{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "Итог: Простейший анализ данных.ipynb",
"provenance": [],
"collapsed_sections": []
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "eJYTR2X3nCEb"
},
"source": [
"# Phystech@DataScience\n",
"## Занятие 1. Простейший анализ данных.\n",
"\n",
"*Примечание.* Подробнее про работу с различными библиотеками Питона можно посмотреть в наших туториалах."
]
},
{
"cell_type": "code",
"metadata": {
"id": "oFfC6Lu--5QR"
},
"source": [
"import numpy as np\r\n",
"import scipy.stats as sps\r\n",
"import matplotlib.pyplot as plt\r\n",
"import seaborn as sns\r\n",
"sns.set(palette=\"Set2\", font_scale=1.3)\r\n",
"\r\n",
"%matplotlib inline"
],
"execution_count": 1,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "clYTo6ZQA5lc"
},
"source": [
"## Выборка\r\n",
"\r\n",
"Пусть нам попались какие-то данные.\r\n",
"\r\n",
"В данном случае — 100 различных значений какой-то величины."
]
},
{
"cell_type": "code",
"metadata": {
"id": "hM0Sc5y7_FpY"
},
"source": [
"size = 100 # размер выборки\r\n",
"sample = sps.norm.rvs(size=size) # генерируем реализацию выборки из N(0, 1)"
],
"execution_count": 2,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "KFkaJ1N6CGHn",
"outputId": "bd8f3cb0-fb60-4dc4-dd00-6c74acce02d6"
},
"source": [
"sample"
],
"execution_count": 3,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([ 1.4694579 , -1.16372156, -0.09200531, 2.04196851, -2.19080146,\n",
" -0.80042021, -0.09050994, 0.48966552, -0.6104091 , -0.23999291,\n",
" -1.0702828 , 2.45909955, -0.76026938, 1.23184359, 0.35449563,\n",
" 1.11132924, -0.01478976, 0.90131138, 1.94285115, -0.00437244,\n",
" -1.42938627, -1.583689 , -0.75399546, 0.23257718, 0.46463143,\n",
" 2.28289654, -0.63173655, -0.48240007, 0.36558808, -0.33946135,\n",
" -0.69307015, -0.5291172 , -0.26940048, -0.06775259, 1.61811464,\n",
" -0.64247747, -1.25793313, -0.57045423, -0.51259108, -1.84549721,\n",
" -0.0415442 , -0.75694694, -1.16910648, -0.64347649, 1.84542019,\n",
" 0.64610333, -1.18691192, -0.53480103, -1.3065839 , 1.29200288,\n",
" 1.39365151, -1.14226594, 2.39410418, 0.54798032, -2.14564358,\n",
" -0.9717628 , -1.12055401, 0.56010414, 0.33707968, -1.18463322,\n",
" -1.00506209, -0.8777773 , 0.58858336, 0.89087633, -0.83548504,\n",
" 1.07012296, 0.56872595, -0.12620446, -0.57501254, 0.16505806,\n",
" 1.23542741, 0.70491768, 1.23909053, -0.54949278, -0.00982067,\n",
" 0.56001076, 1.37450317, 0.66161703, 0.51650303, 0.66760551,\n",
" 0.20649633, -1.46794604, 1.3694449 , -1.15321223, 0.85429712,\n",
" 1.13053514, 1.00234837, 1.11106 , 0.74626235, -1.0790917 ,\n",
" 0.64532372, -0.47783549, 2.17570268, 1.47756022, 0.4606338 ,\n",
" 0.80496987, 2.40936153, -1.45828842, 0.51503181, -1.61893947])"
]
},
"metadata": {
"tags": []
},
"execution_count": 3
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Ito8HkWme9Jo"
},
"source": [
"Нарисуем их график. По оси x — значения реализаций случайной величины."
