main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Простой пример модели детектирования акустических событий"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Импортируем необходимые библиотеки"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from librosa.feature import melspectrogram\n",
    "from librosa.core import load as load_wav\n",
    "from librosa.display import specshow\n",
    "from librosa import power_to_db\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from utils.preprocessing import extract_log_mel_feats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Скачивание данных и подготовка к обучению"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Скачать архив с данными можно, перейдя по [ссылке](https://www.kaggle.com/c/freesound-audio-tagging/data). Для скачивания потребуется авторизация на [kaggle.com](https://www.kaggle.com/). \n",
    "\n",
    "Затем необходимо положить данные для трейна и теста в папки data/audio_train и data/audio_test соответственно, так же в корень папки необходимо положить файл train.csv, который также можно найти в архиве. \n",
    "\n",
    "Далее задаются различные параметры для извлечения фичей из файлов - частота дискретизации, с которой мы будем работать, размер окна и перекрытия для быстрого преобразования Фурье и число мел-коэффициентов. \n",
    "\n",
    "При работе со звуком в основном принято использовать не \"сырые\" данные - значения амплитуд в каждый момент времени, а извлекать из файлов какие-нибудь признаки (хотя, конечно, существуют успешные примеры работы и на прямую с сырыми данными, например, [WaveNet](https://arxiv.org/pdf/1609.03499.pdf)). Обычно для этого требуется переход в спектральную область, который осуществляется при помощи [преобразования Фурье](https://betterexplained.com/articles/an-interactive-guide-to-the-fourier-transform/). Затем можно извлекать различные признаки - спектрограммы, мелспектрограммы, mfcc-коэффициенты. Мы будем работать с мелспектрограммами, о чем подробнее чуть ниже.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_folder = './data/audio_train/'\n",
    "test_folder = './data/audio_test/'\n",
    "\n",
    "\n",
    "sample_rate = 11025\n",
    "fft_size = 1024\n",
    "overlap = 4\n",
    "hop = fft_size // overlap\n",
    "mels = 64"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Визуализация фичей, используемых при обучении, для одного файла\n",
    "\n",
    "Для начала посмотрим на признаки звуков, которые будут использоваться для классификации. \n",
    "\n",
    "Для извлечения фичей будем использовать популярную библиотеку [librosa](https://librosa.github.io/librosa/). Она позволяет производить множество операции с аудиофайлами - от загрузки и ресемплинга до добавления разных звуковых эффектов, например, [pitch shift](https://en.wikipedia.org/wiki/Pitch_shift) и [time stretch](https://en.wikipedia.org/wiki/Audio_time_stretching_and_pitch_scaling), которые кстати [можно использовать для аугментации](https://arxiv.org/pdf/1608.04363.pdf).\n",
    "\n",
    "Мы будем пользоваться мелспектрограммами, а точнее логарифмом мелспектрограмм. Подробнее о том, что такое мелспектрограммы и мел-фильтры можно почитать, например, [в этом посте на хабре](https://habr.com/ru/post/226143/). Кратко - [википедия гласит](https://en.wikipedia.org/wiki/Mel_scale), что мел-шкала - шкала, основанная на восприятии звука человеком. Дело в том, что высота звука, воспринимаямая человеком, нелинейно зависит от частоты и считается, с помощью такой шкалы удается поточнее отобразить эту зависимость. \n",
    "\n",
    "Чтобы понять, зачем брать логарифм, нужно увидеть изображение спектрограммы до и после."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApoAAAEYCAYAAAD8nFaxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsvXn0bclV3/fZVXXOucNvemPP3WqphSaEIIRhxcbIsb0YDQQCmFkBEogXAYxlwMR2MDHBBlawY8MCDGYQYDAiFhg7IQyWQxaYeZBAiEFqqec3/aY7nKlq549d9/5u//Re91NLr1t0n+9bv/XuvadOnTq7dtWpU3vv7xZVZcCAAQMGDBgwYMCA9zfcc92AAQMGDBgwYMCAAc9PDAvNAQMGDBgwYMCAAbcEw0JzwIABAwYMGDBgwC3BsNAcMGDAgAEDBgwYcEswLDQHDBgwYMCAAQMG3BIMC80BAwYMGDBgwIABtwTDQnPAgAEDBgwYMGDALcGw0BzwgoWIvE5EoojM8l8tIv/fc92um4WIPCgif/W5bseAAQMGDBhwIwwLzQEvdPyqqm6p6hbw5c91YwYMGDBgwIDnE4aF5oAXMgog3uigiLxCRN4sIgci8gci8ikbx35QRNqN3dCZiOyKyBkR+VkRuSwi+/nz3RvnvVlEvkVEfl1EDkXkp0XkbD72IhFREQn5+0fm7/8ofxcR+Q4RuSIiPwgI8Kki8pCI/J6IvHLjOioiv7fx3YvIoyLy8FPc79OeIyJ3ishP5ft7p4h85ak6Tu8Sq4g8kI+dFZEfyHXui8ib8u+vFZF0SpZRRF63Ue8Xi8jb8nk/JyL3nWr36hqfkuXxYhH57FP11avvuew3isgbReQnRORYRH5bRF7zPvb/a0XkYRH5htxPD4rI522c90ki8jsicpTb+Y0bx1b9/9Mbv505vdN+k/103d1uEfmRzWte5/h/n+V8LCJ/KCL/xen6sl4+JiIfeZNy3Dx3S0SeuM79zDfk2Gb9vt6Y+Ju5L87l728WkS/Nn52IvOWpdHzAgAHPPoaF5oAXMsZAc70DIlIA/w74f4CLwP8E/KiIvGyj2LeudkPz3yE2pn4AuA+4F1gC/+JU9V8IfDFwJ9AD/8cN2vetwCMb3z8D+ETglcBP5vPPAS8DfhR4w6nzSxH5iPz5k4CDG1znps4REYfJ5PeAu4C/Any1iHzcxvkO+JWNXeJNvAGYAK/CZPodG8ce3ZQl8Ksb1/004BuATwcuAL8M/OvTDReRjwW+G/hEVX2Hqv7ERn2/DHzFddr1qZgszwI/BrxJRIr3of8BbgfOZxl9EfC9G+fNsf7fy/L9H/P9beLFInJH/vwFwDtO3yvPrG+fEiLymcA35vbtAJ8CXD1V5hXAvwU+X1V/fePQdeV4ncv8HaC7zu+v2eibb71B+/4G8Hrg41T16nWKfBFw5oY3OGDAgOcEw0JzwAsZZ4H9Gxz7aGAL+Meq2qrqLwE/C3zOU1WoqldV9adUdaGqx8A3Ax97qtgbVPWtqjoH/j7wWSLiNwuIyCdj4/MXNn7+68CPquolVf332CL0X6nqAlu0vVpEXrRR/vuBL82fvzR/fzo81TkfAVxQ1W/KMnkH8C+Bv7FRpgTa05XmhdMnAF+uqvuq2qnqf7qJ9gB8GfAtqvo2Ve2B/w340M1dTeDDgJ8BPk9V33KT9QL8lqq+UVU74H8HRljfP6P+38DfV9Um3+O/Bz4LQFXfrKpvUdWkqr+PLZhP68cPAa/Ln78wfz+NZ9K3T4cvxRbPv6GGP1XVd20cvw9beP89Vf3FU+feSI5riMhtwJfk4+8tPh67x09Q1ffYsRSRETaW/tdnUPeAAQNuIYaF5oAXMu4H3n2DY3cCD6lq2vjtXdgu1Q0hIhMR+R4ReZeIHAH/L7B3aiH50Kk6C2wHbAUHfAvwtaeqvw24fL3r5gf8AbabtsLPAq/NZuU7gN96qrbfxDn3AXdmU/KBiBxgO423bZS50eL9HuCaqt5oYf9UuA/4ZxvXvIa5DWz2xfcBfwL8tfey7nVf5L5+GOv7Z9T/Gfv5JWLzvDsBROSjROQ/irkeHGJ+wedPnf8G4PNE5KNy+564zjVupm/flGX2kIh8q4jI07T7HuDPnuL4P8fkcz0Z30iOm/jGXMe1p2nH9fB9wIO856J8ha8Cfg54+zOoe8CAAbcQw0JzwAsS2Qz8XwG/eYMijwL35HIr3MuTTdnXw9/GTNkfpao7wF9aXXKjzD2n6uyAKxu/vQ54u6r+51N1X+Y9FyVWuZkp93jyoqTHzJxvBH7wadp9M+c8BLxTVfc2/rZV9RM3ynwQ8MfXqfch4KyI7N1kO06f+2WnrjtW1V/ZKPPVwCcDX7LyK7xJrPsi9/XdWN8/0/4HOCMi01PnPZo//xi283qPqu5ipv7TC8CrwFuB78EWWNfDzfTtp6nqHqbnnw983A3KrfAQ8JKnOP5tmLvER8qGv2rGjeS4wgfl69/ITeTp8DnAZwPfLCL3nDp2FvgK4B8+w7oHDBhwCzEsNAe84JADCf4FMMXMmtfDr2H+dF+bffZei5muf/xpqt/G/DIPxIJ8/pfrlPl8EXmliEyAbwLeqKqbQUn/M/B3r3PefwA+V0QuiMgnYbtr/12u56uw3ZwHT53zvcDbMB/Om8WNzvl14EhEvk5ExjkI5YNXvoIi8heATwN++tR5qOpjwP8FfFcOcClE5C+dLncDfDfwd0XkVfk6u9mfcBO/rKqPYz58P3AD/8Dr4cNF5NNzsMlXYz67/5ln3v8r/EMRKUXkY7AF8E/m37exnd06B9N87g3O/w7gd4H/+ymucbN9e4wtTJ9uvv8+4PUi8uFieOCUe8IvZzeNL8H6cfOl4UZyXOHvAd+kqsunacON8Muq+lZsofo9p459NfD9uf8HDBjwAYZhoTnghYgvx3Zg/rKqzq5XQFVbLBjiE7Ddxu8CvlBV/+hp6v6nWJDRFexBe72FwhuwXajHMV+2rzx1/GdV9U+uc96/AX4J+CPgM7Edo6vYDuLrsAANPXUf71DVz1HVmw4WudE5eTH814EPBd6Z7/H7gF2xiPcfAl6vqr92g6q/ANu9/SPgErZAuJn2/FvgnwA/nt0R3or1y/XKvgHbmfuGm6kbWxR/Nmbu/wLg07P/6DPtf7B+3cf650cxv9TVeX8T+CYROQb+Adan17uPX1PV1516ATld5un69l/nCOy3Am/iqRetqOpPYj7FP4YtTt+E7RaeLvef8rHNYK7rynHj+FXgh5/q+jeJbwHuEJEv2vjNA9/+fqh7wIABtwBy6rk0YMCAWwgReTPwI6p6I5Poe1PXg8CXquovPF3ZAe8JMZqfB1T189+Pdb4W69+7n67s8wW3Qo4DBgx4/mDY0RwwYMCAAQMGDBhwSzAsNAcMGDBgwIABAwbcEgym8wEDBgwYMGDAgOcxcnDqTwAvwoJGP+u9oZsTy9T2bTyZeePLnsInf41hR3PAgAEDBgwYMOB5ALE0uD94nUNfD/yiqr4U+MX8/b3FT6jqh278Pe0iEyA8gwt9wKMMEw3nzpEK0BVD3Q2oil0LONDVklvtsyhID5ppttfHnZWRlI85BacI4JztDjtRFIjJrb8DxCjWIAF6Ac3t8hu7yrJqhCCtIAlIuY0+f8au7zorL5qrWh1TJRVCCpCK3NagJ/ULSCuQ8rnupM51m/I9awBEkV7s+vm4xCfLav17Z6fH6kSea9lvfha7F4mg5UbbFEhZNiu5qCAhoQoioElwPiECsXdWVgWc4nzK8hdSEjQJREF6q3+zva63vxTs9xSsPZKyLMSO68Z50ptMkgMXc3lVNHNhi+bOyPfaj01uq3pdt5K1lROFVAhxlH+PkPxGv57SWw0bcvMberjWndVnhSgnMl39Dyevl/keCYrIiQ7q6rzcZy4kFEF7+915U7SkYn2VNpQwCRKzbJLJlJWq60penOj1Sh+jmqzk5DcALRRxiqrkcaCsaMdXfYuaXE/0dUPZ1M5Z65RkmW3qodOsj7KW64kwWPcTZD1Yne/UZNmfCH6znPVzHvIb48wEsHENb/OH9s7GVb5vwHR3gzLeNydtf5J887zk2zwUTul78vbZxY3janJfy3ulj0Hox/Y9LE1XJW2MnZTns7Gz65P1tjyp4z30dmMMrOaPJ7V9VT4BhZ7I1GU95mSsEvLY8Wo6sBY+iGRdiSZ0yXPIWldWdbrcD92JzFcqvGrfZh+ujq302vVK8mLPmJXe5farB4rcacnGjOkwJ+NFT+YWJJ+z6oN0IpMn6eNqmlzpyOltIn8yBu0+n9QDVqHTk69rRdKTtq7miiLPoyqoisk11+RESat517Meg2Dz4+n2i57Me5qfY5qPuZVur+bMjblJVImVrOeOYh7X86zJIYET1DuSF1yXUH8y50nUJ8/NfcxtFlLwzI4fuaKqF3gO8XF/eapXr92QWOJJ+K3fb/4AqDd++l5V/d734nKfCrw2f/4h4M3A150uJCL/LUad9m4sAci3q+r7xOrwvFxojotd7vvcr2Fxp+YHQlbu1SDbwPTdjjiCbhsQxXVCP1Z8I1RXoduxc+PI6kqjhOsE1wj9ToJxxFc9PiTGVYsIjIuOmByHixECjKoOVZjNxsTOIaLI1QrpII0U3erXiypx9mBNnaN8d0WowS8gTqCbgq8BgbCA6WOK+jxhOQiLhCTwnTK/zbO4TagvKn4htLd19oAOCRHFP1YRFoLrbFFo5+eFVGH/91NoLkY0JMrLBd2ZaA8AUYojj0SIlclMki1qx4/b5HF8f76fzuoDnrTQTFXC1Y7yQFje09liskj2sJ95XCfE7WgTUiuE8zWaHOISfR3Y3lsiohwdTKB30DiY9Ex2aqZVSxs9i2VFtyyQw0B1xaEB+qniWrvv0RUYX0kszzu6KTTnlGImFEdQLEwu4ytK8tDuCGGpVAeJ+qyn3YLRgX2XpKi3RZJvFYmK65QUhCuvLui3IMygOoLx5YgkmzRtgZRYXizZf5lN3OUxNLv24Fr1zeZDuD0b8bWjOBbaHaU4FrpdRZ3ay8TqwVUk/FFAg70kbD6w0zhBAl87UqlwpqWo+vxQge6osr5wijaeybkFSYXm6hhGkelOjSo0TUFsPczD+iHnFp7yUKj2TZ/qc3khH8AvoZjZYqicJXxjMlIP5XHi8quDPbiD3bck6O5oKSYdfeNJrcePekIZSUno64DbL0Fh8qjQT6A9o6Qy5fbYy5qOEm7pIAoaFL9wqFc0YP+Pki0Mlp7i0NPtxZMFZhLTl9b6N04SYWZyi9sRRhF/pVz3j+sF9Uocmf77hSNViq+FOEn2ElEkexDnBXDYbhFR+msjigOPeqW/kN/YZoHi2K0XFbt/CrG0B29YKsvzJr9+bL/tvFPpR8L4SkSd0G0JvlGaPUc/Mv1KASaXTW+L44h6IVZWzvWJ+kzBldc4XAvn3xoJ84hvEv3Y45uEr3sArr1qyvKiLRiKGczvNJ11PU96cQJo9xLlgcMvrX+LY6jPQyqVVNq8SgI/d6TbG9gv17rqZx7pYXRVWF5U0pkOOkd5pqZdFuvVofhEKCN949HjAoJSnq1BoV2UcBxwEeI04kbRXqCeGFEc5g2B7mQhFGYQx9b2MDedTR7GV5VQK6MrHe1uYHant7njrOmrb6HdUbjd1gNxGZCQKCcdfeft+9LjG2F0WYglpAq6nYQWiqsdfiH0Wwn1Sph5+u2VPgquF8Kxzbf95OQFXT2kqZXzR95ehqt0Mu+G/LI1sg7R3iFdXpRVCY2CWzpcI6RKcRdryrInJke3DBTj3l7qgarqmF+dIFGQSY/OC1xj46O65ghze3aomF64HrYfMp3vJkJzRlBnOlAdJ9SZ7qFKHDlQzbqoHN0XSB6KJVz4jUNSFbDdBsUfLdGqoNsd0e4Gxk/UtLslGgSJSrnf4trezgHC5WPoI7o1orl9mzf//Ne/6wZLiGcNV69Ffv3n7r2psv6OP6lV9b98Hy53W+YzRlUfE5GLN7oU8CZV/eLMKLGJzxaRv5g//ybwt3Kq5afE83KhaW85IL3Yy03Q/IZqA2u9a6LgOwj16i3MBrBEwddQHtniyd7UhThSikOPa21SHz/mSYUnNCWxsBd9iXBU2uLKJ9uB6PPkNW3t++hawjeJOBJSEGZ3laQCum1btMVK8asFskB9wSbk8lAIC3uotLvQbdtEtboX13t8vhc0L/weFYoFjK4U+AYWt9vvoba6Vw90gG7L/l/vXiTYe4sHPKFWFrd7kod+Ygte14NrhbC0ujRAMbfFVpjJevGqzibT1Q6xOpDe4ZdCeQR1ZxNY+USJ60zuEkFiYLSfaLcccTTFt7A8D6MGXFuRPGw5a4dvIFaeWFbMCyBZXscCW0D2k6wbCYoj69PZvcrxi6yvfWv60m0r7Z69lPiFox8JzRmTv03ogmii3Hd0u8L+y7w9vEpFkuAb2yVf7Yqmwh6ysXS0Z+DgZQ4NilYOWkeYu7WMJNrCDPLi30MqNL/9Cw5byKwWysXM2p63ctaytYWO4Guhnyqjy4Lr7LhrwTce9bD9cKTZcXRbI2J1cs0zDym+s22UfgT9eAeAnZlSHQnHd+2aDmyDy7vccaR5t8sWQs2e6VNxZA+d6Ox7fV7z7p6QSlvISQ9x5HCtLZxdB3t/bOXad5W0WyX9vYli6ZBUkPKO3GRufQuwuDPvXDmlOPB5J24lS1uomfXC6k+FZFkJfS9ocLjO5KmHfr0zaLtXtvBIFfStw9f24uGWntQ5qmv5wTmzHUCSoEFodvMOoxO6beiCMn3IUxw7RvuJfiSkUuimARWYPqH4OjG/09HMSuLY2pAKezkKC9P/VK3GkOCiyVwDxEI5eMB+O3zAbewi2q6dr2F+X0JaYXaPLZYl2eKimOe5LuSXWYVuR7n04Z5UOGtHZVtYYVZZfe3Ji1Bz1l5oRE/G+OaLpfSy3tWqrkF5vHrJEGIF/dT0rTiG5TmPS2LjKHrGj4u9nBzai2A/Lm1eHW1ROJsLXQeTx/NCtzNdSEHotrZAYHdfKWfK8pyjPudozgZrX2H3EOamH81ZG+sqkl94rJ39ToIozO9Z7V4WuNbGu0S7p9DY/RWHQr8/pp9AyDv3qahwAqGzOkVtLKx2Av3cMX3EdE0ipMLRT6xPENNNl58fK8uSupVw804vJsOwtN9jk7dCnZplK0J0ijSOYu4YXc6L48bbXF2bJQwVundOSBXUdya0VEa/PbKd8M6eSxeOFddDNw12nhfabWjOQX1B1zrgekE7uPJqZxuqrf11uzZHHAeXd22FfppwbX52LvxaX3xt8/WDn7K73iW1/6eEOez9WU+74+jHY1I4sQAsbgvZGiX4TpEXT0jFycbMBwJsykxPW+5mICK/BlTAFpaF7Xfzoa9T1Z97L6ra4sZpYn9CVb8ip7P9Toyv+HrJRZ6E5+dCc8CAAQMGDBgw4AMYitLdOCfDe1eX6kcBKy7f16nq604VeUJE7si7mXdgSTOuh/uBh5/mWioi/w5L/fq0+ABZ1w8YMGDAgAEDBrywkG7y3/sBPwN8Uf78RVwnVbCIlFj2txulZt7EXwTecTMXHnY0BwwYMGDAgAEDnmUoSnz2KCb/MfBvRORLsECfz7xOmR8BXgX8pFnHuR2IIvJj+fjKR9NjKXa/6Dp1vAeelwvNFBzlIZx9W2dBGs78NIpZtKjgHMWdguPaKwqaPZg8Zv4qxSKZn0d2tj/z9oYw7yAlDl65i4sJ11oUZnncs7hY0I+FMkF1mOjH5g/Wj4T5nVDftgoKEIqZ+a9NLuWoxSj4XhldFbopgPnhhCDZB8t8Hs3Xz471E/NR8U2O8NuISlz5k0nP2helOgSE7LuppCofO5KTIBPM0b1oTqIBV3Wm0nyjxtcSoXbsv8x8YM7+oQUTxMoq8Y2e+AwB2w+az874qjl1I+agHZZp/bk46gizFkm7tDtQHkA5U0bXIu2OZ3FRmN/u2H4oUSwTfSWEpWN250kkd1jAzrv7dT+DyRaFg5cEui0Ic7EoW0wPQg1xw2fXddkvasf6KsyEMLdydm9YwJNu+Pbl81Y+WrKKfs7+oiv/tPaCooUS6pX/mBBLIUULKPP5d+mhPWMBStW+4ttVv5lv6MqXVsV8//oxjK5CeZQItcne9RaIVMx6pE8c3zfm8AFHKi04I8yszm5qPrr92OFbxe9b/x3f44gVHL7kRC5+CaN983srjxN+mfCt+TDGMWhcycF88Mw3mrXscCeBIcUc3CogqTjRWbI/XlhYP4WFBQ2UM/PZWvmcSTqJul4xBpQHstZTdXmcdDmwLZnvXRzD4jYbC9NHkwXT1Qnp1fy5BVxrenl0b0EcwWhfGV/qCIveIuKnBQgURy0A+6+YcvQiYftdFsgEEEfWB+VBxPVKP/X0I2FxwfpAojC/L7L7Nk991nw9fa3W3qVSHZk/8srHd6WX6s2P1a9kmiO1XQezB3rzuTt2hLn5DKoDXwjJ5zp0JRsIR84iz51VJGqR1DY4rO5UWNCcpOxvHLP/74oNIZ7UKTGXjcL5t5gcXAeojfWwiHTbgVgKs7vNhzwsre7dB5Pp6zyux6ZE5dGyRJ3NMaE2nU6Fo922gLydd1n5WJqP57WXmz9qCqZ7LsLkiQ4UlhfMh7Da70iF4+oHl9l3XNYMEtYPZN9dSKUd60bmF++XoN7lNprOm7+hrOdg35mu9RPWUdXdmUhxzeM0+8Q7Tq69EemeipMx4zrzfXRROZo6JMEdv2JzbV85JCm+s/Hu60QqhPltgeYMVAfCaD8RFr2xjhSOUCeKow4/b1ncs8Xl1wSKIwvYdK35Zc5vt/7tOuHMn0RiZUFHKQgp2JzQnIHpI9BuW6Ckz4FT5bHiW2V+m9CcN1m4Rtb64essn7XPZpZdc+LvSbIx6mvzJU9R1nPi2rce1kwAa9mljdiKFRtAxKLNE+a3LpCCfQ+1kqKNOd+8f/wi3x9IvH8Xmqr6Ziyi/PTvV4G/8jSnX1TVJ60LReTbgVJVfxD4wWfSpuflQnPAgAEDBgwYMOADGQrE9/NC833EN13ntx8BLr8vlQ4LzQEDBgwYMGDAgOcA7+8dzfcFqvpL1/ntd69X9r3B83ahGUdw9VVmBiuPMv3NPUbj0E/N7Ogb40hUD/sfnIwXLNr2fRybHck1Y8J8jG+NuqQ+Z5QxKKSqII0jEhSNgiw9lNEIsGcezrSUkw4Bus7TZw7Lx89vsfOOQLttbeknSsim9WKW+eWKzCMdYXzJzN7dltA5qA4y51swU8vKpJDKE6JldWbmWNwd164CVAkWHl8bmfuaGiXLQFoziaxI7Gf3KMt7I3RCeeDXpjvfwuGLs7mqPLneigh9RcTsepjfC65x9NvG/5iKFQuyUF0eo2FMt2v8gov7opk194PxwJ01vpCjVzmkMYoPFUWrCKOEC1b+6IERYWEmruaskqbRKK3aiGTO0/JIjIIkU3j4GsqZmVtiYTyqGhS/NNPl+IrJttkTqgMYXT0xWc0vCnFilCZbj5rZ3nVKt2VcmM2OycY3yuKuXG8N1b5RP8XS6Ih8rYz2I82e5+h+MwFPnlC6qaxNgYWDdss4IlcE3c0ZozA6ftnq/owixHVmju12zO/BtUIqlP5Mjzil7h04JUyNN+To0sh0fZLQoLgatExMLs4BUBUScPUw277nAUkeHfX4Y8/u24ViYUTrK2os1Ezkoc4cj87MZKmwMbhyNRhfNZNoLAXfJeo9x/KCmfXrC0apIq2gRVqTlrusm9W+USX1U6iumh6v6Fvmd0JzR09zBxDFeCmzyW5xGyxuN/1buce4TggzMwcubrO+6ifKkUJ1ucTFkmZPiTvROA7npc0NmUV6/1XZtB0hjjN/Z3RG9ZMEdYr0msnqBd3pOXhNMiJxp9A73EEgLB0qbk2uHmrWNGqyNA7W5M31ppibPvZjYb526VCqa0J5bGbb8SXTtRWhersjLC+aS8HK5SOWrLkf1ZnT1eJ20Klm7mHT97CwOTSWpocrirVux3hFXSPmolDaOFBnbhmhUfqxZ3ZHoJ8YdVC7l8n+x8b56+ae8sjjGmj3Mp9mtDntiY8GM++bHjOKUDvcrumvPDYyPslpxO219C8zAndNYkkHAG0T4cjjmhH9VqI8MDeU4hjaXV3T4Hns98kTpkvNGeuv4hiqA6XbFsjuTOTEBSvKq25baS5kHtsoMInI3COZ/igsoN2xuXJFVVXtWz+iRs+0uCtS32Z0QNLbcyiV5ub16Dm35mdNQdZju5h7M9UbtTHHL+9WUyuaTcgaFD+vII7WblXtixJxGk0HGw+TnlCZb9HDLxohnQ1kiRC3In7u4ULNtRdlnyEUd7W0nBDjhGscxfEJzZPrV/RV5soiETg6ob1a3GHVVAcn7l56YPIOjRILQdSuf3yvrJ/hO+9KNNuO6aWY3SmE5IWwTJTHPbM7SyRBNTMdGl+K1OdLwjxSzHv8UYNWNjemcjMrw3MHBboXQBrw5+1Cc8CAAQMGDBgw4AMVin6gmc5vCYaF5oABAwYMGDBgwLMNhfj8X2ci+jzctt3evVvP/6O/he52ljqrc+jVykwbSQgXlvSNZYZwRSLtl0zvmllubLUc2UURcaLUTUFKYlkWFgHphDDLkeW3tZZ3e+7RKuGWnjCXtbmqv701s9ixW0dnqmOd4WcVpTu/Sy0t4ChaKrBRxHkldg56h78Wcrq7hNsv0HONRZ5GwYWELxIpOlIvaO9gkc0CHsKBP4nO8xDvXRKqyKjsEFGCT9RtwfxgbPnEo6X0EKdo56keLpCYTWc5I0m7A+4lM5xTUhK6OpDqYKkOAcaRYtLR1YHdM3OOj8aMpy19dIzKjpBzZV99Yocw6bj97BF99FybTehbD5erdRpeX+d8wn6VmWcjm8bC0kqGRY74bk/MqGmV1aNSdCvLddyj0aHZNKWC3XMytwcdRfy4xxcJ5xOxd8TWo51HZj6nI7S60zjZ/e61uGB5jFdmO5fr1CT4SyXcVeNDpKuDmasUUhPQKIweCzRnLQWiqx1pp8ePe+Iy4MqIWEY2qlGHc8r8cMx0d0nXeaqyp+kCfevRmPu+8RRna5xPdE1ABKpRS0yWbUaOLNxbfU4lh1HgAAAgAElEQVQd2gl+aabAdi/hepPlKpLeUjaa2TCYRZ04skxA/W5EvSJlwpdxnWc69W6dSrWfF0jnoDLZahJS5wkjM/P1dUA7h6xz0guuFqoDsYwqJTTnE/7ikhASk1GLl0QbPYdXtyinFgXed97k2npYeHMnSOCXluGlO9fjd1q8T5YLG4idI8WcmjIK5+44IrjErC4tveZRidQO3evY3l3StIGutnR8VdGzWJbEoxJ/7Nfmy7WbCmaGR83kqZL1NFk6UNdYu9CclrNm7YoSS2jOJXQcceUJmbOqwNKbnuz2aC+ErQ6NQmoChGSy9Lo2HUvtcG12F6gSZ+874PBoYnPLClFwB5ZGsL+jyWmcBLdtaWtj63DBovXjMuAPQ3bVsFSJxZGjvdDjdzo0QSgjqmKpIFd5slXwRWJra7meD1RhOa9ItUc6R3lhyWRkvy+WFd21EdKaC0IapbWu9luWqjYszMUknu/gOJzIdy17mz8s61dmSfDQbSd0t0Oc4oJCzt2tvUOOg12rSMhRoLhjsX4mpN6ZXJP1Q5i5dfauVVrilRtCe8au4YuED5GtSUNwiT45jucjVMHllMix98TGU73b0qmuslLFyuopb1/gXaIIpgvOKU4UEaX0Ee8Sh4sx3ididBQh4iXRJXsOLOqS4CNdF+xZ1nq0dUhh+lJcKejPmOncLT2IPaf8cpWtx1K2ApQH9uwrjuz51e3kjEMjM4Ev7+4QrzavqthzpHfIUTB3gN0emXl0t8OFZOPVQTG2/mjnpc3PjbNn6sJTHDhSkd00WllnoOq30noukygUB86yymVzfDFT2m3LyrVyn9Gcnch38Jbv+Jrfeh9TOr7PePWHFPrT/+H8TZV9yT2PP+ftfaYYdjQHDBgwYMCAAQOedQhxk2fweYphoTlgwIABAwYMGPAsQ4H0/DMqvweelwvNVJh5m1mF6yqcWKRq5iOm3p9Sin1pLvbIJNL9wS5F3naHHMUtMMrRkEWCbkvZeacwvmJEw/M7SnRFipwcxVyJpZlsju4XODZT+6rerYcE3xmRcQpQ7UfaPU993pEmZgpHBe0cMZmZX1e8sslscykozunaLIFgJiASqg5VRUvTXLc04nrIxO0Kh/dB33qOjio74JOZOVdm85WQyKb5zqIrp48l6rOOwweUVCr+wSlRjKh7PMtk3L2Z/dR75vcUUCVmT5xhdM2i90ugnUATTL7bh9BNSx6+cwQRRpc8k7lFg5Mj2H2OqkeEWAj9FI7vU4pjiwgVNbNjqHVtdirmRnD8+EcF+r0cdU+uM568PUoyc7cIUCZQIdaB1GZ5JHMjQBRJOXpSLSJSiwQ+nyu6Ntmt63Zm2kqlorVHK7JbQjb36Ik7A06hcxYJOg/0UZDak1pnfRGFxSKAmql9fsHMxMW7PWWEcQ3FXC36OSnNzpRUgW5DezbSXxkzugbbh7lMNAL9YmFClqg0u45jdfRTZecdso7i9o2RZItCdRjpR45uIixuF/oLye5fIW2YYjU5NCoumG5pSDkS1twJIJsNBXMlSApeUXG4xjG6IuuIVHUQR45+z9PNSmpfmbwUWHq6ENamWZGs970xR0g0vVSfifMfHeP3jXFAetMVSRZBnQo4ODqLBiXMHaND1gkT1FdAxUTNRKgCy4uKlrD7Z0J1ZCTbMZPQF3NFM3l4twWHDzjavUh/3qKRt97pSAGmj1t7u6ngW2V8JdLseuZ3CnVmszhRKBCU5I0pA1GkABHFl4m2NZ0QMX1a9QvOvqsDgtJH66eVe4c4RcVcJDSBryKaEkl8Pt/6yeU+TEUilbqeJ/3S0d5ltl69VJE8tNls7WqLohc1Eu/uTOTocsX0IYe6MdURVHJCwF2f36IOsLy7R1ph7w895cxk2419nmuV0DhiYfN8uwPHFxSCUh44RtdymRpcuxq7UB72dNueRz9WSNOIqOli7LKbUFq5jJhbDSVoldYuGas5UbzpGikn1ejERNzk+ShZ2fYMFKOe7qCid4F2UWSXBocvI5p1W1yeDzq3JinfesTafOU1WXZv3SI5qN0J80QqlDQy/Ugjc+OpLnvCDOqse5ITSBSFuQ0UHczv7ykvB0aXjfkhloKLSixCZkbIbCs9NLvC4Ycl0iRSvbskzI0NwTcwuRyzS5bSbjmO73XUFxQpTACq9iwTScYUkttOSEZ+H80cv9JBye4AiJq7UJlsDq6EODFmDCR7pojNn1om02sx9gbNz5VUZBYFLznZyTpY3vpQTz4/11CgfQFkAn9eLjQHDBgwYMCAAQM+0JFUnr7Qn3MMC80BAwYMGDBgwIBnGZYZaFhovs8QEQ/8JvCIqn6yiNwP/DhwFvht4AtUtRWR+4B/BVwArgGfr6oP5zruBb4PuAfrm09U1QdvdE0Vyxk9uqaZMFYpjy3ntotKepdQHlg+2Ec/pqTfUopji6wuj9VyghdmzlKB3QeV6qDj6itHHLw8sV+ZmVW6hJYWGbeKJO2nCa0SEsz8mhTmFxziE8fLYKaBfMy27yPsl/iZY/qwp90xIuqwtNzf6lf5yz3L2zz9VBn9/thIl1s7HuYQGjNXrXK177/M0e4oi7uMRHqRIxxpvZnCjj3h2MjJl/dZBHB5qaA4zoTtArMXRZb3d+ATx68RaFfhtFBcCdaG2r6bydyIkBHg3UJzxjO6xjo3tW+hOlSSF+LYSKdDrVzF055LdK9Y0CZhNitwc2+mz2OhvhiNLHkrRzQKlC9pWNYFRRHBJxZtoD2u0ChUZ2qKEOkOxtA5/CzgF4KkYLm0c/T0Kj82WGQz5xuLqr1WUF09ybdbLC0/s2+Ucha59GGBycMF5TH4zhsBO5bzepX7vZsIzVlYvLTFFZHy7WOKmZE3IxAWmvOiJ9ptx7VXg+5GykcKxpcdYalG4u0lmwmt3yzHuqcfs86JXR3qmiy73RZG+0bsXB6Z2VlfNqd2SpNNywKkJLRXxoRjR7+TYNyCmnl7/8PAlRFfRvplgWQTeChOol7reYm/XBpJ81UzddfnWefgLg/tQvVZaM6ZKXD0RInLEaqTx83O7Fa5ohvl8odZFO/8vsj8VT3aOlxlxNJV1bOsA9UjBf1YM1m04h8ZWaKD7EZx9JKEnm2hjPSdp629mUMbRzEzYvEV20Ms2cj7DZNHhX7L+s/XJ2a4lUtGl/MuV4eW13zxoo7ZfYF6bqbG+mKeD+Z+HXleXRXacxG3dPgLNaNzHcfbEyQk5vcXFPtmEk4V+DsbiwpeBsrHS9QrcaRrtgPXnbArFH9WEccgH7lEFbrjgF84yjx+1eWkFJ3pxuEDkBK0v32GQmH3z07M9qHO+tMr8Y8nLC/C4kUdXKnY/UNHuwNhWREWNi+GWgkLNbP1vaDjnn5WsPc2BwmKpRILm0v7sdBNhWKuHLzU43OO61iaCdg3UM4S/UiYPgqja5FH9jwalHYvE83fmdCtzlwCvJKWATfqM9MBBIU+CosPiiwUI83PpO0SxYjeFwHGEW0d/tAi1PupEZpLL4TaZAzgrni6HUe/nfAPjhldOYlarg515U2Db5SrHyz0Y6UbJ7o7IvgTU7w4RWpHWDhjqWiFeKEjNh63XzB52K7Z7kK8p2f5QAsJFi8VWLq1GT/Mba4KixOZ2RxqhPD9xNHuKX4JOItWd870IOaEGiG7JDVnPd3ZSLyvZQ6k6Ng9Mzd2geiYHY3QowK/tOQPJvNE98CS5qiESY+vIteagHaCOLKPiKN6LDB+vLJ5M5uni7laTnOg2xJiVdGcAdRTHBeWmKQDF0tcBztHlhijmCXa7YKrr3b4xtyq9v400k9sPqwOE76za0ifmTCqRDHPrggu543vlDg2l7DyoLX8505wXeItN1pAPItQhPgCMJ0/G3f4VcDbNr7/E+A7VPWlwD7wJfn3bwd+WFU/BMu3+S0b5/ww8G2q+grgI4FLt7zVAwYMGDBgwIABtxBJ5ab+/jzjli40ReRu4JOw3UhERID/GnhjLvJDwKflz68EfjF//o/Ap+ZzXgkEVf15AFWdqeriVrZ7wIABAwYMGDDgVmJlOr+Zvz/PuNWm838KfC2wnb+fAw5Utc/fHwbuyp9/D/gM4J8B/w2wLSLngA8CDkTk/wTuB34B+HpVPWEyBkTkfwD+DrAXRlPaPfCtUB2a2U4Uml2LgvStsri9oJhb1FqcRmIVkGSRdqN9pZtYxFo3FtotTnLMXmgIZW/BwNGIqRtfIYUSe8GPekIVCd7MOl3viZ0nlD2tWjRyqCLjqkXVovNmQFRhriVaQNrr6JeefmKmxMkjFt2s3nLPNn2gumqRhNKBX1o7ceCiIL2ZI/q9SLnbEEIk+EQfHcEnvE/4C4mYcgRq51kcjomTxPiSo93GyKQLJUxbxqMO7xNd75k/MQXMpNiPwZc52nyUTZLuxIyJM3NPrMg5ha1MMU/EWugmQj8WRtegvgjb4xYRpS4ibVXALCC9R3bMZL6zs6Rwkbor2J0sEVHOThaMi47jtuKKU2J03HXmgFHoeViUxbJEJ0Jbm6pL4/Bzi+ZOpUXUFgfOCLWLROoT8XzLYsvjlp5y30ihu6mZsZraESszYak3HVOXI0GDs1zzqwjViLkblJaTvZ9kmdU5D3kA1yvzu4S41eMOSnYeVHwLfQXqTUauNzOlRMvJnYpV201H63OSI+9hdGC6CzC6lljcZm4bIkqR9cDnkOEDFbpJoNpqGVdm31o2BV0dqMYdZdEzU2EyNqLzUegJ3si4j4ueIzWS9VQEuh0xImxnJt5YmTmZTFTeT5Q4AXLEd33GjlcHihZYNPsdLQiMthvTt85TFJEiWPKE0R0d3AGlj7TRs6hLmsOKNPKoc0bI7JVy3LM3XRJV6HpP2wXqx6b0E8sz7XqTYRzl3NU5InVtKs8uCr62zytTuwBE+1wdwDIo/r45CjStZzJtGZUdx/MRIZicFtMJ0/Nz2jbQLQvEKbvnZojAkYzpUomOEtVOwz3n9q1f6jFX2z0oE2HSkXpH33tjTBCF1tEdeXwtjDKR92KnJ0pADkw/RaHdhmIuZkrtLEq7vi1SHHj6sTEUxNIirYuFsQ/UZyxa35WROIVuatHQsYJ+bHpVzLPrwG1Cezah1yqmD3t8rfjW3BqKudLsOFKw66fCxrlieloeZwLwY2M8WEUltzue0RVh8dKO5rxN8WXZMx21JBXGRcfBfMxk1DItWwoXuTKfsvTKuGppu0BZ9EYYHx1tG9iaNnQ7nuAjs9mYGBLt2ON2WooqkqLQNZY4AM1zRO2QrY7YCl3r1nnh+2yuNrcXobmtp9hp8ALjUbtOBtB0wUjW7z3Gu8So6HGieJd49Ik9u9ddKPdtHmccGU1bnEt4p7S9p742Jhx5ui1L9pGCzRkrl6luWy2hhc/JKQrJBPLZpWq5QVQOkKDaF/r7llRlT8oy2qpMjgDeJ2ZlRb8s6Huh2jKXmth78Eo56ZiMW7qRp20DZdnjXaLtAk09Rb1bj6/kYfqYrBlEmnPmDgbQTxTfCt3UXAGqAxtzqQBJQrPrTSd6cwEINTR7DnXmCtdPHCyVfmSuHHY9oR/ZM02S0m57fGPucuqExR0jfGNzn3zARJ0LnT7/Q2Vu2Y6miHwycElVf2vz5+sUXXX564GPFZHfAT4WeAToscXwx+TjHwG8GHjde1Si+r2q+lJVvVBsnXm/3ceAAQMGDBgwYMCtwLCj+b7hLwCfIiKfCIyAHWyHc09EQt7VvBt4FEBVHwU+HUBEtoDPUNVDEXkY+B1VfUc+9ibgo4Hvv4VtHzBgwIABAwYMuGVQFaI+/4OBnpVc5yLyWuD1Oer8J4GfUtUfF5HvBn5fVb9LRM4D11Q1icg3A1FV/0GOWv9t4K+q6mUR+QHgN1X1O290ven5e/T27/xyRJR+WUASyxc+TTl3s0N6IRWK7nZo6wlHnvJA6CerKDYz6bnWcqq62q2JisPc8ii7zszEq3Mgk+S2ML9bSVuRcOAZPy7ryNXyWGl3zDw1uZw4vttx9LII40i13eBdYjmv0NZne7/Hzx2psOjTVCbbhy4T5U5jF1Uz469yn/eLAplZLvfqqpkypAMczO9OMLV82iJG/h6jo58VFkHYGxmx5DzYqzzi1b6ZevqJEcL7Rc73frE1UudZgEItR3LriFtxnQM7HpRmslw64sTMq2GeP2/3yLHlSR8/YTngJa6I2JV2V9bE1xKh3zLzSpjD+DIUC12brlcRzP3YzM3H9wrdjuY88RZxKo2juprNMtHMNs0Zuye5rTYTcxGJyZEyuXtfG1l6td0wKjuatmB5MMKPe8qqN9LdR6YUh472Yg9Vsny9bdazaczyhnhUmIw9aJEI05yrel7gjywCubomtLtq+lqaqcdl87t2mcRdQHNfjR8JSG9uCvV56O+tEaeWf37uGT/u1u4LcaSEmZkxkzemBVElVsLhBykkTF+B0LCOHFVnZi6JZkZtzkF7vkeKxGi7oQiRpgs4UZIKzUFl+cuLBK3pip9Y7ueUBOd0/b/Ppv2+93QHFW5mZmHXm2lxfo8S7p2TopV3zq7R1SHn1Ya+CWhjEeZEGF32liu9stzWrhHSSFGn+KWN5TRORiJdZKL3Kudibz3jvZqq7OijZ3Z1YpHkZ1tCGelmpblE5Kji4tBl8upMMp9NgGZeheUdyX4bJ6rHA+NL1hfr54sAyeRdn4XmrJKqZIkXRJHa5gJfu5wjXUmlnaOFooVSnlsyGbc4UZZNQf3wFlolynM1TmyMe5946YXLPDrb4eo7zlre6Dz9x6lFRadxsvuKQnm2ZntSU7jE4WJE2wb6WYFbeHQrwsKjo4ifGZuGXzjCAtPdcYJCLQ91kXLCCSMo12WwBAmto7wc6MdK3O0tWjuK1T9KuKWzhBczi872S3N3wMHoMnTbNic1ZxPFsaM8wJggasVF09VuaskHwPrm4BXGCjI6t2RUdSybgnZRoPMCyW4wlNZfEvJcm8i+FAmOC9jqqbYb67ac051LI8aPCUWOED98ZU/Ybtdz7IqQXERZXJrCaJVFAmgcUjvKA9MjvzA2hPq8XTZOLcd3HCXYzg+atMECUq28yAQaZy4Wkx5tvZHmn23QJOhxgWscqUrWv/uOMDe5lEfQ7pjeFsfm4hMnUN9r+et3/6Cw6PfC2uYym0GxsCj8/ZcL/bbJljLPf53DtYKfWw7y5e2KayFVip5rGU1adic1XXIsm8LcQ4QTt7PO0z02Mb1JlgCgmLHOW9/u5QjzLBM5tn0zLRXJOdE1KNUVRyrtvkJ9Mqf91vf/7ec8d/gHvXqs//xn7r+psh//4rc95+19pngultJfB3yNiPwp5rO52pl8LfB2Eflj4DbgmwGyL+brgV8Ukbdg0/K/fLYbPWDAgAEDBgwY8P6CBQO5m/p7byAiLxeRXxWRRkRef+rYx4vI20XkT0Xk69/bNovIm/P5v5v/fu7pznlWvFBV9c3Am/Pnd2AURafLvJGTaPTTx34e+JBb18IBAwYMGDBgwIBnE7fMdH4N+EpOWH3samYh/k7gr2HB2L8hIj+jqn/4Xtb/ear6mzdb+HkZ7qQeqqqj64KZF9QiYbUVUgLpzXQtThjv1DR1ic6MIHxlSnC9RfD5GlQsWr160Yzl4Qh/1YiqNZPjWpls7s2kv8VMaM5FVHzOhW7m31QK1YGuzy+P1fK6zgLtUchb/R7pzQQjyUyi6oX6nMI44q8WSHK0TVbQVR7XBCko0ji0sLLusZJUmVnOL7PZ44mS8rBkdBWaPejuTlTHRrbsG4s4X93P6IpdYkUC3VxIuL0W3jXGNxCzuaI42mhLAj3fM5q2LA/GhENvpqweLDReCQtwvaOPBWm3gyIhj45OTJDZfB6Wdm9yZHWHBmb3KnJsptxYCluPRtptyx/tV/mxS3ODiJVSXgrEsfWvXwrlYTa/13ad+iKkvZ7CKSk6FtdKxu8ucFhkZ1AgCe1RoM7mVtc6dO6pyxKA8thRzIHLwfKXY/mI48SI+eWRkUVUH5tJuB8B4ul2PeqU8sDTTxQu1rT9yPqgMOL/VbicJoHaI9sdvojow5N1HveVK0d5CN1xgbpMWr3b0c0qYwLoc07mmekCYzPVFgszlatTdJIoFgFfQ3NmZSoXNLMeAOy9I3KYvJHI46nnniYJxZGjHVve8tGhjbtuW0mThD92uH1PrBRU6CpreHRKG8y0yNKv3QLKQ+vH6khZKHTLAPsl5WM2FtvbzDUlPV5Z1HTWGxM8FIdWlb8M9XnH8oFm7SITFkJqhdTZfadSV6eBT8jS086mNM5kUhwb0XhTFHQ+ZEJ2cycYXc7E1mRTuML4irkiqLf5gSQUd8+5/cwRD0/P0DbjNTuDb81MKdmVo9qH+qKZ9+VY8lx0MgZdCxrMxKseupfXpF7oW8/+oZF7SBTKY0eqhVZGrCj47r//Cf748gW6OuBrwTeyzjMuvbe80F1mUSgT7dUR15bZHNl56AS/MLeGHnIbHd1uRLY75HhkOiIreZzkXNfOUZ2p6VtP9Wign5h5PiyMgaF1Zn4nCum85U4vHg8W4d3keclnHT8iRxoLJdDc05trTV8yuaTEYjUZG6m6eku+EGolTaIxhTwypS4UP3MU9YmcJUEsHd22EncVfxAoD8UI5gvFN0KalzRzb4k5OmOxWEV3g7XRzzyxryiveXMH6u347L4Euz3uWkFYyFpnVgk4SFAdWZ72+V2QznX4SwW+FVBH9FlOai4ikgStIr5MxKOC6gnrr27HIZ0997qiNNL2maO/0OGOAuU1R3HEOj94t5XHj1rCkJX7x113X2UUOh567B7K3LfSAxPJOcUlu9aA7nb4SyUSw3oed10miz+ysnFkpnO5XLEsC5bjMWoEGuZSpNAWauOrc+bOtd2TOmH8mLlmpOxVloLJPaXCzOHHpnhxZNH40ttzs9q3+yyPlFCbbFdz2XMNBbpb0BhVvQRcEpFPOnXoI4E/3Yh5+XGMSvI9Fpoi8rPAy4EZRj95p6peeSbtef57oQ4YMGDAgAEDBnyAYZUZ6CZN5+dF5Df/f/beLNay9L7u+33DHs9w57rVVdXV3exuskmJgyKJkmLGtmIbcvLilwQwIgSQHSeAYT/kLW95DuKHIA9KAMEw8mABeTACwwgUOQECJ4rl0JQcUiTVVM/dNd/x3DPs8Rvy8N/3VJEi26JC0qXOXsBB3bPP3ufs801n117/tdYzj//sT/GRt4F7zzx/1mLye2GAvxlj/BKDaPsZ/OZAm39NKfVr/7oP/UTe0RwxYsSIESNGjHjeEf7k1PnZj0AM9HEWk9+LKULBfz/8aozx95RSRwj9/tUY45s/YN9P5oVmsNC+N2f3O4r8whONIr9oiUphK4cv5FZ1PzE8/sUdYh4pHyqmD8W0OFhFtvRy+99FQqqoDwyn84nQwE7oBtMixvADLd3uQb/nwSn0oAi2n1oTP+Oou4TurJDc3Qj2JCVZDXnancY0ip23hI4wrSiBRT09mO52kvsbENPd4gRU0PhUbc2Ok0qyXdsdzeqViO8VIRX6EoRKDlkkmEA0ksWtoqj0XBnJTxT5OWSrwOq2pn4h0u2A3ch27YGJKHOn94ReyS6EzkhqyZKPWoygL0JOO884+PaQoZwPRvJOaBbbBEwnWd/3/7IlzDzVCzLeJXMa6jsOXRmSldA8m5e9KCpnjuqA7ZSpP0rpd8RRQLea7FzUp253UCMGzeS+IruUHN1uImrbpJIs+/xE0fpEFO5BkSwM2dVQsnAiJskyFjTpGi7f0HS7gfxUUzwZyicGs/q9tzxRK5o9TTdVbH6+I80c6sOZZJ3XSIlCL9npKoLLNKe/IM4D5mFOObgUaJdIGcjlkFEeZGwvX8np9gI7H6ptacF1vrStFTvvybi5fN2w/kygP3C46WBG7ZVQs40iPx2oqtngnOAVEVj/OxuMDfSNFboeiEFjH6fkZ/D4ywZfBsr7mmyJlIdcj1MHxIHateAWivVdje4V8/cYjMLBNIZ8EfCZnFc0olztd0Rd3x5ETK24ek3hdnrSzNMVgX6mRbVqrqlXNZSWSH/VR9DvBXSvQUOCqKAJiuzESmmMY1AlK6EroyiifSrG3NFAfiGUNoipvIoQPtTbrO/FpzXulZpwUWDi07F4be6PhqClXXUP3XnBumyZzWpWhzkhCTQ3ZJ0o7w/hEBNRGqOk9MY0shZcf79gBueKdcRl0M8UMXX00aI/KigWUuoSzVDmEBXpyqAinPwsPFnO6B5K4MLs3lA+FKQ9knooHzBQ3VRsXgazMsz+wGJrWRNtE4c+ltKAkMDmFqhZT6wkREI7KE4UwRiiNZhGyh+ihrMvTUg2ivKx0KjpMm7Xa/0uuCGT+uILiczlDpIl9HNRmzc3AtFE7NqQnYvC25VALTR2d6fj9LaCRpOfmO0arTv5npvbsibTambvayJP563pZB7bSgI9Ln4KfJTjTC30r63UtjwjqTQqaBavK5rbPfbC4iaigu9nkXjUUk5a+ssd+slTh4G40xOdxm4U2cXQryk0hxFWaptP3k0VMQmwspSPpa/SFYTEbB1OTB8JFuqjjHY/Ul6KmbxtwHwgtLEKEd6VvPC+VFwcKkIeCKmmnyOuHGVA9QrdKsrHinZXzinkgQcP90mnHd3dlm5Q3Osri13Luu/K6/EesQ8zdt6R84pK2jEkshYk64htJKBjrbW4rjyW8hsJQBhKhNR19nlEOXjyZUX0lmSh2X/TyXqZ68H9xYsZ+6BKt+uebjclu2zRVQfPOOrE1KI3LbQ9YX9K88LkT3198aPEtRjo/yuUUn8H+E+Hp//+YBf5/XAfePGZ51uLye+Dlz7mNQAGJ6CvAj/Nd0eNfxdG6nzEiBEjRowYMeInjIjCxz/Z42PfJ8ZfjzF+aXh83MXh14DXlVKvKKVS4K8D/+R7d1JK/SLwUYzxB93RvN6vBH4GeO/j9vtE3tEcMWLEiBEjRox43hF+DPf7lBsfIDQAACAASURBVFI3gd9DgnKCUuo/Bz4XY1wqpf4u8E+RGsx/EGP89vccewv4baBTSn192HwL+HvA3xie/6ZSqgYy4B9+TwLkH8Mn9kIzJHLbvp9qTBdxpagqe5PgSo1pgqjgikgoPLq3MNDVPlOoaMguHdEqghHKJD035J+/JLOOqkvJ054bWcPZespmnTGb17y2f85BuqENlsqlpMYxMR2n7ZTmyFLanqlt+fDFPT54eIhJPQfziqpJ8Z9xGB2pOyvZyCtRNOvMi+Guk2z1YOHqNZ7mNg9CxPRKkawU3VzUhrxU8eLnz2m9ZS+rmScNjU9wUdN4i1WBB6sd5lnDyWqKfTVwtcnxvYaguHlzwV5RselTLjclAEmb4DrD1VdqlI741sLaoFstCsZUguXV4CV/9ZrQ9xAHFaJQ6CoqUT6jCYnHlD0375yjVOTyUyVl4rhdVISoJGNbBb6w8wCP5kG9S+MT3pg9ZtGX3H91l6NsLZ/XF9xf79A4y00dOJ6sOHtlwulySjmtUCoS+oS2szhnCF6TZj0p8KXjJ3xwtcfldMLqDmgTiINxe/Sa6BR6Y4jGYw8b6iSnPpb/aZpastDrI7M17VYRptOGMu053ZkKbTy4A6AUulfkZ1DfADUVSm16T5GuJHMaJVRmfajIFkIn+Vz6XA/OCSFhSx2pACiojyS73RdQ7tWkiXBtTZdgTaBMO3zUXN0uiQF2dytuz66YJi2VS7lbXnIzu+LrVy+yl1YUpqMNlnfvHrLuMlJn6Z3BfCpQtQnea8K16X+AaCJmYyBIaUgoPaFQrF8U1bmt5XtXN/Sg8h0ouJmECWQ3W+7sLei94aIqOcw6bk2XVMcJly+W7BUVD5dzNlVGW1h6GwibBHpFfrPizu4V3actISpOLmeURcf6wzndoaO9IZQ5AVENmzjQ3kKvo4HaUN1lUMCLi4OKSIjBkHEdlfRt/JLksoegeGG+ZJ60VC7h8WbGG/snrPuMk2qKDxqtIi/sLDn/6YbC9rigCVFx9XpB1aRkaU8JZNZTfrajcQlPzuaos4yQBtKjmqOdNUpFVNAor1FhGMNvXFFXKZvebJ0IopcxSRRatD6ZgI7oTssa0QJaaOF+olBIuUA/A7Pb4n1Gu2fktSCOGckm0uwPed570B55UdVfWEwn9H9xFkmc0MKmg24mx9mNUMPBDq+1AwVvFdFEskWgnypRtR87up/q6ACbOo53Vnxu9zGF6Tlrp7y9OOLkO0ekL675qy+9zdw2tMFykGwIKN7e3MAFg95aEcC6z7hoShpnKV4XM/7c9iTGs5s2VC5h0Rb03jAPGqUi9UFC96IlBEXoNaE16NqgBoW0n/dM9iu6qaWc1RRJT2F7DvMNjU84nVYcFRtOqimXm4LDouX0cgZvtGxa6Tui4vYLF4SoMDpwVRW0bUJuvQRq3AbnDctVChtLutD4PBLSKE4Phx417WlfjMNapcBrUJEYhjHgxMA8mXbkWY870sSoSIDJsD4ALI8mpJOOadGRGk9iPKlx6P3IrXLJC/kV9+o93rk8ZJJ2nG8mrBcFs92a1WXJ+a4RM34Ap7CX4hwQ9eDwkICbBOxGs74rv9PRQLejtup7UeArggZ0JMwdbWpYvmwJBrKriK0jrhjU7V5+64NNREE/M+i+oJtp0lUgW/Q0+wkqTNEuoruAy56PSMcY1Y9Ldf4YocW/32u/BfzWxxyeAv84xvhrz25USv2j4fi/+MOezyf2QnPEiBEjRowYMeJ5RYTnMYLyFPjvv8/2/+ZP+4bjheaIESNGjBgxYsS/AfwoxEA/SsQYN8BXv8/2f/6nfc9P5IWm7iE/0STrZ1RnCrq5Jlt4fCrmt9pHQiLZ2+0+mE7UcMEM++8YXD5QlFbUh+tVzioqQmPYJIELMyUuUlSA5XnG1zc5s2lDajwuaNZVRt9Y1EVKmAtFkc1b2k1K+lGGmwX0fEOWum1O8c2dJbl1bA5SMiPbrQp8tNijqjJe/0vvUdqeyiVCgetA543kc6PYUYG6T7mqch4sdnDesClTEj2TbN6gWdcZAH1jWecZN3eX7KQN7ILVgXWfsmpz3j09FGXyKpNM86Ak4zgJYt7bapRXQuNXSgy5lZjihzQOZvVqMHJ+2h/Bgh2M2U2l8XnC/c2B0Dy9Yp1GLps9Qh6w0x6bOhovWdr3F7sk1nPWlFxUJV1v+SjfpXeGpkvwTnLfY1Q0zjLPG6ZFiw8KayLTrGWvqChtD8DJZsq6yTitJ+wXFZ/eO+WiLblqCi43BUfzNVYFEuNZ1AWrOqfvDQcvXZJbyYyfJB0XVUlqPWXSsZs2uKj5zskNvHV86Zfe+a4xOrVSW/CN01u0D+ciUd5Yrj4tWdNothn30UY2fsifj1G2zRz1cSq5v7lQU9mFqGjb3SFzW0PXJFTnJWZlSK8UfQrnhbyncoqYRC6cpmpS5mXD5arkveyAnbLm/qN98klHYj3Oa5pKjJ/jJoHSkU06jJac8um0Yn4szuVVn3JzssIFzarLuKhKbu9csZvWpNrTBUOImmWf8WQ9k0XIa27nLZ0z+Kj58Gwf7wz+KqU5rDE6cLkp2SxzTtMpwWmSzPHZlx+RG0eqPVZ7ll3Ok82M07MZKDG5b3RCKMWZXPUau5LSEF9KFvJ3IYoZvwpAGIZsUOK4EJ9RD2tYXRUkhSM0MtYeMecycXTeUNUp99JdOme5WJe41hBay0k+E2oTCJ1B2YgygVBbGptJnoH1TGcNifUcHaxIjy/RKqKRrGxA5irw/sU+82nNndkV5XHHoiuGtgi4oDlvSi7WJfvTilWT4bwhsZ5Ee4yOVF2C0pG9vEET6YOhcZYYFa+8+IC9L1doIufdhPO6ZNOlNHVO31pCa0gmPdoE2l3L2kjJRjtkdNt6KDm6gnZH4SYRChmrpoPNLY3uhIJHKapjUSi7Uho+OEUMGrexPHSaeij9cF5zdjEj2Sic03xncQzAZV1wY7pGq8iT9YzeGayR72lUYFXnJNZT1SlxVsu81UHWz2BIjefu7JIQNU+qKU+WMz53/ASrAiEqrJYxpFXkwXqH0/WUtpWf0OA0i2XJRa9RJnK+I6rmdZVxVRVUVYavLU2dcmN/xSxryI2jCwYXNH0wPHhwgE4CwclaEHY7glOE1ohDwlLM8lUUVwKfiTOBrjTeGnGICEpyvoMi6oiOQ0DJUGoTI3TO0C4z9NKiegVLMVL3ZUQH6DaGrsgwhZNgiKDoq4THu3P2Jvss65zlouQyc/SrlOQsYW3kt2Hn9hWp8RRpj1FBxq2KtN6y6VImacflpmRvUrGTNsxTGXcuahZdwYOrHWJUuKhIrOdG0lN1CatlgfvlTt4v7QnApk237a9NpMgkLWXZpvhe6OgYZd4G56ScoNOkZ+k2ZOHfNCKK8K8R+nwS8Im80BwxYsSIESNGjHje8bzd0fxxYLzQHDFixIgRI0aM+Akj8kMZtv+ZxSfyQlNFqF/rKH5phVKRNOkJURG8YTUoPRPtCSjsqiAGzRf+0nukxuGCwWpP5y0uCvW0qApWy4LoNVoh6u/GYK4sfurRnSI/VwQDfl3SrUsWLznSgwZ3b4LpB6XweSq5tzFFF6IK1q3cNp9mLeerCW2dcJUVxKyh94ZECwUftGKSdSgVsVpUfXqg0TRCT2gj1E6qPVpFNm3K5t0dwq7jfJMSncKWjuAV6Xs5ACaD9rAn7Cgql5AaDwFc0HTe0NeWGDT5hwm6g+pFj/Zgjxu0jkyKFqMiMcp32M9rQlQ83sy4WJd8+fZHHGZrTtoZLhhS7XBR0wXLos3ZdBn71rFuMy7Op+Kw3Vvyh5Z+FlFakfxRgW7hnVtCR6kAXaVp2n3CMIJXCtJLmIr4HJ/B1Wc9/dTQ9AnOa6o+xZjALG/RSugaUXpG0sTx6GKHw501Lmgu65KqTulbi5sYtI303tB5oWTS1BGfoTxaZwnP0CBaRSxB6Ps+2R7nokYTSfVTpScR9KXk9YbdHoxkPvtCAt+VjiS59NsrN8RFoOkTfuELHzKxLUuXk2lHGyy1TzipZ0IfbmZcrko8YO5UNPuJlD50Gt1ISEB0kD7KUC7j9M5ElNinE07ULmmAGDO64TTTKIbcdgOrz0QmBx1NZ/Fes6pzjI5Y4+m8Yd2nhKhQKlKmorDuhs66nmOap2PZD8bwqfXUvaY7K9CtJqkVnStZ5x3r0wn5Iyuq5TLijyOb4XOeXclc0JiHOSGNxCLQR1A2ks+kXEEdR6wJGBPwXlM3CUqB7zVZ0fPpo1PulpdMbIuLhj5o2mC56gsaL3256jM+undI8Ar9zgTdQpOWtF7Mrk0DF5sptob+M4E4c1AZzONkm6ttmsGsvBNz+5DIsd2eh1mDD5qoItob0qFE45qKDCisCigFd2ZXAHTB4oKmtD2pdmgMVgWskXGYGk/bJZRph9FCw1sj1HExlJEkUVTGMSr0M4EhqXYUiezTByN0aARjAmniaEsH+w7fWGgMJBHvFKSBLpFSANMm3Nhd8cvHb2NU4NvLW+Smp/FSAnRWT5imLSfrKW2X0KxTqAzJlUE/STgvy22pg/JCBSffmvDRfkk0kfxU87DdJ12JIX/hI/1E4YE+AZ1B/pVT0sRRNTJunNcUiaNIOqwO29Ki65KQdZ9uSxEIbOnz6zmulCjCY1T4jSU5TfBl4KLXpGVPfGfKJpNSFVtr3I2ID4rWX88FcSPwQWOfiMsImcx/ZxOUicyPNvTOEA5kPsUowQqp9VjjSRH18s8cP2A3qah9ylVfsJdu2LiMe+tdlIp03tI4y6bO0KnHlxq7NLSHAd0qkqWi2w1op0juJYQkob/bwEVKttC4j1KeJDska5hV0M8yZqdyyldZRnp3TWb9QM9brPFkxg3t6kl0IDNO5l3QdMHQ+ETG6lAedl2mQ9Ckxm9/49LckVh5ng/j0A/9pwCtA+lQJtH1VszqVcQ7I6VcOhKjBD1cm8I/D4j8eFTnzxs+kReaI0aMGDFixIgRzzv8902F/GRhvNAcMWLEiBEjRoz4CSNGNVLnf2YRIJu1TLOWwvaUz5gjd2GgPgfq3HnNepNjtWdiOmoSUXoPSrhZ0uJzTdtb+tZiE4/XER8hBAtZILaaYBQhEwNalCI5t/RdielE2Rr9YNY9mHBH8ekF4HJVEpxm9i8L8g4iBZcpFKeBtVakK081M5x/XuFudHzrW3u88C8c69t2a84M0E8HlWyE6gXwRWDySKHvJZRPAuk68ODP5+gIu2/HQVEfWZDwaHGT2XtKzHCbSDdTuF3Fy9/ucIUmJAGfAtqwedkxLzoS7ZmkHf3QplaHLY2fGbf9f5oLhsL09CqQaUcfNVYFXKK31OmmS0lyhzaBptdEYwhlINpAMAYytkpp1YkReNSDClgx5HjLIxqGDHnF+qykf7wDEYozoSrrHjYWFkNbre9G/I5DNZqTD0uSjbx/VsPeaSToI9aZGM/rAHoP6uMAJxr1eMjZbiIxV4RNZKngMoHVy4p+Huki3H97V5S4TSSpIuuFJyaKmVaUqeLs86LQta/UZInDB0Vmn6GOrKPpE6YDxdc5ix+oekP8LuXitcH9Tl5zVeXyvtbjrCE6PdBHMhhDIsdEI+pUlN62JzxVWavwtH3zy0h7Yrnwu+SPDfP7EdtCcZpRH1p0pqgXe0QNp18ydEeezVuG87Uo5JUXI+9oIF1JVnGSwdWeYv2aRzeaG//PoJxXkfWLmst8RnJhRVnfQbOv8KuC5T8vqOrIeZS2bfY0PoMb9wLdXOFTw+aWoTuWUgVjRAlrTCAZSkw6a8SbXQuFuewyLpJy275aRULUW3oPA8YF6DXBaXQUGhwtueEhHdrpYui7S02Hhf0OF1J0p6BRGIZxem3iH8VE3S4N/df2SNYQPbhNpM3UNhPeJ3KMK6A9jHzz/TmmVsw+hPwiUPuIrQLtriFYRW7g/GVFcxQwjeLSzdj7IzG9zof1o1nv0k+EZrz8jKa+5Xl4bihOxCAbZN1SAUyimIRIYSQXe3UnwE7PdNLSpY4ut1grJvZaS1tPsg5rArO0pQ12WwIg64YnB8qkI9Ue5w19b0gKRx8UodbgZJ7HJG7bKmpRXauAmNJ3kK5lHFznuPdTGcPBSp+cvXOAckrM43toAdfB2ksuumSKR86/IOUMD/75Dula2tVnartuRw2Zgt0zz+kXLWo3Mr+vyC8jyQYe/7mEttfMLiBdKfpC5ue6sFw+OWJVSfkEEfJLaOewcwEqRlSQtWD5kqV5uSdPehLr6Z35rjEsVHGg94Z1k+KilHj0z1y4WO2ZJKLWrlSkcZa+sygTUWkgJFq+UwJ+qAq4/n6mgd5r2O0Jm0zaevjtMj3oKyjOA910oN0fTlhczbbZ8uq6TSPk52LM/vCGoj6ONAtF+yDyJFdSNmIHo/YWtIGYQQ2cvSSG7WpjMPc1yRrWTvojjzBtAt1UY+uAaQqq44Tp8JvQT6S8QvdP12gZw1GcZ54TPIc+mj9yfDIvNEeMGDFixIgRI55jyP8xn5+L3h8XxgvNESNGjBgxYsSInzjUeEdzxIgRI0aMGDFixI8eEUbV+Z9V+Azc+1MW92e055F1FciuHN3cDqkUgTaCTxXVzxr8vufr3/osphlqSgIkldQP5Ree8qxj76Wc9W2N6cBWYDqpcWp2U9JNRHeBkCqIUouyuano54COlA+GZBEF6VKOC4mi2VVEo9iYkvRKsfO+I2qFCpF+oqVmpfMkS0e6dFy9lhPPEiYPoD40JJtIsg6oCNlFx/LlnGQTcIWk9TRHmmQDTpyMqA8M/qjj6GhJ+W+1xKgwOpB7y4M3j1m+BsUjhW0U/UzsVp78fCq1T0upr6tue1QE/7/v46PUOJlG6mDCJrCO0janXzT0u4Fv/M5PUZyJzUg0kKwjpo8oB65QaBepbmjafcgXMH0YZHsfAY1tFNp5oobq0KAibF64rvu5TiCJRKNIV0Fqs6IkkehWYTcJ+RkkG+nTZBMIBpK1WB+pCNoZlq+KVZU9l4Qd5aWmx+VCa+y93eAmlnbH0O4osgtNfgH5IqD7AEqRrqCda0wPtoXyMVze7dGpp3tS4ArIFoq+lJot20SpZQpix9IcBvwf7aBOxfqqVU/rxkwnNV0P1ocsX1a0h4H/67dvoDxky+uxKIk26SagHFy9omk/16Frg//2nOlC3sMnYJvr2iVFshn6xypCoph+xJBApLB1JF0F2rkmJArTSv2aLzx73zIk60h54iSFaG5J1wFzFqgPZa7tvh1ZtQZfQHE+FCkrSNeBdqZpdqVWzrZR6kE9pBeKZl9slHwGzVHktU89xr4maTdWS43ao9WczR/so7wiXYJ2iuoYQhoJVmNrqc0KCahGY742k74dEn6Cl3beOw9EowgWli8r7hUTLj+8Q7KJ231tHelmGhUiyUZqvLLXFNN7lqSSOlNXyBgEyC4d7Z6lPhhq4Gzk5uEScyNw+i9vgoL6RkR3ClvL/LQbOR+01OnZjdQJRqVQTtrMZ4qEoU51T1G9GCAoiidSb6t9RPkotbI+Mr3f0u0mmNZg1zI+ko2MqWQTqA8M2TJQH1n5DqXUIV7XgzYH8r2SofZxy/JFOTftIOYBvUhw39wnrSFVYtl0XS+eLQP9ZM7i87BqDnj04GW0B91LvfLkQUt1M0MFqBXUX4GQRYp3DdOhrtc0kF0F+tJIf3rILwPNvqbblVpNW8mY9qmMW9SwzUVMB6sXFbpXJCvF7IM4nJcmGEjXEd0Hmr1hTegV/cSjvaGfQF9qTCttEoda1WQdWbxmcdNIulAkm6E9NPgycHx7wexTzZAqJGO2dZYPHhziNpZkYShOodmDbi+CVuRniuoFqVnsdyL6yuJ+74h0GdGprCtRg8sV5lLG2vp1TX0r8Ptf/xyzjyLdVG1/w3wi8y2pZH6df1nsi8r7mnQFPoXyNOAyML2M6zDU3na7wMZQvrCmfjFi3s4hyLm1u7I2rW9rfCJzLDvVzD+QPq0PNaaJshbWAVPLGLM1lA8VbiJjw3Sgq0hx7tGdrDk+ld9FnynafUXoLOlCMXkk62W7q2gO9FB/qdh5t0b5SLefkV15bBWwm56QGvq5xdQBu+5RIRCtoT1IWd96Pi59xmSgESNGjBgxYsSIET82hDEZaMSIESNGjBgxYsSPGjGK8fwnHZ/MC00lD9MyUH+eYBS2ChAjphXKwWpFyDRkHlsZtBfbhnQZMO31bXyNLwzaCV3S/vkVWdmgVWSSNRwnHe+cH9I5wws7S47LNY23nFRTXi3XHOVrdpOaZZ8TkESI+5sdLuuSu/MFpe35zvkRq2XBR7cTVKtQToOJ6FYRrME0FtMKjRgNNIfQtwpbQbtjcAWYriAYodR1D24Cze2e439bohsS7ZkmHbsuwepA46TrQ1RkxpHcqujWKY1PUF7hy8DRa+fcnl1RuYTaJWgi1cUu3mlWXwiS2rGQtokGlBd6KVhwE6ET1y9H1i+B8hHlxcpCPJ5ABeko5aG7IedqVOCyLkm05+Z0tT3H3Do+N3sEwNLlLPqSV8tTqpBy1k0BmbBdsISomNmW2ie8v9rnZDGjjop+kWEvpS1DYuhnAdKAKhuS3HFjWnGxW3K1ykBHoTB1hAjnX9HgIzE4iArlhdbZ3NYoJ9/7muZWQaikqOHmzQUAF9MCgHYPiNDtKrFk6YUabA8CJAFba0IiNCpKqLhrW6FghTJz00goPdEILdvOFS4TiiwaxEpl+FuZwI03TpllzdbmS6tI5RLOqwlGB3aKiv1MEp2O8hUHyYbTbko7JPl0wdJ48UHSRF4IhkfrOReHYssV3OD5UkdUr9GtIT9XJCuhV30h518fiEWUctDuyv/i9WAHFFLF5qc6smnL7qs1B8WGyiXcP98jtlLD1HlD3adS7mF7Ei1lHCrKeFcekk8vee3wbEtVVi7hbDPh/HLKZsewaTS6NkQdtzZOy9ch2iCpSEb6fXGsiL3a9rUMRI9yCt3Jtv6wR/1Ugx4sum7vXrGf1eSmp3IpT+opt4oNjbN8cLlPkXR8ZueE8CuPsdqTKM/GZ7x5eUzvDQflBpC0mKumwEfF5fkUdZES9nrSopcEGqdROmJs4MaklnH/oqVxhiooQlAEr4leETqD8pGoe0zp8JuEflfR7Wh0b1EOTGdQ/dPxJvZM8p2zy6fjOTih1VUUOr/dg+ZASoxMJaUKSSX2aFv7mAh9KXOc4VhbS0lCso7YNlAfSSKO3Xh8rglFRJWO/hcrogkcTDcc5BWvT09IlN9mQ+e6pwkJv3v6CkYF7k5lrjU+YSep0SoQBsufyqXkpue95QEPHu0RP9+xUzak1tE5y7JLtilWSkV2vSagaF5ICEHRdZboFTEMD6chKOysY3++Yd1kmLxlf7KRZLZ6IulzUdEHw6bLtnZOykRUrwhppL6h6Hc9t1895Ua5pvNmO3bfX+xTNSlXcYJpFH4SUJ0ivVL4HBafVoQsEqc9KvHUhaG5Le0eFUPtwtCXQdZcikCxX7Hz2ZrOGVIdmee12M1FzVk1wajIrhU7sM6JFdXnjx+TvuYoTM9VX/Comm1/V7SKnFcTLi8nnLxsoNOk52BqRXYJOZr6wOBKKcXod2S+LEo9WDwpFq9bKSHqB1szLX93NxzkgagTsoWUkJhGSkRMKyUR3W4q1PlM4xNFPzXY2mC6SDCKODF0c4N2kWAVrtTbcrLnASN1PmLEiBEjRowYMeJHDqnRHKnzESNGjBgxYsSIET9iiOp8vND8MwntIDtXZEuhb32mUT6i+0h6XhGyBLQipBrVK9hYTAfTBx6fibJWD+kaKFB9IF16tNMoFYnxqclq5w1NK