqBHU2hfvBQ6OEJjSSNJ3oWglTRGlQ9DmRK38Jp4QaBgYequQFMgpCXhELYHphp+4ttCXsUhU0hNsjmZikWS0+UKSQyMQEbLDFf+MJZOXcdpwKXlPgQiIvM4n93E3roYfxGB9PhaSMZFB+PUlGzc+OnoUzYhf97FMht7GacJQwgq6DUuplp0Xk/L0ZCVCkgXGgBcQzw6ZAHLEfbdeccFWlrIkkYw4VT2FF62ssrjpNxmtOIy5Z1AKxuseTrj0l/9fdp9i2tyPEBI9O8qJiTplbwljNPIxpg4D8NWsTd04jqtPMlJrEHZybTUnGNWyIA268gNQXMcG30EIRcUrt+3mK50KZV3HElksw4RsN73JCd0JG+1bq/hNbE1DVzrZpMV7xipiKEWg2lskntvA1wSaMOcH4jOxRYLw5sU6mrMWSecYiYlOJNSiepdmckWf0635FgkGgSYJO8RIpuz475FXEi3XTcGK626amiCyO8cu6uE4xUhg6Vk2dG2KX0sYiNBdsrr94PT+eJ1S8VgmQ0k2iOxJ8Fcy/8Tl7HRquVyKLNu6Uyn15qdiIT/KRDIiVCEZvUOKbJvtj5SPMNocpebXuLb/2vYFcC2MNafZM/E4Xx3+e376sv9Kf7CNqtOkO9M5bXP/8DRW74N06SVqUZ19E0e5fF0fQdShK2/o01M7xG628L2k17qWOceJyRphp7bjzUnGdBNd5Shmu6F7E31JkLgR1Nt2+QjtClVdYIk8lm5ApjBrkfiymYjCgs/Dr2MLga4MTD3ugyUyXRQdhau8tllR9aS+RF+YdmvmySmF0nwi5iyS5qxFYnS2SIROIEKcBanNTuKGWiQkwsTH7dj6perGvbHyUiFyvZiaiaF0GpqGbHaeY+LKKv1SEZk5+rK9gKCmaQgUOJ0TFkbdo+iayd12hXtyXSgmkNMnIXhmx31m9605KBTbBtoUmaY8rVlJ1tZ2IcT7hBCfF0J8eeb2VCzuR4GG3+CxicfYO7EXuUS2TCsjzREmnUnWFNZQ82qz/u6lmLCneWjyGwA8Ovk9TM1kcolCPIDTzb0Is8yPD7wATeYYdo/iS69z65L6CH3+Of4mv4NACIRSGLlznJ6s43Xq5OtMManp6CpPIZON+2YlotAM6m2ti8ip0NAgq+ViITEymLqFKTUC1WG+iN/E0QQ6JpmZDKxk2mGIS9jG5dJoxp+PvrABo1UgJyWI+bEfezbOodMpRmIInVBjkZDMxjqMzDwhyQoLV1M03PZCUpkREiXRk4wtS2ViIXE6WySBsuPKfrOApltktAKNmcPanQWork6y3be43gtp6hFnDAutfHzZHl1BJDlROcupytkVt9pJefqwEtfWF4FTwF8SZ3DN3FKeJEEUsG9yH5GKsEOb6SW++DMopRhtjs76wruz3Zxrnpvnq1+IF3k8PPowU94I63KbOVY7gBe6SwpJEEnK4jBCGdxQvIpBsRXfOkq1KbGDDu60sz8AYF/JYRCLZ3g+hdxpRiohjt/m5K4UyqlQ0cAUefJGIiRybr5IO5+67ddRArJGjsxMBbZVICMFkfLw20089BrYQkNTFuZMvCNToltGSLxFLUsAmm4cD9JFe4tEioCw5XOfydrShdnRtWUlXYRnAvMzzKQO543sfCHRsjga2F57IZmpP8lLhZ40ezRFlrqmoZYQhBCbbhnGY32FRsEoYc8YcR3+D+tencgY41pfcpmKreevW+sxKyeWjZFUHI9J9xy1YBp3iTY2KU9PViIkrlLqL5RS9yil7p25XfCV/Qgwbo8TqYi8madklThVO7WsVeKEDm7oYiZX1ZrQkFLG9QQdcEOXe4a+QX9mLT9z2S+gUBxtPMKU0+h4vLrtEuhT5OjD1AyuLW1FM2scmZrE7tTxdWgX08KgkhvlmcYGrvZ8ImucuqtoBm18/F4dEfk0dImlFciYZiIk8cnZi2yCNkFmO4zfa87IYhkzQlIkryDEa5+B5TewNYHAmtsnU6IoZTIlsY1guXGKtK4t6AJgzbi25gvJjEViiKVcW/HfreHNz6hyIwdTKXR9gUWiZ3GFoN6h9Us9iU2Y6LNz4i1tpnFjpe0+ABKHHhmgZ7pAMyhZJVwteS8dYisPDu+Ns7UClw16AU0ZPGB1kasejwP0S1zMnJoeQyCxgwb2yQc6HiPl6clKhOTPhRC/K4S4XQhx08ztgq/saYJUkmOVY+yf3M+Jyglqfm3ZbrcAgQw4Uz9DMRnhaukWTugsa5WU3fKiCmxd06l4lY77jDZHGXOGuSWzla31EbbktrCv/CBNL5y9El7IidFpCtYw1/tlTLfK5aV1AIzbZ6i4Ha4+x/Zxr7kFhOSy/Gau9n0iLcSW0+3FpzaCAlw9JKcXYzdVfg1dcibeYeMtyG5CKeykfXvBzGPOdNnNlMhJkAsyqWbxGzhCoIkM5kw8ySomjRudRXUdAHaSWKAvjF2ZBfJSJvUdc/s5iUViaUbnYHvS/LG54AreCx1yUiIWWiRGHiUEjQ7WYz2xSEyRgST2Y+ldNIS2dCdfzYkHe2VLoOl0Z7rw9dhSkG6Ndl0iD07EA7Qu82uIXD85UeKskcUMbVRjtGO2l5SSAxMn6M4VyYY1KtNnYOpY221Tnp6sREiuJ56I+MfMubXSPlwJE/YE5xrncCOXMWeMPeN7OFM/s+x+k84kkYwwFnSVPVs/u6QQjdlj8Xa1s+wa3TW731hzrON+j4w9AkAp2sB7d21gE1sYd4ep+zZ20F5Ijo6MEVo1rnMmuXXX++jxmiglaITT1Jw2QiIjVH2Ux/QeANblerkqiQN4YqT9FXVjlIYQKE1SMAtkDBOKg2QUGFIjok1art/ATs7RXZnCnKhm4iaMUvidXVuahhDZuYaK2a5YSHR3nmUxg5MMedL0BcObzGxikUjq7pxAzmTBme2GWiVYyXPNBZ97ENnxrHcjM0+EckZ8odG0J9q+XjNoYkhAz8726MqaXdQ1jdBptLUSvCACzaVLSoxMDwidvmwPoR7/vaRbbbvfcD1ew0a/hlYcpKjnqCTTtsLpsx2FpO67TDebfPYHNlF1inMqA251yaB+ytOLlQjJa4HLlFIvUErdmdzuutALezoQypCTtZN0WV1kjSxdVhe92V5O1U4xbo933E8qyZnanDUyQ9bI0gga1IP2KbZ2YGMHNq5X5w8eeDd/uuu9HJs4gKEZ+JHfNp1XKcVjE3sB+PLB6zg8ZbHnzBYAKt44Zaf9sfaOHicSMKh1kXOrPP/MPWhRCSeqMO22qYr3m2BPczppO9JdWsv2SKArENlzTNnO4iBrfYRyEq8omcU4CG4VQLfIK40QB2+hP92tUk/6bHVlWxpIZkqUpEQJj6iNu851ynECgJabi5FYBQpSokRA1EZ8ZoTEMBY0qjTzcfovMNFSoOklAWdTdBaSGZGx/flCEkqbvFRgZOcLSZLO63jtT7rNoElGCSI9NyskBatETdMIfadtu5MpuwFCUZQSPRu7tvpzPSg9WZPXvuHjhD0FCvpkRLa0kR4zj6cnQrpEIaMdeOwZGeWQ+Tv8YLzOmN9E6lmYPtnxc0p5erESIdlHPLc9pYVQhgzVhwijcDZeAXHMoifTw+Hpwx27yk47cTGgqZvYgc2nDn2KU9VTQOziOtc413a/ql8FBR9+9C9phjYFLceHH/9rQhkiEPHzC3BCh5OV45jhAFWnwIt2VhmvxnUDFX+SqQ7uj/FyXC9g9lzJaN/VrLFHMGQXHmVqXmOx9VM5jUAxZggssqhcLyLfx6ZIR8+OUrbbZEY1xpnWZ0ShO7ZIhIBsDyUpkMLBWdgvyq3SSIRksNgz93imRElFKOG3dW3Vm/GJWNPz6FpyoraKs7GOpr/4JOgltR2ZBYKPmSOXvP+pFiH2knhFdok07kwitPYC15aUTmyR6PPjMbmkeNLp0NLGDm1yUqHM7GyPri6rREPTiAKvrWUxnmSjFaVCz3WDptGX60XoLg2RjYWkjauv7JbRogw6kMn3sKE4iDRsAgxUc6KjRWL7Pofr9yO0kFPBORqhjS2sOKi/TKfhlKcHKxGSHuCQEOI/0vTfmEl7kl1juxiqD9Hdpn22oRkUrAIHpg7M9kKaQSnF2fpZ8maek9WTvOO+d/ClY1/iw499GKliF894c5xmsLjAa8we49D0QR4pH+JVa+/kpT13M+SM8PmD/0bWzDLRxv3R8Bucrp/GaW7h+dtcXnJlnev6LZS0KPujTCYnlXlISeQdAiDMbSE3eCWZ5hh5VSTUKriRi7+wLiTp3VQ2Q0paFxQH0AqDbArjdiexkCzYpzHOVJJ1NZDvmnM55XroVhKlObgLT/BOZU5IknkdAGS76JIhCHdxXAWoJ7EnQ8/OucOsmcJCNTsPpBU3Scm1Ml3znzBycfovUHHmTvB+sn3W7CwkM64td8HVe4RLTkk0c4GQJBZJELYv+HNCh5xSCDM326OrJ9uFrwn8wG5rkYw34ouHkpQY+bhb8Ew90ahWQHntZ5nU/AoZGa8/k+tla99WhO5xTu+Jg+cdUoCHa6M09HiS4jTnEAIagRf3El9BRXwkVRyzWWZ0cMrqsRIh+V3g1cAf8iOW/ltxK3iRh1QSO7CZcqY4NH2IA1MHyOpZenO9aB2yczJ6BkMzODB1YLYViVKKocYQ9aCOJjTev/v9hDLklZe9klO1U/xg5AcIIcgYGU5WT8676vcij4bf4PvD36XbyHO1dQM391zDJnOQB0e+hyUyVL3qorYnh8uHaYZNQnsLa0oOg2Y/N64TSHcd4/YIFddZ1Ak49G0UIxSkJChtI7/2CgSKDUIHo0rTc3EX9IqicgoA27TpMXsp5HsxShtYF7oIo07Vthc3bmyOM2rE8YfBfP/cCT7XR28UgebQaOPaaiQWRU+2r+UD76IUBaD5eG2qwBtuffbvMrdPcc6ysBe7+PxkfkkxmVg4i6aRTQoe626LRZIIyVKFpVYyB8YL5ycfRMojOyMILeSTNO8wan/CdUOHgozQrNxskL43ubixI7etZTHpVIBYSEgaWPZm4tThEb0AfvvYih3VKMr4GGZxkPVdsYv0mN6DsKc6Vrd/6uAXEVqI5m4F8wy3Hv4sTuACalkh8cKIx85WCKZOwsl7oXJmZYISevEt5SlhWSFpTflN0n4j4Kcv/NJWl7Jb5rHJx3ho5CG+P/x9Hhl7hINTB6l6VfpyfbPuLD/yO7YpyZt5AhlwePowk84k+6f2c7J6kt5sL189+VXG7XH+2+Wv4/XbXsbm4iY+c/gzhDKkaBWZcqaYcubqAGpejYbf4JHJvdxaugqBhqFrXFvcybB7lrGklXfdnzuxKaXYkwTa824fl5sj9GtZrujOEXlrmfRGsL1oUeaW6zTwzQrrQh1lGGQGrgDgSiERQjLRrC/y8VM+TQREZoPubD/rCusQvdtZH7hohk3ZthfXhDTGGU8C2X35FlHI99MrQ4RuM72w27CbWCRK0GW1nOCz3ZRkhNBCKs7iE1ozcQ1lW11HVpx9BbSNFXlJzKkrs7gSO5Ok8VZbTp5+5GJJhWl1HhpmJa6tYEE8S+LHjSAXWCSFbNKEsUM7G0868fu28rMWSX8+FhJX+W0tknLizixKCUntSU+2B4ARkUUEzfZBelmjOwJF7H5cX9wIwDGtG82e6Ni48ZHJ+wibl3F7ZiuRHpCrPI559BvxFEZ36RnxVcfn5OhRGmf3g5mD0f1QH1lyHxoTRCe/hzpxL5zbk1oyTwErsUgQQjxTCPH/hBCngN8DDl7QVa0yoQw5Uj5Cl9VFX66PnmwPvbleenO9FK3ivPTbP37oj3nr19/K3zz+N5yunV70Wl2ZLhpBg0NTh7BDm/5cPzW/xhePfZGbB5/J9Vhok0f56bW3M9oc5aHRh2b3O1I+gh3YRDJitDnKnvFHiVTEDcUbMJNsmWtKVwHw8LkHyRiZee4tN3I5Pn0YoXS+rP6St+z7A6776hu5SRtGemvxVJOqX6XuzT9JnRsZYdoM6KFEycii92wE3eJ6YvGYaE4uLkqsDnHc7EYISXdmgFKmAL1bGAxn2ruXCcIFJ6fmJBNaBqRBT+tVf2GAvtBD6M5iIbHL1DUNITPkMy1B7UwXhZnRue5iUXCTOpus0WItWKVZi2SmQrwVP7lA6M62EZLEReW0uHOC0CGjFKbZOdg+M5kyWGBhRCIgpxTaAoukkLjVQtn+YiWQDkUVgVmYjZH051qEpI0glN2Z/lwCktfvSroGj2pZhN9s69oKaNCvQkKrBFaewcIaAE7pBTRnGrzFn6FUEltOkpebeFMt/t++N7+WwUP/Em+wTLr7wbGT1Mv3ccZRKD0D+V6YPtFZHJpTqOHdnKhCmS5oji97jJQnT0chEUJckdSPHCKuaj8DiCRr6wNP2QpXgaH6EKEMsWZ6MnWoCThSPsKBqQNsKW3h3qF7eft33867vv+uRbUgXZkuenO9FMz4hPSt09/CDV3esO1loGkEVjdX6OtYm+njqye+CsRBd1M3OTB1gL2Te6l6VR4Yvp9t2UG6tYHZzKP11hq69SKPju0iZ+SY9qZng811r86Jygm6wj62iilGNt8Z7zPxIFkZnwRq/hTjzfkZQeeO/YARQ6dgDNBjZeIr3d5tXOPH21XcMnVvwYmtPsJ+I7667S+siz+7nq0MJtlQjaBGY2EtiTPNpGYiVI5865V4vp/eKEBoIdP2gqtcp0xD0xAqMxtvACDbQ0F2dlPZyRV9wZxvkcxkX1XbBLMDGbvIenOlRc9lZ7KvWtwnfuRiqaUtkmwiFHKBC1ISkJUKzZovWsXZQVXtA9mhdMgrBdZcjGSwEAuJR9C2HqSSZN1lhTU7jbErEZQJYcWWxQK3kFQSJZqsiQKiXC8YWQZzgwCcJRv36LInFs0lmWhOgYi4ygy4bfoIptT5XO4KMm4ZDn01yRBrLwoVt8bBsX1k9YCJUNLww3i9odNeHKSEycNMR1kmHcVk048FdvpE29dPOX8sZZEcAu4CXqGUep5S6i+BSz7FYtKe5HTt9OwXayn+/cS/kzfy/K9n/y8+9KIP8XPX/Bwnqyd55wPvZKw51nYfqST3Dt3LdQPXsUFCpOcYLtucbZo8O3sVRytHOV45DsSuMakkXuThS58T9dPc3nM1SsFM4pEQgmvzl3GscQQ3CJBSMtIcIZQhR6cPM+SMsdWJ/8wTV/4MTv+1FMZ2sd5aC0DZH2dyQZXx0NhjSCEoZNcwoLw4kNq7ncvqwwDU/PLsjHEgPnk0xjmU1DwMZPvjQsHeraxJhMQO622FZFro6CpHqfXkWxiYrW6fcsrzM8TcChVhoKnsXBovQK6HQnJCmmqTQODOuqlaMrCMDDNHbXiL3WGe8tGVoiubW/RcLnGR+S1uzVDGcQ7D6GyR5JPq82ihRaKF5JREmPMzxIrJmGLZSUiUS15KhJGftUj6cvH/ri/Ctu3na0mhZV6fK2LstmLxmRIGWtBcJCRVrwpCsS5yUNleMPP05WJ35NhMLVRzclHm1oGJ2Eq/UtXQgFy4njN6k6nSZtSZB2LXWwe33ZGp0+Q8h5yRpyHrVJrJa1sFmDq+WIDsSQKnxpmaoidvUXVCPC0bu8/apazPICWM7oXJY8jaGNRHoUN91fmi5gZU7UtnYvlSQvIaYAS4Rwjx10KIu+nUie4CIIR4qRDisBDimBDiHU/FMe3A5lD5ED3Zno5B9BnG7XEeHHmQu7feTdbIUrSKvPyyl/O/n/2/sQOb3//B71PzF//zHp4+zLg9zvPX3w6Bw2gjwg0k3XmLq3PXkdEsvnryq7PbF6wCRas46/K6oesaNkw8yrMf+CNu/sEf44QOVxevIFAhe0b30JXp4mTtJKdrpzlTPkKgQp7lVDhlbsbIr2Wq/yYy9TPcnPNQYZ6Kd44puzmXMhv6TLlxBlZ3YZBSrieubejeRN6ZQoRFmlGVamtRYuiBPckJzQIl6Mv1xjGk7i30J80aXdmg0WrFBA4EDlUNdJEl1+rSyffPNm6s+rX5sRW3Qk0z0LDmKtQBsl0UEsGptgn6usmJeJ6bSgiyYiYdd/GJI1ABphLkLXPRc/mZoHmLkATSw1IKw+ocbC8k8ZawZT+lFJEIySmFvsD66SnEQhLSvkYjwiO/IGtrpg+bp4VtU6Frfh1NQUbPQiJ6MxdOVU1HyBAWWGjjSSfhTVETcn1g5jA1k4woUJ35MzSnFgnJ3rFTAFwTThNoFgVtI9IcZzi/Ia5ul7LjSXu4OkkpamCaXbiyQcVJPjMzH4tDq6URuDBxmKrMESmFoQs0ATUniGMx1aG2x4g/kGGonWNs6CjH930feW4PjD5+wVKT/VCyd6jK/pElumJfSEIPyqfbuiKfKB3PlkqpLyqlXgdcBdwD/CqwRgjxISHEj523FbRBCKEDHwReBlwDvF4Icc2FPCbAmfqZuA23ZsQf9sSRtpkfxyvHefcD78bQDF6y7SXxg4ELk8fZ6Xm84xn/nYpX4c93//miTrT3Dt1Lzshxa9cOQqmouSGmqXAjj658D8/KX8n3z31/US3JQyMPsi27hm2TZ7n9wMcwmtMM1s4gR0+yLbcZSxjsGnkQXdOxNIvTtdOMVeOCr1d4pzlVugYZ6TjrbwXgBTxG5G5g1Bmm6QVzJ8TApi7jOEu/USTftQ7WXQ/F2IIpRlkcVaHW+k/YHIfQ5ZwhsOiiaMUnGaw8fZluNAU+9fltUurxhMaGprBElowxV+FPvn+2cWPda8yvP3Er1DUNncz8YVOZrlmLpNYmRuIlMYa+3Px07Zngu9/m7+yrAEsKTHOxkOSSIsUZ9xdAKAMybTKvWiklFlEk56Y/uqEPIp7xoVsL1pdkYEVq8dx7qSSRCOIW8kZutpBxVkiERLbJYGsGDXKS+AIhsSZMzYybPSaxN+zJefucqcQW9vrQQeR7IXERFs0ebCP+H49rSeYL3rHpuDP11e4oteImepM42j5rEBHY4Ey2Dbj7kU+1Xuah4SInTlZ51QO/R+/jHyac6dlW6Ifp4zB5DOpjcPYhUJJJV5DRNc46I0Sax0Tdg0w31IZiN9pCAhcmjzCtipy2LSp0MUVPIlRLF0w2vJAD56p879gE47WVWTBKKY6Nx1M8/VAyXF4+/fm8Yk/Dqftg4jCc24M4T8bBSrK2mkqpTyqlXglsAh4F3n4+Dr4EtwHHlFInlFI+8GngJ3+YFwhliBd16LvUBjuwmbAn4jiGlPjjB/HqozB2kIY9xbHyMUIZ8vVTX+d3v/+7ALzzOe9kIDcAUQSTR8AtgwrZ6Xr84hU/w/6p/fzprj+lkVRIHy0f5YFzD/Cs9c8i61RwVIZqWEOh8JWPJuD23I1YmsmnD316dm3T7jRHK8d4VvdO+od+QNMocofzfmyVIdh7gD1jgiuyW9g3vRclFUWryGB2gGPl4+SUyfbIp7nmGThByPotVxAWN3CtuwfpbWDCG6Hp+7O9n6TXpKY1KEYaeTTyXesh2wX9O4F4NGsgKtitc84rsfuibEbktZ65inPdRCuuoVcKAq0x2xcKgFqceeNpEZaWw2wVklxfSwfg+jyLRDlxZbspFri2MqXZGInTJqXUU/EJbqCw4ESdpB+7so1FQoSpBHpLG5sZZlxUfksQPMTHUmpRLUgrM6nEkfJmpz+WnaQJpVJoCywSwyphKkVEsCjcMdPJIC8lemau+n4mFtfU4imPC7GjOkWpUC1CApAzSjS15CAL3J1na3Gnhj4p0QuDsy6x3mwvUrdxtDzKnlx04TVUH0FJnR3OKKp3B9uLcd3KAa0n3qA6vEi0IO5F5tSqfLl2D2OTn8AMPTad/Q/Ev/1abPUILU5drpyOrQddxzeL1N2QhnsS68zXsWsHsIMIL1KgZ2J3WCtKwdRRAgmnyh6ljEkxYzJUdvAzPbHF0yGrLJKK/cNVys2AnGGw91yVoemlh3sFkeTQaJ3Rmkd3zqQnZ3FiosnZaRu302iG80kUwOg+sIpQHAQZULDofNXzQ7DsYKtWlFJl4KPJ7UKyEWgdsjEEPGulO4cy5LGJx3ACB0Mz2Nmzk4HW4rU2nGuew9AMhBC4U8f53/s/yohfYWOmn2F3igiJqZkEMuCZa57JL934S7NXflTPQuihrC6aQYRm6DxfgHf1z/HxQ5/kt777W1zddzUPjz5Mb7aXn9r2CqgOM+ZJBs1eripsYdSfZNSbIJfp5yV9N/Gl0Qc4NH2Iq/qu4sGRBwG4rbSTwYnP8SXugEyRytrbefnYD/jkqZ/m5muuYN/ECQ5PHuGqNVeihS7HmufY7hs0VI7c+mspZU1KWYNw87PYcPjfUMbdSKI44N6YZjDfjzt9lgkzpCfqRlOKYin53PouA81ks5QcNaoEYVz4l7dMmIqv3GzDY72xiVKrn78wyGBwkjG9Ttlx54ZiNWKLJNR9ckZ+rh08xDGSxOR3Qhs/Cikk