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 164
},
"id": "V-rjyv20_U1h",
"outputId": "889972a4-0ff3-40a3-eb8c-fd0b3401b4a2"
},
"source": [
"plt.figure(figsize=(15, 2)) # объявляем график и его размер\r\n",
"plt.scatter(sample, np.zeros(size), alpha=0.5, color='purple', label=\"Реализация выборки\")\r\n",
"plt.legend() # добавляем легенду\r\n",
"plt.show() # печатаем график"
],
"execution_count": 4,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA4gAAACTCAYAAADMSTcLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXwV9b3/8deZsyU5WckOyKYwLnilFuqGCyKWW662VmrrVuty1auNttrW2lpFq+Jafyp1QbFaFKi1tvXWutRe27oVi9QFLVMoVBBZkkBClpOzzu+Pc3I45+Rk5YQcyPv5ePDQzMx35jPf+cxkPpnNYds2IiIiIiIiIsZQByAiIiIiIiK5QQWiiIiIiIiIACoQRUREREREJE4FooiIiIiIiAAqEEVERERERCROBaKIiIiIiIgA4BrqAPYwLzAN2AxEhjgWERERERGRPc0J1AJ/AwLpIwelQDRN81TgFmASsAGYb1nWY7208QC3AecARcAbwDcty1qdNM2fgOPTmr5hWdb0PoY2DXitj9OKiIiIiIjsq44FXk8fmPUC0TTNI4BngZuBZcBJwELTNLdblvWbHpreDZwBnAdsAm4C/mCa5kGWZbUmTfdz4Jqkn4P9CG8zwI4dbUSjdj+a7fvKywtpbGztfUKRQaZclFyifJRconyUXKFc3LsZhoOyMh/Ea6N0g3EF8dvAa5ZlzYv/vDpeNH4XyFggmqZZDFwMXGxZ1gvxYecCW4EzgUeSJvdblrVlgLFFAKJRWwViBuoTyRXKRcklykfJJcpHyRXKxX1CxkfuBuMlNUcDL6cNexGYZpqmu5s2UwFPcjvLslqI3WZ6TNq0c03TrDdNc7Vpmg+aplmZpbhFRERERESGtcEoEGuIXflLtgVwAxU9tLGBbRna1Sb9vAQ4F5gB/AA4Afhj/PlFERERERER2Q171VtMLctamPTjKtM03wHWAXOAX/d1PuXlhdkObZ9QWVk01CGIAMpFyS3KR8klykfJFcrFfddgFIhbgOq0YdVAGGjooY0DqCL1Yclq4NPuFmRZ1semaX5K7G2pfdbY2Kr7ptNUVhZRX98y1GGIKBclpygfJZcoHyVXKBf3bobh6PGC2WAUiG8Cs4D5ScNmA29blhXqps0KYm8jnUXsLaWYpllI7PnDb3W3INM0a4ndgprxDTwiIiIisnfz+9tobW0iEgkPdSgSt22bQTQaHeowpBtOp4vCwlLy830Daj8YBeI9wBumaV4P/AKYCZwFzO2cwDTN04gVkDMty9pkWdZO0zQXArebprmN2FXDG4HtwNJ4m/2JfSPx98SeVZxE7LuJnxD7rIaIiIiI7EP8/jZaWnZQWlqJ2+3B4XAMdUgCuFwG4bAKxFxk2zahUJCmpnqAARWJWX9JjWVZy4HTiX3T8APgauCStG8glgAmsRfXdLqa2HcTFwPLgWJgVtI3EIPEXk7zAvBPYCGwEjg67TuJIiIiIrIPaG1torS0Eo/Hq+JQpA8cDgcej5fS0kpaW5sGNg/bHlbP4o0D1usZxK50L7nkCuWi5BLlo+SS4ZiPW7Z8THX1GBWHOUZXEHOfbdts3bqBmpqxXcYlPYM4Hvh3l/GDHp2IiIiIyACpOBTpv93Zb1QgioiIiIiI9EE4vO+/LGmv+g6iiIiIiIjInhCJRHj22V/y6quvsGXLZpqbmzjhhJn86Ec3DXVog0oFooiIiIhIlt1yyzxeeOF3ADidTioqKjnyyKO5+OLLKCkpHeLopDe2bXPttd9h06aNXHjhpRxwwAEYhpMRI0YMdWiDTgWiiIiIiMggOPzwqdxww81EIhEs6x/cfvvNbNu2lTvvvHeoQ5NevPLKS/z73+tYtOhJioqKhjqcPUrPIIqIiIiIDAKXy015eQVVVdUce+wJfOUrZ7J8+VsEAh0ANDQ0cNNNP+ILX5jJ5z9/PHV1l2BZqxPtd+7cyU03/Ygvf3kOJ554DGee+WWWLn2S9K8QrFy5gunTp3b519jYAMCiRQ9z1lmnp7T5n/+5kOnTp7Jq1QcAvP/+u5xzzhmcdNJ0Zs06jssv/29Wr/4oMb1lrebqq6/gv/5rFjNmHMNFF32dv/71zZR5zp17CosXP54ybPHix5k795QusXbGBvDwwz9l+vSpKW03bPiYq66qY/bsGRnXKZPk6WbNOo7LLruI1av/kRh/yy3zuOqquoxt0+N67bU/s//+B3Dddd/j5JOPZ86cmdx2249pa9v1dT3btlm8+GfMnXsKJ5xwJGee+WWef/65LjE988wyrr32O8yceQynnfYFfvWrX3SZ5pVXXkr8vHDhA3zxi59n48YNAPz+9//LiScendLmxhuv69IuW1QgioiIiMg+reHDet6+403+78qXePuON2n4sH5I4vB6vUSjUSKRCIFAB1dccQmRSIR77vkpjzzycw466GCuvPLSRJESCgUZP35/5s+/myeffJrzz7+Yxx5byO9//78Z5//EE8v47W9f5JZb7ugxjldeeYl//WttyrCKikquvPIqnnhiGY8++nNqamq5/vprE+Pb2lqZOXMW9933EE88sYSjjjqG73//KjZs+Hi3+mTz5k955plleL3elOHz599IW1sr99//UJ/WqdN3vnMtv/3tizz88GMYhsFPfnL7gOJqatrBa6/9maqqah566DHmzbuVd95ZwW233ZyY5plnfsETTyziggsuZvHipznllNO4/fabeeut11Pm9dhjjzB16uf42c+W8LWvnc29997dZZpOTzyxiOeee5Z77nmA/fYbk3GaVave5803XxvQevWFCkQRERER2Wc1fFjPew+sINDUga+2kEBTB+89sGKPF4nr16/j2Wd/ycEHT6agwMcrr7xMMBjk+ut/jGkeyJgxY7nssiupqKjipZd+D0B5eQXnnvsNTPNARo4cxcknz+bUU0/jD394MWXenW/WrKiooLy8gqKi4m7jCAQ6ePDB+zn33PNTho8cOYpp045k1KjRjBo1murqGqLRXd86PPzwqXzhC6cwYcL+jBkzlgsvvIQJEw7g1Vdf2a1++elP7+Xkk/+TsrLUZ/v++U+LU075IhMnmr2uU7LCwkLKyysYOXI0hYWFFBYO7PbQaDRKTU0t11xzHRMm7M+0aUfwne98n1dffYVNmz4BYMmSn3PGGWfxhS+cwn77jeGss85lxoyTulxFnT79OE4//QzGjBnLV796NscffyLLli3pssxly55k6dIn+clPFjBhwv4Z47Jtm3vvvZtzzjk/4/hs0DOIIiIiIrLPWvf8GjwlXryleQCJ/657fg0Vh1QO6rLfeedtZs06lkgkSigU5LOfncZ3v/sDAFav/gf19duYPfuElDbBYJBPPtkIxIqUp576OX/848vU128lGAwSDoeprq5NadPaGrvtMS8vv9eYlixZTFVVFSeddDIPP7ygy3xOO+0/CQaDVFZWceutdybG7dixg0WLHmblyr+xfXsjkUiEYDCIaR7U737p9O67K1mxYjlLl/6a5cvfShlXWzuKv/zlz5x44iwKCnx9nuett97I7bffTCAQoLZ2JHfffX/K+M5tYhgGFRWVzJo1m/POuzDjvA455FBcrl3l0pQpnwFixX5paSn19ds47LDPpLT5zGcOZ+HCB7vMJ9mhhx7GU089kTLsd7/7LStWvM3++x/A+PGZi0OAF198nqamJr761bO6bL9sUYEoIiIiIvus1k0t+GoLU4Z5ir20bmoZ9GVPnvwfXHvt9Ym3mLrd7sQ4244yfvz+3Hxz11sgfb5YQbRs2ZM89dTj1NVdxcSJkygo8PHMM8t4/fW/