cosh_approx.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vypracování - Backpropagation v třívrstvé síti\n",
    "- V jazyku Java (či Python) vytvořte program, který umožní trénování třívrstvých sítí (sítí s 1 skrytou vrstvou).\n",
    "- Vytvořte a natrénujte síť tak, aby dokázala aproximovat Vámi zvolenou funkci z nabídky: sinh, cosh, tanh, coth, exp. (Neočekávám, že si všichni náhodou zvolíte tu stejnou.)\n",
    "- Proveďte analýzu citlivosti na počet neuronů ve skryté vrstvě, learning rate (parametr učení) a počet epoch učení - porovnejte mse.\n",
    "\n",
    "Pro vypracování byla vybrána funkce **cosh**\n",
    "\n",
    "## Teorie\n",
    "### Feed forward neuronové sítě\n",
    "Feed forward neuronové sítě se skládají s vrstev neuronů, které netwoří cyklus. Existují tři typy neuronů: neurony vstupní vrstvy, neurony skrytých vrstev a neurony výstupní vrstvy.\n",
    "\n",
    "Vrstva je definována jako vektor aktivačních funkcí vrstvy předhozí\n",
    "$a^{l} = \\sigma(w^{l-1}\\cdot a^{l-1} + b^{l})$\n",
    "kde w je matice vah $w^{l}(|a^{l}|\\times|a^{l-1}|)$,\n",
    "b je vektor biasů a $\\sigma(z)$ je libovolnou aktivační funkcí, \n",
    "\n",
    "Pro vstupní vrstvu l = 0 platí, že vektor b = 0.\n",
    "\n",
    "\n",
    "\n",
    "### Backpropagation\n",
    "Backpropagation je jádrem učícího algoritmu neuronové sítě.\n",
    "Výstup z feed forward se porovná s požadovanou hodnotou z tréningových dat $\\rightarrow$ k tomuto slouží účelová funkce C.\n",
    "\n",
    "Účelová funkce použitá v této síti: funkce střední kvadratické chyby pro n iterací\n",
    "\n",
    "$C = \\frac{1}{2n}\\sum_{x} ||y(x) - a^L(x)||^2$\n",
    "\n",
    "Chyba ve výstupní vrstvě\n",
    "$\\nabla_a C$ je gradient podle aktivací výstupní vrstvy, který sestává\n",
    "z parciálních derivací podle vah a biasů (3. a 4.)\n",
    "1. $\\delta^L = \\nabla_a C \\bigodot \\sigma^{\\prime}(z^L)$\n",
    "\n",
    "Chyba ve skryté vrstvě $l$, rekurentní zadání pro chybu ve vrstvě $l+1$\n",
    "\n",
    "\n",
    "2. $\\delta^l = ((w^{l+1})^T \\delta^{l+1}) \\bigodot \\sigma^{\\prime}(z^l)$\n",
    "\n",
    "$\\bigodot$ značí Hadamardův součin.\n",
    "\n",
    "\n",
    "3. $\\frac{\\partial C}{\\partial b^l_j} = \\delta^l_j$\n",
    "\n",
    "4. $\\frac{\\partial C}{\\partial w^l_{j k}} = a_k^{l-1}\\delta_j^l$\n",
    "\n",
    "vektory $\\delta$ jsou použity pro aktualizaci biasů a vah."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Výsledky"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from neural_net.neuralnet import NeuralNet\n",
    "import neural_net.activation_functions as af\n",
    "from matplotlib import pyplot as plt\n",
    "import itertools\n",
    "import json\n",
    "from pylab import rcParams\n",
    "rcParams['figure.figsize'] = 16, 6 # resize figures\n",
    "plt.style.use('bmh')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = [np.array([np.random.uniform(-1, 1)]) for i in range(1000)]\n",
    "y = [np.cosh(n) for n in x]\n",
    "training_data = [(x, y) for x, y in zip(x, y)]\n",
    "_x = np.linspace(-1, 1, 1000)\n",
    "_y = np.cosh(_x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Testovací parametry"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "neurons = [50, 10, 5, 3, 1]\n",
    "epochs = [200, 100, 10]\n",
    "learning_rates = [0.005, 0.05, 0.1, 0.5]\n",
    "\n",
    "params = itertools.product(neurons, epochs, learning_rates)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Kartézské součiny budou použitý v jednotlivých iteracích"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "stats = []\n",
    "for p in params:\n",
    "    mse_list = []\n",
    "    nn = NeuralNet([1, p[0], 1], a_functions=[af.sigmoid, af.linear], a_functions_prime=[af.sigmoid_prime, af.linear_prime])\n",
    "    print(p, end='\\n')\n",
    "    for e in nn.gradient_descent(labeled_training_dataset=training_data, no_epochs=p[1], mini_batch_size=1, learning_rate=p[2]):\n",
    "        print('EPOCH: {}\\tmse: {}'.format(e['epoch'], e['mse']), end='\\r')\n",
    "        mse_list.append(e['mse'])\n",
    "    result = {\n",
    "        'neurons': p[0],\n",
    "        'epochs': p[1],\n",
    "        'learning rate': p[2],\n",
    "        'mse list' : mse_list,\n",
    "        'last mse': mse_list[-1]\n",
    "    }\n",
    "    stats.append(result)\n",
    "    print('FINISHED')\n",
    "stats_json = json.dumps(stats)\n",
    "with open('resources/stats_cosh.json', 'w') as f:\n",
    "    f.write(stats_json)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Citlivostní analýza"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Tabulka výstupu analýzy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>neurons</th>\n",
       "      <th>epochs</th>\n",
       "      <th>learning rate</th>\n",
       "      <th>last mse</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>50</td>\n",
       "      <td>200</td>\n",
       "      <td>0.005</td>\n",
       "      <td>4.550588e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50</td>\n",
       "      <td>200</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.576850e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>50</td>\n",
       "      <td>200</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.417064e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>50</td>\n",
       "      <td>200</td>\n",
       "      <td>0.500</td>\n",
       "      <td>4.020041e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50</td>\n",
       "      <td>100</td>\n",
       "      <td>0.005</td>\n",
       "      <td>8.886462e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>50</td>\n",
       "      <td>100</td>\n",
       "      <td>0.050</td>\n",
       "      <td>8.230188e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>50</td>\n",
       "      <td>100</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.015384e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>50</td>\n",
       "      <td>100</td>\n",
       "      <td>0.500</td>\n",
       "      <td>3.694219e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>0.005</td>\n",
       "      <td>6.747720e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.075443e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>0.100</td>\n",
       "      <td>3.170347e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>0.500</td>\n",
       "      <td>5.806157e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>10</td>\n",
       "      <td>200</td>\n",
       "      <td>0.005</td>\n",
       "      <td>1.855490e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>10</td>\n",
       "      <td>200</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.368015e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>10</td>\n",
       "      <td>200</td>\n",
       "      <td>0.100</td>\n",
       "      <td>9.466203e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>10</td>\n",
       "      <td>200</td>\n",
       "      <td>0.500</td>\n",
       "      <td>3.773642e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.005</td>\n",
       "      <td>6.579429e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.769363e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.682562e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>10</td>\n",
       "      <td>100</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.669553e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>0.005</td>\n",
       "      <td>5.037918e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>0.050</td>\n",
       "      <td>9.217336e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>0.100</td>\n",
       "      <td>4.453521e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>0.500</td>\n",
       "      <td>7.630469e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>5</td>\n",
       "      <td>200</td>\n",
       "      <td>0.005</td>\n",
       "      <td>7.471602e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>5</td>\n",
       "      <td>200</td>\n",
       "      <td>0.050</td>\n",
       "      <td>4.374841e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>5</td>\n",
       "      <td>200</td>\n",
       "      <td>0.100</td>\n",
       "      <td>3.293414e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>5</td>\n",
       "      <td>200</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.935023e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0.005</td>\n",
       "      <td>9.049019e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0.050</td>\n",
       "      <td>4.530042e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.582729e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>5</td>\n",
       "      <td>100</td>\n",
       "      <td>0.500</td>\n",
       "      <td>2.491765e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>0.005</td>\n",
       "      <td>5.840269e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>0.050</td>\n",
       "      <td>5.095461e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>0.100</td>\n",
       "      <td>3.508909e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>0.500</td>\n",
       "      <td>2.432748e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>3</td>\n",
       "      <td>200</td>\n",
       "      <td>0.005</td>\n",
       "      <td>1.743977e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>3</td>\n",
       "      <td>200</td>\n",
       "      <td>0.050</td>\n",
       "      <td>5.270341e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>3</td>\n",