/6rKrWHnIdfafBEg2xWnsxJnqy3EI0Io6MvPj0HElkV9XqxjBp8IQ+lo2uLrLdMqYLqKQPhIJdGERiCDpBak8+Am08hjKJUIiUJHzGaMZZWM/f+t6AZZqYgICaMQS5uzjmaKXfNKxe6eBEMz0KWGLTTCBYkUEHdVHpQyTqdtceEWzRL12ZYn81vQD9ViK7UviuITeJL+Ppgf4IhxjrLRx9rm5KIU4Cl3Aj0qko08gsEruTE3yL/V4ag0UEJDlE/FbjQZzbrmAKpOk4fG9qN37+bbUcRtxZ9iTUHjWUc+Cfu/BDe8NhbBlrTxRtMnM/Eg1+3/ALnk4m1s3fOwB3+XTL4LGmOxQM7sUz0LtXNMyS6kchCawpc+UulMNUPW53KxV2LTLYuacJ6rONh+xEDBwqyfpVAb4uyoS8+tL6ZYWNAJgdgS2X+uRtX2GShYCCHQBfTkTU5MNDgzbXPz1l6y5tID6p4U0ydB+mAktUmZbgxjhdP0lmFF6b8XI0KItwohdgkhdk1MzK/oHm4M44QOvbleslqGA6e+w6mj/x77Bdv0A6p6Vc7Vz8W9r3ybjx/+FMPeNHcO3kSXVeLH1z6L/2/zy7h7w/N407Vv4n/e+j/nRCRwoTGGpxU4PtngzJTNqSmXM9WAu7uu4J3PeSdbu7ayf2o/O3p28K7nvosBYSCBsuOxpbAWIQQDZg+gUJrG8wvX0pfp4Q9/8Id85LGP8I8H/pHLcmvYWR3HkAH/4j6H/++WHHLniylhs7O6j63ZzQjgoaHvA+A0JzjnT3OLXeUhcQ15M0sxE5+MjbVXo0uf7UH8e9kfZ7geN32snHqMM6ZBSfRQyBTQZtJy831QXMMO6SFExJRTxp5xM1ROEwKRYdNtdVGwWnpTFdeyPvQRRo2K48yN3G2M4QqB0iJyRjLpcIZc76xF4kl73nxz6VZxNIWlZTFb97GK5JP0X29h8Db0cYXCkDr5BRlV+cTV5LdJr/WRGEpDF4uvt/RMgayCUPqzrfhDgtgiWdibqwVhZshJRaT82UaRMzUsWSVmr/RbySoVdxpeYF3PDMPKSRlfZbZgKZOmJpDuYiHxlU1JRrH4tJy8S2YXnpZ81l5tXjB7vBlbDT2RRMv3QSJoG0qDCKPOuNaHaCMkjXCK7ije1lxzBbd2xeucjly80sbYSlAsKmY8Uz7No8F9FCNFTdf5oLeDY+tuwFl7Ezz6922D57XR41y3509oakX+sOc3Ob7+2awd/R7uuX3JB9kFI4/H8bnaCIwfJMr2MVJxKVoGY/4ER5onsQwYr7tIoxg3llwgql4YcWKiQW/OIFM5SjS9n2P+MGf805w5cD+qjTux5oSUmz59hcy8LFBD1+grZEDBgXPVZWfaezMp9EpB+TS+Pd2xFdMMfigJnHpsvSV1Q14o2T8yhTKXuOr5IfihLJKnkGFgc8vvm5LHZlFKzVpGN930TDXZnMBL/NVnKqfpsaegMowpI3p9h9POJEXPZqCxBdbfCIY1W7G+f3I/xUwRDY0fnPga3ynv59Ubns/PbL577oCRz3OkhHU3zL86cWu4MmCk6YCCUjb+SOtuFrc6zuWbb+Ttt7V4AhUw9RjTkeLWc7u4/NEPgNCYvObnWdO3gxFnCp8u/mDn6/jg6H3cc/YentV7Nb+4+cfoeuifqGglHlFX8X/WGzja9dTNQX4iuo9djTeyObOO3RO7eYP8eR4Y/i4KeI5XYU/uxTwLk3wmOWkMxG6q55sVPi1Npt0zTDs2TbfCqaH9jBsGW+QAxdKaufeaKUFpHVc0YtGeaEzh+B4UClA5xTkjB0LRm+udX8BX2sCakQAtX6PWjFur5NChMU4tudIvWIW5qYXJsXIINCXiVvItmUeuX0OJfrJ6BqPVItF0clpspXvSJYyiuRhKMovEUDqmNf9EnbOKCAmB8lFKzfuS+yIREn3x10RYeXJK4gsfKSVoEKkojpFkFl+RzqJnyCWxjZm5KRU3PolaSp93Yp/9OBSEREQLMrBaLRItO/+YBlbcb6vNCTdU8VArzNw811Z3totQTzpXu7W4liRJc552y5iREQ8dznXPWiRr84MILeCM1ssz7OOLhMSnwk4Zf6ZWcQ1r8wMQFanjUM+vJzt1NPbS+434RJ/wlRNfItJsPnRujDev38Dp8AhOcCtnL/9Zrvjeb8Ajn4Dn/I/Z7SMJpQOfIhIGr7H/L2cqPZg3buH/m3ic3N5/QO14ZtyiP/TgzA8g9CHfS9WVRErh4TLiTqAJjYlwkpIaoOGFdGVLMH4Itt4++7ep2gEKyDTOMj29l9N4WJqFyGYZrpxlzYlHGdh5K7RYsmemm4usjUiGjDhnCSKfnFHEc/o4OdFg59pSyzaKUEoyhs7ZaZsTkw2uWZ9DK+9jcuIAkwQEfTu4fOAaNpY2LvpbT9ZdDo7WscpH6PJtrGKWQCrGnQZnvTMo/fxowMUqJA8DlwshthMLyOuA/9xpY8+vs//kNzB6t4JSFBpjaF49dhNoGlqhlx4ZcjBs0D91EK1+mmbPZpwkgJk1smSEhV0+ycdPfYXLChv4qU0vnH8Q3QKvHF+hFObMabc2xHQI045DxlJ4IeS0HKYuqAWCfPUsDFw5l6/rNyF0cV2bq07/O2FhLVEYsfaRP6d2158zpinsULBJ5Pmdra9keN1z2CQKaGae3rG9fEV7Hpu7DSxdAILq5ju54/i/8PVxl+duupVPjf0rH7j/93i0fpwrVZFnO2f45qbrECoi402DMiDbi9JMbjXP8k/eesacM9TskKn6CKdrx6EExUwffV0t/5hGForr2Dl5COhnqlnGmfGHV4c4nIldYN1WL/nWgryeTQxGEcJwKNv2XJZKY5zJ5Mq9K1OYGzYFIAQiUyIvNXwc/JnGjTKK234AOSM3P0YCGJkCphQEwiOQ4TwhcYRAl8aieIee7SLbVIQqbvZotIiGLxS61NHaxEgw8+SlwtYCAhlhASEhmWViJBgZ8lLhM2eRVGbat6C3HYhlKUEoIqRaICQz43mlnC0snF2esLA1DdWmVbvEpktKhDVXxAjQn+0hSlxb0q2jyRCI/5YVr0Iu0lEIRLZnVoBmakmOaaW4It6eO57t+Si9yrYog5MbIGeaFAa2YcheXOqUc2sYHH0oFh97Cro2xMdWkoOTJ+gKTK7zIrZbl3G4eIBDE8/nqu0D7LzqlWj7vwA7XwRrrgbAnzpF38h9fNp4KaNRNwVL8r3RLl6y5fnccOJreCP7yWy4FrLd8fGy3SAE080GGUPjjH2GzeXjZJXGWCZHvvfZTDRMugaL4E7GySHd8fdhvO5RED6Vyd2cUjZdZje60LCUxbRpc/rsSTLFHkobrgSg6YVMNjz6C3NeJD9oMlTZh+1NYpndVPUpei2HU9MbKGVN1nZnkVJxYKTK2fIU9WiIuh9Q0nVO/mAffYZLLSoxaCi6xDj3VAMu63F4zradBJHk6FiDiuPjBhE9hqQ3msTNdmMHEYEMmIiGKVid09R/WC5KIVFKhUKIXwb+A9CBv1NK7V9iB0r2FBmrG7xq7G5KCrlmMDSDLqOAo0KkW8acCulZcw3CMCAMYPwgnzn5Zaqhzf+88mfnGgi2kinG7rFMCQwTApe6PUVO7uTW3i6kFtKIbE44Z+jSS1RCi0G7gl4bhp5NsTVSPYsnBOvPPIIeuXxlw3/nzx/X+ar1Dvoe+Si9N72JsaCOLXrp1hSbc2tBM8gPPYouA74cPJsr1sytLbzsRegnPsPlk3vovep2nmvfyv3Vh8lqFr8xLTkhN7Jm3Vr0sInZfzV0rYHh3ajebVzhniFyrmHE2Y2GxcGxY9SCOKi6Lpunq9jStsTIQtd6Nvs20E/FK1NzvdhVWB/hiN4D1OjO9mG11lF0b5qtbi87FdwwADLQHGcsqTvpyxUXp1tnuihKGMeZm67o1Wb7bOWM3KJ9hFUgI0mEJJqLIiazSDRlzLdiAJEpkK0rQnxCJed9IXwBmtLR9Tb/C2aevIqQyp8dihWyEovEIq8kDuFsE8tqkgGXUZ2ERCMUcpFFMtPYM68UYpGQZGkKgWpTaCmFQ0mquGal5fPoz/fEs1nQEV5tXnV7M6zSFQkCs4Bl5mct1ZnY4ymRuPOaY/H/hG6yf3wYISQ7whp+9zZygDBzFPR+KmqIscJOrgConI2FUCkQAjd0GXXHeKbvMN53NTf3Xc3R4BiH6hP4shvnpv9C4ez34d73wKs/AjLAeOAvCDWT/9f4SV51XZ2zTpWHT27i8aufy1Vn7kM8+vew4T2zfzuID1d3QwaHv8YVB/+eTDhnTT3+zP9Oec2L8EJJJtcNk0ehMEgoDKabPgPN0xyOanRl+tCTv5kuNIQGY3qEefgx1utdGMV+Do7WyBoGzbBOw6/QqBwibJzGUIJ+I4dwyhSMPBNFRcmM2DukaHhd+GHA/vHjeEwglEV3xiAzfYi80PBZQ29WpxlGuOOTGF0ldjsHMA2fvLaeiXpAT86ilDExqydQQmAYOqiIU/YwQgiy2nkJjwAXcYxEKfXvSqkrlFI7lFJ/sOwOma44lVDTF4nIDIamk9Uz5PMDmJGHGN8fX0GNH2C4McR/TD3Gi9fexo7iYhMRSMz5JEPLd7CrZzGiLBlyGIbA0kz6zG42ZzZQixpIqXC0YpzDXhmOA312hYqU7Bi6n4nCFfzaY5swejbyV7yWNZMPMTh5DjRJ2QliK0iPW6sXzj6Cq2e5N7iay/vmvvxBcQPHrCt5vvM9nNDjlWvu4o6uZ/OGwZdzc/0435U3cFWPhaEkVvfa2Edq5KDvMtb7p4ncDQT4+OE0o5NDNER8dbw2UyDfmkGkadCznaxS6GGWRlil6sTWlWqMc0zEp+3eXO/8ivOujbPV7TW/QcNNRKE5yagRu8B6Mos7KMfdfEHiJuLDvM6/JatNHMIqkZcQKi92N83gN7GFQFMWxsLAuVWgoCQRwfwRvVGAJ0BXBkZLgLt1v7xUKLzZccChiLCUQjc7x0gwkhiJCGfnpsykUmc7jOi1lCAQclEqeeu8di03X0gsrUBT05ALOvLGFeoeRSkRmfy84/XnehEiYlwvzrm2EpyoRq+MiKzSbOovwJpc3CFhKJnh3lqU+NhInISx06siuzcDCswcfZk1YFQ4kdsW9+2aPBqLT5JtN9ks41Lmer/J5PqbuKqwHYARbwQ3cvG0PNzxG1A+Bf/0WviXN2OM7eGTxZ+haige1z/GPuuDSGuYxyYHOLnlbqzhB2Hi0LzPwg0jMtUTbNj315Tz63h34df4ux3/i1pxI5fv/3uEX6VmB8n3XsHQwzSnR8iWjxA2TuLrmUUXnCW9QE3WGNdczhzexZ7TU0jlMe4d42h1P9PlfdTHxvjYnh0cq29CWt1EmR70wKZPKppRlYnoCA8O7+V7Qw8RiCm6rB66s0VykU9ORZj5HgoZA10T5C2DbLGb3mCMdbk+dg2f5IHhR8lnInRNIEKXTO0MUaaLetjkYPMYbuSTXzis7Uly0QrJD43QYktBb/Olb0e2K76qmjwCMuJLk49gaQY/tfGFS++XKUHo4J97BLtyGp2NFDMmmt/AaI4hAps1Vh+DVi8uTSaafpLHPhz/I2cL5IcPUbDH+Nvgx9hY0vh/dxa4+nmvZkJ103/0K/RnSkzajbluuVLSPfoYB/PXEmJwRZ8OSqH5dVCKobV3s0OcQ42cQxPwot7ncUu1jqUCHjGvpWTo5DJWLLZCQPcmtO5NZIIa66JYLCbLD9Fbm2ba8CiEFhlhUlxQc8GauK9Xd5TBpkrVrUHlLCLyOaOZCGXRl+lBb73q7940W91eC2o0g7msoFERn5B6k4aB8/8+PfTICKnZNGfqT1pmkZRyba76M0XyShGq2CKZxa/jaAKhTIyFlubM2FzlzQ9mBw6uEGhKR9PbfE2SmSQKf/ZYIRJLgWYscaVnZOI1tgzTmpm7ntPMtiN6LaUTCLmoSn0mtpJVCs2Yf2KwtDxNTcCC9vMNvwEi7vyrWaV5x+vJ9AAwrJeSYPvc8XxVZ0CGRJnSvMyyNUm/rdGZz7U5N+DqyFSceLkhCmK3lRVbQBtL6xFayNHIQHVtgvGDsYstKTA8MLIHhGKnHzI2eD1RJDBUliYVPHucuu3Cltvhx98Lm26F4lpOPucP+Qv7+ZQu+xATwSRZLUPX2m/y+EiOI5tuJzSLqF2fmPdZuK7Htr1/gW8W+AXv1/m7qVv5w4PX8LUtbyITNNl65GOM1t047TrXA0LQPLWbnDvOqC5QyuKrRzJ8/kB2NpVbCEG3UWKCJpXoLHn/KKfqe7GjBj1ajh67zMcPbuLAhMUHHyzymX05ggiiTBGzMURJz2MIDcPw6csX6LLmiqN1ZwrZZvqm0kw06ZOPqvRle8lbcLjyGHunHubsyHeoSIdT7giHmycxJTzz1D3c/sAfwfSBRa/1RLl0hKQFqSRn7TFG3ClCGcWFf07ASNXBC1uuOo0M5HqYJOD+qce5a83NdJkFQt+j/4GPUfrBPzA0Xot79rTgGhZNw2Jb1/UYokDOHmbb19/M9m+8hcu+9nPkpg+yObuedbl+xpwKNT+MA5T5PqoyYN25fUR6jo/VbuGFmw3ypmB7f5Z/VC9lS30PV/geCEU5qR7PTx7D8hvsMm/D0GB7KcKwJ1FKoAUNSlc+n4bKctnQblzpkTcNdk48zATdjHRfjuF7ZHvXzbkx8n3QsxWAu8w6Iuzl8cZBBsI6p02TvOpBoVHMLugxNXglCI21EfiiSt2uEA0/CsC4DhZddC1scJjtoX+mejxq0HBnrItpxpJJgr1JJsk8cr30RyFoNraXpPO6VerJiW+mw+08rCIFKZH4BK0TD70GttAQWOgLRSFpJa/wka2FGoGDIzQ0zPZuTjOeSSKFRxhFKKUIhURXGqJNuvAsRo68lEgRzbq2mrNCkmlrkZjo+JpalLVVTdKGTWUgjPkXUBmzRFNoRN58i2SiOTeLRFvggptpJT+sF2IhSQQhrrxvsjbyUJnu2KKdt4+gjIyti5ZJiWeqcVHtujBCL62LT8bAZb3xYLUxx8bv2wHjB+KYRW0YyqfZf/Y+AAqsZVxF1JpTdOt9CGuSQ9E67EqSRbXpFrj7/6B+4gPsjrZQUcNEeoXXr3slL+i6iSh/iJPNCepCY2TLKxFnH4DTcVYjMsT63nvIN07xfuvn2F/t4a7rHqanUOb391/NgU0vZGD4XvSpo3GPL6AemYyERch3cWDa5T3fWcOXDmX5+gnJlw7NXTwIIeg1u5jWNE5PPEBv5Ti99WGMqcN8+UQvJ8omP3ejzZ3bPb59IsN77isx1rRARujOJJaeIaNn51vCKkT3K3HtTxukWcRonCNrCApmgS6rl4LSCOxhDodlqkGdNRJe8PD72Xni39HcST5+7rNtX+uJcFHGSJ4MUkref+yfebgcm7FrrF5e1vtitmW2xMkhIWzty82r5/zyue8Bgpevew66V2fDt95Hb+0UAKWpY3z/yp9H37SBjCEIVYRAcH33FZwa9yhaOoMP/y0Ap677ZTac/BzrH/wDzr7w/WzJrUdFcGRqisvXdBOqgJzIsXFyH6e7bsZrWty0Lv4TaELweO9LsKtfZM3xL3Pldf+VY7VTDOSz5M7sIhI63wpuZlu3Rj6sU+u+kkpg0F1+jFLver5r3cHdze9wzrsbS8+xZnI/HwlfztY+gQh9sr3r595wphSLAnCrNcQ/V27gqPFdhkOTxzMWO9lKPtfT5qRbgsIAW2TIAb2J507gTe4jD9SNgJIo0bNQSHSDvmwvugJX1mOLJArAqzNZMEBp9Lbra5btYTAKEHqTyUZyMmyxSNYU2ohPpkSpHhHhzXdT+U1sTSCizHxrCZIpiRLFQovExtEEOiZ6O2Ew84kABfhRQKhClABDiaWFxIr3C7Vw1lXVCJoIBTndbO/awsAXLmqBRTJT4GliLcr2yprd2JogdJ3Z2APAeHOuhby+YCjbTJuUEZFD+OXZ4kI7tEFErI3c+ITf4trSNZ2MKOHrTbxMD1l7CpKxBJPuOLrS6JYSr7gu3he4bu02OA6TXgO7ZxvZU9+B+jkoDMLYfvZND5ORiii/ExHpXFHopdcZZMI7xalGgWbJJQg8zGSAmBdKHhgJ0fPH0dC4IrOBy81+vl15hKD3Xs5WX8SZLS9l3fRDmN/5I3jhb8PBr1A4+332bf3PfPTwc7j6iq/zcHQPg9vWcWLv/+CfjZ/kcvNBNh35BGf6L8c0NKp2QCFjcM6t8ve7B4iU5I4r3sMevc4uCd0HruSua149+9mUzBKYJVAh+0c1/v7xdVQ9nZs3+Dxns4/arNg2MM1nHlvDRx4u8L+eH2A2Rwjzg/OSIAB0vxH3JRMa00GFA81jnHKGKYdV1lkDvLjvefRLieFVCbPxd8OyR9GMIpZZQIt8bt7zQQrNMXY/4638ZXiS/fbhjv+mPyyXhEUSqpAjjSFONM/xDye/zcPlQ7xy3XP5uY0vJ4gUnxj7DIe8gxQyBk0vpGLP1ZJ8f2ovXx9/mLsGb2Ig083ah/6ervoQD17/Jn5ww3+h4IzzkkfeQ9eBe/A9g0bD5JriDnxPI4gUXVOPUBh7mD/1XsULdz2HV5R/HdcPKNz3+2gqYlthAwNWP75rcE1pO4OjU5hBjXvFLeQMuKp/7h9m+9pu/iF8EaWh77Lem6ZXG2SsOkb/qfsY7r+O/dU8l/cIAnSm9X6uv3wra3pK1Jo25za+lAwB687uYf25B9GQ/Ev0Aq7uN0AIMsWWE68Q0LsVWVzHdfI4Qe0GJIo/006QUfDs4nWUCm0KOI1MnLkVOggtpFofwZk+i0QjNJsUjG4K1mLfq1YYpCcCXzRxPRtZjl0eU0JHUzkK7brl5noYCD2EFjHSSDKBks6/0MEdlilSkhFKeIStA528OrbQgCzGwhO1VUxiHf488Qn8BqEQaJjt+64lri0pApzQnR1UZSgNsdTXKsn2CkVElFhAzaAZt0cxOlkkBr5g0Sz6mt9AU6Bp1qL98lYXjqbFc+RbBHJiVkgU+gKrbqaV/JiWRXh1SIo0y0ktyoAMELm5GpLZYxldaEaDhtkHzbm5JPVgmu5II9CzWIWe2SD3M9bsAKAa1Kl2bYtfZOxA/P/VtZ5xZ4jLgoBa/za2aD3sWHclvbk1aGaV09M+9eJGvPpcdpgXSvaMK/Jdx9ia3UAuCtD6ruWqnpvQC8c5MQmRITh702/FtTH/8TuooYc5c9V/4febLyPT8whn9XvYkt3ARDjK4KZvcM9wH/u23U3X9F6KE4/gh4ruvIVpaLzvwZCqp3Pn9g+xR6/zGk/nCh++YRxi7OT3Fv39FAafO9SFZcDbbm3yX26yCVTAR4Y/yWcbf05h5x8zad3Ht0/mEEpiLKj9OVh+lH85+TH+YfJe3nvqb/mjUx/mSxPf5Ix7DkPo7K7v509Of5Tv2UfQ68OgQAuaGF6Z/eVuPvRQHuN7n6Wrepr35d/GH9Qku+3D+OMvWrTWJ8olISRnnHH+4Ogn+Z19H+Grk/dxXf4qXpp9BndMe7zPuJqrtB6+MP4fjPtT1NQ0j06c4dR0la+NPMRfHf8CV5W28vNbX0p+/DC9w7s4vOVFnBy8iuzWl3Hmrr/C7b+OG478C31jkwyItZyecBmqOBRMnf4Dn2DCWMfHwpfy1hsz3HjFdv7Q+G+scU7Qc+CTCCG4orSRQjiI3TTJDj+A1Aw+WbuBZ6wxMLQ50+i6AZ2/Cn8SX88zcOBjbMoNsvnEw5ihy6mNb6DuwzN6A6bNdVy1vpu+Ypb1W64gKx12bN/JLnkFzzr9Fa489mUOGTsZNdZyS59JZBXJLuxgm+1GrbuBzfZ+cmE/luyjrimeVelhfS6kpzC4+IPWM1Bax+XJbPipZgW/Ps6U2Q96k26ru30QvLg2jpPodRrNCsF03FKlIkAn1340ba6X3tkOwEmh1oyQKOhuZ8VkuuiWIVL4swFwALxaHCPRcouti8RCWDjr3U6+zBoW2uKwReLaUkgtxPad2V5lxnKuLd2MCxm1uSwsO2zG7VE6CIklLDwB0YJi2obfxJIi9psv2K+UFCg2FozbnWqZjiiy8626mXG7E8JEhM5sU7/xZGhaXIzYv8j66bZ6EUadstkfx0iSoLmrKgxGkkZ+LZom4roVYE2xH2QGO6owXtgUC8x47K8PZEhZldnpB9gDm+nRYf2GK9mUJMCcq03StEq4WLOC1fB8hho+kXmOK6y1+F1bqUZZNhe3oxlNjld8AuEwbawhfNmfwJ3/i/FXfZpH+17Og0MaXWvvZVNmHb+06Q3cXLoOt3Qvw06FB0t30cz1s+7QJ8gYCiHgaMXhwESGV2+/h2+LYV7lm9x+7W/wsm3/jUIk+KJ9D0Z5/lyik2Wdc3WdH9vpceP6AFc5fGLk85x0znJn77PZkhsks+ZrfK32eY7YOkbjHChFza/wD4c/zD8d+yD76id5tDrCRL3E84sv4pXZ/8bzvF/gF9a/gd/a+otcltvC5ye/zafH/w0xvRe9OsQ3zvTylz8oYJdrPN/7Pp/iJXxS9nJEfYOgdj0vGXxOm3/QJ8Yl4draEEa8zbiO6e7LyLvneN25/WzcP9cw+AO6zk9tXM+fnv5rZr9SSZf3ywub+M0rXoeJxrpHPomd6eXg1ufTZ3ZT0HOofI7RZ/0O277+Frac+iznNrwL149AKbqm95CtnuBd4Vt5wbY8r70qPiE+vPYOPn3/bn762Odw19+G2381xYzJcMXm5omHqPTdyNGhLC+6fP4X8sp+HVsr8o2u1/CKsX9g8/FPUjx9D9W+G7jX3QF43DwY4eYG6crFV4WitIbenCCMFB/IvoWXqu/x7HVN/u/J5/PsTQZFGWIX12MsdFOZefSNN8Oxr/Pavr18bfpqSj3f5TL/epSI6C31sQhNg77L2HrqHqDEZCgx3TEOWgMI0aQv20u23VCnrvWsq+xiv1FjulFHGfGH39QkpsiSa9d2Pd8Xt+MgnugXSonuxA0bURny7TrsZkpxxbbw5glJ4NYIhUCILObCmhCzQE5KlAgIW064TlItrAmr/TwaM0+PjFBCUvGq+FHcQ0pHi0+aS5BBRwpFIwmWu5Ed9wkzLNr10LM0i0ATi0YIN4MmWQVSbyMkSfzDCZ1543bLyYz5kpSzMYsZZoRkaqaSP2msOVxLqtplhJbvXZTQMpDt54RxlolwIzurjyCcCjKKCEWVDaGHV9wc150kSQiGrmHIPlxVpiL0uI9bIiQj9jR1PaIr6EaKDP3FHFq+j1vW38AXToEtJ5lym0yuuZwB/xBSZNg9UkdmToFQXNZ1LXWjD1MI+jPxKOChYJpq06G7uIYhcxu5tTs4PW3z8FiEMKo42jjPKN2JLnRe2v98dtf3keneww/OPY8bL3sJz9n/SUpnv0N9y93ce9ZBoBHJr1OSkufsfCsIjTWlbp5T+Am+6n2ZL575NC/v/k1EIrj3n7HI6AEDfUf58sQxHqo+hq98/odcw4unJjm6/RX8a3k/96hv86Hxw1w2fTl9lfvZU9lLKCP8qRdTn3whVw5IRuo6D5xs8FfWn3OjOMb/Gf1VijuuYLP7s6zt+g731b7LkRP/hH3uNVSq27hhXch7er/O9BnBI9eY6PYn6dMH+emdL2Frt8/89IMnziUhJF3o/LdjX8PN9JJ3p/DMIkd3vIKp/qtwMz3knCneeeqLfJsqO7MbGN3wXCaiGhuNzTxn3Q6KhkX/ga+Sq5zh/qt+nsgwWG+tmX19ZeSo7Hw1Awc+QaZ8GHrj+ELv7s9T03v5vPtcPnT13MnwmWt13qS9kRdrexnc+xHOvuB9GLrG1vojWPYYD/S+DoCb1xlofgOlWyjdwtIFV/TpfMR9MXf376bvyGcAGLnyf/LIgZDt3dDT1c05ozBXJZspUezuZ2ysxo7LdvKbezbz8jUmD4UB796cJQgrlHrbuKnMPKy7FqmZ/Lj1GBMj1/D++jB/uWEnQgTzU39b2XQbGx76CABTkUfRmeLx7AagSXe2l3w766K0gTVRiJapUak3ieQpAHw9oqQVsdpl2uX66E2KF+t+nTCSZNwKZc1EyAy5dplRM1MStZBGS8uNhluJ37KeQV94krcKiYtqfst1O6m/0Ml0tEj6opmJjFO4UVxMZygdscwIgpl57zPxCi9y6JYSkelgkST5/o0FNSHNwCYridPEF+zXnQhJM3LnWSQzxY95ybz+XBB3QtYw4pM7zDZTnBGSbinRsr3zquEB1hQHEHqdEW0gbkHfHGN8Yhxh1NjsOMju9XF8rfVY2gANbYyy1ON43b7PQW2EI5Vj8QZiM1mg2LsODItnrImbf2vWJKem6+zo3UbQu4Pm8AEOTIToheMYmAz034YbKa5cV8AJNiLQUZkhTtWu4vIeSbkZMCF9ujIme8Zd+gcO4QJX5rcjIo9+qbgss56hvj3sOnknQ1fdRLXrfvoP/gPVtbfwg2HJz5Xu5WuW5HZrI6Llfd215RrGHznIvaWjZE/8I3fu+FlON8vskQ+Qvfwofzfqo6HxXJnl18+dZnt0DqEkg+N7yd34X7k894t84tQPOC5PcGZa4TWvZF3wUu5cJ8nvcLh2TYhfKfPc3e+jW9Womb38nvdnvPmx3+IBeS3d2Zeybd02jpufQ9vwUfo35FjfdRXfHXqIv9m8Ea/xONcUdvKTgy+i1zy/p/5LQkia+TUMD2wlo0JOXPYSRtbdStRyonFzfdD7G7zx5H9wxbF/ZcoO2HXTLxMowVjNQy8Pc9XezzPUfwOn1lzFzuwGsgt6mVW2v5zeo59nzeMfYfg5v0emeoL8xB7eG72eZ23OsblrzrowNMHNm4r8ydmf5j2VD1Ea+i71TXfQf/Af8Yub+Kvys9jerbE17wECx/UInUmKfeu5db3Bx/dG7HruH7PBstG9CvXcRvZN1HnVdomdXUMho887IeYHt6MPP8CrLi/yrdMB/3Y8IGvAzWsFYQNKPW0C00acDhyuuYFrK4/z383juCpHYd06lDq7OPV3hg3PJGfmyUcadTlJNrQ5qOIAe3e2h2w766I7rm7HcKjaNfxwjAIQah55c0GfrRlyffQkJ/ZG2MQLAwpOmbJmoi1sIT9Dtmu2A3C1pWV4I+ngamkZtIXWhRXHLCIRErZketWTuea6lu1okcxYTJP25Gy7ErNDk8d5y0zmvU8lLiNfupRUFLsO28VItDiGVFnQHt8JnXhUcBuXWHeScecvGLdbSSytrLAW9fUSQpDRijRaOwArxUhjrs+WyPcsipFsLK1BaBGnSE6qbp2jZw4htIA1YYToWreo8r7HXEP9/2/vzePkuK773u+tvaqrep99MBsGxM4VpETKEkWR0WZJlGXFoi15lSPZVpznxH6JFWVxkqfEWxI/Pz/Fsf1k+yWOZUd+iuVYtqxdcrRLpEiRFEiAG0Csg8HM9F5dVff9UdWDHkx3T4NYCdX388EHPdW13L5dXefee875HR5nud0m2n4vymN/Dl//IEfbZ0GBSuYmvEjBLMZKSaOZIo6apW0scWS5RTAf8Gi9SKTM8+zKI6jOU8x4i/iRRtmTtOQajhUyYk5xzDrKk5Xv4R5/CTcTL5H5oeTgGcn44kEM1WUSE6TEz81zY+s2nmr9T5rKKZ5d9vjq9tdy70O/x+Tnfp7XV+5mofg3fETx2DV2z6bv6lU7vp/MwX/Px3PH+dIT/4k10UA4BrvtvRzIz/Hmpz/P4okHOTz3aj6x8HrylSPc/K3f5cZHfp/qne/ll/a9nr96wuKvnjAZcUJ+7o7nsTIOUglQQp9Xfue3cZQaX7vtH9GwS9z+lf/Ab8vf5UP7/xW//XWPh5/Zw8vnJ9g19RjfqR3mq2sP8cWsxU3aCK+degtlo8ez4BJwXfhIpBB8bc8P8c1b38PR6e/ZYETWEQpPLbyOR/b+MKXlg+w6+GE0VSFbPc6er/0nAs3im7u/jwm7TNHIb76G7nDylp/FXDnMts/+HFP/65+xopX5w/a9vH3v5uvdPaPzp+2XsWQvUH70g4x98zcxK8/xxOwP8ugy3Deno7Xr1HM3cKZ0G+7EIs3Vk9w7G/9IP/Vsm8hwaXvTPHI6pB3BgdE4wTFnb/wh694IGVMjDEN+4Q4bVcAdExpW1KRllTbWNu/GyhNN3UqmucQO8Tz/e/T3mcoHCFQy52k3rWO6yJE9zLaD9ZnFo0oGJBStYlyH5HxyU+vZ7UutGu3aWdrCALWJd748SodMaX3E3wiq+GEbmmucVlV0mdmo/LveNm+9Jslql8ZUZyRvKsbmZSc9gy0jEFDrymyuJktASr/Era4ZyYq/ysl6vFyXlb2TCjc0MzEky4lYYyjrSZVDq+exZlIzZaWyUTwwXhKLENrmGUk++f5a0t8wI1lrVVAj0DSzZ86Vpbo01GT/xgpEQewjkR0HfXGDhhTEhgTgcEdLoFXhOyfjZMTRMETzxmNViC4mMpMIJeBks0nDyMG+74dDn+Ts2pMoUqDn5vB0I3bSA6ZmUrZHMOwlji1LTCNgrRmwqo/wpG+gmieZz+6kEdTxbI395f3kMzrjziyafYynVl2q7cp66eHHlkLakaSuPs1OexvSyFHJ7WRFZNk1/ioUBHb+Ib553KBZ2sMXbnkXSnONf6L/CX+RzVFSXebsbZzPqAdF92d43/GAQrBCuTHNO0ffzY9tew33nzrE4okHObjjzXxz9nupBAor+e0c3Pn9ZKvPM3HiG2gKvHFXk391zxr/5KWnMd0sMvlN7f7On5KtHOXh/T/OWm6WtuHy1OLryTVPcSB4mPe+vMLff0mVH9onuC27l7ePv5GPLEt+f6nBO2Z+9LIZEbhODMmF8PzUnTw9ey+zRz7HPZ/9Re7+xq+jhS0evPndNAyHESNe6z5aCfnNrzd4y/+3xuePxE7O2uRdHHvJP0NrnmVl/GW8vvVvuXEqy0J+80PtplGVsqPyK+IniFST7JFP0czv4EP12+OkwamA0MxxJsqwMOIyc8ONtJ0RxpQ19o+ofOqZ9noho2+cCNAV2Dei0RQWOee8B4Bm4I3MEDRW2V5Q+Y/3OfzMrRYiaBI4o9j9pKntPOq2O6hmF/nS/rfzqnvm8IMKnjOOrvU5RjUQ43u5p7HGGWOFx6wpTmZOU9anKVr53qP33Mx6dvtKu4nqr3FWySHUFlkjg9HLKNiFWIwwUmiGVZphgGyusqwKDOFu0MRap6u41ZlkOQugnkiKm70Usw1nvW57pauiZc2PH/Kq2idLXbMprmtlrXKsGtdXyUqNrWoFdaQpOtIogWwkVQ6tTQ9pADPRJKueJwnfDOtxGG+PGUkxmZG0CDYYkkq7gh0JpGZuWqICyGgefkcBuHkWooCzrRWsSCHUMihmZtMxY05c9OwIyX1ZO83hpNzxSBii5qdjiZ0u5nOx/+JkfYWaloG93wemy5OmS+iXKdo+WSe/HmqsCIVxZwKhL3HsrKQR1Ck4Bhkr4kTjWRCS6cx2UNrsLi1SsArsKy9SMkdBafLs6hKnNQs1+V6/edJHtY7g02SXPUPd3cZyq0ErWkXBYbu3EyP/DR46roDUOZZd5G28nweMn+chI+KW3E2973XgZbtsRm9/Dx864/Ppk3/LW4//Gbsf/xN2HP5Lnp94CYdm7kNXBDlbo94OOD5+G2veNDsO/UW8NAiMuBInc65YWaZ2gunnv8Qzs/dyemT/+rWOj95Ew8wz++xnyFqSfWPBeo7pyOlHmFt9DnPmNXGVyPNIRG4Hyw0PyXedIQF4YsebeXT3A5we2cuxidv54p3/lOPZKbKqSxQZ/NY3Gvzkx2r89VNtNEXwnx9s4idZ5vXx2zn8vR/i/3T/N4613c2zkeThryqC77vB4L8vL/DJ236bQ2/4MM++/Ff4xLMht4ypjOoNGt48qqIwVXAwdZ2xhf00Wy3um1U5Uok4uBwRScnnj7S5dUSiZcdAiJ6GwR2ZQUmkKXaXNEoWBFJgusXNfoEOpoeeKfLU9/w60cTdoFZp+XVmR/f17zzNQozu5e56AykE/9LciWofY3d29+ZkxA7eBCPJyH4l9NGDCk+p8YMuo2cwtB6zGCuPAOxIpR3V8YM2YWOFVU3iqC5Gr6Ut08NNjMJy45y0dmfZydZ6zC6SfBCAalfVx1onR0PrMzPT7XUfztlmhWNJfRUv2jw72PTRkhnGWmLgIprJzKJ3slnHkFTOK7LUCht4UYCi9zAkTlK3nWDD0la9XY39I5rVU2k4a2QJ1RYR6vqMZK21ghdCYHjrDvNuOtnty1FEqBhQPcGzSSCBo3qYPY7bVY4TYpcby9T1TGzU7v8tHjRHCP0RPLWKV9g44p/LzRApNVabVU4mIeHH1s7ia88gUCjpMxQdAy/xW4y742zPx0rXgXaUw6087bAFUZtvngwoluI8itmxV7DcCpjKa9wytY1Ka407Rl9NoFRoZx7is0+b/M/HijxaH4X5Z1FQeGnuZiBOfl5qrrLWrrAaVNYHf75d4Et3vZfntr2C6ee/yPTzX+RUeS+P7vlBGkHEiGsxmo2/70YY8vj2N+A0lpg58vlN/Qsw//TfECkaT83/HQCaUYuVYI2a9Dm87S7Ky99BXekq3CUjFg//JTV7hGMT55VykiFh/SRB8yxKvxrOF8h1a0j8MOC/P6rxa3/r8KtfcPnnn/T4d593WW4IpKJyZNsr+PbeH+bRve/AN7O0pM+oWeL3H2nxF4favG67zh99r8k/vdPmVF3ykSfORczUQ5U/O+jzkkltg+6VCNtojdOoyWj4dQsGtgZ/dtBHahZfPSE4WZe8Zl5DCo0qGcreOQdwLpuj7s1yz2gdQ4WPPunz2FLI6brk3qmQ0IpnS70MiZMtopoZQj9up9KuU9NL5NwBuk+6DUg8S8dVskgiCorHZGmi/zGqAYU5dkiDkSDi8fwzAMx7u8n3GK0CoBmUs/GDox5Vyaw9xZfV+OFTNEf65mmg6niRSiBqNNo+q601QhGPmjdFXwGYHtl1afZz/oRaGIejZnqVwDViiRSA5UbXjCQRRDT0PkEHmoUtJVqksNqscLJ2Ci+UiD4yJ910DFrNjxP3QuGTiSRKr9BpwEpKRjTbG30kbdkgK8OehqSUBEv4ItwQtdUIa0l1RLunwcuZWYTaoK55RM1ViAJqQYVcFBFaufUQ3m46CsAtWaXmbiM69hBnWon/x5vHVpXY/9PFLRPbkVKw6p+lqmggJXV3jJpYxYiKGIBXGN9wzI3leICj2M/xxMkVWmGTx06cRrWfoWTMEEkYdbPrM09FKNw4sgtNmKj2czy9pHHGnWB5eYVDyxIz8zTj1gRoZXSjyUu33ci+kV3MlFxKxgIT9jackc/wZ4+Z/K/nTO5ZXOVQ8BA3ebvJaR6RlByvrzCqltgmJiipLpXwXJBHqNk8vvtt/M19v8En7v0Nvnnre/DRUAR4toaqSBy3TTPwedKd5VRxJ4uH/xLd31hLXa2fZPL4Vzk8eQdLCqwFFQSCXZnt3JTdjX3DDxPqGW568s/XC7TNP/NJcpUjHNr+emT3gEGGhI1lKpkxdrgLCHkukPViuC4MyZm6SiglzbDNwyc0zjbb/PVhnU8ddgkiBaEEbMu3OVlV+Y0vuqw2z/3QpZRUghqOaiEim795yudV21R+fk+NEbXBrYUGd0xo/MEjLX7641V+71tN/uTxFhVf8va9Xb6HoIVfPctJawctvw1hiGsIXrdg8Nnn2hw+G/JnB33KtuDuMZ/AGaElYcQ79wPLGCqhO4WjK7xxu8annm3zX77dwlThZZMKTTWO1toUygsIRSE7sUi7vgKAErSoWyPknQFS0ZoFqoFnAFJlwZ5hRCvguD1yNDooCpguJ/f8ECPRDEIJsKISWWME1+zjTxCCYnkXmpRk5VHUoMHX1Ngolp1ybwkSIcDwyEUQijp1v81SFM8ssobbO9vccNcNSXdN+UpStCrbSxxSNbCSpaiVLkPSKddrJvpTvfohVEzsUKXarrDUWKIYSmQf4cVuOoakETQJogApAhwZIfqIPVpJTkjL31iwK5BNslGY1GvfeM1Cx5AQEnX0zYgLhWVlhNR7+2MKVh6hNqiqWWRzDcKARlihFMVLsecbBEjChqWCLyscG7kVZfkw2eAJvDCiWtyFqlubllam8wUIstTCszxbXwMkh6tHkSLEU4o4ioF1Xlnk2yZuQaCg2s9xYkXwTOUJHjqyimofYSG7gzBqMZPbaHwmvDLj9jYM9zmOLCk8I6v82ndcFCWkrh1jNruLqt9kvjBKyS6hqzo3je2iHTW4e/INSH2ZN938OR7YX2dk/Cu0Ip+X528HCScbqyy6I7xyNEvezjDWNjAQNM6r0CmTwUWrHeKHEdsKDqoiOOtX2ZOb5i2Lt7EjM8dXFl6HGrbYcegv4uOkZCVYY8fTnwAE/s63s8OZY3dmkV2ZBTLJLD8yPM7s/hFGzx5i5PiXyZ7+NjsOfZQTY7dwfOKO+D4IImrNFmurJzmhF5kdewWUbyOUIjUkHWpthT8/6PLfvuXyga+6/ItPlvjr7+S4e5vG776mwK/e4/ADN53mPS+psNpU+I0vuVRagiAKWA3WyOtZFu05PvVMSD2At8w2aZb3URu/HSEl/+hWeOOiQdYQ/OnjPv/tMZ/bxtU4axwgCmmsLSOmD7BtfhFRXqS6ehop4YE9BnlL8E8/V+ehUyH37zDQpU9gjyIl61ULIY6aKeddVqwp3r4QG5Bvngy5azxCLc7SCASlTH/DkBudRMoItbVKJFTaeo6sNUDEUghwR/HUuB54XthoRo6MvYUyqOkRzr+SqdLNACwaC0RC2VgZ8TyU8X2UwpBpLZ5+f0cxEVJlxB7tu9aMlaUoJVLUqTbWWCZeP86Z2T7O9nOGpOpX1ysXVpOCZ6VekWhCYCQ1NzraVXCu1odr9JmRAJFm4YYqjaDGmeYZykGIVHrngnTTqaneClrr14lnJL1ndHZSujgIzxmSdtQmoh1XhezhpDc1E0UKmorE7wobbst6bHzOq47YoewUEIrPkpIIN7bWaMsKo2ELaRU2yKN0UISCITzaVDlcvhmAEXGI0TCkVrxhU+gvgKXrKGGRplzmSHWNulB5cPlJAApqFlvPYVob7yfP8Bh3pjDd53huSbDSqPOt40cRSsCsu4hlKhTOy40pOlnG7BnQT3D4VJ1PPBLw9aNt7th7irZsMZPZQShbbC9Mrh9TcnLkbJ35zH6mnFm+2v4cc2NP8qnl/8UOe45pc5xKq41nadyeczEmb2Lupntol/cz4dv4YZv2eUrNkYRASubLLpYhWG6tUTQ8pp0RTE1hz9gIhncjh6fuYvro38LKE5wN1phrtZg99mXW5l6HYk8TtHR0DBQU1hpt2ol24Or8a2nmF7nzsT/hrgc/QN3M86Ub7me1XWet0UYISUY9S3FmJzcv3EfTN1iNLCqNYGXTl/MCuC4MSdaAv322yFeP2rxhB7x8RuHAhMbP32EjhCCruUxZo4zkVnjX7aucrin8hy/aLPs+251Z5uwpkCp//qTP7gIsTo8TOKNIzaYxchMjZsTP7mnyK690+MBrMrxmXuenbjn3g1Kay9RyO1iYnWaq4LBr524KnkurWadgKfyLl9ms+RJThdcv6AgEDTWDZ2mbqqaVXZO6OULBiHjLDbEReNW0JMiM044iigMMieu4NHPbqTtTnM3vpeBaGNoWX7FTxiSklDGoVtaw8qNbJtThlLBFxB5vgrtzt/PmiRuJhMAzB6jeTtzEaBBiacuc1QrUjDqOUugpqbKOlacURaDWWFl6jqUkAKBgeZi9Ir0Md71ue63dWM8LqSYGaDSJANp0WKKoWu2SXO/Ufff6PNwB0GzcQNCK6qy2ViiH4brs/yAyyYO1FbbW/TeZKNqQk9CNk4TO+sG5ke654yRCt3s66fVIpaYohF3+ok51xHhpa7MhGUuKth1TM3ERttUjhFQpRCGRN7Ep9LeDp+dBrfIdStTdGVZViRsZWFZmUxngDqYo43OGKIJnpODRM88AMKrZuNnpTfehoRrMefMo1lEePdrkQ38rkGYcHTZmzeKZGhlt4/eVszJMZWZBRLTUI3zm2xHTJRgbjQUlR83tlD1tgwEyVIO54gj1dovvX3gnLRnwgec/hCIEPzD2ehrtAMuU7MgYOPlpyE5gGyo37rqByYlFJkKXalRbl+mRUlLzfQqORkPWWWlXmXHG2OnNrC/repbGntIElcV3E+kuLzv8CRatbew+9DEixeTU4tuo+QETeZu1ZpuVRpusrVPzg9hdLlRO3PbzLC++hZM3/wMO3/mrFMxFDEyy2TY5p86OiV3cved13LZtkjvmS7xkoYQM2xuzXF8g10UeyWhG4a4ZmC2YvG23CTJCbywhgwa0I0I9w6hRohLUcEcCfv6lIf/+SwYffnCaX35lhkeXAn7ja02OViJ+6fY27ezM+rkjw6MxdgBj5RBaY4nFQp5feMm5h58IGlRwGZnejtmJdFJ1snO3sPbI58Ey2VPW+Ncvd2gGkrxSJzDK1AOFufLm0Z1raYSKSdud5B0Lx5g3JbftmCNUTaRs4Vr9vzJFEUzO7+LJk1UQkr25IWoOmB4gGcmanDnlky+MbnkIhoutaWhC43sLdxIqFquq0t/ZDpCbpSQVjmqSrwQ7UfQzZLUyXq8M9Q52gdHVMwhVUD37FJXEwT7hjWwWXwRQVIRqo4cKraBOOwzQhUJVSPRQJef07o9OFFWta+moGbYwkHhW//YJwyYbwTOyRrPdZDQMEH2EF7txzSzU4jrx1WQ93JESYfdeUrQTscN2dO433/HhZKIIpZfvB9CkTl0oRPXl9W2RaJCP2vGMpEc7y04cInpMsRH1NVrtOlIJ4hyS