pEzf0FBPUVFxl9s009XXb2PJksXcd9+DGccXFBTws58tobW1hV//+hnuuecO7r9/IS6Xi1tvnce2bdu47LIrGD16NC6XhxtvvI5wONS/Tkmsv819993N179+IWVlZV3GX3vt9cyffxOf//wJ5OXlEYlEM8ylq0svrePoo6fj9/t5+uklfO973+Lxx5fi8XiAXdskGo3w0Ucfcscdt1BRUcnIkaNS5lNcXDKg9RqoVave5447/h/33HMnjz22kEsuubzLNH6/n4cf/ilXXnl1Yn0Gg24xFREREZF9VuGoIoI7AynDgjsDFI4a/DdTer15jB69H7W1I1OKQwDTPIhPP91EUVERo0fvl/Kv83bLd9/9O0cdNZ05c05l0qQDGT16PzZu3NhlOR99tIpJk8xe43noofs5/vgZHHTQIRnHG4bB6NH7ceCBB/PNb36bDz54n7Vr1yRiOf30M5g+/XgOOGAi5eUVfPrpJ/3tkoTnn3+O9vZ2vvKVr2Ucf8ghkznxxJMYO3Y8jz66mO9//7o+zXfEiBGMHr0fEydO4rzzLmTDho9Zt+5fifGd22TMmHHMnj0H0zyINWusLvMZO3YcH374QeL2XYj1AcD48RPw+QqprKzivff+ntLu739fyfjxE1KGffjhByk/r1r1PuPGjU8ZdtllV3L00dP54Q9vYOnSxYmXByV78snHGT16P2bMOKlPfTFQKhBFREREZJ81Yc5Egs0BAk0d2FGbQFMHweYAE+ZMHNK4Tj55NlVVVVxzzVWsWPE2mzd/yqpV77No0cOJomPMmLGsXPk3/v73d9iw4WMWLnyAVaveS8yjtbWVJUsW88c/vsycOV/scXnbtm3lL3/5M5de+s2M499663VWrlzBli2bWbPmn9x3390UFPjYb7/9ErG8/PILrFu3ln/+02LevB+kFE+dIpEwgUAg8S8SCWPbNsFgMGW6J598nMsvv7JL4dzp9df/wi9/uYz58+9i7NhxVFT07Xbg1tZWGhsb2LTpE55+egkej5fa2l235Np2lEAggN/vZ+XKFaxdu4YJEw7oMp/TTpvLjh3bueOOW1i/fh0rVrzNXXfdxowZJzFq1GgAzj776zz99BJeeOF3bNy4gaVLn+TVV1/hnHPO67Iuv/71M2zcuIGnn17Kn//8f5xxxlkp0xQXx56xnDLlcObO/Ro333wDHR0dSf0a4emnl3DFFVf1qR92h24xFREREZF9VsUhlRx22VTWPb+G1k0tFI4q4qCzDx305w974/XmsWDBIzz88E/58Y9/xM6dOykrG8Ghhx7G7NlzAPjGNy5i69YtXHPNVTidTmbOnMUZZ5zFCy/8DoA33vgLL774O773vR9y8smze1xeR0cHF110abeFVmNjI0899QRbt24hLy+fiRNN7r77Pny+2O25P/jBDdx5561cdNF5jBhRxte+dg6hUNcC8dFHH+LRRx/qMvyss07nmWd2vX314IMnc+yxJ2SMZcOGf3PzzTdw/fU/7vZNnt2566753HXXfPLy8hg3bgLz599FSUlpYvzf/racmTOPwel0Ul5ewdy5X+WUU77Eu++uTJlPVVU1d955Lw8+eD8XXHA2BQUFHHfcDOrqvp2Y5vTTv4rf7+fRRx+isbGB2tqRXHPNdRx11PSUeV1wwX/z17++wYIF91BYWMTll3+LY445ttt1uPjiy1i+/E0eeOBerrrqGiBWIJ544iwmTTqwX/0xEI7076js48YB6xsbW4lGh9V696qysoj6+sG/F1+kN8pFySXKR8klwzEft2z5mJqasUMdhqRxuQzC4b49EwixT1nU1V2SUiAOF9OnT2XevFs46aTP7/Fld7f/GIaD8vJCgPHAv7uMH/TIRERERERk2HI4HN3eSiq5RwWiiIiIiIgMmpqaWpYufXaow5A+0jOIIiIiIiIig+D111cMdQj9piuIIiIiIiIiAqhAFBEREZEcNsxeqCiSFbuz36hAFBEREZGc5HS6CIWCvU8oIilCoSBO58CeJlSBKCIiIiI5qbCwlKameoLBgK4kivSBbdsEgwGamuopLCztvUEGekmNiIiIiOSk/HwfAM