       "      <td>200</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.071298e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>3</td>\n",
       "      <td>200</td>\n",
       "      <td>0.500</td>\n",
       "      <td>6.158491e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0.005</td>\n",
       "      <td>3.549025e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.619495e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0.100</td>\n",
       "      <td>4.027127e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>3</td>\n",
       "      <td>100</td>\n",
       "      <td>0.500</td>\n",
       "      <td>2.054036e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.005</td>\n",
       "      <td>9.048959e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.050</td>\n",
       "      <td>5.309683e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.394260e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.383072e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>1</td>\n",
       "      <td>200</td>\n",
       "      <td>0.005</td>\n",
       "      <td>6.989904e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>1</td>\n",
       "      <td>200</td>\n",
       "      <td>0.050</td>\n",
       "      <td>9.277064e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>1</td>\n",
       "      <td>200</td>\n",
       "      <td>0.100</td>\n",
       "      <td>1.276824e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>1</td>\n",
       "      <td>200</td>\n",
       "      <td>0.500</td>\n",
       "      <td>6.849134e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>0.005</td>\n",
       "      <td>2.093512e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.151596e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>0.100</td>\n",
       "      <td>5.833758e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>1</td>\n",
       "      <td>100</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.753175e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.005</td>\n",
       "      <td>1.963039e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.050</td>\n",
       "      <td>1.112857e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.100</td>\n",
       "      <td>8.990957e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>0.500</td>\n",
       "      <td>9.465807e-03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    neurons  epochs  learning rate      last mse\n",
       "0        50     200          0.005  4.550588e-04\n",
       "1        50     200          0.050  1.576850e-04\n",
       "2        50     200          0.100  1.417064e-06\n",
       "3        50     200          0.500  4.020041e-02\n",
       "4        50     100          0.005  8.886462e-05\n",
       "5        50     100          0.050  8.230188e-04\n",
       "6        50     100          0.100  1.015384e-05\n",
       "7        50     100          0.500  3.694219e-03\n",
       "8        50      10          0.005  6.747720e-04\n",
       "9        50      10          0.050  1.075443e-03\n",
       "10       50      10          0.100  3.170347e-05\n",
       "11       50      10          0.500  5.806157e-04\n",
       "12       10     200          0.005  1.855490e-05\n",
       "13       10     200          0.050  1.368015e-04\n",
       "14       10     200          0.100  9.466203e-05\n",
       "15       10     200          0.500  3.773642e-06\n",
       "16       10     100          0.005  6.579429e-05\n",
       "17       10     100          0.050  1.769363e-04\n",
       "18       10     100          0.100  1.682562e-05\n",
       "19       10     100          0.500  1.669553e-05\n",
       "20       10      10          0.005  5.037918e-06\n",
       "21       10      10          0.050  9.217336e-05\n",
       "22       10      10          0.100  4.453521e-03\n",
       "23       10      10          0.500  7.630469e-05\n",
       "24        5     200          0.005  7.471602e-04\n",
       "25        5     200          0.050  4.374841e-06\n",
       "26        5     200          0.100  3.293414e-05\n",
       "27        5     200          0.500  1.935023e-05\n",
       "28        5     100          0.005  9.049019e-04\n",
       "29        5     100          0.050  4.530042e-04\n",
       "30        5     100          0.100  1.582729e-04\n",
       "31        5     100          0.500  2.491765e-05\n",
       "32        5      10          0.005  5.840269e-03\n",
       "33        5      10          0.050  5.095461e-04\n",
       "34        5      10          0.100  3.508909e-04\n",
       "35        5      10          0.500  2.432748e-02\n",
       "36        3     200          0.005  1.743977e-05\n",
       "37        3     200          0.050  5.270341e-05\n",
       "38        3     200          0.100  1.071298e-07\n",
       "39        3     200          0.500  6.158491e-07\n",
       "40        3     100          0.005  3.549025e-04\n",
       "41        3     100          0.050  1.619495e-05\n",
       "42        3     100          0.100  4.027127e-05\n",
       "43        3     100          0.500  2.054036e-07\n",
       "44        3      10          0.005  9.048959e-02\n",
       "45        3      10          0.050  5.309683e-04\n",
       "46        3      10          0.100  1.394260e-03\n",
       "47        3      10          0.500  1.383072e-05\n",
       "48        1     200          0.005  6.989904e-03\n",
       "49        1     200          0.050  9.277064e-03\n",
       "50        1     200          0.100  1.276824e-01\n",
       "51        1     200          0.500  6.849134e-02\n",
       "52        1     100          0.005  2.093512e-02\n",
       "53        1     100          0.050  1.151596e-03\n",
       "54        1     100          0.100  5.833758e-03\n",
       "55        1     100          0.500  1.753175e-02\n",
       "56        1      10          0.005  1.963039e-02\n",
       "57        1      10          0.050  1.112857e-02\n",
       "58        1      10          0.100  8.990957e-02\n",
       "59        1      10          0.500  9.465807e-03"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_json('resources/stats.json')\n",
    "df.rename(columns={'avg nmse': 'avg mse'}, inplace=True)\n",
    "df[['neurons', 'epochs', 'learning rate', 'last mse']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pro aproximaci byla použita síť s konečnou nejmenší střední kvadratickou chybou *mse*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "neurons                                                          3\n",
       "epochs                                                         200\n",
       "learning rate                                                  0.1\n",
       "mse list         [0.012408447090325, 0.052534303985128004, 9.88...\n",
       "last mse                                                1.0713e-07\n",
       "Name: 38, dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index_min = df['last mse'].idxmin()\n",
    "lowest_mse = df.iloc[index_min, :]\n",
    "lowest_mse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EPOCH:\t200\tmse: 1.9022531636549555e-06\n",
      "DONE\n"
     ]
    }
   ],
   "source": [
    "nn = NeuralNet([1, lowest_mse['neurons'], 1], a_functions=[af.sigmoid, af.linear], a_functions_prime=[af.sigmoid_prime, af.linear_prime])\n",
    "nn.learn(training_data, lowest_mse['epochs'], 1, lowest_mse['learning rate'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### grafické znázornění"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f1106231f10>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAAFzCAYAAABWytaSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAACrWUlEQVR4nOzdeVhUZfvA8e9hRzYRUcA1JTXNyLTU0nIpM9tttc3K3jJt0cqlzGzTNFts19LSMi0tW91azMxcMjWy1AxNURkQERlAYGA47x8gDpsrMPcZ7s91/a7f65zD8EzfjvFwlscwTROllFJKKaWUUqq6eLl7AEoppZRSSimlPJtOPJVSSimllFJKVSudeCqllFJKKaWUqlY68VRKKaWUUkopVa104qmUUkoppZRSqlrpxFMppZRSSimlVLXSiadSSimPZhhGe8MwHjcMw7+K3/d2wzCuqcr3tCLDMAIMw3jEMAxvo8gwwzBCq+l73W0YxlXV8d5KKaWql048lVJK1SjDMJoYhvGjYRjZhmGYxa/tNAzjyRN4D9MwjNuOY796wBfAv6Zp5p38qMu9b39gLLDqBL6mR/G4G5/i925e/D7dTuBr7jQMo+BUvm9lTNPMBRoAE4AxQCvTNO1V/X0Mw7gTeBT4tarfWymlVPXzcfcAlFJKyVE8URsBXA00B/KAXcBCYKppmrur4Ns8QdFE5Wwgs/i1c4FDVfDeJQzD8ALmAC+ZpvlZFb5vG+AF4FLTNPdV1fta3FPAb4A/0LGq39wwjM7AaKC3aZppVf3+Simlqp9OPJVSSgFFZyKBlUAB8DQQD2QApwE3A48BD1fytX6maTqO81udDvxmmua/h18wTTP15EdeMdM0C4G+1fC+W4HWVf2+Vlbc/uxqfP+1QJvqen+llFLVTy+1VUopddjbgB/QwTTNj0zT/NM0zV2maS43TXMwMOzwjoZhLDcMY4ZhGM8ZhmEDEotfv8UwjLWGYWQYhrHfMIyFhmG0cvk6E+gN3F18uejM4tdLXWprGIaPYRjjDMPYbhhGnmEYew3DeKOygRuGcZthGJmGYdxQPLZ3y2w3it9rbCVf/7FhGN9V8PpiwzBmG4bRwjCMQsMwzi+z/ULDMJyGYTQr/vM9hmFsMQwj1zCMA4ZhrDjapbWGYcQahvG5YRgHDcNINwzjO8Mw2pfZ50bDMBKK33MVcNbJvE+Z/QMMw1hgGMYmwzAaVbJPuctzDcNoXNyth8trLQ3D+Kz48x4yDONPwzCucNnesXg8WYZhpBZ/32Yu258u/nxXG4axtfgS7OWGYZzusk+VXKaslFLKfXTiqZRS6vAltv2ANyq7P880TbPMSzcCkRRNJC8pfs0feB44p/g1J7DQMAy/4u3RwGqKLoGNppIzqMAMYChFZ17bAtcBOyoZ+0jgTeAq0zTnA9OAAYZhBLvs1gtoVvy+FZkF9DYMI8blfaOLP8OHpmnuAL4H/lfm6/4HfGea5i7DMDoCUym6DLc1cBHwYSXfD8MwGlJ0hnkf0B3oAvwDLDcMI7J4nw7AXGA+EAe8BLx2ou9TZv9w4AcgAuhumubeysZ4LIZhRFF0n2td4CqgPUX3vhYWb28L/ExR804UdXAC3xuGEeDyVtHA/cCtwPlACPD+yY5LKaWUPHqprVJKKYBYin4ZucX1xTJn2HaZptnOZbMNGFJ8SSsApml+UObr7wTSKLqH81fTNJMNw3AAOaZpJlc0EMMwYoE7gBtc7s3cDqwps6uXYRivATcAF5mmGV/8+gLgdYouD55e/No9wELTNJMq+fw/AMkUTXwmF792a/FrPxT/eRrwkWEYD5umaTcMoy5FE+Jbi7c3BbKBL10m75sq+X5QNNHaaZrm/S6f/SGKfgFwKzCFoofprDFN8/HiXf4pnhy/cYLvc/j1JsASiiamtxQ/GOhUDAVM4GrTNLOLX9vusn0k8K1pmuNcxnAbkE7RZdBfFr/sD9x++JJrwzBeBOYahhFQBWNUSiklgJ7xVEop5coo8+ebKLp3710gqMy29a6TTgDDMM42DOMLwzD+Mwwjk+JLcCk623i8zin+/+UufS3jeeAW4HyXSSfFT6+dSfHZScMwIoBrgfcqe6PizzEbuN3l5duBj10+49cU3fN6eKJ5W/Gfvyn+8/cUnZX9zzCMTwzDuNcwjPpHGf+5QMfiS1CzDMPIouhhS80pug8Wis72ln1y7sqTeB8o+m/+auAv4PoqmtB1BFa5TDrLOhe4tszY0oCAMmNLKnOfbxJF/y42qIIxKqWUEkDPeCqllAJIoOjyyDNcXzz8FFvDMA5U8DWlJhuGYdShaLK4ErgLSCne9DdF945WtR8omhjfDEwss20a8KhhGGdRdHlnKrD4GO/3ITDSMIyzi/98FjDg8EbTNAsMw5hB0YT2HYrOon5gmmZB8fYswzA6ARcAFwODgRcNw+htmub6Cr6fF/Aj8EAF2zKOMdaTeZ9C4FugP9COo5+NPbx/Wb4nMK7DY/uI8n2gaAJ6WNkHUx2+rFt/Qa6UUh5CJ55KKaUwTfOAYRiLgQcNw3jTNM0TmfgcdgZF93yOMU1zC0Dxw3jKnkU9lg3F/78PcLRlUJZRdG/mQsMwfE3TfO7wBtM0EwzDWEbRJLEn8L5pms6jfVPTNP82DGM9RWc6DYrO6G4us9t04AnDMAZTNDHtX+Y9nMAKYIVhGOOAzRSdla1o4vk7cCew5yhnHzdTdM+jqwtO4n0Oux/IB34yDONi0zT/OMq++wBvwzAamqZ5+JcI55TZZz3wP8Mwgio56/k7Rf+ctldwj7BSSqlaRH+TqJRS6rAhFE1KNhqGcYdhGGcVP831MuAKih4KczS7KFr388HiJ532puhBOCc04TBNMwH4GHjbKHpabUvDMM41DKPcg4hM0/wZuBQYYRjGs2U2TwPupWhCPL3s11biQ4omigMomtSW/X67KLpH8jXgx+KHDgFQ/FTW4cVPcW0KXAM0oWjyWJE3AW/gK8MwuhuG0dwwjG6GYYw3jjw991Wga/FrrQzDuJai+z5P9H0Oj980TfPB4s+2rPgMbWV+o+iS3YmGYZxuGEZfitbrdPU2RT9LfGUYxgWGYZxmGMYVxf/OAEyg6J//bMMwzive3tMwjNcMw2hxlO+tlFLKw+jEUymlFACmaSYCHSh6gurjwFqKLpN9maJ7A3sf4+v3U3Tf4yXFX/cSRWt/VnTJ5rHcRdHE8XmKHnj0BUXriVb0fX8t/p4PGYbheknnlxRdarrk8CXDx2EORU97jaDoabIVeZeiS4ffLfN6OnAlRRPTbcCLwPOmaVb4JN3is4hdgf0UPRDpH4om3M0oenATxZfo3kLR5cSbgNHA8BN9nwq+96MU/fP9wTCMLpXsc4CiCXgX4E+KnlY7ssw+NqAbRRPURRR1H0/xWe7iM9/nA8HAUoom4e8BgcDBir6vUkopz2TolS9KKaU8UfFDhfYAN5um+VUVvu8QYBzQxDTNsvcmKqWUUqoCeo+nUkopj2IYhi9FZyyfBvZy5Kmzp/q+wUBjis76vaWTTqWUUur46aW2SimlPM0FFF1i2gcYWHbJl1PwJkWXnP7NkbU+lVJKKXUc9FJbpZRSSimllFLVSs94KqWUUkoppZSqVjV6j+fy5ctNf3//mvyWJ8Q0TQzjRJebU9VNu8ijTWTSLvJoE5m0izzaRB5tIpMVuhw6dGh/7969I8u+XqMTT39/f9q0aVOT3/KE7N+/n/r167t7GKoM7SKPNpFJu8ijTWTSLvJoE3m0iUxW6LJhw4ZdFb2ul9q6yM7OdvcQVAW0izzaRCbtIo82kUm7yKNN5NEmMlm5i048XURFRbl7CKoC2kUebSKTdpFHm8ikXeTRJvJoE5ms3EUnni6Sk5PdPQRVAe0ijzaRSbvIo01k0i7yaBN5tIlMVu6iE08Xvr6+7h6CqoB2kUebyKRd5NEmMmkXebSJPNpEJit30Ymni7CwMHcPQVVAu8ijTWTSLvJoE5m0izzaRB5tIpOVu+jE08X+/fvdPQRVAe0ijzaRSbvIo01k0i7yaBN5tIlMVu6iE08XVv4NgifTLvJoE5m0izzaRCbtIo82kUebyGTlLjrxdOFwONw9BFUB7SKPNpFJu8ijTWTSLvJoE3m0iUxW7qITTxc5OTnuHoKqgHaRR5vIpF3k0SYyaRd5tIk82kQmK3fRiacLK6+L48m0izzaRCbtIo82kUm7yKNN5NEmMlm5i048XVh5XRxPpl3k0SYyaRd5tIlM2kUebSKPNpHJyl104nmYaRK6fbu7R6Eq4Ofn5+4hqDK0iUzaRR5tIpN2kUebyKNN5PHasgUrV9GJJ4DTSfBNN3HaTTfh/eef7h6NKiMkJMTdQ1BlaBOZtIs82kQm7SKPNpFHm8hipKcTcs01nH7TTRgWPeupE08Ab2+cLVpgOJ3UeeghKChw94iUi7S0NHcPQZWhTWTSLvJoE5m0izzaRB5tIkvgU0/hlZpKXlAQZoMG7h7OSdGJZ7GcJ5+koFEjfP78E/+33nL3cJSL8PBwdw9BlaFNZNIu8mgTmbSLPNpEHm0ih8/PP+P/8ceYfn5kTJ4MXtacwllz1NUhOJikZ54BIHDSJLwSEtw8IHWYlR8b7am0iUzaRR5tIpN2kUebyKNNhMjOps6wYQDkjhxJZqNG7h3PKdCJp4sDHTuSN2AARm5uUeDCQncPSQG5ubnuHoIqQ5vIpF3k0SYyaRd5tIk82kSGwAkT8N61i4IzzyT3wQct3UUnnsVs9jw+31HAyAvvIjOsHr6rVuE3a5a7h6Ww9npFnkqbyKRd5NEmMmkXebSJPNrE/bzXrcN/6lRMb28OvfEG+PpauotOPCmadA7/+h9yMvazKtOb8Vc+AEDAuHEYe/e6eXTKyusVeSptIpN2kUebyKRd5NEm8mgTN8vLI+ihhzBMk7wHHsAZFwdYu4tOPIEpv+ziQK6Tg/kGAD+2v5Cf2l2Ad1YWdR57DEzTzSOs3QICAtw9BFWGNpFJu8ijTWTSLvJoE3m0iXsFvPoq3v/8g7NlS367fQi3f/I3186K58P4NOKTMt09vJOiE0/gr5RDAKQ5iv9xGAYvXvMQmQFB+C1diu+CBW4cnQoMDHT3EFQZ2kQm7SKPNpFJu8ijTeTRJu7jtXkzAa++CsDKEc/y2LI9pGQ5yM4v5L/MQkYvTrDk5FMnnkBh8RnNlkHOktf2h9XntcvvA6DO6NEYupaR26Snp7t7CKoMbSKTdpFHm8ikXeTRJvJoEzdxOosusc3P58BtAxmVGYXrtZctg5w4TXhpRaLbhniydOIJhAf6ALAt07vU61+d14/fW56NV1oagWPGuGNoCoiIiHD3EFQZ2kQm7SKPNpFJu8ijTeTRJu7hP3UqPhs2UBgTw8Red+Isc8ff4flKVl6BG0Z3anTiCYzq0RyA6MAyy6cYBuOve4RcX3/8583D5/vva35wisxM611K4Om0iUzaRR5tIpN2kUebyKNNap7Xjh0ETpgAwJ7nJ7HyQPnnzByerwT7+9To2KqCTjyBuJgQRvdoSohP+bh76jdiWp87AfB9eDjoQVjjHA6Hu4egytAmMmkXebSJTNpFHm0ijzapYaZJnWHDMHJyOHhNf4bkNie/sPzcJMTHxAAeu7BpzY/xFOnEs1iv2AiMOuEVbpvb7To2N25NQHIS5tina3ZgytLrFXkqbSKTdpFHm8ikXeTRJvJok5rl9+GH+K5ciSO8HredcwcpWRVP/Ncf9GFUj6bExYTU8AhPnU48XVzd3IdAn/L/SJze3jx3w6MUeHlT78MP8Fm1yg2jq72svF6Rp9ImMmkXebSJTNpFHm0ijzapOUZSEnWeegqA8VcMISWg8knl1U296RVrzftvdeLpIrJuMM/2aVHhtoTolszsOQAAY/AQyM6uyaHVavo4b3m0iUzaRR5tIpN2kUebyKNNaohpUuexxzAyM9nUsTuLzuxR6a7RIX50bdmg5sZWxXTi6cLPz4+4mBA6RAdVuH1G79v4N+o0gvckkv34uBoeXe3l5+fn7iGoMrSJTNpFHm0ik3aRR5vIo01qht/8+fgtWYIZEsLzVz4IhlHhfg2D/Zh4WSwNwurU8Airjk48XWRkZAAwrHuzCrcX+PjyzI0jKfDyIubjD/BZvbomh1drHe6i5NAmMmkXebSJTNpFHm0ijzapfkZyMoGjRwPwz8in2FGn4ktovQx4sV8s0aH+lu6iE08X9evXByA61J/RPSp+UtQ/jVsxq+cteJkmPoOHwKFDNTnEWulwFyWHNpFJu8ijTWTSLvJoE3m0STUzTeo88gheBw9iu+AibvfrUOmucQ2DiA71B6zdRSeeLlx/g9ArNqLSyef04ktuA3fvolCfclvtrPybHU+lTWTSLvJoE5m0izzaRB5tUr385s3Db8kSCkJCGdTj/kovsfUGhl145GpMK3fRiaeL/Pz8Un/uFRtB2wblr6Mu8PHl2eJLbsNnziDjh59raoi1Utkuyv20iUzaRR5tIpN2kUebyKNNqo9hs5VcYvvi5feTGhZZ6b7nNQ0tOdsJ1u5yzImnYRjvG4axzzCMvyrZ3sMwjAzDMP4o/r+nqn6YNaOi9YpG9WiOVwW/gNjqcsmt95AH2LR9Xw2MsHbSdaTk0SYyaRd5tIlM2kUebSKPNqkmhy+xzchg3Zld+KLDJZXuGujjxeAujUu9ZuUux3PGcybQ9xj7/GKa5tnF//fsqQ/LPSparyg61J+RFx39ktsm+/eyd/gY4pMyq3uItZKuIyWPNpFJu8ijTWTSLvJoE3m0SfXw+/RT/JYuJadOMOOuHlbpJbYAz/ZpUepsJ1i7yzEnnqZprgAO1MBY3C4oqOJlVHrFRjC5Xyxl/7VwveT2pl8X8O30r6p/kLVQZV2U+2gTmbSLPNpEJu0ijzaRR5tUPcNmI/DxxwGYe+vwo15iO7pHU+JiQsq9buUuPlX0Pl0Nw4gHkoDHTNP8u6Kd9u3bx6BBg/Dx8cHpdNK/f3+GDh1KcnIyQUFBeHt7Y7fbiYyM5MCBA5imSWRkJCkpKQQHBwOQlZVFw4YNSU1NxTAM6tWrR2pqKqGhoTidTrKzs4mKiiI5ORlfX1/CwsLYv38/YWFhOBwOcnJySrb7+fkREhJCWloa4eHh2O32Ul8fEBBAYGAg6enptIyI4LrTvEnNzGH9QR861i0gzeFFZquW/NTvJi75di4j50xkyzXn0KJ1SzGfKScnh9zc3Ao/U0REBJmZmTgcjpLtgYGB+Pn5kZGRQf369cnIyCA/P79kuzs+U0FBAQEBAR71mazeydvbm3379nnUZ/KETmlpaTgcDo/6TFbvlJaWhre3t0d9Jk/oFBAQwO7duz3qM1m9k5+fH7t27fKoz2T1Tjk5OezatcujPpNbO9WtS8SQIXhlZGDv0YtdF3Yj7lA+aQ4vWgY52ZbpTXRgIeF+0OfsWOqZmezb5yz3mTIzMwkJCZHxmSrpVBnDNM1jzioNw2gOfGua5pkVbAsFCk3TzDIMox/wmmmap1f0PqtXrzbbtGlzzO/nLrt27aJZs4rX8ASw2fO4a/5mCsv8I/MtcPDh60OITf6PL3veQKsPXi93WlydvGN1UTVPm8ikXeTRJjJpF3m0iTzapGr5zZ1L0NChZAcGc8Mj0ys82xkd4sfEy2KPOo+wQpcNGzas7927d6eyr5/yU21N07SbpplV/L8XAb6GYVhygZnIyMpPd0Pl93vm+/jxTPElt1ct/4x5by+oriHWSsfqomqeNpFJu8ijTWTSLvJoE3m0SdUxkpIIGF10ie3kK4dUOOlsGHzsSSdYu8spTzwNw4gyjKK7Yg3DOK/4PdNO9X3d4cCBY9/KWtn6nlsbt2JWjwF4mSZ3z5jAz5t2V8cQa6Xj6aJqljaRSbvIo01k0i7yaBN5tEkVMU2Chg/HO9POL206s7Bjnwp3iwrxO64rJq3c5XiWU5kLrAZaG4axxzCMQYZhDDYMY3DxLtcDfxXf4/k6cLN5PNfvCnS8w+4VG8GsG9viW2adlRkX30ZC1Gk0TdtL3uPj9Cm3VcSi/zp5NG0ik3aRR5vIpF3k0SbyaJOq4Td3Lr7ff489MJgJ1z1S6VNsI+r4Htf7WbnL8TzVdoBpmtGmafqaptnYNM0ZpmlONU1zavH2N03TbGeaZpxpml1M01xV/cOuHidy6jo61J9OjUo/aSrfx49xN42iwMubm1Z9yWevfaqTzypg5UsKPJU2kUm7yKNNZNIu8mgTebTJqTP27CGg+Cm2L181lP1hFd+N6GXAwI7Rx/WeVu5yypfaepKUlJQT2n9w18blXtvW6HTevWQgAGPnTeadJRU+4FedgBPtoqqfNpFJu8ijTWTSLvJoE3m0ySkqLCTogQfwzsxkebvzWXTOJZXuOvKipsf9YFIrd9GJp4vDjxw+XtGh/gzuXP63Ex/2uJlNTc+gYUYqA+e8is2eV1VDrJVOtIuqftpEJu0ijzaRSbvIo03k0Sanxv+99/BdsYIDQXWPeont5H6x9IqNOO73tXIXnXieov7to2hbv/R6NU5vb565cRS5vv702/A9Hz41jWUJlnzeklJKKaWUUuoEeG3bht+4pwEYf/0jpAeHV7jf6B5NiYsJqXCbJ9KJp4usrKyT+rpRvVoQ7l/6H+WuBk14vd+9Rds/f5X3vvlD7/c8SSfbRVUfbSKTdpFHm8ikXeTRJvJok5OUn48x6F58HHl80+lSVrS7oMLdYuv5n9CZzsOs3EUnni4aNmx4Ul8XHerPlKvb0CEqqNTrn3W9it9izyE8O4PHP3+V53/YXhXDrHVOtouqPtpEJu0ijzaRSbvIo03k0SYnJ3/iS4T+/Se2ug14+cohFe7j5wVjL255Uu9v5S468XSRmpp60l8bHerPpCta0avlkVPpppcXz944gsyAIC7avIruqxbrWc+TcCpdVPXQJjJpF3m0iUzaRR5tIo82OXHeGzdS7/VXAHjmxpFkB1Z8P+b4vrHH/TChsqzcRSeeLoxKbvo9EQM7RuPt8jYpdRsw+eoHAXj0q7d4+eNfWbAp+ZS/T21SFV1U1dImMmkXebSJTNpFHm0ijzY5QTk5OO/+Hz5OJ3O6Xcf62A4V7ja4c/Qp3ddp5S468XRRr169U36P6FB/RlzUtNRri8+5mJ/O7EZw3iHGfTqJaav36uTzBFRFF1W1tIlM2kUebSKTdpFHm8ijTU7M3mFPEL5rB/81aMrblw2qcJ/BnaPp3z7qlL6PlbvoxNNFVZ267hUbQXigz5EXDIMJ/YeTFlyXTjviufnXL5i61lYl36s2sPIlBZ5Km8ikXeTRJjJpF3m0iTza5Pjt+nIpZ86fRYGXF+NuGk2eb/nLaOv6e53ypBOs3UUnni5CQ0Or7L2e6Nm81CW3B4OL1/ABhiyeTvOUXXq/53Gqyi6qamgTmbSLPNpEJu0ijzaRR5scJ7udmMeGATCj9+1sadK6wt3G9G5RJd/Oyl104unC6XRW2XvFxYQw8bLYUq+taHcBX3fqS0CBg2c+ncTTi7dhs+dV2ff0VFXZRVUNbSKTdpFHm8ikXeTRJvJok+NT5/HHaXAghc2NW/NBr1sq3OdU7+t0ZeUuOvF0kZ2dXaXvFxcTQrBv6X/Er1w1BFvdBrTd8w+3L5nF6MUJOvk8hqruok6dNpFJu8ijTWTSLvJoE3m0ybHt/egz/OfOJdfHj6duHo3T26fcPlVxX6crK3fRiaeLqKiq+5fisHGXlD6tnh0QxNM3jabQMLhz2Rwa/LmeWev1fs+jqY4u6tRoE5m0izzaRCbtIo82kUebHN2va7bS6IkRALzV7x52NWhabp+qnnSCtbvoxNNFcnLVP2k2LiaEwZ2jS722oWUcH/a4GW+zkGc+mcjWBJue9TyK6uiiTo02kUm7yKNNZNIu8mgTebRJ5WwHc4gY9jDh2RmsPf0cPj3/2lLbGwb7MblfbJVPOsHaXXTi6cLX17da3rd/+ygm94slwOfIP+5plwxkS6PTaZSezL2fTtFLbo+iurqok6dNZNIu8mgTmbSLPNpEHm1SuT+feomu29ZxsE4IT984CtPryM/4Qb5efHRzuyq7p7MsK3fRiaeLsLCwanvvuJgQpvVvQ3SIHwAFPr6MHfAEub7+XL7he878ZSkjF+nksyLV2UWdHG0ik3aRR5vIpF3k0SbyaJOKJfz8O9d+8iYA469/jP1h9UttD/Yvf59nVbJyF514uti/f3+1vn90qD8TL4ulboA3ALsaNOXVK+4HYPQXU2D3bgYv2KrLrJRR3V3UidMmMmkXebSJTNpFHm0ijzYpL3lfBvWGDMa/IJ8vz+vH8jO7ldvnsQvL3+tZlazcRSeeLmriNwjRof6c0+jI+jsLulzBijO6EpqTxTOfTiLPkc+TS7frmU8XVv7NjqfSJjJpF3m0iUzaRR5tIo82KS/5kTG0tO1gV/1GvHLlkHLb20YGVtsltodZuYtOPF04HI4a+T4DO0aXXHKLYfD8DY+SFlyXjjviue3neeQ5Tab8sqtGxmIFNdVFHT9tIpN2kUebyKRd5NEm8miT0hLnL6TnojkUeHnz1IAnyPEPLLXd38dgVM/Tqn0cVu6iE08XOTk5NfJ9Dl9y2zC4aPKZHhzOszeMBGDwdzNpvWcbG23ZesltsZrqoo6fNpFJu8ijTWTSLvJoE3m0yRH7dibRaMQwAKb1GcjmJm1Kbfcy4Pk+LYkO9a/2sVi5i048XdTkujjRof682C8Wo/jPq87ozLzzr8bXWcDzcyfg78hl7Hc79JJbrL1ekafSJjJpF3m0iUzaRR5tIo82KWLLyOXAXfdT357GhtPa82GPm0ttN4BJl8VW+yW2h1m5i048XdT0ujjRof6cHRVU8ufXL7+PHQ2a0jx1N8MWTiO3oFAvucXa6xV5Km0ik3aRR5vIpF3k0SbyaBOw2fP4fsyrdI3/hayAIMbd/DiFXt6l9unSJLTGJp1g7S468XTh5+dX499z2IXN8PcuOu+Z5+vP2FvGkO/tw/Wrv6bb5tVstGUzauG2Wn3m0x1d1NFpE5m0izzaRCbtIo82kUebwOcLfuXe+W8AMPHah0kOb1hqe4CPF4O7Nq7RMVm5i048XYSE1NxvKw6LDvXn+UtbUjz3ZFtMLG9fejcAY+e/RETmATbasnns29o7+XRHF3V02kQm7SKPNpFJu8ijTeSp7U1saVlc/8oYAvNzWdyhN0s79C61PdDHi+f6tKiR+zpdWbmLTjxdpKWlueX7xsWEMPGy2JI/f3zhDayL7UC97IM8/clEjMJCUg8V1NrLbt3VRVVOm8ikXeTRJjJpF3m0iTy1uYnNnsffDzzBGXv+ISm8IZOueajUdl8vg6n929ToJbaHWbmLTjxdhIeHu+17x8WEUL+ODwCmlxdP3TSa9KAwuvy7nttWzAOotU+6dWcXVTFtIpN2kUebyKRd5NEm8tTWJssS0nhrwmyu++5jnIYXT938ONmBwaX26dQ4pMbPdB5m5S468XTh7scTj+rRvOSS2/1h9XnmxqIlVoYseZ+