3FTP6B+AEafIqbVTHKsaPJPdz0n1QbaFObKa1nNGApDVRjmlfg1DGBxs+TxVXUWGBqOGilee37S/oRgs5Bf40snPMzN5mkPHxiktPknR3oYqTEbdwqb7wlAMdhZ3wHPw6gOnuKN8gJyj8MeHD1EyxzAVm7JrbwrEWMhP8e3jxxm1Jnn11Jv52NEP89bR15LXs1RbARk9YMIoJorYscHTNJXxhX2snH4e3chyxD9GJCUCQSsERVeYsMqMWyXsKIyDGRBxaLRmMpW3yZhjHN/xg2x77He46QvvxaweYXXvj1BVXeYKDiOeiaYImu2Q2WKGZ8/WOFP1MVQF356ive8nWGu08SydWc/EMSephzWM5hrO1EvWBwIZ89I++q8LQ4Jq8gs3tlCSB4XWXKGZW6CdnUdpV7HOPIrRPMtOvUykZ9iZERD5/MqXm/zEx6qcqElGbMG/fWnEgflxWvrGB6JUDfz8drTGUlLH+ly3KX6Nqn0DNxQ2TsPdwii13CJe41kCp8ztE4nKb32VVnE3YSjJnx/GC5iaimdpVJRxcu0aL7t5ksAewQ8iXFM9Z6z6MJmzObnWZKXeXpdRGYhuge7ghnWKWQc330Ma5XyMDIaqkVNcWuEyDcVl3O2TQ9J1naKW4SGlwRfau1CMz1MyF3sLNnbIbWN86UEgi187xoqqokRqb9n5rrbZkUIzbMZ1Qvwqa6qCHmnk+izZxXkaLepdI/5m5GMKcIz+n0nRbXLtgDZVkDAetDGcrWcklp1FkxI/anE2ySXJRBFKLx8O4CaF2oI+hqSXAxxAkwZVRRAm5XZjf4wf62yZmZ5LW/lEDPN5LJSgwdnE0OWjEOFN9Z2RTGVHefzMIY6v6nwrezNLzYcZVyZxNbNnpBfAqDXJqTa0qfJ8VWEpbBK1y7hFh0KutGl/VVHZV94FB+G2XSexxQgn9GfYkfs7BLLNRA+hUSEEC4VpCsYop/2nKboqoQx5tvIEuwu30qbJbHZh0/Jq0cmTdwwafsBdE6/lgDlD3q8TBAGKACuqkR1/1aa+122P7NgcnD7KHncHa+0KRuRQyNtsH0lmZn49VsAe3x9H1C0/Da0qSqZEMWPgveQHqJ/4W/Cr1G97N7lb/i43odFxjU7krPi42immbRsZGQgBdT9gueaTs3UWRjLrOn6GkoG2v66+fDm4Lpa2pFCpF/aitdbQ6meINIe2NwtCxDOK0VtplvYS6R5aIx6d3Tdn8OP7TYII3nmTyQdfrfCSSR0/v9D7GqpJK78DrTsEU4bU21AcmcAxNtpkQ1NwRmapKS5K8qMXoU+k29RVF1tXN/hHupnM2VSkRWPsVoJMXEek7geMeAMeugmKItg1nmW25GxaNuuLO4YIG8zvfzluZsCsYv0iKnqmQFm18AOfNUVnLj8++BhVp2SXWFFVvqLGhY3Kdmnw0lZhjnKiqtv2V1lSVfTQxusXHQZJFBa0okacxNiqsqYoaKGB02fpzUpyRRpdTummbGNGAsfs34fCcCh2io8BI1GIoW9tvHXLxYkiZNRcjxRzpEQ5v/5LgpvMEgJ5LlJz3ZBI2TdsWMOM67bXz2w4xoskiu72XNrq6G0dl/HnOFmP73dTGuh2vmfuCcBsdhtCq3BktcXfrLrxUqdXwlKUTTkkHWayiRBj9SSOUmBFniRqjuEWC7h97ovZ3AwZLcuxxhPcc+txIkK2Z/eBiNZDns9nNFNkzJ7hSO0ppJQ8cuarNMI6270bcU113Xh2Y6om23IlaskSolHYiZ+bw/cbOKySdcexCptnTQDl6R1EQRsTlYJWBDQmOqsDfi02ItMHwBuP68DP3hXPSjrqCLqG+ZYPEL7lgzi3/SAo54wIEKcY1M9AcQGdNgsFjflyht3jOXaOeWwfcTeIweLXwB0dvujfC+C6MCQAgTNCbeJOGqM30xi5ccNoS6oGgTNKs7yXdmYMrXEGpOSH9pr88f0eD+xUcWnRLO9H9kpa61wjM0ZoZlGSUZrqV6haY2wr934AjOcczroLKEEjLlrjV/C9Wap+HLnRz8mcz+jr8egdQikpDIrA6iJjaiyO9ncSbyI7CdteitKnRHEvFHeUOUvSECahajLqbnGsbjOXxLM/X4wr4WXNMvk+D08AivMUEp+H79dYUlW0KEPGHODUtzwyURzRFEQh+LXYkEgDvZdfhXPhta2wte6gb8oAXQoy/VQBAKFblLtyNMphiDNgKWy9jYnicCSbLNdjQ5KJIkQfX4JtZ9GlJJDnklSHmZGowqamCGQSCr2WJFy6UYSwMj39KutS8okhObIaZ+xHIoeuqn0fRvuT0NxTzed5aDn2SZUME8vy+rZv38geZGjw+NmHWJPPEVBDbe3Bs3OYfQZBWTPL/tKtPLbyDT53/GMYismktYCjazh9ot7KTpYpZ556UOWrpz/LZ4//JeP2NLPOXrK23ve4neV5Alpx9UOhEthlVgt7YXof0/P39uw/gIyXw5neR3VlicbaEvNFB8dQoVWJZxLTBzZm+itKLFzZOhc6rCqx32QTURBL9efnoLwI4zdCczWumKpA3tHR1fOeK+0mZPuUD79EXDeGBEBqFqFV6FtFDKHQKuyk7YzHxiSKR3hac4VmYSdR1w85jCRnai2Way3O1v3145ulPYgoQG2codmWZEbn8PpERuUzBtLwqHvzqH6VUMsQWCWkhKLb/wHlGBquqdFKZEUiKVGEIDNgdHxRGM4mJdgtsbKUbQvVmUMRGsV+Io8dNItXTNxBVnXR8l8GIGeOkOkl8tihuEAhyYiX/mlOqyrIHN4AQyKSEOBANvHDEPwKa4qCEpm9Nb0Aw3BQpcQPW4SJg7RFiB4pgwMIdIeRruickSDEHuTzWb9gXO89kk2WE00wQ6p9R+665WBFkoBg3dB1DIkTSegzQ1NFhppQwF8DKTco/yp6b6PVmZE8HcXvnzn57aQRpbjkcC/fFHDjaJxtLY0jVNWjAEwoeYzcZN+lvu3lEkFlH4drD/LIma+C1CgqO7FUu2+QSM7I8bq515E3Rjhae4p5byd+GFF0bAy1928qYzjcOno7i9l9/MWzf8RS8wSvnHwDkYwo2V7fJdmilWfM9agkVSarzQDbEORth6Kzeemtmx037Gbh9tcwPT1DSaxC9VScg3W+EelgF2ODG7Q2v9chCqB2Bkb3wMgNSdTlCHgTsZHqRdCKzztoOfgScF0ZkqFQVFrFXTRLe1D9GnrtJIFdjpeQuliu+2wfcbl9voihKTTbyQNNs2mW9xO4U5wq3MK2sR7KugmmpjJTzLCkT1KbfjmNiTuohSol19hy2Wkib1NrxQ+pWitgLGv2zB+5ahgudmkb46UbGHdHeutldSMEmcwItzp7QEg0oVIyC5iD6pvnZ8knS0cyXOGUpmEoWTKDHtamGyfPiWYswpjMSNTI7DsjUQwXKwI/8tcf1C1CdKngmIP8Pjaj0bnlppEwROklo9/jOEfGJYE7iZM6el9HtmXHVRxDGawbuo4hceXmkrkdNNU7NyOJQpa6y+z2WS93dReBYE1VWLa3IZcfTto82tepD1CwCmS1ERT7CJp7iBG1QE4xcbL9lzz3TxXQmzcTyAZfO/15wtoikwULR3fR+9zrju7gWRleO/XDcVJe4WZaUZNxb8DvUDUpOy6vnfxRxuwpJpxt3JC7Bam0mBhwnBCCfSOLrLWqrDTaWIbCaC5iNj/bO4n2PHKex+iOAzB1G0zeAtN3bKh3vwFFgeJ2aPYxCADNNSjvhPy2jca5OB9XpTxvFSM+phKfdwsh0Yvl+nC2XyhCEGTGCa0iWv0UgV3e8MWsNHzGPJPpQhw+vGPU5eGjK+sP/9AqUFOzuLBJQPF8JvIWzy3XCCOJqgga7ZCd41svOxUzBoeIZ0atIGIsu7V/5Iqi6jB5C/sqdWrNYOv9Ac0Z4YCzyBerD5LTC1iG3ncUCYDhULCLuKHCg0aVhqJSVDK9a550MD0KUUBEk2bor/tISti9s+EBxcyQaUraUSsOGVahQYQWKYN9HrrDeNgCdBSpk5ESXxnGkMSKw5HqrydOGsLou2xkmCZWJGmLkCAKMFSja2lLblLWXb+MnmM5UpCtNYgCTtfPldkVfdSdFaHganl8bZXH7JvxW58D8mjOOFq/hFNAExrT3gyrze8g1BY3WLvQVQXd6R+8YekqN43t4sEgg6LVaK/t5dY9cnO55u6+UA1yZobpzDz/cP8v4+p5lhsrTGf7GwSAqWyZoyt1fnrPPyeSIXU/JO8o6+WF+7FQHGPfRBkhFBxDJ0RQtgZfaxO9qo32wh2FpYOxD+X8eyEKiSu6TW4+zvTiWUljeaNDPWzHKt/e2IW19wVwDQ1xLz1hJFmutVhttKn7AX6wMYlTqgZtb3rDUpgfRAgBi2Puug+jmDEoZEyWqi3qfjJL8ANmy/39HB1MTWWulGG53qIVhDiG2jNa63wcQ2O+lOFMrYWuKoMTC68WQjCezbAwMvjHuL67laWgm9yf+x5eWrqXrOUMHtkpGoo7zg9UWzxsx0Y8o2YwBzkNDZdi1EYqPnW/SbN5Fl8RCJzeSYyAanlkwxBfNs8tbQmJJlXUPrMEAHSHsU5lxDBZbhswau8+Lq6B4nO8egIhQVfMns5viHXbTCkICYmSgl0dmfw4/6R3/1uaS10I2q01kOG6/pgXSdRM/6WZcWcSYZzhS+atnFUVnChCc0c2VVTsRgjB9ux2hFYDETDn7sCySwMdvJoiuGuxQLB2E1IqzGVupOgJcudX8zyPEadE2VNQpEfDjxjPmmT7zMrWP5NXJowCVKHR9BVKGYNRz8bRBkgIAZqqccvEHoTSohFW2VXYhX65nNaqBqXtsc/jfJqrUJjvGwFHcQGC9gZdNZqrse+lz31/KbnuDIkfRNT9ACklyzWfbUWHybyFY2pEMuJ0tbluDHqx1vTZOeZtCLMVQrBvMsveySy6pnCq0sTQFEqZIdbDgemCw5hncbraYq60tfFZP67okLV0pgrW1kmCV5FhPw+6Tc402W3OsZC7dV2hti+KCt4471w5sa7qm9VdrEGzBCtLMQmTXWmdZa0RRywpIhM7i3ugWi75KCSQDfzkQd0UoEoNtc8sJv48Dp5sIUMTPYgHI3KLEW7nOCeKCAnwQx8ziqMCB/3gDQltEeIna+grzQpKJNABpY/kf0bPIIWg3qrGKr7NruqIA9b45/LbUIwzPME2ntUzlIMIOzfSe22/iz3l3QCoQmU0eyNWbmbg/kIIFoo5JuRrqD/993nV7hHCMKJgD36458wcni1i36SIGPUyvQU5u/CMDK6lcrbeopDRmcpbaIrWWxG6x/WmvCnmsnM9I7wuKd5EPKDozozvvB7kMDddKO+AJNQbvx5v8wbo5l1CroulLV0VVFpthIBKM8CzNU5WWkwXbObLmQ0PutV6mydPV1iqtchZ+oa12OVai7Gs1TPMVlMVRrOxYudqPV4XH/bhriiCG8Y9dE2h5A5nfCAeie6fzm0uxvRiRbcpWjbPNRTaROvFlwaSmyYbSd62VuGD+Rw5w8UauBzmUUj8FsvNs6wl4d6iXyVGQE2kVQLq6yV6mwLUSEMbNPq082gE0M7jtOJ7RnWHyMPRbRwZF9Nqhi1sSc/iVN2YUlATUaxeSzwjMSIViegrreImDvV6uw5RyEpiSHSMgTOn2ewMilblVKXNGdtlvKVg62bfYIAOt4zuRxEq29zthGoWL99jGeY8snaGV+6xOHx8gcVxwdmGjrtF5JujOQgBc6UMSzWfCXd8y8GMpVlM5B0ymkfWNqj4FUp2aehB0EJuc67JZUHV42is09+BTCyGSf0sjO7uKU+zgfw2qJ6IHfIyhG13XHbfSIfrxJAozJczPHW6xk3TOUquyWqjjWtqm778nKNz20yB05UWjx5bpeDE6rsrdZ+srbNjbOuH26Z6IEO28YYhzn0+Q+eCvBhQVEazE2j+En4QkO1XPKubJFb/bVXBB2o/yMytY73lUTqYHl7ioF9prbLajA2JpnhofX5UIjEkIc3YkEQRDUWgSK133ZMOyQ/dPXI/+8UhUL+AGCY6JqmBEioBtXY9jrzawpBkQsFRLaQdxEay4tcwpKAt9L4GsrPcUw8aICNWW2vokSDSMn3zQQDmEmWH0+1l1twmKN8Tz9C3MCTjXpmXj72R+ewibdmmNEROUtHymCoJbtrm0Ahq5I3SlrI+lmZRsArU2jUcU6EwYMmtgyIUJtwSjbCBlJJ22GYiM/xo/YoYkQ7ZaaichOZK7EC3vOHCdxU1duz7VYiiyx6pteHSV+xKl5m5UoY75ouUvbg0at4x+kY5CSEYzVpsH3FZrrc4U4ulR/ZN5fpGi6RcGsrlXYyVJ1BViTsoq339gBsACNxxfummHKPZLZb5rNw5BeDmCmeT9WZLy/QfnBmZJNKrRTPwCfwqbSFQ0FEHzUiS5aFyCGOJwrAY4HtYR3dwZEQgQhpBI85qN3qr8a5fKlKpK5Jmokpc9WuYkSBQeheoAijZsZO3go8M21T8Ck4Eoe4M9F3MJIZkKTqIEJIdnoep2bHjdgC2ZnPn+MsZtxbRVcGEu/WDzLNMbCWHHzYJZYCnFzGG+A0u5BbwIx8hxPrMaysmM5PU23XW/DVKVgm3T97OVUdRYGwPaHZsQMZvHH5moeqxARnmPryEXBczEoiNQ798jn5MFxyqrQDX1JguONe0H+J6QS3Os9fIsNR+aHAOSYeRnUihsuRO4RptwkEiigBWNvYBAMuNFVYSp7RnuP2XCHWHbBgRiYBqu0ajFe+nSJ2Bt0QSjVMUFYqiSihU9GEeTknUlhSSNXmI7TKMDckAH4ktNWpKm3oiLFlrV7EjQaj01/aaTqJ1Vglot5tU29W4FomRGcqQaN6jAOzzdPQhklUtzcKzNI4utZgsm2S2+q4AU1dxtQJr0TOoaJiqPZQhcXSHWW+WRtDorbvWg4JdYGdhJ4dWDq1/xmsWIwMzL73arRia68aQvBAURbBn8vLpz6T0ZiRT5mUzeweH/nZwClRf9W84WG8jojbZrZydTnl9RrLarLGSlFbN6h5qP0PSVcdkpbFCIykap2AMXtJIZiQFKuSp4KsZ1CEkUtBtXl2r83F1kZPZGd6++gnERO9CUx1sqSNFm9WkaFo9qONGkmiAk34+KQy1qoQ0WhXqQYVCFMZqvAOWtvJmHlN4tMzTCKkyioKR2dr3owiFkp3nmH6C6ewU2qBAhQTP0nAMj5VGRFYvkrPNoQd027xt61F2wzKWGcMzvL7Z7CkvjKuyjiOE+LtCiEeFEJEQ4sB5771XCHFICHFQCPGaq9G+lMuLIhSmvKnh1p1VE3NkByVrkkrYxOujkLuOO3auJkmrympS7yNnev0fUIazfszZ1lkarZW4nRiDZySJISmJCkVRIdQz/Wvdd6OoTIaC+05PMtN+G69u1MDoLevewZTxMuDZxOfTDBpkIpkUTer9M96WL6JEgrOKoNVYpRHWyEcB0vIGXktVVMpWrAJdMsZAqBhDKh+MuwU822cqO4SKNHF4/GTWw1Ty6CLLTHGI8OkEIcRQxup8UiNy6blaDoFvA28BNhQoFkLsAR4A9gKvBT4gxIA7PuX6RzMxVMHObB5EAdfa4iHgjWFJiSIFFb/KSlDHDsEboJmFfs6QrLRWaSR+FYExOGLOyiOFQkFUKIgq6C5iyAdbW7ExpY/wE0FPrbes+/qlRPy5z7bitjXDOq4MkbrTV4KklDGxQpPTmkpQXaId1chFARjZLXML5pKIqylvGgwPa0jl2LydZSKXXdfsGoaJnEVRnyVr5CgOGVKfcm1xVQyJlPJxKeXBHm/dD3xIStmSUj4NHALuuLKtS7mmUFTQbMbMiMnsTnKDlH8BDJdINTFDBT8MqMkAO1IG1hXByJybkTSXaSQVBRVhDlaEVxQws5RYoyQqWy4ZdRNpFg4tzizFM6Z+FQs7WErse1nt1EKPGnhRGCv/9jFAmqpghDZLqkpQWyKI6rGE/AAV3w7TXuxDKFkTqN4i6qBaM13Ymk3RKm6Z19FNztaxdYOZYgY19VO+KLnWQpSmgCNdfx9NtqV8N2O6WPjsnLyB/FZLW0IQWSWcSNAM61RFiBkpFDIDIsR0h1wYG5JKa42Gf86QbJnDY+cTZ3sFaXqxMN8Q2HaGvNYgbCaGxOg/swBwtPhzr7ZWiWREWzbJyzaKMdi3okcup1WVdv0MgWjiRTKW0djCkGzPzQFQMkfQnGm0rbTUOp9Ls9lR2DGUFlUHTVXYNZ5lLHeNyQClDM1lc7YLIT4J9FJse5+U8s8vwfnfBbwLYGbmGo/ASLk4DBcUhYXJMmJAoal1nBLZqMLZqEFNSIxQI2sPeEh1zUiqrQq1xM+hKPaWhkTYRcaUVTyq1E1v6BmJYjjs9Cpk23ExLaEPHvFnjDiUtt5cXa+qWIh8FD0zOJFRyXNaVTm5dgQpwlhY0spv2c5Xzd7Dp5/5OhP2AnnLw9yqZHOCEGK9Jv2FMHqtacmlXBCXzZBIKe97AYc9D2zr+ns62dbr/L8D/A7AgQMHesheplw3mB5oTqw+O4SDXnHL5OornIyaVATokU7OHry01QkZrvjVdUOiKi5bXs0pMqccRpUyFkIcVtdIt8m2VnjTvjocjIUjB5ExSxBCrVmhkkSi5SMfYW7hW1FLHFUVnmgcB2DR9+OlrUGJloBneNw5dh+VVpuFUv7KJuSlvOi41pa2Pgo8IIQwhRDzwA7gq1e5TSlXG82IczaGTMpSvVHKUYumssQJTSEKHexBznZVRwgNPVSo+HVWE2e7oWe3DkXNlCnIxDk/xEh/Hd3GQnJXOc6OF33qmncwMnncKKLmV84VqJIRwhg8I8mY8aLAt9unAdjht1GcrdVoDdXANASaIpjIDu84T/nu5GqF/36fEOIocCfwl0KIjwNIKR8F/hR4DPhr4D1SXmCgeMr1h5mLVVGHxR3jJ1dXKLdzNBSFdlDG3CIrO1QtrEih1q6zlhgSVR/iAdolfiis4Ze2MBzUsIlVeTY5dnBUlG7nyIURtaDBShKe7EZRErXV/2ecd2IX47dlDSvUKEQSZYisZ03R8AyTUdcmM6i4V0oKVykhUUr5EeAjfd57P/D+K9uilGsaRelfEKgX3gR7fZ93nxrhldHD/Hj4M5hbZNEHRo5MJGgEDdaiWIbE0IcIee0e3VvZC5iROIiojZaE82INNlq6kyUbRTTDBkuNpfjSUiZO+v6GZNybhCo8r7SZqdvUlAymMVwexVQ2j7gAp3nKdy/pXZJy/eHFyzl7g0NkQ5UVtbilIYnsItkwflDXgwZGJFGHEZXsKlqkWlvnZ6yjO4jQJ5Lx0tlWMxLTzZOPQhrSZzlJSsxEUWxgB1xzNn9OmHDKFzTVDGIrFdmEvO1R6lMAKyWlm9SQpFx/eHEy3U75LEfEBIYu0LdwLkunTD4K8KMWy2GDXBRhDCoD3CFzLoNbyZSGCgYAwMighi1k0EAiULdwtluZPLkwokl73UeSiSRsIVg4VxhFJKEoC35IaDioxnA5HgWzcPnrb6RcF6SGJOX6IxsbEo2Qp5nA0qK+EvIdRGaEcuTTlk0elxX2NQPMYST8E0MihYp6IbLdugOhj0WLSLVQtpgxOV6OXBTRJOB45RhIKIdh35K5HSZyDtkwNm572g2EmUHZwl/UoWgXyZnpjCRla1JDknL94Y0jk8DdR/xpLG3rImRaNo708kWNMyJge0MfzpB4sSEJDS9W8B0WI4OQIZ6oIzVrS2mVjFsgG0U0RMiTK4fItl0yUsIWM4YRzyITxOe+MzgNRhYxZNJkSsqwpIYk5fpDM5CJz+GQnMLWlC2lN4zsGPkklwRgvJ7DGUbJN3G2K1YWhijbus7ITgD0I1+Kl5q2qAPuuhncUCAFHF49zGgzXgoTW2T6Zy2N08E2igGMRC2keQG5LikpQ/JdLSOfch1jF6G5wsTEGPtn8oNVfAHFG13Pbi+EEZFfJmsNMXLXLdBsFNONqxwOy643QnERlg8h3LGBkVcQl102w9gAtKM2E8149qNs4aQXQhCt3c+xM2v80cTTHNh2IxMvQDE3JWUQ6Ywk5fokU0YieMetY9y6bWJrzSz3nCG5o9HkJCUywxZKc4pJ9v0FGBJFgdt/Mn6tDZaQ72BG5wzbbEOjKayhIrAK2gzt+g5Wp29GzZSGD1FOSRmS9I5KuS6RI7uory0jFJ1As7ZWlXXHySWG5PZmk2/KEncMMyMBuPdfQn15aMHGdSZuhH1vjUN4h4j2MqUJRJiKxawf0jQtvCEc54WMzrGVOlOujqEoqSFJueSkd1TKdYl4+T/mUPbLZIFIMftXR+zgTbC/2eL2qsnfqdX5mCySHTJxj6kDcPyhC39AKyrc8fegVR3Kb6FLC6gz7cyQFWu0FQtF23rW9IYbxymrB7Eo4lpqakhSLjnpHZVyXaIYFqoAKQFtiPKtdh5Lqvwfp89SJOK4LGKbQ/48NDP2cQypBbaOUEEGyeutjzWVDFBn1JjG5RsEijVUBNbr909jVM5iiVEMw7jwdqakbEF6R6Vcn6gaOgFtLYOqDhGlJAQVJccksfzICVnAHlZjSjNiNd0LnpFoICNADuUjcYTHz51us2jeTlbUCVVnON+KpqPrBWy/hrlVhcmUlBdAakhSrk8UnZylsRxY6ENW3atpcQRUQ5i0FAdNH9LnoWhx6O+FGhK1Y0gYakYS6i5vrtSotxRcGkjNGmpJTFMFVukuLG8MfasKkykpL4B0aSvl+kTVGcu7rIgSrSGLMjX0AvhwWhQoWAIx7BKQosaO9gs1JHYJ1k6AYCiDEJkuGRpU6k08UaelDa6O2EFTFGyjjDE6BSNpAamUS086I0m5PlF0FMNhcWqUhdEhxBeBlhnLqx8Jytw0qm6Ze3LuWtoLMyTeWJyHIhnKIGB4WKJN25e4NGKZlSFnJJoqyLsW6MPXUk9JGZbUkKRcnygKmFl0wyY7ZD5IYMdZ6seiIjePb11m99y1tPgBfaFVBBUVyjtiH8sQBkFJpOa1SoWMaCX12ocwJIrAMTScYYMHUlIukPTOSrl+Ke+AYUN4iRWAAU6JInsn3OHtgmpAce7C2wex6GM0XO02LRFobK3FDbMdbygDJIQgb+vYw2iHpaS8AFJDknL9Yg63pNVBZOKyt1qmgGHaw9cpFwIuRPm3G0WB3NRQu2qZxJBUVkEHxRxuRgKwayItl5ty+UiXtlJSEtRcLD9fKhSJFH14H8kVwnBiQzJOXNhqWB9JSsrlJjUkKSkJ87e9hj9038n84h6kog3vI7lCGMmMZFwkhsRwL9wvk5JyGUiXtlJSEjzP5aUvfTmKgKrQrzlDYrl5AOb1ZZCgDFMKOCXlCpDOSFJSOgiB0HTCSCI0A3PI/JMrhePlAbhVOQyATMvgplwjXFu/lJSUq4xi2DQijaJrba3PdYUplieoKh4T4TFC1QJ37Go3KSUFSJe2UlI2oGoWNdlih3vtlaNVDJult3yYI6eeoqnYLNj5q92klBQgNSQpKRtQDAtFb+PZQxa1usKolkegOSBVLDOVO0m5NkgNSUpKF5qRwXTkNZu8p+k6QTvAtBTM1JCkXCOkhiQlpQvVyVLQnOGTEa8wum7SDEImbR2Ua3PWlPLdR2pIUlK6yJfGyF/tRgxAqCqaKsiYAoaojpiSciVIDUlKShfWNbqk1UFVNTK6hqOltddTrh3S8N+UlBcRqqaTdzSUyB9aZysl5XJzVQyJEOLXhBDfEUI8LIT4iBAi3/Xee4UQh4QQB4UQr7ka7UtJuVbx8iUm9r0CxvdfkLJxSsrl5GrNSD4B7JNS3gg8AbwXQAixB3gA2Au8FviAEOmwKyWlG2HnoTCbFqlKuWa4KoZESvk3Usog+fPLwHTy+n7gQ1LKlpTyaeAQcMfVaGNKSkpKynBcCz6SnwD+Knk9BRzpeu9osi0lJSUl5RrlsoV9CCE+CYz3eOt9Uso/T/Z5HxAAf/QCzv8u4F0AMzMzF9HSlJSUlJSL4bIZEinlfYPeF0L8GPAG4F4ppUw2Pw9s69ptOtnW6/y/A/wOwIEDB2SvfVJSUlJSLj9XK2rrtcA/Bt4kpax3vfVR4AEhhCmEmAd2AF+9Gm1MSUlJSRmOq5XR9FuACXwikaL4spTyp6SUjwoh/hR4jHjJ6z1SyvAqtTElJSUlZQiuiiGRUi4OeO/9wPuvYHNSUlJSUi6CayFqKyUlJSXlRUxqSFJSUlJSLgpxLmDqxYsQogIcvNrtGIIysHS1GzEEaTsvLS+Gdr4Y2ghpOy81O6WU3sWe5HqRDz0opTxwtRuxFUKIr6ftvHSk7bx0vBjaCGk7LzVCiK9fivOkS1spKSkpKRdFakhSUlJSUi6K68WQ/M7VbsCQpO28tKTtvHS8GNoIaTsvNZekndeFsz0lJSUl5epxvcxIUlJSUlKuEi8aQyKE+LtCiEeFEJEQom80hBDitUl1xUNCiF/s2j4vhPhKsv1PhBDGZWpnUQjxCSHEk8n/hR773COEeKjrX1MI8ebkvT8QQjzd9d7NV6udyX5hV1s+2rX9WurPm4UQX0ruj4eFEG/reu+y9We/e63rfTPpm0NJX811vXfFKoEO0c5/JIR4LOm7TwkhZrve6/n9X6V2/pgQ4nRXe36y670fTe6RJ4UQP3qV2/kfu9r4hBBipeu9K9KfQogPCiFOCSG+3ed9IYT4zeQzPCyEuLXrvQvvSynli+IfsBvYCXwWONBnHxU4DCwABvAtYE/y3p8CDySvfxv46cvUzl8FfjF5/YvAr2yxfxFYBpzk7z8A3noF+nOodgLVPtuvmf4EbgB2JK8ngeNA/nL256B7rWufnwF+O3n9APAnyes9yf4mMJ+cR71M/TdMO+/puv9+utPOQd//VWrnjwG/1ePYIvBU8n8heV24Wu08b/+fBT54FfrzFcCtwLf7vP964jpQAngp8JWL6csXzYxESvm4lHKrpMM7gENSyqeklD7wIeB+IYQAXgV8ONnvD4E3X6am3p+cf9jrvBX4K7lRBflKcKHtXOda608p5RNSyieT18eAU8DIZWpPh5732nn7dLf9w8C9Sd9dyUqgW7ZTSvmZrvuvu2LplWSY/uzHa4BPSCmXpZRniUt5v/YaaecPAn98mdrSFynl54kHqP24H/h/ZcyXgbwQYoIX2JcvGkMyJP0qLJaAFXmuvO/lrLw4JqU8nrw+AYxtsf8DbL7R3p9MN/+jEMK85C2MGbadlhDi60KIL3eW37iG+1MIcQfxSPFw1+bL0Z/DVPNc3yfpq1XivruSlUAv9Frv5FzFUuj9/V8Ohm3n9yff5YeFEJ3aRddkfyZLhPPAp7s2X6n+3Ip+n+MF9eU1ldkuhqiqeC0wqJ3df0gppRCib1hcMgLYD3y8a/N7iR+YBnFo3j8B/vVVbOeslPJ5IcQC8GkhxCPED8RLxiXuz/8C/KiUMko2X7L+vN4RQrwDOADc3bV50/cvpTzc+wyXnb8A/lhK2RJCvJt4tveqq9SWYXgA+LDcWArjWurPS8Y1ZUjkFlUVh6BfhcUzxFM3LRkZ9q28OAyD2imEOCmEmJBSHk8ebKcGnOoHgI9IKdtd5+6MvltCiN8HfuFqtlNK+Xzy/1NCiM8CtwB/xjXWn0KILPCXxIOOL3ed+5L153kMU82zs89RIYQG5IjvxaErgV6hdiKEuI/YcN8tpWx1tvf5/i/Hg2/Ldkopz3T9+XvE/rPOsa8879jPXvIWnrvWsN/dA8B7ujdcwf7cin6f4wX15fW2tPU1YIeII4oM4i/yozL2In2G2B8B8KPA5ZrhfDQ5/zDX2bR+mjwsO36INwM9oy4uAVu2UwhR6CwFCSHKwMuAx661/ky+648Qr/l++Lz3Lld/9rzXBrT9rcCnk767kpVAt2ynEOIW4D8TVyw91bW95/d/Fds50fXnm4DHk9cfB16dtLcAvJqNs/wr2s6krbuIndVf6tp2JftzKz4K/EgSvfVSYDUZdL2wvrwSEQSX4h/wfcTrdS3gJPDxZPsk8LGu/V4PPEFs5d/XtX2B+Md6CPjvgHmZ2lkCPgU8CXwSKCbbDwC/17XfHLH1V847/tPAI8QPvP8KuFerncBdSVu+lfz/zmuxP4F3AG3goa5/N1/u/ux1rxEvm70peW0lfXMo6auFrmPflxx3EHjdZf7tbNXOTya/qU7ffXSr7/8qtfPfAY8m7fkMsKvr2J9I+vkQ8ONXs53J378E/PJ5x12x/iQeoB5PfhdHiX1fPwX8VPK+AP7v5DM8Qlck7AvpyzSzPSUlJSXlorjelrZSUlJSUq4wqSFJSUlJSbkoUkOSkpKSknJRpIYkJSUlJeWiSA1JSkpKSspFkRqSlO9KhBAlcU6F9YQQ4vnkdVUI8YHLdM2fE0L8yCU834eEEDsu1flSUl4oafhvync9QohfIlZl/fXLeA0N+CZwqzynUXax57wbeIeU8u9divOlpLxQ0hlJSkoXQohXCiH+Z/L6l4QQfyiE+IIQ4lkhxFuEEL8qhHhECPHXQgg92e82IcTnhBDfEEJ8/LwM7A6vAr7ZMSJCiH8gztUA+VCyLSPiOhJfFUI8KIS4P9muCiF+XQjx7WT/n03O+QXgvsRIpaRcNVJDkpIymO3ERuBNxJnxn5FS7gcawPcmxuT/Iq55chvwQeD9Pc7zMuAbXX//InCLlPJG4oxjiLPdPy2lvIO4RsivCSEywLuIlRBuTvb/IwAZC1MeAm66dB83JeXCSUcyKSmD+SspZVvEqscq8NfJ9keIH+47gX3AJ2I5L1RiaYrzmeCcNhTAw8AfCSH+B/A/km2vBt4khOgIS1rADHAfcYGsAEBK2V1n4hSxTFC3kUpJuaKkhiQlZTAtiEf/Qoi2POdUjIh/PwJ4VEp55xbnaRAbhg7fS1zF7o3A+4QQ+5Nzfb88r4BbYqD6YSXnTkm5aqRLWykpF8dBYEQIcSeAEEIXQuztsd/jwGKyjwJsk1J+hrg+Sg5wiVVWfzZRKu6o8kJcpe7dHV+IEKLYdd4buHwK0SkpQ5EakpSUi0DG5VbfCvyKEOJbxOq5d/XY9a+IZyAQL3/912S57EHgN6WUK8C/AXTgYSHEo8nfENfeeC7Z/i3ghwCEEGNAQ0p54jJ8tJSUoUnDf1NSrhBCiI8A/1gm9eUvwfn+IbAmpfx/LsX5UlJeKOmMJCXlyvGLxE73S8UKcbnZlJSrSjojSUlJSUm5KNIZSUpKSkrKRZEakpSUlJSUiyI1JCkpKSkpF0VqSFJSUlJSLorUkKSkpKSkXBSpIUlJSUlJuSj+f4hCI01BxTO0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x324 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "sp.plot_average(center='Peak', time_before=1, time_after=1);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Check the agreement between channels\n",
    "\n",
    "Do the detected events overlap across channels? We can test this with the [compare_channels](https://raphaelvallat.com/yasa/build/html/generated/yasa.SpindlesResults.html#yasa.SpindlesResults.compare_channels) method. In the example below, we calculate the [recall score](https://en.wikipedia.org/wiki/Precision_and_recall) between each pair of channels. To allow for slight variations in the onset of the spindle, we allow a tolerance of 0.5 seconds before and after the onset of each spindle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Channel</th>\n",
       "      <th>Cz</th>\n",
       "      <th>Fz</th>\n",
       "      <th>Pz</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cz</th>\n",
       "      <td>1.000</td>\n",
       "      <td>0.643</td>\n",
       "      <td>0.743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>0.782</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>0.716</td>\n",
       "      <td>0.519</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Channel     Cz     Fz     Pz\n",
       "Channel                     \n",
       "Cz       1.000  0.643  0.743\n",
       "Fz       0.782  1.000  0.655\n",
       "Pz       0.716  0.519  1.000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp.compare_channels(score=\"recall\", max_distance_sec=0.5).round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In other words, 64.3% of the spindles detected on Cz were also detected on Fz. Conversely, 78.2% of the spindles detected on Fz were also detected on Cz. We can also use the [F1-score](https://en.wikipedia.org/wiki/F-score), which is order-independent (e.g. a symmetric matric)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Channel</th>\n",
       "      <th>Cz</th>\n",
       "      <th>Fz</th>\n",
       "      <th>Pz</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cz</th>\n",
       "      <td>1.000</td>\n",
       "      <td>0.706</td>\n",
       "      <td>0.729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>0.706</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>0.729</td>\n",
       "      <td>0.579</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Channel     Cz     Fz     Pz\n",
       "Channel                     \n",
       "Cz       1.000  0.706  0.729\n",
       "Fz       0.706  1.000  0.579\n",
       "Pz       0.729  0.579  1.000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp.compare_channels(max_distance_sec=0.5).round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find spindles that are present on at least two channels\n",
    "\n",
    "Using the `multi_only=True` argument, we force the detection to keep only the spindles that are present on at least two different channels. Spindles are considered the same if their start or end times fall within the same second. In other words, start and end times are rounded to the nearest integer and compared across channels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1260 spindles detected that are common to at least two electrodes.\n"
     ]
    },
    {
     "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>Start</th>\n",
       "      <th>Peak</th>\n",
       "      <th>End</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "      <th>Channel</th>\n",
       "      <th>IdxChannel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>585.51</td>\n",
       "      <td>586.09</td>\n",
       "      <td>586.38</td>\n",
       "      <td>0.87</td>\n",
       "      <td>59.435</td>\n",
       "      <td>13.185</td>\n",
       "      <td>2.255</td>\n",
       "      <td>0.473</td>\n",
       "      <td>12.945</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.659</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>598.06</td>\n",
       "      <td>598.62</td>\n",
       "      <td>599.54</td>\n",
       "      <td>1.48</td>\n",
       "      <td>79.922</td>\n",
       "      <td>15.977</td>\n",
       "      <td>2.374</td>\n",
       "      <td>0.421</td>\n",
       "      <td>12.889</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.376</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>604.36</td>\n",
       "      <td>604.73</td>\n",
       "      <td>605.12</td>\n",
       "      <td>0.76</td>\n",
       "      <td>60.649</td>\n",
       "      <td>12.520</td>\n",
       "      <td>2.210</td>\n",
       "      <td>0.350</td>\n",
       "      <td>12.646</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.481</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>655.06</td>\n",
       "      <td>655.50</td>\n",
       "      <td>655.88</td>\n",
       "      <td>0.82</td>\n",
       "      <td>47.596</td>\n",
       "      <td>10.832</td>\n",
       "      <td>2.149</td>\n",
       "      <td>0.369</td>\n",
       "      <td>13.297</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.530</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>725.60</td>\n",
       "      <td>726.50</td>\n",
       "      <td>726.88</td>\n",
       "      <td>1.28</td>\n",
       "      <td>77.591</td>\n",
       "      <td>18.836</td>\n",
       "      <td>2.371</td>\n",
       "      <td>0.288</td>\n",
       "      <td>12.368</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.698</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Start    Peak     End  Duration  Amplitude     RMS  AbsPower  RelPower  \\\n",
       "0  585.51  586.09  586.38      0.87     59.435  13.185     2.255     0.473   \n",
       "1  598.06  598.62  599.54      1.48     79.922  15.977     2.374     0.421   \n",
       "2  604.36  604.73  605.12      0.76     60.649  12.520     2.210     0.350   \n",
       "3  655.06  655.50  655.88      0.82     47.596  10.832     2.149     0.369   \n",
       "4  725.60  726.50  726.88      1.28     77.591  18.836     2.371     0.288   \n",
       "\n",
       "   Frequency  Oscillations  Symmetry Channel  IdxChannel  \n",
       "0     12.945          11.0     0.659      Cz           0  \n",
       "1     12.889          19.0     0.376      Cz           0  \n",
       "2     12.646           9.0     0.481      Cz           0  \n",
       "3     13.297          11.0     0.530      Cz           0  \n",
       "4     12.368          15.0     0.698      Cz           0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp_multi = yasa.spindles_detect(data, sf, ch_names=chan, multi_only=True, remove_outliers=True)\n",
    "all_spindles = sp_multi.summary()\n",
    "print(all_spindles.shape[0], 'spindles detected that are common to at least two electrodes.')\n",
    "all_spindles.head().round(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Count</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cz</th>\n",
       "      <td>486</td>\n",
       "      <td>1.031523</td>\n",
       "      <td>77.564127</td>\n",
       "      <td>16.878661</td>\n",
       "      <td>2.427572</td>\n",
       "      <td>0.408903</td>\n",
       "      <td>12.619925</td>\n",
       "      <td>12.705761</td>\n",
       "      <td>0.522530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>356</td>\n",
       "      <td>0.954298</td>\n",
       "      <td>68.662420</td>\n",
       "      <td>15.007348</td>\n",
       "      <td>2.320736</td>\n",
       "      <td>0.384774</td>\n",
       "      <td>12.466655</td>\n",
       "      <td>11.629213</td>\n",
       "      <td>0.522715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>418</td>\n",
       "      <td>1.056986</td>\n",
       "      <td>70.409223</td>\n",
       "      <td>15.256941</td>\n",
       "      <td>2.336045</td>\n",
       "      <td>0.409350</td>\n",
       "      <td>12.646223</td>\n",
       "      <td>13.100478</td>\n",
       "      <td>0.504384</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Count  Duration  Amplitude        RMS  AbsPower  RelPower  Frequency  \\\n",
       "Channel                                                                         \n",
       "Cz         486  1.031523  77.564127  16.878661  2.427572  0.408903  12.619925   \n",
       "Fz         356  0.954298  68.662420  15.007348  2.320736  0.384774  12.466655   \n",
       "Pz         418  1.056986  70.409223  15.256941  2.336045  0.409350  12.646223   \n",
       "\n",
       "         Oscillations  Symmetry  \n",
       "Channel                          \n",
       "Cz          12.705761  0.522530  \n",
       "Fz          11.629213  0.522715  \n",
       "Pz          13.100478  0.504384  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp_multi.summary(grp_chan=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Compare two detections\n",
    "\n",
    "What is the agreement between the regular detection and the detection after automatic outlier removal? To check this, we can use the [compare_detection](https://raphaelvallat.com/yasa/build/html/generated/yasa.SpindlesResults.html#yasa.SpindlesResults.compare_detection) method, which will calculate the [precision, recall and F1-score](https://en.wikipedia.org/wiki/Precision_and_recall) of the two detections for each channel. Note that by default, the other detection is assumed to be the ground-truth. This method is also useful if you want to compare YASA detection against ground-truth human annotations. Here again, we can allow for slight variations in the onset of the spindle by defining a tolerance period before and after the onset of each spindle."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>f1</th>\n",
       "      <th>n_self</th>\n",
       "      <th>n_other</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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>Cz</th>\n",
       "      <td>0.725373</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.840830</td>\n",
       "      <td>670</td>\n",
       "      <td>486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>0.646098</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.785006</td>\n",
       "      <td>551</td>\n",
       "      <td>356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>0.600575</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.750449</td>\n",
       "      <td>696</td>\n",
       "      <td>418</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         precision  recall        f1  n_self  n_other\n",
       "Channel                                              \n",
       "Cz        0.725373     1.0  0.840830     670      486\n",
       "Fz        0.646098     1.0  0.785006     551      356\n",
       "Pz        0.600575     1.0  0.750449     696      418"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ground-truth is `sp_multi` (other)\n",
    "sp.compare_detection(sp_multi, max_distance_sec=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>f1</th>\n",
       "      <th>n_self</th>\n",
       "      <th>n_other</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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>Cz</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.725373</td>\n",
       "      <td>0.840830</td>\n",
       "      <td>670</td>\n",
       "      <td>486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fz</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.646098</td>\n",
       "      <td>0.785006</td>\n",
       "      <td>551</td>\n",
       "      <td>356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.600575</td>\n",
       "      <td>0.750449</td>\n",
       "      <td>696</td>\n",
       "      <td>418</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         precision    recall        f1  n_self  n_other\n",
       "Channel                                                \n",
       "Cz             1.0  0.725373  0.840830     670      486\n",
       "Fz             1.0  0.646098  0.785006     551      356\n",
       "Pz             1.0  0.600575  0.750449     696      418"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ground-truth is `sp` (self)\n",
    "sp.compare_detection(sp_multi, max_distance_sec=1, other_is_groundtruth=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find spindles that are present on ALL channels only\n",
    "\n",
    "The code below show how to apply an even more stringent selection: only the spindles that are common across all electrodes are kept. This method is not natively implemented in YASA as it is certainly too restrictive when data has one or more tens of channels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "714 unique spindles are common across ALL channels.\n"
     ]
    },
    {
     "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>Start</th>\n",
       "      <th>Peak</th>\n",
       "      <th>End</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "      <th>Channel</th>\n",
       "      <th>IdxChannel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>604.36</td>\n",
       "      <td>604.73</td>\n",
       "      <td>605.12</td>\n",
       "      <td>0.76</td>\n",
       "      <td>60.649</td>\n",
       "      <td>12.520</td>\n",
       "      <td>2.210</td>\n",
       "      <td>0.350</td>\n",
       "      <td>12.646</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.481</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>735.12</td>\n",
       "      <td>735.90</td>\n",
       "      <td>736.39</td>\n",
       "      <td>1.27</td>\n",
       "      <td>80.165</td>\n",
       "      <td>16.962</td>\n",
       "      <td>2.544</td>\n",
       "      <td>0.579</td>\n",
       "      <td>12.672</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.609</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>744.01</td>\n",
       "      <td>744.98</td>\n",
       "      <td>745.47</td>\n",
       "      <td>1.46</td>\n",
       "      <td>79.723</td>\n",
       "      <td>16.769</td>\n",
       "      <td>2.603</td>\n",
       "      <td>0.636</td>\n",
       "      <td>12.774</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.660</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1321.01</td>\n",
       "      <td>1321.76</td>\n",
       "      <td>1322.28</td>\n",
       "      <td>1.27</td>\n",
       "      <td>98.536</td>\n",
       "      <td>21.469</td>\n",
       "      <td>2.546</td>\n",
       "      <td>0.299</td>\n",
       "      <td>12.360</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.586</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1364.50</td>\n",
       "      <td>1365.71</td>\n",
       "      <td>1366.22</td>\n",
       "      <td>1.72</td>\n",
       "      <td>71.419</td>\n",
       "      <td>15.509</td>\n",
       "      <td>2.284</td>\n",
       "      <td>0.436</td>\n",
       "      <td>12.870</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0.699</td>\n",
       "      <td>Cz</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Start     Peak      End  Duration  Amplitude     RMS  AbsPower  \\\n",
       "2    604.36   604.73   605.12      0.76     60.649  12.520     2.210   \n",
       "5    735.12   735.90   736.39      1.27     80.165  16.962     2.544   \n",
       "6    744.01   744.98   745.47      1.46     79.723  16.769     2.603   \n",
       "9   1321.01  1321.76  1322.28      1.27     98.536  21.469     2.546   \n",
       "11  1364.50  1365.71  1366.22      1.72     71.419  15.509     2.284   \n",
       "\n",
       "    RelPower  Frequency  Oscillations  Symmetry Channel  IdxChannel  \n",
       "2      0.350     12.646           9.0     0.481      Cz           0  \n",
       "5      0.579     12.672          15.0     0.609      Cz           0  \n",
       "6      0.636     12.774          19.0     0.660      Cz           0  \n",
       "9      0.299     12.360          16.0     0.586      Cz           0  \n",