3NDUQiXT/KLkPDMAyi0b5/B1H2LKfTRVFRWWL/6S8ViCIiIiKSs/LzfQM+0ZXBUVlZRH19y1CHIYNEt5iKiIiIiIgIoAJRRERERERE4lQgioiIiIiICKACUUREREREROJUIIqIiIiIiAigAlFERERERETiVCCKiIiIiIgIoAJRRERERERE4lQgioiIiIiICKACUUREREREROJUIIqIiIiIiAigAlFERERERETiVCCKiIiIiIgIoAJRRERERERE4lQgioiIiIiICKACUUREREREROJcgzFT0zRPBW4BJgEbgPmWZT3WSxsPcBtwDlAEvAF807Ks1UnTOIAfAJcClcB7wLcsy3prMNZDRERERERkOMl6gWia5hHAs8DNwDLgJGChaZrbLcv6TQ9N7wbOAM4DNgE3AX8wTfMgy7Ja49NcBVwLXAS8D3wLeMk0zYMty/ok2+uyJzR8WM+659fQuqmFwlFFTJgzkYpDKvs8fm+Qi+vQXUzZjLXhw3pW/exdtq7cAjZUT61l8jcOA8hqf2Q7h9Y+Z/HBondp39KGu9iDp8RL6ycthFoCOL0u8srzKawtpGJyFSMOLGf76sZu+7FhVT2B5g68JV4qJlf1uOzONuHGDlzleUyYM3G3+qqn9R7I9l/7nMXK+/5Gy4ZmDLdBzZGjmXbVkRnj6a3PO/u4ZcNOcEDRmGLKJpaDbRMNRbvEtOrx99i6YjM4oHhcCfnlBbRvbaN1cwvBliDuAndieDQYSbQHMuZgNva//uZVpukhO/tCd7Ek53JBjY+xJ48n3BZOTJcpf5v+tT1lO487fhyHXj51UI9Z/cnV7va5gS7P8DjBtmnf1kagOYC3JI+KyZX93teSc7T68BpGHzcmJc4RB5bzyWsbU6aZfP4UIJYDm97cSNPa7YT9EVx5TkYcVEHJuNIu+0O6XftSM5FQBMPjJK8kPyXXB3Is6M826ml8pr6ZfP6UAW+z5ONjb8fS7o5jieN7kYeisSV4Cz0DzqW1z1m8c9/btG7YieF2UnvkKKamHRcH8nu1y3FvfCn5ZXm95kP6MhtWbRtwXvdmbzq32VvaD/Z8sz2f3vIrF3Okrxy2bWd1hqZpLgOqLcuakTRsMTDBsqxjumlTDNQDF1uW9UR8WBGwFbjSsqxH4lcPNwEPW5Z1Y3waB/Bv4EnLsn7Yh/DGAesbG1uJRrO73gPR8GE97z2wAk+JF0+xl+DOAMHmAIddNjXxi6Wn8dlUWVlEfX1LVucJva/jUOguprGfn8DHL63LSqwNH9az/NbXaF7fjLvQA9iEWkPkVxbgKXRTPLY0K/2R7Rxa+5zFX3/8Gm6fB5wOdq5vItQewpnnwuk2CLWHcHld5Ff6KB5XzM51TVR9diQl40q69GMkYrP9H/U4HA5s22bEgeU4Xc6My06Os6y2mB2bd7JzQzPY9oD6qqf1Bvq9/Zv+tZ03rvtToi/saJRIIEr5IRUcN39ml18IPfV5Zx/jNAg0d0DUJhqM4vK58BbnUXvUKFxeVyIm6+mPaF7XhLvQTag9TNvmVlwFLgyXQagliMPpwFPipaMxVojvN2McLq+TnRuaCezsoKOhIyUHSyaUcsS103f7jxL9yatM0+/O9u1LLGUHl/PR4+/j9nlwF3lo39aGf1s7NUeOouqwaprX72Dbyi0p+Vu/ahtNa3cQDUYS2zkajDLi4K7bOVv6k6vN/25m2zufUnV4DSXjywbUZ8nLCwcibH7rEyKBMA6nA3eBJ7avHlSJ0+no8762fP7riRwFBx2NfiKhMCOPGk3J+DKa1+9g81ubMNwGeeX5gINQa5C8iny8RR4i4Shb395MOBQBBzg9TiL+MHnl+Yw7eUJif0iPZ9e+5MDf0E64IwKAr7YQl9tJyYRSzDMO7na/Tu/f7vpzoMfZ9P23c71LxpdwxA+OHdA26zw+drfde4q16V/bk47v0PLxTmwbRh03hqLawn7n0trnLF6/7lXC/gjOPCd2OEokGKFichXH3nrigM9j0nMq3BGhdVNL/Pg2ttt8SO+DSDjC9tWNu34H9SOve7M3ndv0Nab+tk8/bxysPsnWfLM9n97yKxdzJJlhOCgvLwQYT6yWSh0/CMs8Gng5bdiLwDTTNN3dtJkKeJLbWZbVQuw2086icjxQmzaNHf85Y+GZ69Y9vwZPiRdvaR4Ow4G3NA9PiZd1z6/p0/i9QS6uQ3cxfbDo3azFuu75Nfgb/LiLPLjyXbjy3XiKPDSvb8Lf4M9af2Q7hz5Y9C5uX+yqYWBHB7G/HzmIhiLYUTBcTmwbwv4QTWt24PZ5aNvckrEf2za34Mpz4Snx4sxz0balrdtlZ4qzvb59wH3V03oPZPt/sOhdIuEorgI3To8TV54bp9dJ8/rmLvH01uedfRwNRnC6nbH/j9qE2sK4izw0rd2RElN7fTueIg+ufDfh9hBOr5NQa4hAUyAWj9dFR2MHTq+TaChK09rtif7buX5nlxxsr2/f7f2vv3mV7e3bl1hWLXovkcsOw0E0FMXwGOywGnEYDtq2tHXJ3+b1zYTaQinb2ZXnyrids6U/udq2uSUW85a2AfdZ8jyb1m7HU+QhEooS8kfwlHhx5blo29zSr30tOUdd+S4ioQh22E7E2baljUg4SiQUTUzjLvLQvL4Zf4OfpjU7iEZtnG4nTpeTaCCC4TQItYZS9of0eHbtS1HsCDjdTgynQWB7RyLXe9qv+5rHAz3Opu+/nevtb/APeJv1tt17mjb1+N55/HDS8N7WAeXSB4vexQ7buPJdON3O2Dp6XTSvb9qt85j0nAq1BnF6nURC0R7zIb0P2ra04Yz/DupvXvdmbzq36WtMQ91+sOeb7fn0ll+5mCP9MRjPINYQu/KXbAvgBiqAzd20sYFtGdrVJk1DN/M+rj8BxivmIRdu7KBsdDEOw5EYVpDvYecnO6msLOp1fLYNxjz39DrsTkyb/rSBstrsxBpu7MARhbwSLw5HbH6220nz+iYcUfD5vLu9jJ7WZaA5FKj346v1YRgGdigK2DgMsCM2RG2cbiPx/+H2EGXjSwm2BBPrk9yP21ZsJq84tv5ut5PAzgBltcUZl50ep8/nxRG1ceAYUF/1tN5Av7d/oN6PYThweZyJ7enIdxFqDRFu7EiJp7c+7+zjjkY/Lm/n/GyIQH6Rl8DOAD6fNxGTy+ciryQv9lfKUBR3votQaxBscHmcAASbArhLvUSDUaL+cKL/osEI+UWpOdjR3NEl5v7qb15lmn53tm9fYrGaA5SNL8UwYn8HtUNRPIUeQq0hfD4vUX+YgvL8lPy1gxHsaDR1O3vJuJ2zpT+5minm/vZZ8vKi/jB5xd7YraYOBx6PC7uXfTXT/BxRO5GjsUBtHE5HIhej/jCG4QDbxuOJnXa43U5ags04ohBuD+FwgOGMtY902DgLYn/w6JxHpvVM3pccgMPpwDZi2zqvyEtHcweBen+3+3V6/3bXnwM9zqbvv53rHWgO9CufMh0fu9vuPcWafnx3eZ3YboNQW6jHeXYnUO/H4Uw9LhoFbgI7d63fQM4B0nOq87iXfHzry/Fm24rN5Md/B/U3r3uzN53b9DWmgbTvz34yUNmab7bn01t+5WKO9MegvKQm1+XKLaau8jx2bN6JtzQvMSzQ1IG3PI/6+pZex2fTYN1iuifXYbdjqszPWqyu8jxsA/wtAVx5sd0s0hGO3bZmQFtbYLeX0eO6DDCHvJX5tDf6Y1dd3AbgwI7GTrwwHERC0djBznDgKnAnpu1cn+R+NPJddLQEcOa5CHeEceW7YrFkWHZynD5fbH52/KA6kL7qab2Bfm9/b2U+7U1+wsEIhitWcIQ7wjg8Tlxp8fTW55197HAbSfNzxG6TawngynfR1hZIxBTuCCf60eE2CPnDOJzxGIKxW+o6hxtOAyPe3jYcGB5nlxy0DUeXmPurv3mVafrd2b59icVT4k3kJ8T6KNgajF2Bawtg5Lu65K/D48TRYaRs52gwknE7Z0t/cjVTzP3ts+Tlde6jGA6iNgSD4dhxqod9NdP8bMORyNFYoA7scDSRi0a+i2jUxnAZBINhYNf+YxvgKnAT8oeJRmK/lx1OB9FgFIdrVz5nWs/kfckm9ocsO2pjuA06WmL7QE/7dXr/dtefAz7Opu2/ifU26Fc+ZTo+drfde4o1/fgeDkaIxu+M6Gme3fFW5uNPOy52/p5z9dY3PSwnPaeSj3s95UN6Hxj5rsTxr7953Zu96tymjzH1t336eeNg9Um25pvt+fSWX7mYI8mSbjHNPH4QlrkFqE4bVg2EgYYe2jiAqgztNidN0zmsu2n2KhPmTCTYHDvQ2VGbQFMHweZA4uUNvY3fG+TiOnQX06EXTslarBPmTCS/Ip9QS5CwP0zYHyLYEqRkfCn5FflZ649s59ChF04h1BYk2BzAW5ZH7I/CNobbicOAaDiCwwGufDelE8sItQXx1RZl7EdfbRHhjjDB5gCRjjC+Gl+3y84UZ0FlwYD7qqf1Hsj2P/TCKThdBuH2EJFghHBHiEggQsn4ki7x9NbnnX1seJxEQpHY/xsO3D4XoZYgpQeUpcRUUFlAsCVI2B+79TESiOAudOMt9cbiCYTJK88jEohguA1KDxiR6L/i8cVdcrCgsmC397/+5lW2t29fYpl84WGJXO4sGqLBKGVmOXbUxlfj65K/JeNLcPvcKds53BHOuJ2zpT+56qstisVc4xtwnyXPs/SAEQRbgjjdBu58J8HmAOGOML7aon7ta8k5GvaHcbqdOFyORJy+Gh9Ol4HTbSSmCbXEnsXLr8indGIZhuEgEooQCUcwvE6ikSjuQnfK/pAez659ycDhhEgoQjQSxTsiL5HrPe3Xfc3jgR5n0/ffzvXOr8gf8Dbrbbv3NG3q8b3z+BGh4rDqAeXSoRdOweFyEPaHiYQisXUMhCkZX7pb5zHpOeUu9BAJRHC6jR7zIb0PfDU+IvHfQf3N697sTec2fY1pqNsP9nyzPZ/e8isXc6Q/nPPmzcvqDBcsWHAE8B91dXVPJA37LuC3LOuRbto0AlcDH9bV1b0HYJpmIXAf8EhdXd3KBQsWNAEXAzvq6ur+HJ/GAdwPPF9XV/d/fQivFPiW3x8ky+/mGZCCKh/F40pp2biTtk9bKajycdDZhyYeXu1tfDb5fF7a24NZn++eXIfdjWn0sWOyFmtBlY8RB1bgb2yndVMLdgRqjxzFtKuPYvSxYxJ91kgAAAoaSURBVLPWH9nOoRFmBQU1Pho/aiCwvYOC2kJKJ42AKNiRKN6SPApHFVEyroSqKTVMPP1A7IidsR/9De1EglEAfNWx/uhu2clx+re24RmRx6EXfGbAfdXTeg9k+48wKygcVcR2azuB7X4Mp8HI6WM4+kfHdYmntz7v7OMd/9xOuC2M0+OkzBxB7RGjKdqvGDtsp8Q0wqzAv91P6yctGC6DyinVVP5HNR6fB4fLwGEYeHweKqdUU3FIFXYoSkGVj0Mv+AxjT5rQJQcPr/vcbu9//c2rTNPvzvbtSyz7z5mUyGX/1nZ8NYUcdO5kCsoLaPu0leKxpV3yd8r/TKX6szUp23nCzAl8bjdf6jOQ+DPlavHYkljMUQbcZ8nzDDYHKB5fSsm4Ulz5sVcE+KoLGWGW92tfS85R27YZefRoJp8/JRFn8djYy2IcTiMxTe3nRjLtO0cz+tixhFpDsQKjPQS2A3eei6rDa6iZOjJlf0iPJ3lfinSEcXoMPMVe8orzErne037d1zwe6HE2ff/tXO/DrzhiwNus8/jYl2NppuPYruN7gIJqH5VTqskr8g4olzqPi41W7PeF4XIyavoYjrru2N06j0nPKYfTQeVnaqg4uKLHfEjvA3+Dn0j8Lov+5nVv9qZzm77G1N/26eeNg9Un2ZpvtufTW37lYo4kczgcFBR4AO4FmrqMH4S3mB5B7OUyNwG/AGYSK/Tmdn7mwjTN04D5wEzLsjbFh90PzAXOBz4FbiT28prEZy5M07wamEfsMxcfAFcAZwGHWJa1sQ/hjSOH3mKaSwbrFlOR/lIuSi5RPkouUT5KrlAu7t16e4tp1p9BtCxruWmapwO3ANcBG4FL0r6BWAKYxF5c0+lqYrehLgYKgTeBWUnfQAT4CZBH7JuJlcB7wOw+FociIiIiIiLSg6xfQcxx49AVxIz0lyDJFcpFySXKR8klykfJFcrFvdtQfAdRRERERERE9kIqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBFCBKCIiIiIiInEqEEVERERERAQA12DM1DTNU4FbgEnABmC+ZVmP9dLGA9wGnAMUAW8A37Qsa3XSNH8Cjk9r+oZlWdOzF72IiIiIiMjwlPUriKZpHgE8C/wKOAy4F1homuaXeml6N3A2cB5wBNAK/ME0zcK06X4O1Cb9OzV70YuIiIiIiAxfg3EF8dvAa5ZlzYv/vDpeNH4X+E2mBqZpFgMXAxdblvVCfNi5wFbgTOCRpMn9lmVtGYS4RUREREREhrXBeAbxaODltGEvAtNM03R302Yq4EluZ1lWC7HbTI9Jm3auaZr1pmmuNk3zQdM0K7MUt4iIiIiIyLA2GFcQa4hd+Uu2BXADFcDmbtrYwLYM7WqTfl4CbIz/m0TsOcc/mqY51bKsYF8DLC9Pv2tVACori4Y6BBFAuSi5RfkouUT5KLlCubjv6lOBaJrmPOCGXiarsyxrwW5H1APLshYm/bjKNM13gHXAHODXfZ1PY2Mr0aid7fD2apWVRdTXtwx1GCLKRckpykfJJcpHyRXKxb2bYTh6vGDW1yuIC4BlvUyzJem/1WnjqoEw0NBDWwdQReoVxmrg0+4WaFnWx6ZpfkrsaqKIiIiIiIjshj4ViJZlNdB9cZfuTWAWMD9p2GzgbcuyQt20WQEE4+1+DhB/e+kxwLe6W5Bpmp1vMs1026qIiIiIiIj0w2A8g3gP8IZpmtcDvwBmAmcBczsnME3zNGIF5EzLsjZZlrXTNM2FwO2maW4jdtXwRmA7sDTeZn9i30j8PbFnFScR+27iJ8Q+q9EXTohdVpWu1C+SK5SLkkuUj5JLlI+SK5SLe6+kbefMND7rBaJlWctN0zyd2AtkriP2QplLLMtK/sRFCWASe3FNp6uJ3Ya6GCgkfiXSsqzW+PggMAOoA4qIFZGvADckTdObWoCyMt8A1mzfp5f3SK5QLkouUT5KLlE+Sq5QLu4TaoF/pQ902PawelmLF5hG7JbUyBDHIiIiIiIisqc5iRWHfwMC6SOHW4EoIiIiIiIi3TCGOgARERERERHJDSoQRUREREREBFCBKCIiIiIiInEqEEVERERERARQgSgiIiIiIiJxKhBFREREREQEUIEoIiIiIiIicSoQRUREREREBADXUAcgucU0ze8BpwMHAhFgJfAjy7LeGtLAZNgxTfM44LvAZ4Fa4EzLspYNbVQyHJimeSpwCzAJ2ADMtyzrsaGNSoYrHQslV+gccfjQFURJdwKwEDgWOIbYydEfTNM8YCiDkmHJB7wHXD7UgcjwYZrmEcCzwK+Aw4B7gYWmaX5pSAOT4UzHQskVJ6BzxGHBYdv2UMcgOcw0TQPYBtxoWdb9Qx2PDE+madror+ayB5imuQyotixrRtKwxcAEy7KOGbrIRHQslNyic8R9l64gSm/yAC+wY6gDERHZA44GXk4b9iIwzTRN9xDEIyKSq3SOuI9SgSi9uRNoAn471IGIiOwBNcDWtGFbADdQsefDERHJWTpH3EfpJTXDgGma84AbepmszrKsBWntrgPOAk6yLKtlkMKTYWSguSgiIiK5Q+eI+zYViMPDAqC35xW2JP9gmuaNwBXALMuy3hmswGTY6XcuiuxhW4DqtGHVQBho2PPhiIjkFp0j7vtUIA4DlmU10I8TG9M0bwf+m9hfhbTjS9b0NxdFhsCbwCxgftKw2cDblmWFhiYkEZHcoHPE4UEFoqQwTfN+4HxgLrDJNM2a+Ci/ZVnNQxeZDDemaRYCya/OHmua5hSg1bKstUMUluz77gHeME3zeuAXwExit1HNHdKoZNjSsVByhc4Rhw+9pEbSfZPYN5deADYn/bt3KIOSYWkq8Pf4P4Db4v//6JBFJPs8y7KWE/sQ9BnAB8DVwCWWZf1mSAOT4UzHQskVOkccJvQdRBEREREREQF0BVFERERERETiVCCKiIiIiIgIoAJRRERERERE4lQgioiIiIiICKACUUREREREROJUIIqIiIiIiAigAlFERERERETiVCCKiIiIiIgIoAJRRERERERE4v4/DIbm5j5598MAAAAASUVORK5CYII=\n",
"text/plain": [
"