2iVsBeKoWPunW3V1UedpEJu0ijzaRSbvIo03kqY1NliWkMe3beJ6Z+wJepsn0i28n/rT2pfZpEOTL4C41e1+nKyt30Ymni9zcXLd+/8OX3B5+2NCvZ3RhTrfr8Cl0Mn7O8wTlZpNTUMjUNXvcOs6a5u4uqjxtIpN2kUebyKRd5NEm8tTGJlN+3sW4T1+kfuYB1reI4/3et5ba3jDYj8mXn+62s51g7S468XQhYV2cuJgQ3r3ujJLJ55v97mFrTCyND9gYvWAKmCZrEu216pJbCV1UadpEJu0ijzaRSbvIo03kqW1NbPY8+v/8GRf88xsH64TwVJmlU6JD/HixX6xbJ51g7S468XQhZV2cw0+69TIg38ePMbc+ySG/APr+sYwr1i/FpHZdciulizpCm8ikXeTRJjJpF3m0iTy1rckPn/7IA4unA/DcDSPZVzeyZJuXARMvc/+kE6zdRSeeLgICAtw9hBJxMSFMuiwWA0iMbMKLxU/TGvHlGzTbt5ucgkJmrbe5d5A1RFIXVUSbyKRd5NEmMmkXebSJPLWpScre/Vw3eSS+zgI+ueBaVrQ7v9T2kRc1FTHpBGt30Ymni8DAQHcPoZS4mBC6NAkFYGHHPiw5uzd1HLk8P+d5fAscrE3MYOJPOz3+zKe0LkqbSKVd5NEmMmkXebSJPLWlSfxeO4kDh9BofxLbolvyRr97S21vGxlAr9gIN42uPCt30Ymni/T0dHcPoZzBXRsT4OMFhsHE/g+zp140bZISeGDRdLLzC1m2PZ3h32zz6MmnxC61nTaRSbvIo01k0i7yaBN5akOT+KRMfn7+bfps+IEc3wCeuHUsDl+/ku3RIX6M6tnCjSMsz8pddOLpIiJCzm8zDosO9ee5Pi0I9PEiOyCIMbc8SYGXN7es/JwLtqwB4EBOAZOW73TvQKuRxC61nTaRSbvIo01k0i7yaBN5PL2JzZ7Hu7N/ZsQXrwMw+ZoH2dWgScn2ugE+Yu7rdGXlLjrxdJGZKfNJsXExIUzt34ZeLcPZ3LQNb/UdBMC4eS9SP2M/AJv3HfLYJ91K7VKbaROZtIs82kQm7SKPNpHHk5vY7Hk88dVmnvrwWeo4cllydm++6XRpqX3OaRQibtIJ1u6iE08XDofD3UOoVHSoP6N7NsfbgI8vvIHVrToRnp3Bs5+8gFehE4ARixJYsMm6T7qqjOQutZU2kUm7yKNNZNIu8mgTeTy1SXxSJvfM38xN89+iTVICuyNimNj/YTCMkn0CfbwY2DHajaOsnJW76MTThRXWxakb4IPp5cXTN40iLTicc7f/wd0/flyyfepaG8sS0tw4wqpnhS61jTaRSbvIo01k0i7yaBN5PLGJzZ7H44sT6PL3KgasXECBlzdP3jKG7ICgkn28DHi2TwuRZzvB2l104unCCuvijO7ZHIADIfUYO+AJCg2D//3wIef+u6Fkn5dXJLppdNXDCl1qG20ik3aRR5vIpF3k0SbyeGKTKSsTqX8ghXGfvgjAW5fdw+YmbUq2B/p4MemyWOJiQtw1xGOychedeLqwwuOJ42JCGN2jKQaw7vRzmN77drxMk+fmTiAi8wAA+YV41P2eVuhS22gTmbSLPNpEJu0ijzaRx9OaxCdlsikxnQkfP0dYTia/tOnMx92vL7XP1P5tRE86wdpddOLpws/P79g7CdArNoIX+8UCMOPi21gX24GIrHSemzO+5H7P0YsTPGbyaZUutYk2kUm7yKNNZNIu8mgTeTypic2ex1Pf7eCBxe/RPnELyXUb8MxNozC9jkyF6tfxEXt5rSsrd9GJp4uMjAx3D+G4xcWEEFvXj0Ivb8YOeKLkfs97fvgIAKcJY7/b4RHre1qpS22hTWTSLvJoE5m0izzaRB5PaWKz5/Hot9s4748V3PrL5xR4efP4rWPJCAortd+oHs3dM8ATZOUuOvF0Ub9+fXcP4YSM7ROLD5AWUo8nbxmD0/Bi0I+zOW/b7wDkFhQydc0e9w6yClitS22gTWTSLvJoE5m0izzaRB5PaGKz5zFkwWb89uxm3Lyi+zrf6Pc//mrWttR+o3s0FX+J7WFW7qITTxdW+w1CdKg/L/SLxduA32M7MP3iw/d7vlCyvufvezItf9bTal1qA20ik3aRR5vIpF3k0SbyeEKT0Qu34sh18MLs5wjJzWZ5u/OZ43Jfp5cBk/vF0is2wo2jPDFW7qITTxf5+fnuHsIJi4sJYeJlRZPP93vfytrTz6Fe9kGenzsBb6eT/EKT+xZstfT9nlbs4um0iUzaRR5tIpN2kUebyGP1Jgs2JWPLLuShhe/Sds8/7A2P4tkbRpas12mA+CfYVsTKXXTi6cKq6+Icnnzi7c1TNz/O/pB6dNwRz/9++BAouuTWyg8bsmoXT6ZNZNIu8mgTmbSLPNpEHis3sdnzmLrWRq8/f+bmX78g39uHJ24bS2adI5PMLk1CLTfpBGt3OebE0zCM9w3D2GcYxl/H2O9cwzAKDMO4/mj7SWbldXHiYkLo3DiUAyH1GFN8v+ddyz6mc/H9nlZ+2JCVu3gqbSKTdpFHm8ikXeTRJvJYucmUlYk0Skti7GcvA/Da5feVWq8TYHDXxu4Y2imzcpfjOeM5E+h7tB0Mw/AGJgHfVcGY3CYoKMjdQzglg7s2JjrEjw0tz+bdPgNL1veMzEgFis58WvGyW6t38UTaRCbtIo82kUm7yKNN5LFqk/ikTP7edYCJHz1DcG42P7bvzqcXXFtqn8Gdoy2xdEpFrNoFjmPiaZrmCuDAMXZ7EPgc2FcVg3IXb29vdw/hlESH+jPxslgCfLyY2XMAa07vSHh2Bs/PGY+3swAomnyOWbrdUmc+rd7FE2kTmbSLPNpEJu0ijzaRx4pNDq/XOfzbd2iTlMCeetE8d/1jJfd1AlzVJpz+7a17uaoVuxzmc6pvYBhGI+BaoCdw7tH23bdvH4MGDcLHxwen00n//v0ZOnQoycnJBAUF4e3tjd1uJzIykgMHDmCaJpGRkaSkpBAcHAxAVlYWDRs2JDU1FcMwqFevHqmpqYSGhuJ0OsnOziYqKork5GR8fX0JCwtj//79hIWF4XA4yMnJKdnu5+dHSEgIaWlphIeHY7PZsNvtJdsDAgIIDAwkPT2diIgIMjMzcTgcJdsDAwPx8/MjIyOD+vXrk5GRQX5+fsl2d32mpy+KYkn8TqbeNZLWLw7hnP82MXzpe6y7YxAH8w3SHF4sWreZmzqfbonPlJWVha+vb0mnnJwccnNzLd/J9d89q32m3Nxc8vPzPeozeUKn3bt3k5OT41Gfyeqddu/eDeBRn8kTOhUUFJCVleVRn8nqnRwOB3a73aM+k9U77d+/33Kf6cvNB7j3nyVcv/pr8n18mf/gExghdbioroOUPC8iA325sonBoUOHLPOZynbav38/derUEf3vXqXzRtM0jzZXLNrJMJoD35qmeWYF2+YDL5umucYwjJnF+31W0fusXr3abNOmTUWbRDh06BB16tRx9zCqRHxSJmO/20GrhD+ZNvURfAqdjLrtKZaddREAvl4GC+8+272DPE6e1MVTaBOZtIs82kQm7SKPNpHHik1eeXcpjz91F3UcuUy65iE+O//qkm1+XvDe9W0te4ntYVbosmHDhvW9e/fuVPb1qniqbSfgE8MwdgLXA28bhnFNFbxvjTtw4FhXFFtHXEwI0/q34c/mZ/LaFfcB8NS8yTTbV/Tb9/xCk6tmxrMsIc2dwzwuntTFU2gTmbSLPNpEJu0ijzaRx3JN7HYefP1x6jhyWdyhN591vapkk7cB4/vGWn7SCRbs4uKUJ56maZ5mmmZz0zSbA58BQ0zT/PJU39cdjufsr5VEh/oT5u/FJxf057u4HgQ5cnjxo3EE5uUARfd7TlyeyIJNsp+O5WldPIE2kUm7yKNNZNIu8mgTeazQxGbPY+JPOxnx7Tb23HQ3DZJ2sjOmBROuG15yX2egjxcTLbheZ2Ws0KUyx7OcylxgNdDaMIw9hmEMMgxjsGEYg6t/eDUrMjLS3UOock/2boFhGDx//WPsaNCUFim7GPPZy+DyL+3UtTbRDxvyxC5Wp01k0i7yaBOZtIs82kQe6U1s9jxGL05g2fZ02n/6Ae3XLiM7IIiU997n/DNiiIsOplfLcKb2b+Mxk06Q3+VojueptgNM04w2TdPXNM3GpmnOME1zqmmaUyvY987K7u+0gpSUFHcPocrFxYTwYr9YQiPCGHnHM2T7BXJp/E/cuOrLUvtNXbPHPQM8Dp7Yxeq0iUzaRR5tIpN2kUebyCO9yaRlO7BlOjhnezxDF78HwLgbR7IgN5TRPZsz+fLTGd2zuUdcXutKepejqYp7PD3G4Sc/eZq4mBA+urkdiQ2a8tyNIwAY/s07nLXz75J91iTaxa7v6aldrEybyKRd5NEmMmkXebSJPJKbLNiUzOb9uURmpDLh42fxKSxkZs8B/HxmN9IO5bt7eNVKcpdj0YlnLXJG/QB+POsiPu5+HT6FTl6Y/QzhWekAmMCIRQni7/dUSimllFK117KENKauteFTkM8Ls58jIusg62I7MLXPXQBE1PF18whVZXTi6SIrK8vdQ6hWo3q1IMgb3uh3Lxubn0kDexrjP34eb6ezZJ+pa23innTr6V2sSJvIpF3k0SYyaRd5tIk8EpvEJ2UycXkiAA8vnEbcrr9JCavPmFvG4PT2BmBgx2h3DrHaSexyvHTi6aJhw4buHkK1ig715+3r2lI3OIDHb3uKtOBwzt3+B/cvfb/UflNWyrrf09O7WJE2kUm7yKNNZNIu8mgTeaQ1sdnzGLNkOwB9Ni7j5l+/IN/bh9G3jSM9OBwAf2/D4+7pLEtalxOhE08Xqamp7h5CtYsO9eeVK1txqF4kj9/2FAVeXgxc/gkX/bWyZJ/cgkIm/rRTzJNua0MXq9EmMmkXebSJTNpFHm0ij6QmNnseD3+9DUehSYvknTz52csAvHrl/fzVrG3JfsO7N3HXEGuMpC4nSieeLozi9X48XXSoP8/2acHGFmfx1mX/A+DpTyfRbF9iyT7Ltqdz7+dbRDxwqLZ0sRJtIpN2kUebyKRd5NEm8khpYrPn8ei32ziYW0BQbjYvfjiOwPxcFp1zMfO7Xl2y3+DO0fSKjXDjSGuGlC4nQyeeLurVq+fuIdSYuJgQBneOZvaFN/D9WRcRnHeIl2Y9RVDOkevG85wmY5YkuP3MZ23qYhXaRCbtIo82kUm7yKNN5JHQxGbPY8TCf9l/qACjsJCnP51Is/17+DfqNCb0Hw7Fk7DRPZrSv32Um0dbMyR0OVk68XRh5VPXJ6N/+yhG92zG5AGj+DfqNJqn7ubZTyZiFBaW7OMohCm/7HLjKGtfFyvQJjJpF3m0iUzaRR5tIo+EJlNWJrIvu2h5lEE/zqbH36uwBwYz8o5nyPMLAKCuv1etONN5mIQuJ0snni5CQ0PdPYQa1ys2gk/v68Jz/xtPRmAIF25ZzT0/fFRqn422bLdeclsbu0inTWTSLvJoE5m0izzaRB53N7HZ8/gjqehKvAv//pX7vp9FoWHw5C1j2FO/Ucl+Y3q3cNcQ3cLdXU6FTjxdOF2WFalt6rY9nSdvGYPT8OLeHz7kwr9/LbV95KIEt00+a3MXqbSJTNpFHm0ik3aRR5vI4+4ms9bbMIHmKbt45pOJALzddxCrW58HgEHRJbZxMSHuG6QbuLvLqdCJp4vs7Gx3D8FtBndtzPo25/J237sBeHbuCzRPOXKJrQluu9+zNneRSpvIpF3k0SYyaRd5tIk87m6SdiifoJwsXpr1FMF5h/j+rIuY1eNmACICfZh5Y9tadYntYe7ucip04ukiKqp23JRckehQfyZcFsvsnjfz/Vk9CHLklHvYkKMQHvlmW41PPmtzF6m0iUzaRR5tIpN2kUebyOOuJjZ7HhN/2kli2iGemzuh5GFCz944AgwDPy945cpWHr9eZ2WsfKzoxNNFcnKyu4fgVnExIUzqdzrP3vgY/0adRrP9e3jukxdKPWwoLaeAYV9trdHJZ23vIpE2kUm7yKNNZNIu8mgTedzRxGbPY/TiBJZtT+emb2bQfetaMgJDGDHwWXL9Agn08WJ839haO+kEax8rOvF04evr6+4huF1cTAhnNI9kxMBnyQgMofuWNdz7/axS+6TnFTJlRc096Va7yKNNZNIu8mgTmbSLPNpEnppuYrPnMXJRArZMBz03/cKgH2fjNLx44tYnyW7UhF4tw5nav02tu6ezLCsfKzrxdBEWFubuIYgwrFtTcho3ZcytT+I0vLjnx9n0+GtlqX3+SM6usbOe2kUebSKTdpFHm8ikXeTRJvLUZJOZ6/YwcN5mUrIctEz+j6c/LXqY0Bv9/sdvrTrRLDyQ0T2b1+oznYdZ+VjRiaeL/fv3u3sIIkSH+vPqla3YfnYX3rzsHgCe/mQip6XsLNnHpOhJtzUx+dQu8mgTmbSLPNpEJu0ijzaRp6aavLlyJ3Pii9amDDmUyeRZT1HHkcuSs3vz8YU3ABBRx7pn+aqalY8VnXi6sPJvEKpadKg/L1/Rivm9b2JpXM+ihw3NfIrQQ/aSfVKyHIxeXP2TT+0ijzaRSbvIo01k0i7yaBN5aqLJsoQ0vt6aDoBXoZPn546nSVoS/8TE8vz1j4BhEB3ix8CO0dU+Fquw8rGiE08XDofD3UMQJTrUn+cvjWXCjY/yT0wsTdP28sLs5/B2FpTsY8t08FA1P2xIu8ijTWTSLvJoE5m0izzaRJ7qbmKz5zFpeWLJn4cuns75/6zjYJ1QRtzxDHl+ATQM9mPiZbX7YUJlWflY0Ymni5ycHHcPQZy4mBCeubo9YwY9T1pwXc5L2MAj37xdap+MvMJqfdKtdpFHm8ikXeTRJjJpF3m0iTzV3WTq6j2Yxf/7it+XcMfP8yjw8mb0bU9hq1e0ZMiL/XTSWZaVjxWdeLqw8ro41SkuJoRnB13EC/dNwOHty42rvqL/6m9K7ZOeV8igzzYTn5RZ5d9fu8ijTWTSLvJoE5m0izzaRJ7qbGKz5/H73qKfGc/a+RePfz4FgBeveZD1sR0AuCUuUiedFbDysaITTxdWXhenukWH+nP3g9cy4brhAIz46g06JmwstU9BYdEDh6p68qld5NEmMmkXebSJTNpFHm0iT3U1iU/KZPCCreQXmkSlpzB51jj8nPl8ev41fNHlSgCig7y589zG1fL9rc7Kx4pOPF34+fm5ewiiRYf6s/bCy/nwohvxKXQycfYzNEpLKrWPCUxavrNKv692kUebyKRd5NEmMmkXebSJPNXRxGbPY+x3O8gpKCQwL4eXZz5JveyDrDm9I69eOQQAH2Di5a2r/Ht7CisfKzrxdBESUrsXpD0eo3o0551+9/BLm87UPZTJKx+MISg3u9Q+6TkFlXz1ydEu8mgTmbSLPNpEJu0ijzaRp6qb2Ox5jFyUQG5BIUZhIc988gKtbDvYVb8xT9w6Fqe3N/4+Bi/ofZ1HZeVjRSeeLtLS0tw9BPHiYkJ44fLWjL1lDDsaNqPFvkSenzMer0JnyT5ehlGl31O7yKNNZNIu8mgTmbSLPNpEnqpscvjy2pSsoieyDv7uA3r+/Sv2wGAeuet5MuuE0DDYj3f7n0FcjHUnVjXByseKTjxdhIeHu3sIlhAXE8KDfc/gkTuf52CdULptXcsDi6aXbA/282LEwn+Z+NPOKnnSrXaRR5vIpF3k0SYyaRd5tIk8VdXEZs9jzJLt5BQUAtB3ww/cvWwOBV5ePH7rWBIjmxDo46VPsD1OVj5WdOLpwsqPJ65pvWIjuLJfR0bdPo4CL29uXzGPK35fgheQnusk3pbFsu3pjF6ccMqTT+0ijzaRSbvIo01k0i7yaBN5qqKJzZ7Hw19txVFYtHBKu8QtPPnZSwC8cuVQfmvViQAfL57t00InncfJyseKTjxd5ObmunsIltK/fRQPP34LX/9vFABPLJjCmTv/KrWPLdPB1DV7Tun7aBd5tIlM2kUebSKTdpFHm8hzqk1s9jxGLPyXg3lFZzobHtzHS7Oewr8gn8+6XMn886+mYbAf0/q30ctrT4CVjxWdeLqw8ro47hId6k/PFx4j93//w7cgn8mzxpV70u2aRPspLbGiXeTRJjJpF3m0iUzaRR5tIs+pNpmyMpF92fkA+DtymTzrKepnHmBdy7N56eoHwDD08tqTYOVjRSeeLqy8Lo675Ywfz7a4LtTLPsirH4whOCerZJsJjF588ut7ahd5tIlM2kUebSKTdpFHm8hzKk1s9jz+SCr6WdCr0Mlzn7zAGXv/ZXdEDI/f9hRObx/q+nvppPMkWPlY0Ymni4CAAHcPwbp8fCj8aBY7oppz2r5EJn30DN7OI8uqOM2Tn3xqF3m0iUzaRR5tIpN2kUebyHOyTQ4vm2IW//nBRe/R86+V2AODGX7XeDKCwvAyYEzvFlU32FrEyseKTjxdBAYGunsIltawcSSznnidtOBwzkvYwOgFU8A0S7af7ORTu8ijTWTSLvJoE5m0izzaRJ6TabIsIY275m8uWTal/+pvuG3FfAq8vBl5+9PsatAUP2+DSZfF6n2dJ8nKx4pOPF2kp6e7ewiWd9NV5/L4oPHk+vpzzbrF3LH8k1LbnSaMOsHJp3aRR5vIpF3k0SYyaRd5tIk8J9okPimTicsTKX6ALV3/+Y0RX70OwPjrH2F9bAf8veC963StzlNh5WNFJ54uIiIi3D0Ey4sO9eeW/13OMzePBuDBxdPp9efPpfYpNGHM0u3HvcyKdpFHm8ikXeTRJjJpF3m0iTwn0sRmz2PsdztK/hxr286E2c/hU1jIjN638W2nvgT4ePF8X32Y0Kmy8rGiE08XmZkn/+RVdURcTAiXjrqbN/v9D4BnPplIu8QtpfZxOE0e/XbbcU0+tYs82kQm7SKPNpFJu8ijTeQ53ibxSZnct2AruQVFy6ZE2NN49YMnCc47xNK4nkztc6cum1KFrHys6MTThcPhcPcQPEZcTAhnThrDl+f1I6DAwcsznyT6QOmncO0/VMBjxzH51C7yaBOZtIs82kQm7SKPNpHneJrEJ2UyenFCyaQzwJHDKx88SdTBfcQ3a8ezN47Ey0uXTalKVj5WdOLpwsrr4kgU1yiUpfc/wdrTzyEi6yBTPniCIJdlVgBSDxVw57zNR73nU7vIo01k0i7yaBOZtIs82kSeYzU5POl0Ft/T6VXo5Lm5E2i7dxu7I2J4bOCzOHz9GHlRU510ViErHys68XRh5XVxpHqwZ0vG3jGOHQ2b0SJlFxNnP1tqmRUoWudzxKLKHzikXeTRJjJpF3m0iUzaRR5tIs/RmtjseTz53faSSSfAQwun0ePvVWQEhjD8rgnk1q3H5H6x9Iq17j2JEln5WNGJpwsrP55YquhQf5649mxG3j2eA0F16fLvekZ/8VqpZVYOm/jTzgrfQ7vIo01k0i7yaBOZtIs82kSeyprY7HkM+2YbeQVHfpa7ftVX3PrL5+R7+zDyjqdJiWmm93RWEysfKzrxdOHn5+fuIXikuJgQnr2vN88NnkCujx/X/LaIQT/OLrdfWk4ByxLSyr2uXeTRJjJpF3m0iUzaRR5tIk9FTWz2PB7+ehvpOUeuXrtgyxoe++pNAJ6//lHiY8/muT4t9PLaamLlY0Unni4yMjLcPQSPFR3qz/8evJYnbxmD0/Bi8HczueL3JeX2m7g8sdzkU7vIo01k0i7yaBOZtIs82kSeippMWZnIwdwjk862iVt5YfZzeJuFvHfx7Szq2IeJl8Xqmc5qZOVjRSeeLurXr+/uIXi06FB/zhh0Ay9fPRSAMZ+9Qpd/1pXbb+LyxFL3e2oXebSJTNpFHm0ik3aRR5vIU7aJzZ7HH0lHHhLZeP9epnzwBIH5uXzdqS/vXjKQDlFBOumsZlY+VnTi6cLKv0Gwiv7to2j5xEPM6nEzPoVOJn70DK33bCu334hFCSVnPrWLPNpEJu0ijzaRSbvIo03kcW2yLCGNu+Zt5vBdneFZ6bw+YzTh2Rmsan0uE64bjp+3wbALm7lnsLWIlY8VnXi6yM/Pd/cQaoVesRGc+/7LLOnQmyBHDlM+GFNujU+AScVnPrWLPNpEJu0ijzaRSbvIo03kOdxkwaZkJi5PpLD49QBHDq9+MIYmaUlsaXQ6o28bh9Pbh/F9da3OmmDlY0Unni6svC6O1UTXDeTQm6+zLrYD9TMP8PqM0YRll/4NjknRk261izzaRCbtIo82kUm7yKNN5ImKiiI+KZOpa20lr3k7nUyY/Rztdv/D3vAoht89gRz/QEb3aKqX2NYQKx8rOvF0YeV1cayo5xnRzHjkRf6NOo3mqbt5eeZY/PPzSu2TllPAG9/FV7rGp3IPPVZk0i7yaBOZtIs82kSe5ORkXlqReOQF02TUF6/RfetaDtYJ5eFBE0kLqUeHmGBdq7MGWflYOebE0zCM9w3D2GcYxl+VbL/aMIw/DcP4wzCM3w3D6Fb1w6wZQUFB7h5CrfNQvzN5esiLpIRFErfrb56d+wJehc5S+yQeMhi5KEEnn4LosSKTdpFHm8ikXeTRJrLY7Hl89182qVmOktfu+eEjrv1tIbk+fjxy1/PsatCEyDo+DOvW1I0jrX2sfKwczxnPmUDfo2z/EYgzTfNs4G5g+qkPyz28vb3dPYRaJzrUnzF3dOO1Ea+SGRBEr79+Yfg374B5ZFHi/MIjl90qGfRYkUm7yKNNZNIu8mgTOWz2PB77dhsbbNkl93VeuW4x930/C6fhxZhbn2RTs3a0jQzkpSta6X2dNczKx8oxJ56maa4ADhxle5ZplswSggCzsn2ls9vt7h5CrRQd6s+IIf2Y9shkHN6+3PzrF9z+86cl2xsHFv21dyCnAJs9r7K3UTVIjxWZtIs82kQm7SKPNpFj0k//kXqooOTnr/O3ruWJz18BYPI1D7Ki3QUM7hzNlKvb6KTTDax8rFTJPZ6GYVxrGMZWYCFFZz0tKTIy0t1DqNX63XstE24ZDcBDi97jynVLAPjL7gMU/UZj5KIEnXwKoMeKTNpFHm0ik3aRR5vIEJ+UyebUHKDo56+2iVt5Yfaz+BQW8kHPAay85Hom94ulf3vrPuDG6qx8rPhUxZuYpvkF8IVhGBcCzwEXV7Tfvn37GDRoED4+PjidTvr378/QoUNJTk4mKCgIb29v7HY7kZGRHDhwANM0iYyMJCUlheDgYACysrJo2LAhqampGIZBvXr1SE1NJTQ0FKfTSXZ2NlFRUSQnJ+Pr60tYWBj79+8nLCwMh8NBTk5OyXY/Pz9CQkJIS0sjPDycvXv3UqdOnZLtAQEBBAYGkp6eTkREBJmZmTgcjpLtgYGB+Pn5kZGRQf369cnIyCA/P79ku4TPlJOTQ25uriU+kyN9P90eupm5jv0M+PQdxnz+Mo2j6pDa+Tx+TfOlVYiT7dlOZvz8N9e2rktss0biP5Mndtq/fz8Oh4OwsDCP+kye0Gn79u1ERER41Geyeqft27fTuHFjj/pMntCpsLAQHx8fj/pMVu9UUFCAl5eXR30mq3VKz8pl0toMLqrvICXPi84H/+OumY9Tx5HLb117s23gQN7r05KUlBTS0hyW+Eye2CktLY1WrVqJ/kyVMUzz2FfGGobRHPjWNM0zj2PfHcB5pmnuL7tt9erVZps2bY75/dxl9+7dNGnSxN3DqPVs9jw23jeKW5d+SK6PH+898jwf1u9Yap+GwX682E/Xi3IXPVZk0i7yaBOZtIs82sS94pMyGfvdDnILii6vbZiewkdTHyI8fT+/tOnMiIHPct5p9XimT0s3j1RZ4VjZsGHD+t69e3cq+/opX2prGEasYRhG8f8+B/AH0k71fd3ByqeuPUl0qD8dpk5kcferCShwcPfrT3N60vZS+6RkORi9WC+7dRc9VmTSLvJoE5m0izzaxH3ikzIZvTihZNIZlp3Bm9NHEZ6+n43Nz+Tx257C9PZhcJfGbh6pAmsfK8eznMpcYDXQ2jCMPYZhDDIMY7BhGIOLd7kO+MswjD+At4CbzOM5jSpQSkqKu4egikWHBdBs5lv82uEignIP8cb0UTRKSyq1jy3TwV3zN7MswZK/57A0PVZk0i7yaBOZtIs82sQ94pMyGbU4AWfxT+51cg/x2vuP0zx1N0mNm/PIXeNx+AUwskdTvcpMCCsfK8e8x9M0zQHH2D4JmFRlI3Kjw9dBKxmiw+tgfDKLf665ltZbNvLG9FH8b8hrpIXUK9mn0ISJyxNJTM/hznP1N3E1RY8VmbSLPNpEJu0ijzapecsS0pi4PLHkz74FDl78aBztdv/D3nrRTBg6gQL/UF7s04K4mBA3jlS5svKxUiVPtVWqukRFhpL1/nv827Q1TdKSeH36aIJyssrtNyc+lQWbkt0wQqWUUkopayk76fQqdPLMJxPp/O8G0oLDeeCeSWSHRzKtfxuddKoqoxNPF1lZ5Sc0yv28A7wpXPAZexo0oZVtO6/MHItfvqPcflPX2vSy2xqix4pM2kUebSKTdpFHm9Sc+KTMUpNOTJORX7zOJX/+TFZAEA/dMxFbZCMGtA3Vy2sFsvKxohNPFw0bNnT3EFQFGjZsSIMWjcj5/HP2hUZwzn9/Mn7Oc3g7neX2ffHnRH3gUA3QY0Um7SKPNpFJu8ijTWrGsoQ0Ri5KKPXa4O9mct3ab8n18eORO58noVEsEy+L5axY2U9Ora2sfKzoxNNFamqqu4egKnC4S/12scRP+5CMwBB6/L2KJz97CaOwsNS+hSbMWm9zxzBrFT1WZNIu8mgTmbSLPNqk+o1bso2JyxNxfQLoTSsXMOjH2RR4eTHm1rFsbHEWky6LJS4mRJsIZeUuOvF0UbwqjBLGtUunS7uyeMLb5PgGcMX673j067egzEOU0w7l1/QQax09VmTSLvJoE5m0izzapHpNWrad1XuyS712+e9LeezrtwB4/vrHWNHufEb3aFpyT6c2kcnKXXTi6aJevXrH3knVuLJdLh14OcsnvonD25ebVn3JkCUzSm2PqONbk8OrlfRYkUm7yKNNZNIu8miT6mOz5/HjDnup13r/+TNj578EwKtXDGZhp0sZ3aMpvWIjSvbRJjJZuYtOPF1Y+dS1J6uoS5e7+vP7xNcp8PLirp/mcueyOQBEh/gxsGN0TQ+x1tFjRSbtIo82kUm7yKNNqofNnsej324r9doFW9bw/JzxeJuFTLtkIHMvvIHJ/WJLTTpBm0hl5S7HXMezNgkNDXX3EFQFKutyxqCbSPYziRk2lKFLZtC4UQSnjx3OviwHD3+1lYy8QkwgPMCbJ3qdpo8Dr0J6rMikXeTRJjJpF3m0SdWLT8rkyaXbyXMeuSWpU8JGJn30ND6FTj668EamX3x7qctrXWkTmazcRc94unBW8JRU5X5H6xJ0+83kvPIKAFfPeBGv2R8zYlECB4snnQDpuU5GLErQpVaqkB4rMmkXebSJTNpFHm1StZYlpDFiUUKpSWf7XX/z8swn8S/I57MuV/L65ffStUlwuTOdh2kTmazcRSeeLrKzs4+9k6pxx+riGDiQQ88/D8DpY0fS+8+fK9xvsi61UmX0WJFJu8ijTWTSLvJok6qzYFNy6XU6gVZ7/+W1GY9Tx5HLwnMu4cVrHqJ3yzCe6duq0vfRJjJZuYtOPF1ERUW5ewiqAsfTJW/IEHJGjcLbLOT5OeM5f8vacvs4TZi6Zk91DLHW0WNFJu0ijzaRSbvIo02qRnxSJlPXll5WrnnKLt6cPoqQ3Gx+bN+d524YQZfmdRnVq+VR30ubyGTlLjrxdJGcnOzuIagKHG+X3JEjWdD7ZnwKnUz66GnO2f5HuX1WJ9oZ9e02PfN5ivRYkUm7yKNNZNIu8miTU2ez5zFmSUKp1xqlJfHWeyMIz87g19bn8eSAMdQL8Wdwl8bHfD9tIpOVu+jE04Wvry7DIdFxdzEM6k2ZyOedryCgwMErHzzJmbs2l9ttY3I2w77aqpPPU6DHikzaRR5tIpN2kUebnBqbPY/Hvt2Go/DIaw0OpvLWuyNoYE9jfYs4Rt3xNO0b1+WlK1oRHep/zPfUJjJZuYtOPF2EhYW5ewiqAifSJa5RKBHvvcH3HXsT5MjhjRmjabt7a7n90vMKeVgnnydNjxWZtIs82kQm7SKPNjk1U9fsIfVQQcmfIzIP8NZ7I2iUnsympmfwyJ3PM/yS05l0nJNO0CZSWbmLTjxd7N+/391DUBU40S5xjcPotHguGztfTHBuNm9MH0XrPdvK7Xcwr5Dh3+hltydDjxWZtIs82kQm7SKPNjk5NnseoxZuY3WiveS1epkHeHvaozRP3c226BYMu3sCd1zUstKn11ZGm8hk5S468XRh5d8geLKT6uLjg//s9/m5fXdCc7J4672RtEpKKLfbgZwChunk84TpsSKTdpFHm8ikXeTRJicuPimTez/bzEbbkaec1s06yNvvjqDFvkQSok5j6P8m0zI2hv7tT/yBNNpEJit30YmnC4fD4e4hqAqcbJfoiGD48H1WtO1KWE4mb707gpa2HeX2S88pYMTCf3XyeQL0WJFJu8ijTWTSLvJokxNjs+cx9rsd5Lnc0xmWncHb7z5Gy5Sd7GjYjCH3TsavYSTDujU9qe+hTWSychedeLrIyclx9xBUBU6ly1nNIjBnz+LXNp2pe8he9FvA5J3l9tuXnc+s9bbyb6AqpMeKTNpFHm0ik3aRR5scv/ikTO79fAu5BUdmnUWTzhGcnvwf/zVoyv33voR3g0gmX376cd/TWZY2kcnKXXTi6cLK6+J4slPtclbz+jg//pDVrc6lXvZB3n73MZrtSyy3X9qh/FP6PrWJHisyaRd5tIlM2kUebXJ8liWkMWJRAnlOs+S1kEOZvPneSFrZtrOrfmPuv/clDoXX5+UTeJBQRbSJTFbuohNPF1ZeF8eTVUWX9qdFkj9nNmtPP4eIrHTemfYYTVN3l9pnV3oOE3/aqZfcHgc9VmTSLvJoE5m0izza5NjikzKZuLz0L86Dc7J4872RtElKIDGiEfff9xJZ4fV5rk+LU5p0gjaRyspddOLpws/Pz91DUBWoqi7tW0SSN2cOG1rGEZmZxjvTHqXx/r0l2w/mOlm2PZ27529mWUJalXxPT6XHikzaRR5tIpN2kUebHN3hezpdBeVk8cb0UbTdu4099aK5/76XaXzGabx73RnExYSc8vfUJjJZuYtOPF2EhJz6QaqqXlV2ad+yAYHfLCChTQca2NOYOu1RGqUlldrHacLE5YkM+2qLnv2shB4rMmkXebSJTNpFHm1ydLPW20rd0xmUm83rM0Zz5u6t7K0XzeD7XqZl+5ZM6nfy93SWpU1ksnIXnXi6SEvTs1wSVXWXqKhwIpZ+SUHnzjTMSGXa1OE0Sd1Tbr/NqbkMnKdnPyuix4pM2kUebSKTdpFHmxyd63MognKzeW3G45yVuIWk8IYMvu9l0sIbMrhr46r9ntpEJCt30Ymni/DwcHcPQVWgWrqEhJA5bx7/nXE2DTP2M23qIxU+cAiKzn4u2GTd6+mrgx4rMmkXebSJTNpFHm1Sns2ex8SfdjJi4b8kZxYtoRFcfHlt3K6/Sa7bgPvve5nU8Ia80C+2ys50HqZNZLJyF514urDy44k9WbV1CQkh97PPWF98z+e0qY9UuNQKwNS1Np18utBjRSbtIo82kUm7yKNNSluWkFb0zInt6cTbskjJchCeY+etd0fQPnELe8OjGDz4FRq2i+X9G9tWyT2dZWkTmazcRSeeLnJzc909BFWB6uwSFV2P3R98xG+xh592+yixtu0V7jt1rY34pMxqG4uV6LEik3aRR5vIpF3k0SZHLEtIY+LyRFxWTCEsO4M3pz5G273b2NegMR++MIMJg3sz6fJTWzLlaLSJTFbuohNPF1ZeF8eTVXeXi85szN73Z5Ws8/nOtMdolZRQ4b4vraj4ctzaRo8VmbSLPNpEJu0ijzYpEp+UyaQyS6bUyzzA1GmP0sq2neSopvgtW8LgG7tW24TzMG0ik5W76MTThZXXxfFkNdGlx5mNqPv1PP6IO5+6h+y8Pe0x2uzZVm6/1CyHrvWJHitSaRd5tIlM2kUebXJkyRSXE53Uz9jP1GmPEpv8HzsaNuO956djxsTUyHi0iUxW7qITTxcBAQHuHoKqQE11iWoQRrMln7OrWy/CcjJ5+93HaJe4pdQ+haBrfaLHilTaRR5tIpN2kUeblF8ypeHBfUyb+gin7Uvk36jTGDr4Za67uH2NjUebyGTlLjrxdBEYGOjuIagK1GgXf39CP5vDfxdeQkhuNm++N5Kzdv5dbrfDa33OXFd+GZbaQI8VmbSLPNpEJu0ijzYpvWRK9IFkpk19hKZpe9kaE8v9973MPZfHVfvlta60iUxW7qITTxfp6enuHoKqQI138fOj7vzZ7Lv0CoLzDvH69FGcsz2+wl3nxKfWyqfd6rEik3aRR5vIpF3kqa1N4pMyuf2Tv7l2Vjz/pB4CoFFaEtOmDqfRARt/N2nN0Htf4v7L29MrNqJGx1Zbm0hn5S468XQREVGzB7Q6Pm7p4uuL70fvk3fjjQQ5cnhtxmjO37K2wl1r49Nu9ViRSbvIo01k0i7y1MYm8UmZjF6cQEqWg+z8QnILCmmesotpU4cTfXAffzZry6ODX2Ls9R1qfNIJtbOJFVi5i048XWRm1q7Jg1W4rYuPD4feeov0AbcRUODg5VljuTh+eYW7jliUUKvOfOqxIpN2kUebyKRd5KltTeKTMhm1OKHUkimt92zj3anDaZixn22tz2b++Gm8cmunalmj83jUtiZWYeUuPu4egCQOh8PdQ1AVcGsXb2948zW2efvTavYMnp8znkBHDt+ce1m5XaeutVE30Nctv5WsaXqsyKRd5NEmMmkXeWpLE5s9j0nLd7J536FSr8f9t4kpH4whODebtW3Oo9UPC3ikTh03jbJIbWliNVbuomc8XVh5XRxP5vYuhkHkay8y56pBeJuFPDX/JQb88nmFu05cnlgrLrt1exNVIe0ijzaRSbvIUxua2Ox5jFj4b7lJZ5d/1vHm9FEE52bz/VkX8cKQieDmSSfUjiZWZOUuOvF0YeV1cTyZiC6GwdmvP8/r1wwF4JFv3mbQDx+BaZbbdfTiBI+ffIpoosrRLvJoE5m0izye3iQ+KZP7FmxlX3Z+qdd7blrBKzOfJCA/jy/PvYxxt45hWK+WbhplaZ7exKqs3EUnni6s/HhiTyalS3SoPxdPeZLJA0biNLwY/N1MHl44rdzk02nC2O92YLPnuWmk1U9KE1WadpFHm8ikXeTx5CaHHyLkukYnwOW/L+WF2c/h6yzg4+7X8cINjzLh8tZuu6ezLE9uYmVW7qITTxd+fn7uHoKqgKQu0aH+XP7CcJ69/UnyvX24bcV8nvj8VbwKnaX2yy0o5M55mxnw8SaPPPspqYk6QrvIo01k0i7yeHKTSct3lnqIEMCNv37B0/NexNssZNolA5lyxf1M6ne6mEkneHYTK7NyF514usjIyHD3EFQFpHWJDvXnpmfv5/UHJ5Lr48e1vy3k2bkv4O0sKLWfCaTlFDBykeddeiutiSqiXeTRJjJpF3k8sUl8Uia3zNnE/kMuPx+YJnf/OJsRX70JwCtXDmHGJXcwumczUZNO8MwmnsDKXXTi6aJ+/fruHoKqgMQu0aH+3P3U3WyZPpts/zpcGv9T0T0ajpxy+5rAmCUJHnXprcQmSrtIpE1k0i7yeFqTZQlpjFiUUGrSaRQW8vDCady/9AOchhfP3fAYO2+5i5k3thX5RHxPa+IprNxFJ54urPwbBE8muUvzqy5my4fzSA+uy/n/rOOdaY8Rll1+vI7CoocOecrkU3KT2ky7yKNNZNIu8nhSk/ikTCYtTyz1mrezgHHzXuS2FfMp8PJmzK1P0n3cgzxzaUuiQ/3dNNKj86QmnsTKXXTi6SI/P//YO6kaJ71Ly0vOJ+3rb0mJiObM3VuZ/vbDRKWnlNvPlungkW+2ecTkU3qT2kq7yKNNZNIu8nhKk2UJaYxclIDrLZ0BjhxenjmWyzd8zyG/AIbdPYEDffqJu7S2LE9p4mms3EUnni6svC6OJ7NCl4iz25KxaBHbY1rSPHU3099+iJbJ/5XbLy2ngHvmbbb8PZ9WaFIbaRd5tIlM2kUeT2iyLCGNicsTS006w7IzeGfaY1zwz2+kB4Vx/30vk9jxfIZd2Mxt4zxentDEE1m5i048XVh5XRxPZpUukac3I3/xQv6MjaNhxn7efWcYcf9tKrdfPjDC4g8cskqT2ka7yKNNZNIu8li9SUWX10alpzD97Yc5c/dW9oZHcc+Q1/A/tyMTL4sVe3mtK6s38VRW7qITTxdBQUHuHoKqgJW6NGzSAN9vvyD+3B6E5mTx5nsj6b55VYX7Prl0u2Uvu7VSk9pEu8ijTWTSLvJYucmCTcmMKHN5bUvbDma89SDNU3ezLboF9wx9nVtuuIBJV7SyxKQTrN3Ek1m5yzEnnoZhvG8Yxj7DMP6qZPuthmH8aRjGJsMwVhmGEVf1w6wZ3t7e7h6CqoDVukQ1qEvThfP4tdc1BBQ4ePHDcVy5bnG5/fKcJnfN38yyhDQ3jPLUWK1JbaFd5NEmMmkXeazaZFlCGlPX2kq91mHHn7z3zjAa2NNY3yKO+wa/yqibO4t8cu3RWLWJp7Nyl+M54zkT6HuU7f8BF5mm2R54Dni3CsblFna73d1DUBWwZBcfH8JnvM3Mi2/Dp7CQp+a/xJ3L5oBZegXpQhMmLk9k1Ddb3DTQk2PJJrWAdpFHm8ikXeSxWhObPY+JP+0sd3ltj79W8sb0kYTkZvNj++48NGgiD1zWVvyDhCpitSa1hZW7HHPiaZrmCuDAUbavMk0zvfiPa4DGVTS2GhcZGenuIagKWLVLdFgAzaeM56WrH6DQMBi6ZAajvnwdb6ez3L4bU3J5c+XOmh/kSbJqE0+nXeTRJjJpF3ms1MRmz2P04gSWbU8vdXnttWu+YeJHz+BfkM9nXa9izK1jeeSSWMud6TzMSk1qEyt3qep7PAcB5a8ptIgDByqdXys3snKXuJgQOo0fwbhbx+Lw9uX61V8z+cOnCHDklNv3663pjFpojeVWrNzEk2kXebSJTNpFHqs0sdnzePTbbdgyHSWvGYWFDF08nScWTMHbLGRqnzuZdM1DTLyitWUnnWCdJrWNlbv4VNUbGYbRk6KJZ7fK9tm3bx+DBg3Cx8cHp9NJ//79GTp0KMnJyQQFBeHt7Y3dbicyMpIDBw5gmiaRkZGkpKQQHBwMQFZWFg0bNiQ1NRXDMKhXrx6pqamEhobidDrJzs4mKiqK5ORkfH19CQsLY//+/YSFheFwOMjJySnZ7ufnR0hICGlpaYSHh5OZmcmuXbtKtgcEBBAYGEh6ejoRERFkZmbicDhKtgcGBuLn50dGRgb169cnIyOD/Pz8ku0SPlNOTg65ubmW/kzZ2dlkZWVZ9jM1CwvjikdvZFazMG589Wm6b1nD7PeGM/ae52jQKIzt2d5E+BVS19dkfWoh7y/7gwta1Oec5pFiP5PD4WDfvn0e/++e1T5Teno6Xl5eHvWZrN4pPT2doKAgj/pMntCpsLCQ3bt3e9RnsnqngoICdu3aJfoz7d13gKnxdk7zd9Cujsn6gz50DjrE1e+9Stffl+P08uL1G4ex4+I+PNIkgNb1fMV/pqN1OnToELt27fL4f/es9pnS09OJjIwU/ZkqnS+aZe45q3Anw2gOfGua5pmVbD8L+AK4zDTNbZW9z+rVq802bdoc8/u5S25u7lH/YSn38JQuNnsek6YtZcK0x2l0wEZSeEMevvsFdjaseC2vwZ2j6d9e5lpNntLE02gXebSJTNpFHulN4pMyGfvdDnILCkteC8vOYPKsp+iw8y+y/Osw+vZxxLc9j2HdGlv6TOdh0pvUVlbosmHDhvW9e/fuVPb1U77U1jCMpsAC4PajTTqtICUlxd1DUBXwlC7Rof6Muu9SHnroDf5q0oaY9BRmvP0Q52yPr3D/qWttYp946ylNPI12kUebyKRd5JHaxGbPY9hXWxmxKKHUpLNRWhIz3nqIDjv/IiWsPkMeeI0hT97B13fGecSkE+Q2qe2s3OV4llOZC6wGWhuGsccwjEGGYQw2DGNw8S5PARHA24Zh/GEYxu/VON5qdfh0tJLFk7pEh/oz/aFefPPKDJa3O5/QnCzemD6KSzf+WOH+E5cnEp+UWcOjPDZPauJJtIs82kQm7SKPxCbxSZn87/MtbE4t/VyGdolbeP/NB2m2fw/bolty1wNvUrfT2ZZZn/N4SWyirN3leJ5qO8A0zWjTNH1N02xsmuYM0zSnmqY5tXj7PaZphpumeXbx/5U7raqUKm1w7zNgzmw+7XYtfs58np87ocLlVgDGfrfDEg8cUkoppTxFfFImoxYn4HCW/u9yj79WMnXqI9TLPsjqVudy7/2vYsTEMLirZRd1UKrGVPVTbS0tKyvL3UNQFfDULnFN6tJw6hTeunZoyXIrT3z+Kt7OglL75RYUMnJRgqjJp6c2sTrtIo82kUm7yCOpybKENEYuSqCwzO+CB/zyOZM+epqAAgdfntePR+56nrNaRTP58tM97mwnyGqijrByl+N6uFBV0YcLqZNRG7psePNDLnh2JAEFDn6LPYfRtz1FZp3yi03Hhvsz9pKWbv8PXG1oYkXaRR5tIpN2kUdKkwWbkpm61lbqNW+nk+HfvM1Nq74E4K2+g5jZcwCTLz+duJjy/632FFKaqNKs0KXaHi7kSVJTU909BFWB2tDlnAfuYPW02aQFh3NewgY+ePMBmqTuKbdfQnoeQ77Y7Pazn7WhiRVpF3m0iUzaRR4JTSqadAbnZPHqB09w06ovcXj78uSAJ5jZ6xZG92zm0ZNOkNFElWflLjrxdGEYhruHoCpQW7qcfe3FbPtiIQnRLWi2fw8fvPkAnRI2ltsvOx+mrEx0wwiPqC1NrEa7yKNNZNIu8rizic2ex6hvt5WbdDbev5cP3nyArtt+50BQXe6/7yWWdezN5H6xHvPk2qPR40QmK3fRiaeLevXquXsIqgK1qUubjm3I/+E71pzVjbCcTN6YPopr13xbbr8/krLcetazNjWxEu0ijzaRSbvI464m8UmZ3D1vMxuTs0u93jFhIzPfGErz1N38G3Uadz74Ft5dOzP9+rYef6bzMD1OZLJyF514urDyqWtPVtu6REXXI/SLT/joohvxKXTyxIJXeeTrt/B2Okv2MYG752/m9k/+dstyK7WtiVVoF3m0iUzaRR53NLHZ8xi9KAFnmdevXfMtb04fRVhOJr+c0YV7hrzOwP7nMenyVm5/xkJN0uNEJit30Ymni9DQUHcPQVWgNnaJDq+D49mnefaGEeR7+zBg5QJenvkkQTlHnmTmNCEly8HoxQk1PvmsjU2sQLvIo01k0i7y1HSTZQlp3DV/c6lJp7fTySNfv8UTC17Fp9DJhxfdyGMDn+WOi1rWiktry9LjRCYrd9GJpwuns+zvvJQEtbVL//ZRNHrgLobcO5mDdUK54J/feP/th2iUllRqP6dZ82t91tYm0mkXebSJTNpFnppssmBTMhOXJ5ZaLiUoJ4tXPxjDgJULyPf24ZkbR/DG5fdxb9fG9G8fVWNjk0SPE5ms3EUnni6ys7OPvZOqcbW5S//2Udw6tD/3DX+HHQ2b0SJlFzPfGFruoUO5BYXcOW8zoxb9WyMT0NrcRDLtIo82kUm7yFMTTWz2PIZ9/U+5hwg1Skvi/bcepOu2daQHhXH/vS/xbae+TO4XW2snnaDHiVRW7qITTxdRUbX3LxfJanuXuJgQxv2vF7Mmf8jq1udR95CdN6aP5OaVn4PLOrwmsDEpixELq3/yWdubSKVd5NEmMmkXeaq7ic2ex/BvtrF536FSr3f5Zx2zXh9Ci32JbG/YnDsffIv409ozukfTWvMQocrocSKTlbvoxNNFcnKyu4egKqBdIDrUn2FXtCdv3ifM6jkAn8JCHv36bcbNm4R/fulJ5r7sfEYuSqjWyac2kUm7yKNNZNIu8lR3k1nrbRzIKTjygmlyx09zmfL+E4TlZLLijK4MGvo6aZExjO7RtFbe01mWHicyWbmLTjxd+Pr6unsIqgLa5Yi4JnVp9sZEnrrjKXJ8A7hi/fe8+85wGh7cV2q/lCwH9y3YWm0PHdImMmkXebSJTNpFnupuknYov+R/B+blMOHj53hw8XS8zULevfgOHhv4LA/2PYNv7jpbJ53F9DiRycpddOLpIiwszN1DUBXQLqXFxYRw87P3M+ShN9gbHkXbPf8w6/X76bDjz1L75RYUMnJRAuOWbq/ys5/aRCbtIo82kUm7yFPdTSLqFP2w3igtiRlvPcglf/5Mln8dHh34LO/1Gch9XRvphLMMPU5ksnIXnXi62L9/v7uHoCqgXcqLDvVn4F19uGfY2/wWew4RWQd5+93HuGHVl+Xu+1y9217lZz+1iUzaRR5tIpN2kaeqm9jseUz8aScjFv7LxJ92clnrCC7ZuYFZrw/h9OT/2BnZhDsffIsV7S5gdI+mtfohQpXR40QmK3fxcfcAJLHybxA8mXapWFxMCC8P7Mq7Ld8i4b1XuOXn+Yz88g1a7/2XSdc+TL6PX8m+uQWFjF6cwMTLYqvkYQnaRCbtIo82kUm7yFOVTeKTMnnqux3kFBQWvWCanPHRu4z/+l2MwkJ+aduVp28ejW94XSb3bF7rHyJUGT1OZLJyF514unA4HO4egqqAdqlcdKg/4/q1xtbtdZ4Z04pRn07m6nVLaJm8k9G3PUVKeMOSfZ0mjFiUQNvIQEb1PI3oUP+T/r7aRCbtIo82kUm7yFNVTWz2PMZ+t4Pc4klngCOHp+a9xCV/LgcgZ+RIzhw5ks+89KK/Y9HjRCYrd9GjzkVOTo67h6AqoF2OLTrUn4sev5cHHnoDW90GnLl7K7NfG0znbb+X23dzag7Dv/7nlO771CYyaRd5tIlM2kWeU2nielntyEUJJZPOZvsSmfnGA1zy53Ky/Ovw5rAXyR09GnTSeVz0OJHJyl30yHNh5XVxPJl2OT5xMSE88tBVvP7ix6xudS51D9l5fcZo7vn+Q4zCwlL7Hsh1Mmu9rZJ3OjZtIpN2kUebyKRd5DnZJjZ7HqMXJ7BsezrxtixSsorOBl3yx0/Men0ILVN28l+Dptz54Fvs7n5xVQ7Z4+lxIpOVu+jE04WV18XxZNrl+EWH+jPy+o7kzf+U9y4ZCMB938/itfefICw7o9S+axMzmPjTzpM686lNZNIu8mgTmbSLPCfTxGbP45FvtmHLPHLpoU9BPo9+9SYT5jxPkCOHpXE9Gfjg2+yLac7AjtFVOWSPp8eJTFbuohNPF35+fsfeSdU47XLi4prU5fRXn2XY3S9wsE4oXbetY/Zrg2mbuLVkn+z8QpZtT2f04oQTnnxqE5m0izzaRCbtIs+JNolPyuTezzaTllNQ8lrD9BTenTqcm3/9gnxvH1685kGevGUMZlAQz/ZpcUrPNqiN9DiRycpddOLpIiREn2omkXY5OXExIVx033Xc9vBUNjU9g6iD+5j+zsPlllyxZTq4e/5mbv/k7+NeckWbyKRd5NEmMmkXeU6kSXxSJqMWJ5DnchdJ139+Y/Zr99E+cQu2ug0Y9ehbbLvuNnrF1mNa/zb65NqToMeJTFbuohNPF2lpae4egqqAdjl5vWIjeOzWC3jikTeYf8E1+DoLGPnlGzw3dwKBeUduTneakJLlYNTihOOafGoTmbSLPNpEJu0iz/E2sdnzeOq7HRQW//7Uq9DJvd/NZMr7T1D3UCa/tj6Pu4ZP454hVzH58tMZ3bO5nuk8SXqcyGTlLjrxdBEeHu7uIagKaJdTExcTwod3nMPF37zPnOEvcMgvgL5/LGPmG0NoadtRat9CE8Ys3X7MS2+1iUzaRR5tIpN2kedYTRZsSqbvjI0MnLe5ZH3O8Kx0Xp8xmv/98BEA71x6F8PvGk+r1o11slkF9DiRycpddOLpwsqPJ/Zk2qXqnP3wnTw2+j12NGhKi32JzHxjKP1Xf1Pq0luH02Twgq1HPfOpTWTSLvJoE5m0izxHazJz3R6mrrWVnOUEOPffDcx59V46/7uBA0F1efCeSbzf+zYiQ/wZ3KVxDYzY8+lxIpOVu+jE00Vubq67h6AqoF2qTnSoPw/cfTHvvzqXr869jIACB49/MYUXZj9LcE5WyX45BYWMWpzAuErOfmoTmbSLPNpEJu0iT2VN4pMymROfWvJnb6eT+5fM4M3pI6mfeYANp53FbcOm8tvpHekQE8zky0/Xs51VRI8TmazcxTBdznRUt9WrV5tt2rSpse93ovLy8vD317+spNEu1WPUwm1ELvqGxxe8SnDeIZLCGzLmlif5q1nbcvuGB/rwRM/mJQ9n0CYyaRd5tIlM2kWeiprY7HkMXrC15NLaqPQUnp8znrhdf+M0vJh+8e283/tW/P18ebZPC32AUBXT40QmK3TZsGHD+t69e3cq+7qe8XRh5XVxPJl2qR7Dujfj966XcPvDU9ncqBUx6Sm8984wbl/+CUZhYal903MKSj14SJvIpF3k0SYyaRd5Kmoya72tZNLZc9MKPp5yL3G7/iYlrD5D7n2J6ZfcQY/T6zNVn1pbLfQ4kcnKXXzcPQBJAgIC3D0EVQHtUj2iQ/159arWTF0TxD11X2fo4ve49ZfPeWjRe5yb8Afjbh5FevCRG9gLTRixKIG2kQHcHxfqxpGryuixIo82kUm7yBMQEEB8UiYvrUgkK6+AYH8fgnwN/PPzGPbNO1y/5hsAVpzRlWdvHEFGUBiDO0fTv32Um0fuufQ4kcnKXXTi6SIwMNDdQ1AV0C7VJzrUn2f6tCQ+qQFjAx5gXew5PP3pJLpuW8ecV+9l3M2j+e30jqW+ZnNqLq+vyebJevX1Phph9FiRR5vIpF3k+WlnJu9vSir5c3a+g9NSdjLz4+eJTf4Ph7cvr19+L59ecC0YBrfEReqks5rpcSKTlbvopbYu0tPT3T0EVQHtUv3iYkKY1r8NhZf24dbh77K+RRz1Mw/w1nsjGfbNO/jlO0rtHxPgZOqaPW4araqMHivyaBOZtIsc8UmZXDvrD7bbXNYmNE36r/6GD18fQmzyf+yq34i7HniDT7v1J9DXm8n9YrnzXH1ybXXT40QmK3fRM54uIiIi3D0EVQHtUjNKzn6e2YCHQyZz27I53PPDh9z6y2ec9+96xg54gu3RLQDYlulNTla2m0esytJjRR5tIpN2kWFZQhoTlycCRf9dAaiXeYAn579E961rAVjSqQ+/DnuKQNOXXnV8GdgxWq+2qSF6nMhk5S468XSRmZlJcHCwu4ehytAuNSsuJoR3b27PlJjBDGp1Ls9+MoHTk//jw9eH8NZlg5jb7TqiAwvZcNDJpdM3YgLhAd480es0fbiDm+mxIo82kUm7uFd8UiYTlv1Heq6z5LXowEJO37CGJz97mfDsDOyBwUy8dhh/druUj/qVf9q6qn56nMhk5S468XThcDiOvZOqcdql5kWH+jPpilYsaxPBbVHNGPbNVK5b+y3Dv51Kt61rWTJ4OBtoxOHFmNJznYxYlMDoHk3pFWvd38RZnR4r8mgTmbSL+yzYlMzUtbZSrwXm5XD3R2/Q+ZelAKyL7cDTN45iX91IJl/Y1B3DVOhxIpWVu+jE00VUlN6kLpF2cZ9esREczGnBRL/h/HpGZ56c/zLnJmykzdghHOo/nB/iepTaf+LyRBLTc/TeGzfRY0UebSKTdnGP+KTMcpPO9rv+5plPJtIkLYk8H1/e6nsPn3Trj+nlxS1xkXoljRvpcSKTlbvow4VcWHldHE+mXdyrf/soZt3YFt8rL+f5SXNYeUZnQnKyeOHj53jmkxcIyskqtf+c+FRmrtMHD7mDHivyaBOZtEvNik/K5JY5mxixKKHkNW9nAfct/YD33h5Gk7QkkhqfxsCH3mHuhddjenkxuHO0/hLTzfQ4kcnKXfSMpwsrP57Yk2kX94sO9Wd0z+ZAc243JnPDr/O5af679NvwA+fs+JPnrn+U31p1Ktl/Tnwq/6XnMbhLY30IRA3SY0UebSKTdqkZNnsez/2wnYQDeaVeb7YvkWc+nUi73f9QaBh8eNGN/HL97WzPqUPbBnUY1aO5/rdDAD1OZLJyF514uvDz83P3EFQFtIssj13UjDfNK/kmpiPPfPoC7Xb/w1vTR7Gg8+W8fvl9ZAcEAbA60c7O9AQmXharP0DUED1W5NEmMmmX6heflMnoRQk4XV7zKnRyy4rPGPzdB/gX5GOr24CnbxrNhpZxNDOcTO4Xq5fWCqLHiUxW7qKX2rrIyMhw9xBUBbSLLHExIfzvrDDSGzdl0JA3eKvvIBzevvRfu5C5r9zDef+uL9nXlungvgVbiU/KdOOIaw89VuTRJjJpl+oVn5TJyDKTzmb7Ennv7Yd5eNG7+Bfk83Wnvtwy/D02tIzDAP53VphOOoXR40QmK3cxTNM89l5VZPXq1WabNm1q7PudqOzsbIKCgtw9DFWGdpHncJOZ6/YwJz6Vlsn/8dSnL9J27zYAPutyJW/0u5dDAXVKfZ2XAfeeF03/9ta9MV4yPVbk0SYyaZfq47o2JxSf5fzlcwYvfR//gnxSwuozof8jrDqjMwZwdlQQwy5sRqh3gTYRRo8TmazQZcOGDet79+7dqezreqmti4yMDPEhayPtIs/hJocf/DAHuPuBN7hj+af874cPuX7NN5z/z288d8MIfo/tUPJ1hSZMXWvDnlugD42oBnqsyKNNZNIuVc9mz2PKil1sTM4uea3Zvt08Nf9Fztq1GYCvO/Xl1SvvJyswmHB/L6Zc3abkVoykpDRtIoweJzJZuYtOPF3k5+e7ewiqAtpFHtcmd57bmEtbRzJ19R4+7nMbK9p2Zdy8Fzlj77+88+5jzO96FW/0u5cc/yM3w8+JT2VLyiGGXdhM7/+sQnqsyKNNZNIuVWtZQhov/pxIYfFFdF6FTgb88jmDl35AQIGDfaERjL/uUVad0RmA2Lp+jO1T+v5/bSKPNpHJyl30UlsXeXl5+PvrD8HSaBd5Kmtis+dx34Kt5Oc5uPOnuQz6cTa+zgKSwhsy6dqHWdWmc6n9/bxgfF99mERV0WNFHm0ik3apGjZ7HlNWJrIx6ciyWi1tO3jys5c5c/dWAL7pdCmvXnE/mXWK/p4f3Lni2y20iTzaRCYrdKnsUlt9uJALK6+L48m0izyVNYkO9ee5Pi3Ax4cZF9/OHQ+9zdaYWGLSU3jt/Sd4bs54wrPSS/Z3FMKIRQks2KSNq4IeK/JoE5m0y6lZlpBGvxkbGThvc8mk0y/fweAl7zP7tcGcuXsrKWH1GXbXeJ69cWTJpHN0j6aV3uOvTeTRJjJZuYteauvCqtdLezrtIs/RmsTFhDDxslheWpHI9uiW3Png2wz45TPu+34Wff9YRpdt63jtisF82/FSMAyg6L7PL/9KpW1UCAM7RuvltydJjxV5tIlM2uXkvblyJ19vTS/12jnb43ni81dotn8PhYbB/K5X8dZl95Qsr9U2MoBRPVsc9e92bSKPNpHJyl104unC29vb3UNQFdAu8hyrSVxMCB/d3A6bPY+hX25ldo+b+Kl9dx5f8Cqd/93AuHmTuWzDj7zQfxh76jcCIDm7gOTt6WzZl61rf54kPVbk0SYyaZcTZ7Pn8dyPO0hIyy15LeRQJg8tepdrflsEwI6GzRh/3aP82bwdAPXr+DCqR/Pjup1Cm8ijTWSyche91NaF3W539xBUBbSLPMfbJDrUn7euaUOHmGCSImJ44J4Xeeqm0RysE8p5CRuY+8o93PHTXLydBSVfY8t0MHX1nuoaukfTY0UebSKTdjkxyxLSuHPe5iOTTtOk158/M+/lu7nmt0U4vH2ZdslAbnt4asmkc3SPpsy5pf1x38OvTeTRJjJZucsxHy5kGMb7wBXAPtM0z6xgexvgA+AcYIxpmi9V9l7SHy506NAh6tSpc+wdVY3SLvKcTBPXh1DUzTrI8G/fod+GHwDYFt2S8dc9wuampf9+8PWCRy9sSq/YiCobuyfTY0UebSKTdjk+Nnsek5bvZPO+QyWvxRyw8dhXb9J9yxoA/mh+JuOve4SdDZsBEBHow+iex3eW05U2kUebyGSFLqeyjudM4E3gw0q2HwAeAq452cFJceDAAfEhayPtIs/JNIkO9WdSv9NZsCmZqWth3M2Ps7jDxYxeMIVWtu188NYDfHHe5bx92SDsdUIByC+EicsTOZiTX+kDKdQReqzIo01k0i5HZ7PnMeWXXfxhy+bw6QnfAge3/Tyfu3+cTUCBg6yAIN647H980flyTK+iC+hG9zj5XxRqE3m0iUxW7nLMS21N01xB0eSysu37TNNcB1h3UZliNbm0jDp+2kWeU2nSv30UgztHA7Cm9bnc/Oh0ZvW4iULDi+vWfstnk+/kynWLMQoLS75m6loboxZuw2bPO+WxezI9VuTRJjJpl8otS0jjrvmb2egy6Tzv3/XMfeV/DFn6PgEFDhZ36M31j33Agq5XVsmkE7SJRNpEJit3Oa51PA3DaA58W9Glti77PA1kHe1S26+++sp8/PHH8fHxwel00r9/f4YOHUpycjJBQUF4e3tjt9uJjIzkwIEDmKZJZGQkKSkpBAcHA5CVlUXDhg1JTU3FMAzq1atHamoqoaGhOJ1OsrOziYqKIjk5GV9fX8LCwti/fz9hYWE4HA5ycnJKtvv5+RESEkJaWhrh4eHY7XacTmfJ9oCAAAIDA0lPTyciIoLMzEwcDkfJ9sDAQPz8/MjIyKB+/fpkZGSQn59fsl3CZ8rJySE3N9fSn8nLy4vw8HCP+kxW7xQYGAhwSp9pT2o6GxIPsHSvSas6efjv3M2AuW9x5r/xAOxoeQZf3j6Ub0Jb07FuASl5XuQXwjWxdeje7jTtVMFn2rNnD4GBgR71mazeyWazUa9ePY/6TJ7QqW7dutjtdo/6TFXRKf7fRL7Ymk5SjhetQpwcSDrI4C+nct7vPwOQEt2EDwc8wKrTOtAyyMm2TG/i6ntzacswYps1OqXPFBoaysGDBz3+3z0rfSan00lubq5HfSZP6JSTk0Pjxo1Ff6a9e/dWeKltjU48pd/juWvXLpo1a+buYagytIs8VdnEZs9j9OIEbJkOME0u/WMZw76dSv3MAzgNL+affzVT+9xJdmBwydc0DfXlub6n65Nvy9BjRR5tIpN2KS0+KZNJy3ey/1DRg968nU5uXPUl9343k+C8Q+T6+jP94tv5uPv1FPj4AkX3ar3QL/aE7+WsjDaRR5vIZIUup3KPZ61x+LcCShbtIk9VNokO9WfiZbElDx5a2qE3K9t05r7vZ3Hjr19y869fcPGfP/P65fey5OzemF5eJNrzGThvM3X9vRjTu0WV/eBjdXqsyKNNZNIuRxTdd28r+XPHhI08+vVbnJ78HwA/tbuAV64aSnJ4w5J9jmddzhOlTeTRJjJZuYtOPJVSbnf4wUPLEtKYuDyR7MBgXrlqKN906suoL14jbtffPPvJRK5f9RUvX/VAydNvD+YVMmJRwinfW6SUUrXJ4TOcBw4VcPhu+pgDNh7+dhq9/voFgL31onnpqqGsbNu15Ot8DZhwWdWd5VRK1S7HnHgahjEX6AHUNwxjDzAO8AUwTXOqYRhRwO9AKFBoGMYwoK1pmpZbZCYrK4uICP3hVRrtIk91NekVG8EZDYKZsmIXfyRn829MS/53/xT6bfieBxZP56zELcx6cyjfdryEt/rew/6w+kDRk2+XbjvAsG5Na/Xlt3qsyKNNZKqtXeKTMhn/4w4O5h15eFtgXg4Df5rLbSvm4V+QT45vAO/3voU53W/A4esHgL8XnNM4lMFdGlfb37G1tYlk2kQmK3c5rns8q4r0ezxzc3MJCAhw9zBUGdpFnppoYrPn8dyPO0oWLK+Te4i7fprDLSs+w8+ZzyG/AD7oVfqHI18vg06NQhjctfp+OJJMjxV5tIlMtbHL4StKSpgml238gQcWvUcDexoAi865mDcvu4fUsEgAAny8eK5PzdzOUBubSKdNZLJCl8ru8Tzmciq1SWpqqruHoCqgXeSpiSbRof68fe0Z3BJX9APQoYA6vHXZPdz42Pv8dGY36jhyGbrkfea9dBc9N60A0yS/0GT1bjt3zdvMsoS0ah+jNHqsyKNNZKptXeKTMktNOtvu3sqMtx/i2U8m0sCext9NWnP30NcZd/PjJZPOegHeTOvfpsYuq61tTaxAm8hk5S56j6cLwzDcPQRVAe0iT002ufPcxnRoFMaEn3aSnlPA3ogYRt7xDJ0SNvJI8QMwXvzoGTacdhavXXEfm5u0oZCiy29fWrGbxy5sUmvu/9RjRR5tIlNt6GKz5zFlZSJ/2rJwFl/cFnPAxpDFM7g0/icA0oLDebPfPSw8p0/JepwAseH+jL2kZY1eOVIbmliNNpHJyl30UlsXhw4dok6dOu4ehipDu8jjriY2ex6TftrB5tSiy2+9nU6u+W0hg5d+QN1DRbeVL43ryTt972ZvREzJ19WWhw/psSKPNpHJ07ssS0jjxeWJJQ8OCj1k5+4fP+aGVV/h58wnz8eXT7r154Net5IdEFTydfXr+DCqR3O3PDzI05tYkTaRyQpdKrvUVieeLqywLk5tpF3kcXeT+KRMxixJwFH8U1VwThZ3/jSXm1d+jn9BPvnePszvejXv976VjKCwkq8L8ILn+nruExnd3UWVp01k8uQurvdy+uU7uHHVl9y17GNCc7IAWHjOJbxz6V2kuCyPUtVrcp4MT25iVdpEJit00YnncUhPTyc8PNzdw1BlaBd5JDSx2fOYunoP6/ZmUlBY9PdYw/QUBn83k34bvsfLNMkMCGJmzwF82q0/eb5HLhkL8oWnL/G8CaiELqo0bSKTJ3Y5vETK/kMFGIWF9In/iSFLZhCTngLAb7Hn8Nrl97Kt0emlvq5rk1ARD2TzxCZWp01kskKXyiaeeo+nC6fT6e4hqApoF3kkNIkO9eeZS1tis+fx4FdbsecVkhLekGduGsWc7tfz4KL36LptHQ8uns6Nq75i6qV3srjDJTi9vcnOhxGLEugQE+xRS7BI6KJK0yYyeUqX+KRMJv60k/TcAgpNwDTpum0d9y95nzP2/gtAQtRpvNHvf6xqfR643BvWISqIYRc2E/P3n6c08STaRCYrd9Gn2rrIzs529xBUBbSLPJKaRIf688bVbWgbGVjy2r8xLXnonokMvWcS/8TE0jAjlXHzJjP31Xu4OH45RmHRNbobk7I86gm4krqoItpEJk/oEp+UyYhFCaTlFE06z/7vT6ZNHc7rMx7njL3/khoSwXPXP8qtw6axqk3nkkmnnxdM7hfLpCtaiZl0gmc08TTaRCYrd9FLbV3k5eXh7y/nL2FVRLvII7WJzZ7HpGU72LI/l8N/sxmFhfT940fu/W4WjQ/YANgW3ZJ3Lr2LlWd0KflhrG1kIKN6nibqB7ETJbVLbaZNZLJyl8NPq92YVHTPZps927h/6fuc/886AA7WCeHDHjcz7/xryPMrvdaftLOcrqzcxFNpE5ms0EXv8TwOVrhZtzbSLvJYocnMdXuYE39krStvZwFXrVvMoB9n0zBjPwB/Nj2Dd/oO4vfYDgAYwNmCfzA7Fit0qW20iUxW7RKflMmYpdtxOE1OS9nJfd/NpPemXwDI9gtkzoXX83H368kODC75Gm8DRlwk/8neVm3iybSJTFboovd4HgdfX193D0FVQLvIY4Umd57bGKBk8un09uGLLleyqGMf+q/+hjt/mstZiVt4593HWBfbgXcuvYtNzdqxMTmbgfM2E1vPn7EX1+w6dqfKCl1qG20ik5W62Ox5TFq+ky37DmECTVL3MOjH2fTd+CPeZiG5Pn7MP/8aZvW8udRTvAE6RAcxrLs1fpFmpSa1hTaRycpd9Iyni+zsbIKCgo69o6pR2kUeqzVZsCmZqWttpV4LzMvh5pULuP3nTwnJLbpfYs3pHZlx8W38cdpZJfu1rR/AqF4tLPGDm9W61AbaRCYrdLHZ85iyYhd/JGdjAs32JXL3so+5dOMyvM1CCry8+bLz5bzf6xZSwyJLfW1EoA+je7pnPc6TZYUmtY02kckKXfSM53HYv3+/+JC1kXaRx2pN+rePomVEEBN/2klaTgEAOf6BfND7Vj7rehW3rpjPTb9+QZd/19Pl3/X83iKOGRffzu8tz2bz/lzumb+ZCZfJX37Fal1qA20ik+Quh+/h/CMpCxNokbyTu3+czSV/LsfLNIsmnJ0uY2avW9gbEVPqa/28DMb3bSn+76qKSG5SW2kTmazcRc94ujh48CB169Z19zBUGdpFHis3sdnzmPTTf2xOzSn1eughOzevXMDNKxeUnAH9o/mZzOh9G2tadQLDIMTP4MyoEAZ3cf+adxWxchdPpU1kktjFZs9jyi+7+MNWdIYz1radu3/8mN6bVuBlmuR7+/B1p77M6jkAW72ocl9vpctqKyKxSW2nTWSyQhc943kcHA6Hu4egKqBd5LFyk+hQf6Zc3abcZWz2OqG82+dOPu5+PTet+pIBv3zG2Tv/4o0Zo/mrSRtm9L6NlWd0YXWindWJmxndQ97DOqzcxVNpE5kkdVmWkMak5YklT+Juu3srd/40l55/rQTA4e3LV+ddxqyeA0ip26Dc19f192JM7xaWPMvpSlITVUSbyGTlLjrxdJGTk3PsnVSN0y7yeEKT6FB/Jl3RCps9j+d+3EFCWi4A2YHBvN/7Nj65oD/Xrfma236ez5m7t/LqzCdJiDqNjy66kaVn92Li8kTeXr2HsYJ+4POELp5Gm8gkoUt8UibP/bgDe14hmCZdt63jjp8+odOOeAByffz4svPlfNjjpnL3cAK0bVCHUT2aW/YMZ1kSmqjStIlMVu6il9q6sMK6OLWRdpHHE5ssS0hj8vJEnGVeD3DkcO2ahdy2Yh4N7GkAJNdtwJxu1/HVef04FFCHIF+Ds6JCGNzVvZfgemIXq9MmMrmzS9Hl/jvYnJqLt9PJxX8u547ln9LKth2ArIAgPu9yBXO7XUdaaPmrKnyAxwRecXGq9FiRR5vIZIUuuo7ncbDCuji1kXaRx5ObxCdlMmHZf6Tnlp6C+hTk0/ePH7l9+ae02JcIgD0wmPldr2beBddwIKQe4N4F2j25i1VpE5lqukvZJVH8HblctW4Jt66YT6P0ZAD2h9Rjbrf+fN7lylLrcB7WISaYYd2aeswZzrL0WJFHm8hkhS56j+dx8PPzc/cQVAW0izye3CQuJoRPbzsLmz2PqWv2sCbRjgkU+Pjybae+LDynD922ruH25Z/SYedfDFr2MbetmMe3HS9lbvfr2EhTBs7bTKi/N2N7n1ajl+F6cher0iYy1VSXw/eSb0wuemBZhD2N61d/zXVrviE8OwOAXfUbMfuim1h0ziU4fEuPywDOduMvs2qSHivyaBOZrNxFz3i6yMrKIji4/G8ZlXtpF3lqU5P4pEyeWJJAfmH5bWft/Jvbfv6Unn//WvLa6lbn8km3a1nd6lxMLy9i6/oxtk9sjfzQWJu6WIU2kam6uyxLSOOVFbtxFBb9jNVmzzYGrPycS+KX4+ssWtLpryZt+LDHzfzc7nwKvbzLvUdsuD9jL2np8RPOw/RYkUebyGSFLnrG8zikpaWJD1kbaRd5alOTuJgQpl/ftsIlWP5s3o6RzZ+l2b5EbvnlM/pt+IGu29bRdds6dkY2Yd4F1/Btx0sZOG8zBnBGZCCjep5WbT9I1qYuVqFNZKquLkUPDPoPe54Tb6eTXn+vZMDKBZy98y8AnIYXP7bvzifd+vNH8/ZgGOXew8cLHrvQ8+7hPBY9VuTRJjJZuYue8XRht9sJDQ119zBUGdpFntra5PA6e5tSDlFQWP7vztBDdq5Zu5AbVn9N1MF9AGQGBPH1uZcx74JrSKoXDYC3ASMuqvofLGtrF8m0iUxV3cVmz2PKykQ2JmUResjOVeuWcOOvXxDt8vfAl+ddzvzzr65wDU4vA0ZWw98JVqLHijzaRCYrdNGHCx2HlJQUGjZs6O5hqDK0izzapOjMxpgl20supXPl7XTS4++V3FzmTMeqNufxeZcrWd36XAq9vImt58/Yi6vuUjrtIo82kamqusQnZTL+xx0czHVyZuIWrlvzDRfHLyegoGidvV31G/Npt2v5tuOl5PgHlvv6Oj4GcTEhDO7i3idiS6DHijzaRCYrdNFLbY9Dbm6uu4egKqBd5NEmRZfgvnf9GSXLIrhyenvz41kX8eNZF9FmzzZu+nUBff5YTvcta+i+ZQ22ug34ovMVfH1uXwbOywMgyNfg6UtantLDiLSLPNpEplPpYrPnMXX1Hv6w2TGyDnHpHz9y3ZpvaZ2UULLP6lbn8skF17K6ddG93mX5GvBoLT/DWZYeK/JoE5ms3EXPeLqwwro4tZF2kUeblFZqIfhKhGelc8XvS7l27UKapCUBUODlzc/tzmdBlytZ17IDppfXKS3Hol3k0SYynWwXmz2Px77dRtj2f7huzbf03fADwXmHAEgPCuPrc/vyRecr2BsRU+HXe/qSKKdCjxV5tIlMVuiil9oeByusi1MbaRd5tEnFDi+dsGlfNgWVzEGNwkLOTdjAdWu+5cLNv+JTWLRjYkQjvj7vMhaecwn2epF0jAlhcNcTu/xOu8ijTWQ63i6Hl1Xasi+bOjnZ9N64jJ6rFtE+cUvJPhubn8nnXa9iWfvu5PtUvMxB2/oBjOrVQiecR6HHijzaRCYrdNFLbY9DQECAu4egKqBd5NEmFYsO9WfSFa2Aoh9Yn/sugYSDjlL7mF5e/NaqE7+16kT9jP1cvW4x165dSNO0vTyweDr3L3mfNa068W2nS7nvv/MZdnHscV+Op13k0SYyHavLsoQ0Xv1lN458J522/8Ejvy+h56ZfSu7dzPKvw6KOl7Cgy5VsjzqtwveICPRhdM/mNbqWr5XpsSKPNpHJyl104ukiMLD8jf/K/bSLPNrk2KJD/Xn7+nbY7HlMWr6TzfsOldtnf1h9Zlx8OzN73kLXf37jivVLuXDzai745zcu+Oc3MgJDWPptL97uey3/NW9DRJAfAztGV3rWRLvIo01kqqiLzZ7HrPU2NidnYuzewx2/L+WK9UuJSU8p2Wddy7P5plNflrXvTp5fxT/8+XkZjO97avdr10Z6rMijTWSyche3TzxN0yQrK4uavOS3MtnZ2e4ewikxDIPg4GCMCtYFs7L09HTxj42ubbTJ8YsO9WfKVa2Boh9sJy3bweb95R9GtLJtV1a27UpYdgaX/rGMK9ctoU1SAjeu+gpWfUVC1GksPOcSRq7vyb66DYio48OoHqXPpmgXebSJTIe7HL48/s+UbIKy7fTa9AtPb/yRjjviS/a11W3AN536srBTn5IlkSrTtprX6vVkeqzIo01ksnIXt9/jmZmZib+/P35+Fd8XUZOcTife3t7uHsZJczgc5OXlERLiWb9lzcrKsuxCuZ5Km5ya+KRMnlyaQJ7z6PudnrSdK35fymUbfyA8O6Pk9Q2ntWfp2b358awLGdCrDf3bF60LqF3k0SYy7Uw+wJTfUtm+5wDdt6yh78YfOf+f3/B1FgCQ6+PHT+27802nvvze8uxyT6aNCPQhv9AkJ7+QQF8v2jUIOuF7slVpeqzIo01kskIXsQ8XkrQIqsPhEDEBPhWS/nlWFZvNRnT00X/LrGqWNqka8UmZvLQiEXtuPoUm5Dkr/vvYpyCfC7au5dI/ltF98+qS+8wKvLxZ06oT9e4cQPQt/bFlZmoXYfRYkePw8ZaVlUP/vb/R8pef6fHXSoIcOUDRWrvrYjuwtEMvfmrXjezAin+wi6zjw0tXtNJJZhXTY0UebSKTFbrow4WOg4TLfVV5Dofj2DupGqVNqkZcTAgf3dyu5M/LEtKYuDyx3H4FPr78fGY3fj6zG0G52Vz49yr6/vEj5/27nm5b18LoteSOHUV+3HlMOaMH59zdnwvjZD/xrrbQY8W9bPY8pvyyi793H6Tj9o3c8+cKLtq8qtQVBH81acPSDr34Pq4naSH1Kn0vfy84p1GontmsJnqsyKNNZLJyFz3j6aKwsBCvChZ6thJJ/zyrihXWK6pttEn1KTkrk1dAgI8XJnAgp6DCfetmHeTiP3+mzx/L6LDzr5LX83x8WdeqE6s7XEhaz0u44+J2+oOym+ix4h7xSZm8tGQrrTetpdemX7hw8ypCco88xyGxQRMWx/ViSYfe7Knf6JjvN7hzdMkl7ap66LEijzaRyQpd9FLb43AyIefMmcMff/zBiy++WG6baZpcc801fPTRR5V+xv379zN48GA+++yzkxpzWZL+eVYVK6xXVNtok5pzeB3BdbvtFBzlr+uo9BTu+W85TdesIm7X33gV/91e4OXF+pZns+zMC/m53fk4IyMZ07uFPnGzhuixUjMOP7grcfd+uv6zjh5/r6TbljXUcRx5kFdC1Gn82P5CfmrfncbtYvg57ej/vW8R7k/zenWO+iRpVXX0WJFHm8hkhS56qe1xqOqznd9//z3t2rU76kSwfv36NGzYkDVr1tClS5cq/f6ewsqPjfZU2qTmRIf680yflkXrgn6/nYT0vAr3Sw5vyJfNruOvc24iwp7GRX+votdfK+i4/Q86/7uBzv9u4PEvprC5cWtWftOZn86/iJvu7Et0XW1ZnfRYqV42ex5z5v9CxM/LeGDLGjr8twmfwiNP7drcqBU/te/OsvbdSYxsUvJ6YH7FVxGAPpnWXfRYkUebyGTlLjrxBD755BPefPNNAM4880yeeOIJHnzwQdLS0qhfvz5vvvkmjRs35ssvv+TFF1/E29ub0NBQFi5cCEBycjLXX389O3fu5PLLL+eZZ54BYP78+QwcOBCADRs28NBDD/HDDz/gdDq5+OKLmTFjBm3btuXyyy/ns88+04lnJaz+wCdPpE1qXnSoP29f17bUpbjB/j6kH8rHUVh0djOzoGgppbTQCBZ0vZIFXa8kLDuD7ltW02vTL5z77wba7vmHtnv+gR8+ZP+r9VjR5jxWntGVbWeey/B+7fRMaBXTY6Vq2ex5vPvLTli9inM3rabbljU8v39PyfYCLy82nHYWK9p25af23Std/uTwseJtQKEJXsBZ0UEM695MJ5xuoseKPNpEJit3EXWpbXi9ym/qPxXpBw5Uum3Lli3ccccdLFmyhODgYA4dOsSQIUO46qqrGDBgALNnz2bJkiXMnj2bCy64gPnz5xMTE0NGRgZhYWHMmTOHyZMn8/PPP+Pn58d5553HokWLaNy4MWeddRa//vpryfIm48ePJzc3l9zcXGJiYhg+fDgASUlJ3HDDDfz666+n/Fn1UltVE7SJHK4PJLqovoOf91f+HyR/Ry7nJmzkgq1r6b5lDQ0zUku25Xv7sKHFWaw9vSO/xZ7D7qaxPNO3lU5ET5EeK6cmPimTCT/uIHhPIl3+/Z3z/l1Pp+3xBLvcr5kRGMLq1ufxyxldWNO6E/Y6x/5v4EX1HXQ9M5ZesRHVOXx1AvRYkUebyGSFLnqpbSV++eUXrr76aiIiInA6nYSHh7Nu3To+/PBDAG666SaefvppADp37szQoUO55ppruPLKK0ve48ILLyyZ7LVu3Zo9e/bQuHFjDh48WGpNzREjRtC7d2/8/f2ZOHFiyeuRkZEkJyfXwKe1pvr167t7CKoMbSLH4R+cp6zcw9bMo/+VnucXwMq2XVnZtiuTTJPY5B1027KGblvW0j5xc8kluQDpQWGs+6gDX3ToTPdB11G/XWy1fxZPpMfKibPZ85i/fAv+K36mzV/rmPnvemLSU0rt81+DpvxyRhdWntGVP5u1w3mca3B7G3BWdDC3dqxP84bh1TF8dZL0WJFHm8hk5S6iJp5HOzNZE5xOJ95H+Y/XK6+8wu+//7+9Ow2Oq7zzPf59etO+S5Zka7Mtb3jBxo6xITjYEDNJTEzMfu/EDuHmJjDcKopMBUhmQkjVZAJTVDYm8Z0iuZB7KZxAIIjM5BJszIBvMDHG2PISy5LlRda+WKu71ctzX/SRaMmS1ZJa6qdb/0/Vqe5+zjmt5/RPT3c/fc55zof86U9/YuPGjezduxdgyIBEdrsdn883eD90pNz29nZ6e3vxer243W5SUlIAcLvdJCYmTtVmxbzOzs7B10qYQTIxy6byHDaV51BfX49KzWHn+3Uca+6l2+Nn1GNalKK6cD7VhfN5ftN/JaO3k7WnDrL21Edce+oghReb2Xz4HTj8Djz/FE35RZxYupb6qz/FmrtvIW9h2bRtXyyTtjK2t6vbeO7NY1xVXcmq2iNcc/oI/3ihanCALICLyekcKL+GDxZcwwcL19CYlR/28ztssHxWCg9v+OQw2vr6ekA6niaRtmIeycRMsZyLUR3PaLjhhhvYvn07Dz74ICkpKXR0dLB27VpeffVV7r77bl5++WXWr18PQG1tLWvWrGHNmjXs2bOHCxcuXPG5y8vLOXPmDPPmzQPgkUce4dvf/jZnz57lySefHBwJt6amhiVLlkzthsYwr9cb7SqIYSQTM3m9XkrTE3jylvmANdLn3tMcb3GPsSZ0pmTw1spNvLVyE2hNcesFrj11kLXVH7Gm+hD5TXXkN9XB26/Cjx6nLruQQ/NWcHTeClbdfQvX3rgKlJrqTYw50lZG1lx9ngO/+b+kHdjP8lOH+ffG2iEdzX67k8NlS/lg4Ro+WLCak7PL0eMcAPCqvEQe3ThvxHM2JRfzSCbmkUzMFMu5zPiO55IlS3jkkUfYsmULdrudFStW8NRTT/HQQw/xs5/9bHBwIYAnnniCmpoatNZs2LCBZcuWUVlZOepzb968mX379jFv3jx27dqFw+HgjjvuwO/3c8stt/Duu++yYcMG3nvvPTZv3jxdmxxzCgrk2mmmkUzMNDyXwvQEfrx1yeAlWY419dLb78c/1qn9SnE+r4jzeUW8ct1W7H4/iy9U8anqQ6ysreTqM0cpam+gqL2BWz98E377L7SmZfPx3OUcK7sK98pVfPG/3EzBrMwp29ZYIW0FGtp6+MPL/4nzo4MsqD3O8rPHWdRax6KQZTwOJ8eKl3Bo3goOzV3O4bKluF3jG7kx1WljeWEq31hXNOYAQZKLeSQT80gmZorlXIwaXCjaIn1B1sbGRh544AFee+21Ky73hS98gRdffJHMzMxJ/02TXs9IiYWTqGcaycRM4ebydnUbz/znObwTfPu3BfwsqK9h1ZlKVp2uZGVtJdm9F4cs47PZaSgpp2beUpqXrOCa224kZ9UyiPBlq0w349qK1vz1oyr2vPQnyqqPsvTcCZZcOEWid+hlgPpciRwpXWp1NFdwrHgx/c7xj9RoV5CV5ODRG8vGNRDWjMslBkgm5pFMzBQLucjgQmGI9HU8CwoK2L59+xU7g62trTz44IMR6XTGq1g9jj2eSSZmCjeXgXNCQz277wwVf+0Ia/2Azc7JooWcLFrIrk/fDlpT2nKelbWVLD1/kqXnTzC/8QzFZ05SfOZk8PDcf4WehGSqZs/n1OxyuhcuZtO2z5C1egXE8Tnucd1W/H5O/fkwe373DnPOVrGovpqF9TWs7+1k/bBFz+XM4WjpEo4WL+FYyWJOzi7Hb5/YV5CrZiXz6I1lk7rsSVznEqMkE/NIJmaK5Vxkj2cIn8+HwxHbfXGTXs9I6ejoICtLBoEwiWRipsnm0tDlYef7dVQ29dDTH5hUXZI8l1h84RRLz51g2fkTLD1/koKLzZct57fZOZdfQs2c+fQsuoprbl5LxqplBIqK4mLvaFy0Fa1RTU0cefsAB/ccpLT+NAvrqylvrL1sTyYEL29yvHgxR0sWc7RkCceKF9OZkjGpKjhtimUFKTz86ZKIXGczLnKJM5KJeSQTM8VCLrLHMwx+vz/mO57xqKury/gGNtNIJmaabC6FwwYmeuFgA219XpKdNtz9fiqbevGF+VvlpYSk4GGU81YMluV0t7OgvoaF1p6xhQ01lLTUMbehlrkNtfDhbnjRWt+ZyPlZxXSVzaPk2hUkL78K/8KFBObOjak9pDHVVrxebOfP01n5V468cxBX1UkK6s8wt+ksae5eNgAbhq1Sn5VP1ez5nJy9wLotpylz1qQHmproIbThiqlcZgjJxDySiZliORfpZYWQTqeZ8vLyol0FMYxkYqZI5lKYnsBjG8uGlA3sEf3oQheeCewQbUvLpm1RNvsXfWqwLKHfzfzGMyyqr2ZBQw1zm85S1nKe3O52Fl44BRdOwf97c3D5gFK0pefQlDeH3KULyFhSTqCsDH9pKYHSUnR+vlGj6xrVVrRGXbyI7dw5bLW11B48zukPT1DY1sCctnoKLjZj1wEygJJhq3YmpVGbX8qZWSWcnlVK1ez5VM0upzs5cp3CRIeNhz9ddNlh4FPBqFwEIJmYSDIxUyznMmZPSyn1K2AL0Ky1XjbCfAX8BPg80Ad8RWv9UaQrOh3Guo6niI729naSk5OjXQ0RQjIx01TnErpHdEBDl4cfv3uWI429BAC7DRbmJFHT4cYTxu5RjyuR4yWLOV4y9DSMtL5uylrOUdYcnOZat4UdjeR1tpLX2QrVh+H1oc/Xb3fSmp5DS0Yu/fn5zFs2n6TSIgKFhejZswnk56NzctDp6dPSQZ2WtqI19PRga2tDtbZia2rCVl+PamjAVl+Prb6eQN0F1IV6XCGHxi63pgEBpWjInEVd7hxqZ5VQO6s0eJtfSntqVkRfLxvgsisykpz8/YaSKdmreSXyHmYeycQ8komZYjmXcHbxPQ88C/x6lPmfAxZY07XAL6xbISJiOs9DFuGRTMwUjVwK0xN4asvCy8oP13fzD2/W4Bnz2i0j605Oo7J0KZWlS4eU2/1+8jubmdPWwOz2BuaETm31ZPZ1M7ujkdkdjXDmKHywZ8Tn99nsXEzJwJORRWZRPs78PAI5OeisLHRaGjo1FVJTB+/rgfspKZCQgHY6weUKTlfokF0xE63B6wWPB9XfH7z1eFC9vajubujuRg1MXV2f3G9vp7e+mfZzDaR1XySz9yIJvvCu69aTkExT5izqcmZzIbsweJsTvG3IysfrGP/IsmNRQE7y1B02OxHyHmYeycQ8komZYjmXMTueWut3lVJlV1hkK/BrHXwV9iulMpVShVrrhkhVcrpM16G2v/jFL9ixY8eEf62orKyksbGRz372sxGumZli+ZCCeCWZmMmkXK6enca/3b6Ene/Xcax56PVDFTA/y0Vbn4+OcR6z67fbqc8upD67cMT5if2XyOtsI6+rlVmdrUNu8zpbyem5SEZvJ6mePnK726G7HepqJrWtAYcTlZiA3+GkFxu+wCdfCoq0JuQhCo3T58Xp8+Lye7FN8AtEApAd8tjtTKAjJYOOlEza0rJozsijOTOPlvRcmjJyac7IoyUjl97E6RkNMd0Ja0qy2LG6MCKDAUWaSW1FBEkm5pFMzBTLuUSipzUHOB/yuM4qu6zj2dzczP3334/D4cDv97Nt2zbuu+8+PB4PNpsNpdTgAD9+vz9YQYcDr9c7eAis3+/H6XTi8/kAsNvt+Hw+7HY7WmsCgQBOpxOv14tSKuz5DoeD/v5+bDbb4HybzYbNZhuc7/f70VoPmT+8zmPNB9i5cydbt27F5XJNaJsOHz7MoUOH2Lhx42Xz+/r6sNlstLW1kZWVxaVLl3C73RQUFNDY2EhiYiJJSUl0dHSQk5NDd3c3/f39g/OTkpJwuVx0dnaSm5tLZ2cnXq93cH5KSgp2u52uri7y8vJob29Ha01eXh5NTU2kpqYC0NPTQ35+Pi0tLSilyM7OpqWlhfT0dPx+P729vYPP6XQ6ycjIoLW1lYyMDPr7+7l06dLg/J6eHkpLS+Nqm1wuF2lpaTG7TW63e7De8bJN8ZBTVVUVubm5Rm3T15YmoK9yjbhNKiWLNw/VUNftparHzrJ0H3WXbDhtkJ8Q4OBFB6szffT5FWf77CxJ83Gmz06aQ5Pj+mR+t0/RcMnGwjQ/Nb0usvPySXPOYp81/4hX0dZvY36Kn6puO4VJAbK0h5oLfayzteO/2I3zYidl3ov0tveS7+8ltf8SfZ2XKAz0ovrcJLj7SPP0YXO7SfR7sft82H1enH4fNp8Xerw4gPGO3eq32eh3uPA7nfgdTvrsLnRyIn0JyVx0pWBPS6LFkYJOSYaUJKp1Ktm5KTQlZ1KXmEVWXiofk0V2ZgJpDj34mrT12+j2KcqS/ZzpdlCa7KfUrjl4McDqTB9NHhveABQlBTja5WBBqh+H0lR2OViZ4aPBHRxRuDAxwMedDpan+/BpxakRcjre7eSaLB8ZyYncvLQY5e765H+voxFPgnntyefz4XK5Ztx7hMnb1N/fj8PhiKttivWcWltbSUpKiqttioecWltbWbRokdHbNJqwLqdi7fH8wyjneP4B+KHWep/1eA/wqNb6w+HLRuJyKqEjLeYkOyf9a+q5c+e48847WbduHX/5y18oLCzkxRdfJCkpiVtvvZXVq1ezb98+Ojs7+elPf8r69UOvTqa15oknnmD37t0opfjmN7/Jtm3b2LdvH88++yy7du0C4Fvf+hYrV66ku7ub7373u5SXl5OTk0NFRQXFxcV8+ctfZu/eveTn5/Pcc8+Rm5vLrbfeyve//31WrVpFW1sbmzZt4sCBA6xevRq3201hYSEPP/ww27Ztm/DrGQva2trIyZn6wSZE+CQTM8VqLg1dHn687xxHGnqY4JG50aM1Tr+XBL/VEfV50SGH3s5P8VHT5xyySr/dSb/DidfhJGCLvXEFyrNc/ONny43ckxmuWG0r8UwyMY9kYqZYyGUqL6dyASgOeVxklUVcQ5eHx/5YTUN3/2DZieZefvi5yX0Anj59mueee45nnnmGr33ta7zxxhvcddddQPDanrt37+att97i6aef5rXXXhuy7htvvEFlZSXvvfcebW1t3HTTTVx33XWj/q2vf/3r/PznP6eiomLwn6a3t5dVq1bxgx/8gKeffnpwGonL5eLxxx/n448/HnUZIYSIJYXpCTz1+QWXlQ+MonuipQ/QzE5P4HxHH93hnc44PZTC63AFz40c4WMoJ9VHmz12RkxPT7ABCptSLMlL5hvri2K6gymEEMIckfg0rAAeUkrtIjioUOdUnd/5wsGGIZ1OgIbufl442HDZsP/jUVpayvLly/F4PKxcuZJz584NztuyZQsAV1999ZDyAfv37+f222/Hbrcza9Ysrr/+eg4dOkRaWvgDKNhsNr70pS8BcNddd7F9+/YJb0s86unpMf6XnZlGMjFTvOUy0ii6EBy46J/2nKbTE0AT7CzNz06iy+3jXKcH7wQu9TJZNsWQczkHFCYGqOqZ/voMl51opyDVSVW7G18gOKqs3QYOmyI9MTojy0ZTvLWVeCCZmEcyMVMs5xLO5VReAm4EcpVSdcATgBNAa70T+A+Cl1KpJng5lfumqrJtfSP/zD1aebgGzrV0Op2D53QOSEgI/tI7cN5luBwOB4HAJ99+3G532Osq6zCt0OcYz/rxJj8/P9pVEMNIJmaaKblcPTuN33756lHnD99TOjc7iUS7jQ63l8YuD11WhzWSR/X+97WF/K6yhZa+oZ8TH3dO3d5OxeXboAh+sOvgKZokO+0snZUiey6HmSltJZZIJuaRTMwUy7mEM6rtvWPM18DfRaxGV5CT7BxX+XiNp2M5YP369Tz//PPce++9dHR08Oc//5knn3wSr9fLyZMn8Xg8uN1u3n33XdatWwdAamrqkF8rAoEAr7/+OrfffjuvvPLK4HLFxcUcPnyY1atXU1FRMfg3B9afKVpaWiguLh57QTFtJBMzSS5Bo+0pHW7wGqRNvSOeW2oDxtp56lTwzc+UsKk8h/WlWYMd3oDWoANcne5jX6trsKNrA2w2RSCg0QQ7igl28GrwhfyxFKfC49P4NNgVJDoUl7zBdewKlhem8vCnS6QzOUHSVswjmZhHMjFTLOcSOyeeADtWF3KiuXfI4baFaS52rB55WP3psGXLFg4cOMANN9yAUorvfe97g79E3HbbbVx//fWUlJSwYsWKwXV27NjBnXfeSUFBARUVFaSkpPDRRx/xzDPPkJeXxy9/+UsAHnroIb761a/ywgsvsHnz5sH1b7jhBn7yk5+wYcOGywYXikdqGi7yLsZHMjGT5DI+o12DdDLPN7zDW1dXxz8UFUXsb4jIkLZiHsnEPJKJmWI5l7BGtY0UE0e1DeX3+wcv2zKdiouLOX/+/NgLhiEeR7Xt6+ub8DVPxdSQTMwkuZhHMjGT5GIeycQ8komZYiGX0Ua1tUWjMpNRmJ7AYxvL+JcvLOCxjWURPcxoIofaiqnX0tIS7SqIYSQTM0ku5pFMzCS5mEcyMY9kYqZYziXmOp5TKRp7O4GI7e2MV/G2BzceSCZmklzMI5mYSXIxj2RiHsnETLGci3Q8Q0znYccifH6/P9pVEMNIJmaSXMwjmZhJcjGPZGIeycRMsZxL1DueSin6+/vHXnAahF7+JBb19/fH9AnHo+nt7Y12FcQwkomZJBfzSCZmklzMI5mYRzIxUyznEvVRbQcuDWLCdSr9fj8ejyfa1ZgwpRSpqanRrkbEFRQURLsKYhjJxEySi3kkEzNJLuaRTMwjmZgplnOJesdTKUVaWlq0qwHA2bNnKS0tjXY1xDCNjY2Si2EkEzNJLuaRTMwkuZhHMjGPZGKmWM4l6ofamuT3v/99tKsgRiC5mEcyMZPkYh7JxEySi3kkE/NIJmaK5Vyk4xni1VdfjXYVxAgkF/NIJmaSXMwjmZhJcjGPZGIeycRMsZyLdDxDyHU8zSS5mEcyMZPkYh7JxEySi3kkE/NIJmaK5VzUdF5CZM+ePS3A2Wn7g+PU3t6em52d3RrteoihJBfzSCZmklzMI5mYSXIxj2RiHsnETDGSS+lNN92UN7xwWjueQgghhBBCCCFmHjnUVgghhBBCCCHElJKOpxBCCCGEEEKIKTXjOp5KqTuVUseUUgGl1JorLPc3SqmTSqlqpdRjIeVzlVIfWOW/UUq5pqfm8U0pla2Uekspdcq6zRphmY1KqY9DJrdS6jZr3vNKqdqQeSunexviTTiZWMv5Q173ipByaSsRFmY7WamUet96nzuilLo7ZJ60kwga7XMiZH6C9b9fbbWFspB5j1vlJ5VSt0xrxeNYGJk8opQ6brWNPUqp0pB5I76XickLI5evKKVaQl7//xYyb4f1nndKKbVjemsev8LI5EcheVQppS6GzJO2MgWUUr9SSjUrpY6OMl8ppX5qZXZEKXVNyLzYaCda6xk1AUuARcA7wJpRlrEDNcA8wAUcBq6y5v0WuMe6vxN4INrbFA8T8DTwmHX/MeCpMZbPBtqBZOvx88Ad0d6OeJrCzQToGaVc2koUMgEWAgus+7OBBiDTeiztJHJZjPo5EbLMg8BO6/49wG+s+1dZyycAc63nsUd7m2J9CjOTjSGfGw8MZGI9HvG9TKZpyeUrwLMjrJsNnLZus6z7WdHeplifwslk2PL/A/hVyGNpK1OTywbgGuDoKPM/D/wRUMA64AOrPGbayYzb46m1PqG1PjnGYmuBaq31aa11P7AL2KqUUsAm4BVruReA26assjPLVoKvJ4T3ut4B/FFr3TeVlZrhxpvJIGkrU2bMTLTWVVrrU9b9eqAZuGxkOTFpI35ODFsmNK9XgJustrEV2KW19mita4Fq6/nE5IyZidZ6b8jnxn6gaJrrOBOF01ZGcwvwlta6XWvdAbwF/M0U1XMmGW8m9wIvTUvNZjCt9bsEd6qMZivwax20H8hUShUSQ+1kxnU8wzQHOB/yuM4qywEuaq19w8rF5OVrrRus+41A/hjL38Plb4L/ZB168COlVELEazjzhJtJolLqQ6XU/oFDn5G2MlXG1U6UUmsJ/ppdE1Is7SQyRvucGHEZqy10Emwb4awrxm+8r+v9BPceDBjpvUxMXri53G69N72ilCoe57pifMJ+Xa3D0ecCb4cUS1uJjtFyi5l24oh2BaaCUmo3UDDCrO9orV+f7vqIoCvlEvpAa62VUqNe58f6dWc58GZI8eMEv4i7gH8DHgW+P9k6x7sIZVKqtb6glJoHvK2UqiT4BVtMQITbyf8GdmitA1axtBMhAKXU3wJrgM+EFF/2Xqa1rhn5GUSEvQG8pLX2KKW+TvBIgU1RrpMIugd4RWvtDymTtiImJC47nlrrmyf5FBeA4pDHRVZZG8Hd2g7r1+uBchGGK+WilGpSShVqrRusL8zNV3iqu4DXtNbekOce2AvkUUr9L+DvI1LpOBeJTLTWF6zb00qpd4BVwO+QtjIhkchEKZUO/DvBH9v2hzy3tJPIGe1zYqRl6pRSDiCD4OdIOOuK8QvrdVVK3Uzwh5zPaK09A+WjvJfJl+nJGzMXrXVbyMPnCJ7PPrDujcPWfSfiNZx5xvMedA/wd6EF0laiZrTcYqadyKG2IzsALFDBUTldBBtdhQ6ewbuX4PmFADsA2YMaGRUEX08Y+3W97FwD60v4wLmFtwEjjggmxmXMTJRSWQOHayqlcoHrgePSVqZMOJm4gNcIngfyyrB50k4iZ8TPiWHLhOZ1B/C21TYqgHtUcNTbucAC4C/TVO94NmYmSqlVwP8Evqi1bg4pH/G9bNpqHt/CyaUw5OEXgRPW/TeBzVY+WcBmhh7tJCYmnPcvlFKLCQ5W835ImbSV6KkAtluj264DOq0flGOnnUR7dKPpnoAvETz22QM0AW9a5bOB/whZ7vNAFcFfcL4TUj6P4BeEauBlICHa2xQPE8HznvYAp4DdQLZVvgZ4LmS5MoK/7NiGrf82UEnwi/T/AVKjvU2xPoWTCXCd9boftm7vD1lf2kp0MvlbwAt8HDKttOZJO4lsHpd9ThA8dPmL1v1E63+/2moL80LW/Y613kngc9HelniZwshkt/XZP9A2KqzyUd/LZJqWXP4ZOGa9/nuBxSHrftVqQ9XAfdHelniZxsrEevw94IfD1pO2MnWZvERwJHovwb7K/cA3gG9Y8xXwr1ZmlYRcnSNW2omyKiuEEEIIIYQQQkwJOdRWCCGEEEIIIcSUko6nEEIIIYQQQogpJR1PIYQQQgghhBBTSjqeQgghhBBCCCGmlHQ8hRBCCCGEEEJMKel4CiGEEEIIIYSYUtLxFEIIIYQQQggxpaTjKYQQQgghhBBiSv1/ArlL5KM0oF0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "outputs = [nn.feed_forward(a)['z'] for a in x]\n",
    "plt.scatter(x, outputs, label='nn output')\n",
    "plt.plot(_x, _y, color='r', label='cosh(x)')\n",
    "plt.title('Grafický výsledek učení')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f1105de3460>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax, fig = plt.subplots()\n",
    "fig.set_xlim([0,200])\n",
    "fig.set_ylim([-0.0005,0.002])\n",
    "plt.xlabel('epoch')\n",
    "plt.plot(lowest_mse['mse list'], label='mse')\n",
    "plt.title('Vývoj mse v epochách')\n",
    "plt.legend()"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}