       "11     0.436     12.870          21.0     0.699      Cz           0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from functools import reduce\n",
    "grp_start = all_spindles.groupby('Channel')['Start'].apply(lambda x: list(np.round(x).astype(int))).to_dict()\n",
    "grp_end = all_spindles.groupby('Channel')['End'].apply(lambda x: list(np.round(x).astype(int))).to_dict()\n",
    "\n",
    "intersect_start = reduce(np.intersect1d, (grp_start[c] for c in all_spindles['Channel'].unique()))\n",
    "intersect_end = reduce(np.intersect1d, (grp_end[c] for c in all_spindles['Channel'].unique()))\n",
    "\n",
    "idx_start = np.in1d(all_spindles['Start'].round().astype(int), intersect_start)\n",
    "idx_end = np.in1d(all_spindles['End'].round().astype(int), intersect_end)\n",
    "idx_good = np.logical_or(idx_start, idx_end)\n",
    "\n",
    "# Now we keep only these spindles in the dataframe\n",
    "print(all_spindles[idx_good].shape[0], 'unique spindles are common across ALL channels.')\n",
    "all_spindles[idx_good].head().round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test with a bad channel\n",
    "\n",
    "Here, we create a fake channel with no spindle to test the detection."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "14-Jun-22 15:34:32 | WARNING | No spindle were found in channel Fz.\n"
     ]
    },
    {
     "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>Count</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Cz</th>\n",
       "      <td>670</td>\n",
       "      <td>1.041463</td>\n",
       "      <td>78.698054</td>\n",
       "      <td>17.16594</td>\n",
       "      <td>2.428411</td>\n",
       "      <td>0.404349</td>\n",
       "      <td>12.655632</td>\n",
       "      <td>12.856716</td>\n",
       "      <td>0.519365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pz</th>\n",
       "      <td>696</td>\n",
       "      <td>1.036351</td>\n",
       "      <td>70.985229</td>\n",
       "      <td>15.39909</td>\n",
       "      <td>2.326625</td>\n",
       "      <td>0.396186</td>\n",
       "      <td>12.663587</td>\n",
       "      <td>12.811782</td>\n",
       "      <td>0.513046</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Count  Duration  Amplitude       RMS  AbsPower  RelPower  Frequency  \\\n",
       "Channel                                                                        \n",
       "Cz         670  1.041463  78.698054  17.16594  2.428411  0.404349  12.655632   \n",
       "Pz         696  1.036351  70.985229  15.39909  2.326625  0.396186  12.663587   \n",
       "\n",
       "         Oscillations  Symmetry  \n",
       "Channel                          \n",
       "Cz          12.856716  0.519365  \n",
       "Pz          12.811782  0.513046  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[1, :] = np.sin(0.1 * np.arange(data.shape[1]))\n",
    "sp_bad = yasa.spindles_detect(data, sf, ch_names=chan, multi_only=False)\n",
    "sp_bad.summary(grp_chan=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([70440,     0, 72840])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check that no events were marked as spindles on the second (bad) channel\n",
    "mask = sp_bad.get_mask()\n",
    "mask.sum(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**************\n",
    "\n",
    "## Example 2: Using a Raw object from MNE-Python\n",
    "\n",
    "This example demonstrates how to manipulate [MNE Raw object](https://mne-tools.github.io/stable/generated/mne.io.Raw.html#mne.io.Raw). The MNE package has several [functions](https://mne-tools.github.io/stable/python_reference.html#module-mne.io) to load the most standard EEG file formats (EDF, BrainVision, EEGLab, FieldTrip...).\n",
    "\n",
    "For the sake of this example, we'll load a PSG file encoded in the native MNE format (*.fif) using the [mne.io.read_raw_fif](https://mne.tools/stable/generated/mne.io.read_raw_fif.html) function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "\n",
       "<table class=\"table table-hover table-striped table-sm table-responsive small\">\n",
       "    <tr>\n",
       "        <th>Measurement date</th>\n",
       "        <td>January 15, 2016  14:01:00 GMT</td>\n",
       "        \n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Experimenter</th>\n",
       "<td>Unknown</td>\n",
       "    </tr>\n",
       "        <th>Participant</th>\n",
       "<td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Digitized points</th>\n",
       "        <td>15 points</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Good channels</th>\n",
       "        <td>6 EEG, 2 EOG, 1 EMG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Bad channels</th>\n",
       "        <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>EOG channels</th>\n",
       "        <td>EOG1, EOG2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>ECG channels</th>\n",
       "        <td>Not available</td>\n",
       "    <tr>\n",
       "        <th>Sampling frequency</th>\n",
       "        <td>100.00 Hz</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Highpass</th>\n",
       "        <td>0.00 Hz</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Lowpass</th>\n",
       "        <td>50.00 Hz</td>\n",
       "    </tr>\n",
       "\n",
       "    <tr>\n",
       "        <th>Filenames</th>\n",
       "        <td>sub-02_mne_raw.fif</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th>Duration</th>\n",
       "        <td>00:48:59 (HH:MM:SS)</td>\n",
       "    </tr>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<Raw | sub-02_mne_raw.fif, 9 x 294000 (2940.0 s), ~20.2 MB, data loaded>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import mne\n",
    "\n",
    "# Load the raw object\n",
    "raw = mne.io.read_raw_fif('sub-02_mne_raw.fif', preload=True, verbose=False)\n",
    "raw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Chan = ['F3', 'F4', 'C3', 'C4', 'O1', 'O2', 'EOG1', 'EOG2', 'EMG1']\n",
      "Sampling frequency = 100.0\n",
      "Data shape (channels, times) = (9, 294000)\n"
     ]
    }
   ],
   "source": [
    "# Let's have a look at the data\n",
    "print('Chan =', raw.ch_names)\n",
    "print('Sampling frequency =', raw.info['sfreq'])\n",
    "print('Data shape (channels, times) =', raw._data.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Chan = ['F3', 'F4', 'C3', 'C4']\n"
     ]
    }
   ],
   "source": [
    "# Keep only the channels of interests\n",
    "raw.pick_types(eeg=True).drop_channels(['O1', 'O2'])\n",
    "print('Chan =', raw.ch_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Start</th>\n",
       "      <th>Peak</th>\n",
       "      <th>End</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "      <th>Channel</th>\n",
       "      <th>IdxChannel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>879.76</td>\n",
       "      <td>880.04</td>\n",
       "      <td>880.53</td>\n",
       "      <td>0.77</td>\n",
       "      <td>49.915</td>\n",
       "      <td>9.490</td>\n",
       "      <td>2.017</td>\n",
       "      <td>0.393</td>\n",
       "      <td>13.803</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.359</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1068.45</td>\n",
       "      <td>1069.16</td>\n",
       "      <td>1069.46</td>\n",
       "      <td>1.01</td>\n",
       "      <td>39.570</td>\n",
       "      <td>7.860</td>\n",
       "      <td>1.522</td>\n",
       "      <td>0.268</td>\n",
       "      <td>13.472</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.696</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1078.82</td>\n",
       "      <td>1079.21</td>\n",
       "      <td>1079.38</td>\n",
       "      <td>0.56</td>\n",
       "      <td>41.574</td>\n",
       "      <td>9.711</td>\n",
       "      <td>2.070</td>\n",
       "      <td>0.409</td>\n",
       "      <td>13.175</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.684</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1099.28</td>\n",
       "      <td>1099.61</td>\n",
       "      <td>1100.41</td>\n",
       "      <td>1.13</td>\n",
       "      <td>46.056</td>\n",
       "      <td>9.804</td>\n",
       "      <td>2.064</td>\n",
       "      <td>0.593</td>\n",
       "      <td>13.297</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.289</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1104.88</td>\n",
       "      <td>1104.98</td>\n",
       "      <td>1105.77</td>\n",
       "      <td>0.89</td>\n",
       "      <td>31.216</td>\n",
       "      <td>7.461</td>\n",
       "      <td>1.565</td>\n",
       "      <td>0.248</td>\n",
       "      <td>13.439</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.111</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Start     Peak      End  Duration  Amplitude    RMS  AbsPower  RelPower  \\\n",
       "0   879.76   880.04   880.53      0.77     49.915  9.490     2.017     0.393   \n",
       "1  1068.45  1069.16  1069.46      1.01     39.570  7.860     1.522     0.268   \n",
       "2  1078.82  1079.21  1079.38      0.56     41.574  9.711     2.070     0.409   \n",
       "3  1099.28  1099.61  1100.41      1.13     46.056  9.804     2.064     0.593   \n",
       "4  1104.88  1104.98  1105.77      0.89     31.216  7.461     1.565     0.248   \n",
       "\n",
       "   Frequency  Oscillations  Symmetry Channel  IdxChannel  \n",
       "0     13.803          10.0     0.359      F3           0  \n",
       "1     13.472          13.0     0.696      F3           0  \n",
       "2     13.175           8.0     0.684      F3           0  \n",
       "3     13.297          15.0     0.289      F3           0  \n",
       "4     13.439          10.0     0.111      F3           0  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Multi-channel spindles detection\n",
    "# Note that since we're using a MNE Raw object, there is no need\n",
    "# to manually specify the sampling frequency and channel names: \n",
    "# they will be automatically inferred from the Raw object.\n",
    "sp = yasa.spindles_detect(raw)\n",
    "sp.summary().head().round(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visual inspection\n",
    "# %matplotlib widget\n",
    "# sp.plot_detection();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Count</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>RMS</th>\n",
       "      <th>AbsPower</th>\n",
       "      <th>RelPower</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Oscillations</th>\n",
       "      <th>Symmetry</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>C3</th>\n",
       "      <td>46</td>\n",
       "      <td>0.825</td>\n",
       "      <td>38.203878</td>\n",
       "      <td>8.622970</td>\n",
       "      <td>1.800360</td>\n",
       "      <td>0.351834</td>\n",
       "      <td>13.221299</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.392328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4</th>\n",
       "      <td>50</td>\n",
       "      <td>0.805</td>\n",
       "      <td>44.094940</td>\n",
       "      <td>9.215355</td>\n",
       "      <td>1.842662</td>\n",
       "      <td>0.350846</td>\n",
       "      <td>13.321107</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.552675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F3</th>\n",
       "      <td>45</td>\n",
       "      <td>0.760</td>\n",
       "      <td>43.508128</td>\n",
       "      <td>9.435272</td>\n",
       "      <td>1.884602</td>\n",
       "      <td>0.331049</td>\n",
       "      <td>13.431302</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.420635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F4</th>\n",
       "      <td>51</td>\n",
       "      <td>0.730</td>\n",
       "      <td>42.728126</td>\n",
       "      <td>9.284209</td>\n",
       "      <td>1.869022</td>\n",
       "      <td>0.351274</td>\n",
       "      <td>13.387199</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.481132</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Count  Duration  Amplitude       RMS  AbsPower  RelPower  Frequency  \\\n",
       "Channel                                                                        \n",
       "C3          46     0.825  38.203878  8.622970  1.800360  0.351834  13.221299   \n",
       "C4          50     0.805  44.094940  9.215355  1.842662  0.350846  13.321107   \n",
       "F3          45     0.760  43.508128  9.435272  1.884602  0.331049  13.431302   \n",
       "F4          51     0.730  42.728126  9.284209  1.869022  0.351274  13.387199   \n",
       "\n",
       "         Oscillations  Symmetry  \n",
       "Channel                          \n",
       "C3               10.0  0.392328  \n",
       "C4               10.0  0.552675  \n",
       "F3               10.0  0.420635  \n",
       "F4                9.0  0.481132  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp.summary(grp_chan=True, aggfunc='median')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As earlier, we can also plot an average template of all detected spindles. Furthermore, using the ``filt`` option, we can first apply a bandpass filter to the data in order to get the average spindles filtered in a specific frequency band. below, we're filtering our data in the spindles range, i.e. 12 to 16 Hz."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x324 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# To speed up the computation, we disable the bootstrap confidence intervals by specifying ci=None.\n",
    "# For more details, please refer to the documentation of the seaborn.lineplot function at:\n",
    "# https://seaborn.pydata.org/generated/seaborn.lineplot.html\n",
    "ax = sp.plot_average(center='Peak', time_before=0.8, time_after=0.8, filt=(12, 16), ci=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To get the actual peak-locked data underlying this figure, one can use the [get_sync_events](https://raphaelvallat.com/yasa/build/html/generated/yasa.SpindlesResults.html#yasa.SpindlesResults.get_sync_events) method. The output is a long-format dataframe with the original data values centered around each detected spindles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Time</th>\n",
       "      <th>Event</th>\n",
       "      <th>Amplitude</th>\n",
       "      <th>Channel</th>\n",
       "      <th>IdxChannel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.80</td>\n",
       "      <td>0</td>\n",
       "      <td>-9.274473</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.79</td>\n",
       "      <td>0</td>\n",
       "      <td>-6.832743</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.78</td>\n",
       "      <td>0</td>\n",
       "      <td>-4.507523</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.77</td>\n",
       "      <td>0</td>\n",
       "      <td>-3.998233</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.76</td>\n",
       "      <td>0</td>\n",
       "      <td>-8.483229</td>\n",
       "      <td>F3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30907</th>\n",
       "      <td>0.76</td>\n",
       "      <td>49</td>\n",
       "      <td>9.396885</td>\n",
       "      <td>C4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30908</th>\n",
       "      <td>0.77</td>\n",
       "      <td>49</td>\n",
       "      <td>6.334050</td>\n",
       "      <td>C4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30909</th>\n",
       "      <td>0.78</td>\n",
       "      <td>49</td>\n",
       "      <td>1.008784</td>\n",
       "      <td>C4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30910</th>\n",
       "      <td>0.79</td>\n",
       "      <td>49</td>\n",
       "      <td>-0.327676</td>\n",
       "      <td>C4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30911</th>\n",
       "      <td>0.80</td>\n",
       "      <td>49</td>\n",
       "      <td>2.864102</td>\n",
       "      <td>C4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30912 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Time Event  Amplitude Channel  IdxChannel\n",
       "0     -0.80     0  -9.274473      F3           0\n",
       "1     -0.79     0  -6.832743      F3           0\n",
       "2     -0.78     0  -4.507523      F3           0\n",
       "3     -0.77     0  -3.998233      F3           0\n",
       "4     -0.76     0  -8.483229      F3           0\n",
       "...     ...   ...        ...     ...         ...\n",
       "30907  0.76    49   9.396885      C4           3\n",
       "30908  0.77    49   6.334050      C4           3\n",
       "30909  0.78    49   1.008784      C4           3\n",
       "30910  0.79    49  -0.327676      C4           3\n",
       "30911  0.80    49   2.864102      C4           3\n",
       "\n",
       "[30912 rows x 5 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sync = sp.get_sync_events(center='Peak', time_before=0.8, time_after=0.8)\n",
    "df_sync"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculate the coincidence matrix\n",
    "\n",
    "Do spindles occur at the same time? One way to measure this is to calculate the coincidence matrix, which gives, for each pair of channel, the number of samples that were marked as a spindle in both channels. The output is a symmetric matrix, in which the diagonal is simply the number of data points that were marked as a spindle in the channel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Channel</th>\n",
       "      <th>F3</th>\n",
       "      <th>F4</th>\n",
       "      <th>C3</th>\n",
       "      <th>C4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>F3</th>\n",
       "      <td>3799</td>\n",
       "      <td>997</td>\n",
       "      <td>1386</td>\n",
       "      <td>779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F4</th>\n",
       "      <td>997</td>\n",
       "      <td>4118</td>\n",
       "      <td>1140</td>\n",
       "      <td>1678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C3</th>\n",
       "      <td>1386</td>\n",
       "      <td>1140</td>\n",
       "      <td>4132</td>\n",
       "      <td>1093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4</th>\n",
       "      <td>779</td>\n",
       "      <td>1678</td>\n",
       "      <td>1093</td>\n",
       "      <td>4356</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Channel    F3    F4    C3    C4\n",
       "Channel                        \n",
       "F3       3799   997  1386   779\n",
       "F4        997  4118  1140  1678\n",
       "C3       1386  1140  4132  1093\n",
       "C4        779  1678  1093  4356"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp.get_coincidence_matrix(scaled=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The coincidence matrix can be scaled (default) by dividing the output by the product of the sum of each individual binary mask. It can then be used to define functional networks or quickly find outlier channels."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Channel</th>\n",
       "      <th>F3</th>\n",
       "      <th>F4</th>\n",
       "      <th>C3</th>\n",
       "      <th>C4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Channel</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>F3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.000088</td>\n",
       "      <td>0.000047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>F4</th>\n",
       "      <td>0.000064</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000067</td>\n",
       "      <td>0.000094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C3</th>\n",
       "      <td>0.000088</td>\n",
       "      <td>0.000067</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C4</th>\n",
       "      <td>0.000047</td>\n",
       "      <td>0.000094</td>\n",
       "      <td>0.000061</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Channel        F3        F4        C3        C4\n",
       "Channel                                        \n",
       "F3       1.000000  0.000064  0.000088  0.000047\n",
       "F4       0.000064  1.000000  0.000067  0.000094\n",
       "C3       0.000088  0.000067  1.000000  0.000061\n",
       "C4       0.000047  0.000094  0.000061  1.000000"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coincidence = sp.get_coincidence_matrix()\n",
    "coincidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"white\", font_scale=1.25)\n",
    "\n",
    "mask = np.zeros_like(coincidence)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "f, ax = plt.subplots(figsize=(7, 5))\n",
    "ax = sns.heatmap(coincidence, mask=mask, square=True, cmap=\"rocket\")"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}