/RKQFG5hMol+KBZ9ylX7SEHRcWizZkmHY23XFQlXW9ZdRkualLjOdhbUxz32yQSHzQ+aIwOlEMazDUFePtLZ6Jg9BaN/Nt5w6bOCFGRWI/zmsTJAPZBk2i/Vew2zlL3KX2Q12NU7E5q9g7PST8lVMj1fos2x0c5l6CiUHJBbcsToo14LelHKgCIejcmQttEfR0xIzzlNmlGnm1fU53mqs5JjSRgBK04raZY47f0zHWiyKIvOWmm7Kcb2mBZ9TlaBbog7ZAaTxs8bkggeunwAhc0fl/aweqAVWFLZa37lMu6ZJY1TNKWw9sVs6Sh9gnLTjiW6z5VKuKj5snVjEneYVTAmsA8awCwKkhyTzC4oPnoYo+j+Zov/vJbWC3RGy4YKpfgombZ5tR9Qhbh6mJCc6xQvdCz28SjgFBfQ1sGC7oyNIcD1SQsF6YbJkCQEgXlQT/OONEzLrKSWdmgiRgd6YOW5BYlqkcXNal2hKh40s65V+2ikfSRxls2fcosadEqctXlbNqUnVlNjIoi6cltL1R1n9J7wyRtt+OYumCvqCmSHk2kC4baJUIDD4lddZdwI+3xgxNC4y19MPSNxSSB9x4cyVAJAw3YGCg8n/3KB5S2Z57UuGC4aEu6YOiCQaso/R41wQ9S6oFSjBqw8doxGUwU2vyaxVIRpYcOGIYpcShJsECM6MxTDmMgcK0sfpo45If51QWD85rLuuQ8n1LajpXLBoo9ww/lDNfpZX0wpEO60+yFFl4Ao2Vctc4SUBRW1opVmxMj3NldcJhtCCjWfcbD1RyjI5O0pepTtIqs25TyYEk7lM2UqaR79cFQ9wmJDhgdqPuEi9MZNz5zwa3JktQ43l0cUg308jWqiynaenCGYCPdXNHP5BGMOCkQ2SbF6E5SfZp9ceFodxRqWBCUhzRX4kjRKKI15DsVReIwsrD8MRuY60Sy3huakLB2GS5oumBItdvObxcHGtwnrNsUZSIv7CyZpS2p8dtjcuMorUyiRVtw0ZTs71RMk47SdrhgWPYZdS9lJI1L6IPGmoDWsi503mz7cdnmXJxPwSvSh6mUCO0G5q8suPXiY/azCqtkrVj3Ges+3Z6vVUHWQa/gOgEN5G/LNs0n2mFNaA1KS5IWgHqGTBCmDQAAIABJREFUjY0RFIPjQ2WIe7BqMrreonWg6hLyRNrLeY02fohGlLnYOsuiy5knLZVPWXYZmy5FMzhTqIAP8oGzgw2TrKN4vdt+vlaRPhh80BRBs9gU7E0r9vOK3Uzm7UerXZyXuWpUYNOmxKjgKkddWWIS2fy7G2JQUjaiA95Litm0aFEqUrcJzpttv4coZQ4hSAmMb4wk9PmIap+Xi7sfzx1NpdSvAv/F8HQN/O0Y4zeG1/4q8N8CBvj7Mcb/6od8738GvICENwE8iTH+yscd84m80BwxYsSIESNGjHje8WNKBnof+Asxxkul1L8H/AbwC0opA/w68FeA+8DXlFL/JMb4hz/k+/9qjPH3/qQ7jxeaI0aMGDFixIgRP2H8uFTnMcbffebp/w3cGf7+MvBOjPE9AKXU/wj8NeCPXWgqpf5n4A3kjujngFsxxrM/zfl8ci80FagQ8bkiJJri1NHPDLrLMI3DZwmESMhFttrNob/S6F7Mv4MCpcVQvJ8nRKPQHXTvzXDvzZk+cHR+D9947igxmO0nx1z1katXLFdveE4mjvSjjOptod/i/Y6Qag5bL3ScOuDk1ZzVr6x58eCSm+WKEIXiCVGRao/Vnont0ETmSYuPms/snGCI9FEToqb2CV2wrCdCvZS2Z+1S3j055N6bN4mDATQR4sxBp8keS9crD+1h4Ke/9AH7WcXcNljtMSryqJnz1u++QlQQigBREU0kOxvUgoArhMpVTugwFQal9FLjJpHiRNSOuoekemoczGCY7wrZf/mqpvrmPvZ9uPWHFcTI+u4OwUPvIme54p8lr3D1msKVYoJ/fvES6UoM6tOVp50bbB3IL3ourOb9/1DzwkvnvDS7BNiWJVx/RxcMHsWqz+mD4fFyzp3dBaVtmdkGqzwT02G1p/YJbqBju2DIjGM/E2WrVYFZ0mCIZEYa5ros4YPTA7SK3CkvSZQnoPFR4YKhj5q6EKX/+1f7lPOGJkkJncE/S+1EBYPhPyCUL9Drob1bhQqKkCE0ME9NpUMWiYuUsDK059OtewIItekSOO8OOAdWn4qENLL7h5r8MuBTRXHmMF0gqz1NogmJ5uKLKevP9rz82n1S47dlCCEqpkknLgHGYbUn1Y69vOQg22yp+RD1VhF8XZ9UOSkj+c7pDc5PZ5izlGSpsIWIZ8OdBnWRcvANRb4InPysQb9RcbNYMjUdB+kaTeSh3eHt5RHvvH8TgkLXmpgF7JXBbiQ0wKegFVuVazoY2RPF0SEasPXTkgSfi+PDdZhDHI6tZwkXiz0OvqHJrgLVeodlaegnUn4D8P7uTS4/Jw1evql5y+4//dy1bE9XgWAVlVGsb2muPh347Oc/Ymo7ciMlCYXpSFTAI2Mn0442WH7n8lN86uCcl6YXzG2DIdDmCfOk2R5X+5TKpdwLu3x693Srci1Mz8S2Mo+DQQ/9WPuEd8tD7kwWvFKcY1RgZhvaYDHPUOf3Zrs8vNphdVniJwFfDVT2LIhq2A/jrNFb5TpE+rnatsF1WwJCw1YKU4Gzkepb+yRvR1bA0ig+sK8RjfSTbSKuUNQ3RMWcXGn8vSN8Kq9dNtLPphWFu2kj1Q1D81dWvHTznONyjR1U4RrDPGmY2JZkOBlNZNEWzNOWvbTiOF0CsHQFdUi2c3jRFTxcz2mahPrDGZdqj2SlpQ2mHntmMbXCZ8BAed+YrrlZLDlK15Smow+G+80e/8fX30B1WspmIuhOEdOIGubq9C2NK4exGMB0iqgMKhi6/UB5T+ZSSNka7osbiDxQMpY3RUlybpm/o0hqMfiXQIhIfdfQauiiKMTbA09533CPAwkGWQx95iH6iKsjbaZYfRHyuys+fXhKqh3p0PnSpn471100fJjtsZ/V7KUbDpINfTTMkoY+GNx2PUgJUfFmOKZPAgf7K+7Mr7auGdfrTao906QhRM3aZXQDdR7i05Kgzht81CybHO81bWdx/fOTxvNDUOeHSqln7yL+RozxN/4Ex/0nwP8y/H0buPfMa/eBX/gBxxngb8YY/0+l1Aff89pvKqVqoAd+Pcb4P3zcCXxyLzRHjBgxYsSIESOeU/yQyUBnMcaf+2HeXyn1y8iF5leuN33f0/j+mAIXP+C1X40x/p5S6gih378aY3zzB53H81IRO2LEiBEjRowY8f8bRET89Sd5fByUUn9HKfX14XFr2PYF4O8Dfy3GeD7seh948ZlD7wAPf8DbvvQxr8n5x3gKfBX46Y/b75N5RzNCMFDdkBzbbAGruwkuV3TTXDJ7gWzhRBU9GGd3M0V2Be0EohJFm+4jPlNbw+zZ+4ryJGCaQEw0PjfYtcMse0ybkFzU1Pu72JVGLTImD+VcgCF31VPdTHG5IlhRaLsHE96/N+HyjxQ+FaqwOA9ELdnp7a6mL2H12R6TO/7Xb/4s2TP/zyhPhZrv5pJ33BxAd+TQjWb+thY6qRaq6vzLCtNoihP5HJ9CP1N88w9eJj/RpEu2+eib27D3luTQBiP0z+UbmmTYJySiKDXNkCc9CHRtA64EW4ups/Ly+clGPg+ELutmYuadn8PkgRh8mz7iMwNaMnJtFSR3u1L0M0t2IXnFalBj++w6Fx2SKkg77CQicu/h0ZNdFr9zTHolas1gIL+M6CFrXXm4/OmALwO61rx5PuHN6i6T+5I1HzVDjraY/YYElq8H/MSTnCfMPpBs96gko/76fU0bOf8ioCMfXeRc/PYtbD18fzVQs1GUuMtXnirMj/9QMsBtG7FV5FpoWx8ODgFa+sVNPaZRUrLQg88jdqMwNYTsqXF8NJr8FCZPIv2QLW96KStRDpI6DGbvin6m0b2iPA2SBT/RuEJoz1gqdOdxpSa/iLh7Ce+/9wrZpVB5dgPpJuKygZIvFNlV4OTnFG7XkT1KmDwY1LKDIj4kkKxkrIQENrcjtlaUG5h/EDBtICSK+kCzLDLsWpEtAypAdgab+xO++vtfwNYyN00D67sRNNz5FxG78azvaDYvyLjW/dP8bRgoRp6aRLc74KYRXwa60hF7MaK3SyMG24nsly7k85ILQ7KRzHAi9FMDEXbergiJxpVWKNtjg5tEgpVc+msjaRWGHOg2kJ92VC8UZFeR9ErznT+4S/FIk59LW/nBYLqfss1GX74aMMc1jzcz3rx/E/Uox7Qyps0w300r7VvdkXN8NNkjObdkZ4qkgmZfxotPh0AFE/E3epQJvH//Ll+9gnT1dK66SRTaVIOfe+g05SON3QwUeISdd0U9nl55olEs72p8JtS4z6E5krxtU8t4tdVQHmHkXLUHu5ZxixJz+JBAtpC+96kiu/J0c4PyigqNXQs9Pnkia1V22W+PjUZMuotTz8U644PlEZdfv41PZK3SDnwC2VWk2ZcgjPNf6NGZ58nDY+xast5RkC1krk+eONa3LFevSymQ7hS778iYMk3k7OciOCkF0B1M70k++OrQ88437/Dh6q4EGgzXD+1e5OAt+T1INhFbR1Z3h3CA8DRDvTiNuFIN7SBt1u0CQVHdEteE67mfXirSpdDnwcq25iiSPUrILmVNc5kiX0SI8j0mj4PM3YUnGEtUhvxCTPaLM4crDVHJOq28fF5oFKrX1OuMb7zzGvmJIl1BX8p5y3ogf68+FQgzj17IWp5s2PaD6Z4atbtyGA8vRsLUc3q6w9W/OtrO1WQtc8Gnsr92w3gf3sNuZF3arrU+kjeRdB3oS83qzvNzj+1HoTqPMf46IvIBQCl1F/ifgP84xvjWM7t+DXhdKfUK8AD468B/9L3vp5T6ReCjGOMPuqN5vV8J/AzwX3/cfp/MC80RI0aMGDFixIjnGfGHos5/GPyXwAHw3ynxunIxxp+LMTql1N8F/ilSg/kPYozffvbA4Y7obwOdUurrw+ZbwN8D/sbw/LpGMwP+YYzx9z/uZMYLzREjRowYMWLEiJ8wxMb3x6I6/1vA3/oBr/0W8Fsfc3gK/OMY4689u1Ep9Y+G4//iD3s+n8gLzajh87/yFq9OT0mUqN9E8avwQ/btaTfFBcP64gats/wHf+Hr5Lqn1KLCXIWcNiTb/d/dHFG5hO98dJPL1oBSmLwndIbsoxzd5/RTiDbbqn0hsrkt9FuwcJFpxFV3oJOrgeJqFaYVRXZSCQWkXcRWXoyhe0uaKdwkoT0w2GpQyQ40SbMvyu4w0LLJBqI1uDLicjH2pZE86OPbC/aKit2fb7btdVJPuP/VO+TnDMpxOY/8XOETRbsj/K3LFM1NT/azV3z+xkO0CmTakQ3myJl2g8JQcelK3l0d8ucO3mNmGvpoWPmcPhoS5bcmzE1IuFfv8eb5MctNTvCak05DZbFXBl8GoZhbmYzBRjCRfqYkQ90pihOFaUWR6YprdSswa9FGjg+JfP9oYD1RXPsX615ymkMu5vzmwpIsFekVA/Um46mbS86zaUHXGl1rpveE0oxG+s+2DPtI1vPkPlQ3IV1rEomxxjaDElQqNgiJojiBq896UHD4tz+ktP3WcDzVYtx9rQ4+StcsXc4HmwO+vPcBme5JlORmtyHBDwbATUioQsqTdsa/enKHk2WBvl+gncKuwRcKV0RUsBSPAS10lU+hPtJEK22kIihviCYSdbI1j48mYlqFK0Wd3e1AW6ntmDQ9NHua8j70K6Hq8ssgLge7mvxCaNVmX+aDaWD+nmL1UsRWsj0OqvrF5x2ffeM+U9uR/mWRgk/bkrceHqMeFELlDRTo/H2hFfsS+omV7PdZZP5LZ7y+d7p1ERBFqkET2U2qrVJ16XK0ihwkG2bm6RxphkHVR8OiL2mD5X/7zhtwL2fxaY0rwE0C6IhyBcorkpVQw64Ad9SzftHTvF2QXQoV63KFKzQhBe1SKfcYHBmKx/qpCn5rIi9t6wogQPlQs9EFi3slGLAbJbR6kHZTUd7PNpBeaNqDwPzbCcEKHZ5sIvnl03VTdwHtIhdvpNS3AskGpg//X/beNNaS9D7v+/3ft9az3bX3mZ4e9sxwhiOKFEUiEiVblixHthVDVpxIDoJYAYIkHwwjAQJk8YcksJHY8ocsgA0DCuxEBoI4jhLFcuRYUCTBWqyFIkWJw20Wsme6p2/3Xc9+anmXfPjXPbdJkRQpq6nm6DxAo+89p06dqrf+9Vbdep7/80SChe3XHatLCdFAPgnUWwafJUyeg+XTnpvPPeSZwRmh64o3EsiMhi0k4hkkDb91eoPdfMWHd94gFUcVU9poqUOKIbIMGS5ajuoBH3vwFPW7EpatJZ53pTcWWkN2lJAsE0Kq51u9G2h2wBeGRZVgW0guWa1VURmQz7udXEB2Zmk0qhvX685zC9W+SnLqHcgfpASbKhV7Tk/HTm4TYXU56eZwndvF6/zrCghWsDsNl3en7L57SWICg7TGEJm7nE/+8+cojlV2BDrnFicqt8lmOneEFJZXI7svH/Otl++uTchB3TNSCYyS1dpw/3PLfd49eEjPNPRMjccwcT299hm3HuMz1+MnP/F+3CBj9t4WTMTYoGO7SEiWSveL1/ENLyyYvgDc6ZOsUuxSt8F3biOxu4OQGLFHGclScP0LWUoIOj+cy5byY0PTySbSuV4DE6MSIuO0ZuuRyj/Ks8j82Uh6ZslPEozTdZwPRTKH7LyjHtbuAsEKxkfMkvUcdv5+M7TYKrD9hudJwROYdX4E/N0v8fp///td4TvyRnODDTbYYIMNNtjgSUbkybvRjDEu0AafL379V36/69zcaG6wwQYbbLDBBht8nRERXHhyGpMeF96RN5ohg0vFDEtYU7RWAhY14g0i5MZhJbJdrli2GVNf0kY18AaYu4I6put1FrbtslOVCpcAwRmiM+v8YwmA795LAgwdvs3WudUXBsVKDZime+wftfs2itIR9ZYhWUV67UVncEgNPtdts112sDzy9F88mM5MOgqYQmCg+y7tI+bUiWO/WNDvgrHP89nbXY8/spgVhEQ7ucUrRSJRaR1XwvZTE17ef6B5zkRctDiv1PqCuM7zBjWOP2qGLJMMS1BqrKPVUhNIOpo9M45RXjGZ9jTbN1z8hSeu+znqmIlECIJ4wRD1/Y4aMS1Yw7qDW/NtE/U7Ly6W6XZcf3c69u2WEEctcZxinNJjLr/opD+nuSSAG+o+Lq9ZyiPWBvT1SDpj+oDriXaKWnClBgdAR713xw+jHevKicLoxpRb/VNy69ZpEan4L8jurUNCHRJcMNytdihtS6/bOI/paLUuUz1aXLDUTUpoLEmr21Pvqom16cbO9bttcqoOj0br4px+VXPmR7LXPZzbsYl/JNO67V7tfjatesv7sjPsXwSaoSGb6eBH09HsoevqT3WdrqddpeHcOL3nuFzMyY2awFsiiQQOtwa0Tbk2oRevUgWA/n3tsD+nTJ8ejTFE6pCocXSwNCEhMb6TfqhZeR0Sxk2Pg9UW29kSK5G+rdf1GKJhFVJqn5AWjmTOuhvdtNq9f26QLV0nrBt5bj/zkNvDE169dIk7r19FWtGcahtJzhKyqdKuGHUQsJUgousI+UVNnztFJCvAga2EmHb1kQmsvnB+MF2HuOtHpfUD5NOOqvSQzj0hEZqRISRCNo3dsYk0W9qVbJvI9KbOhbErkGwamD1liUnk1nMPeG7rmNK0GAnrTPLzOfa8ls8lIW/Vu6TiaaNdP81JTcB3jRGp8YzKivuz7fX8ClqD4lQmFLvxCImOSTI365o7HyfXv+jqPh8HbKTZCvQW5hFpCGu3j3Nj82pbu6M5Tsgm5xvwhebnxul3SARfRpqhnv8hBSRyuTfnUqFBAnmXcGEk0ux6xFmV3uTQDKAdddKHpagrhBfMMwtubo1x0ZKgpucBnfeQgMdgYsR3LoUP6hFGIgNbEzC0nbl5zzR41Cx95VMN3jARah3bkAnRGUzU+TaKyq3aUeQ7nrlDP6n5efsCq0mOnSbYSvBFJNpIOjvPqr9wZpEAUp/PwTr2PtHaFq81axod82wa19S28dAMZO0MYTzEPOCskI0tprpwNpDu/fN5mUfpd68rtOvrrda9hE4e4i+WeRLwmCIonyi8I280N9hggw022GCDDZ5oxCePOn8c2NxobrDBBhtssMEGG3yd8SRqNB8HHtuNpog8DfwD4CpKVP5YjPF/FJFd4H8HbgF3gB+KMZ498rkPoSHwPxxj/Inutb8FfD/ak/mzwH8UY/zyz75N5P/73Ltx9/rYlVKA2eSiG1nOqbYI/jsn9POGn/jtD0BlSMYJ6VxNhSVCeWDWebzVfiTtqIWQR+xxQjsK1FcdOCE7tbR7Duk5RODGlTOu3J5T2JZ5m/P5s11EIiIwnZZwp6R9Xmm5VZrRDpR+wCi9EJJEu89Xatruy6hUQs8Qre6TeIhBaYR8HKhHSmW0I6XyVjfUjNwuLAQY393nXn2Z4qHFrpS6Wd1ssZV2v1Y91ubvPLvAmIhrLdlnSkIGt4dTKp/yK596Hiqj3GhQOku8njDRRug78kHDbz18Vo9a2fGr3kBjSKaG/ETNkScve8ywQToKevhKRjsE+eapZtOOC2RscVuBmAS2Ls/pZS1tMKzqlNUbI0wrhDR2huCR4siwuz9jOi+oLlskiBovz3SfDUq7JEtYXdG6yAcNtTfUbWem3xmRI5CfabevBKDnwUCDdmSfd6a6QmlL1zPrZa+/9wFGIvfkmtJGtZAsOspHIJtq1yp54MbWhF+8/y4mpwNi21Htc+UGs1OzpgnrWzXDrRWf/ugtTCO4LQ82kj1MKI9g8kIAq7TspafO6BUN+6M5PAV1m9AGS+ssiwcDpBV8r5NyPL3EO8PNy6dYExhXJYtVjnNaU65KSB7kZC9PSBPP9PPb5Cd6fiSVyhNspTIS45QOb4aiJuNLpa1WezrlpMu4lmiYAL4QTj+g546rLM22wVZa53Ga8sufu018UJBNBHFQv7RCbGTQmVM3A0iX4F5esDtaMlkWVGcF5b0Uv98ybXJ++/A67rTA1OoyIE1H/zshWSntWd1soTFkJxbTmT83u4E4aklLh6stcZLBwCFJgFKlOuKhPJCOitR6sBVUlyDZ0xznK/mU0X5F5VIWdYYPQgiGukypspyQRkgDlJ4mZh39rXV9nrGu64+4pcE2apouPUccZ8R+IFohP9Xj2cVzky40w56exxdWKeS+4CqotxJsFan2pAsbsNR7EQYej8pDJAimvqCKfSGERGvcF4G9csndxTaf/ewNcELMA2alTgWmFdKxGrbb52e4JuETD97FeZb3+f6Ig/xMaEaw//5DfBA4yzBeaV7TCH6/hcTT2IhZqjl+KAN2u8FedzTO4FtLXCZIZYiJfn8UnQt9Gcj3V9STgnZoCElUF4aFyhNMo+ERzUC3CavziTg9r9eUbaO1np9FJi87yp2KZpZTu0y73PuBOE/57OFlPvna7bVsSjy0WxHZr3ELQ72jHdrVfqC4tqB2hsUqpXwjw+cwGqx4uBzw0VeexVTd9raGUKpDRXaondjVzZb+7pLFvSH5qVFa22hdu0EgmxiSpRqcF+8ZY3NH3PfqxuEse5emxCgsqoz2rYE6dOw4+ntLPj/b5Zt2DnjpykMO+iPqy5bFMlc5lYk0WUEyscSbK3xtoeqOy5nF1qzlV+1IL9fJQuu5Hao8xLTyhQ4qe/rz4F6k2hHIPVJG6l1Dujg3tFdngGQB2SzS9gRXQLnsztehYDzYiX5n21MHi6SOasq/bbDnEqonAH8UbjQfpwrVAf9JjPEl4NuAvywi7wH+c+DnYozPAz/X/Q6AiFjgR1Ez0fPXPgx8B/DNaMzRh4DveozbvcEGG2ywwQYbbPBYcZ51/tX8+0bGY7vRjDEexBg/1v08Az4N3AB+APjxbrEfB/78Ix/7K8D/CRw+uiqgQE1EcyAFHj6u7d5ggw022GCDDTb4esBH81X9+0bG10WjKSK30DzMXweuxBgPQG9GReRyt8wN4AeB70GfWtIt86si8gvAAdp/+LdjjJ/+il8YhOaoZPSW0PaVDlGqTrt+o1FaL1rh4bRgZXPMSYpx0Hugd7Y7h4Z2oPRTsookq8BJYXn3977BKK3IrWPlUy7nM65lE9poOWyG7KRLrqVjDtptdpIF78oOGZkKQ2B8vUfo7u3Hvsf9921TiON35jd45fga8yqnbS0iECSyWKbExmJnVumdQmk1V1rarUi9q7SQeLCNMH/arLsum63I3vMnfNP+AaVtOVhtcSmf89p0nzZYDra3qSZZ10mp1PfiWa/G6F4I+w1/7Jk7fOf2awB86sXrPKi2eGu2zUE7QlKPKR2+scTGILHroo0obRuEep7pnzIBWNiL7nCnXYurK0qfpWeW4dNL/vzN36EwLZNvLkmN53b+ECuRmS9YhpyhXdHEhMN2xMvlPfbsnEXIufvSHrZrI2+j5a16j6krGCUVB9UWv9zcJkxTjLOEFOr9QBw5TOYRASRSZo4XLh2x3E25v7vFcp4TxxnNfoDCs2g07zqmEZMEYmtBoN0O1GnUrtda1rTpeXfqs8NTStsyfb5gOilpT3PaoVaytEKzLTTbnr2rEyqXMD4camdtRw+KQzs7OyN+00J7ljKtLeVYOvpdzalNrTTy6PWO6u7D6XDAX3jpt7iZn3AlnWDRbtVlyPn0s9c5bfoEhN10wd3VDoVt+Yv7v862WTENBbNQrLvYH7otTt2A3LS00fIzg/fw5vEu7YMexIuuXrxSV+oIEDvjceH4vRnV5YhdCY3TztOQnbs0wFPPHrGTr3i6f8blbLY+nZc+4xfuP48fLdWZwAmxtUgMWkNOTdnrVnjp2kP+2N7rPJWdsgwZH5s/w9vLbQ7mI3xroecIuQETib2uwzpCayIxCiYJ+JjQXAuwunCYYJbiz5TOTmrBnGYMPnjM1acPcNHQS1qeKsckxrNwede9XgIwSivGTcnDesQL/Qf88M2PkpuWQloCwsyX3Kn2+OjJ0+zkK/aLOXdu7PLW0S5l2TAqK1LjqX3CZFlgTcSYQAiGK/0ljUuY9AqaJsFVCe2uIJXVrtzKIEGINmBSz+LdDWaSXjhWBJ07zjuq6x3wZYC5pTjSjHa70nn0PHM7JFDvCMunA+mlitOq5GC8hTQqoyjup0TppD2hyxNv4ORaTmyF0V2h2YJ83HXme8FWEeMj6auRg6eHDIYVL7z3LokEnhsesZMuuZmdEBCO3JDr6ZihWVHFjHvNLlfSCdt2ybZZ4hEWIaeNlipmtNHy8cVNxm3J4WrI2/kW125PccEwb3IGWc3Kqa6qcQnRWy5bT+MtZj/iXjBUTYp3XfiGE6I3RC9cu3HGs1sn7GcLyvc35MYxtBWH7Yg3F7u8IhFjIvPjHnZhNZyjbGmyjMVN7dzOryz5E8+8xjcP7ul15FtGzH3OZ6ZXeOt0B0pP8EIytbihX3fSN9dbnSdST9MkxMLTvKshjjN1GOgcUGIXxmBamM8KvunmfZ4bHnEzP8VI4FIyW8+drz53lZO2z8Ll5Nbx+dkur00v8W/f+A2qSymZOHqmoY2WRcg5bocsQ0YqnqNmwM985j3Ica4m97udtUeUzoFB5/2QQkwjLoksbpr1NUOvZYKthMntR1IKiPihp95WqUDLhSNFvS3qwrBUaVxSKZ0uQTPUjVPKvB4JVBcuBIurT8YTwvhHpBnosd8mi8gAfUr5H8cYp19h0f8B+M9ijF9g2S8izwEvAU+hT0S/R0T++Jf4nv9ARF4TkSN3dvbFb2+wwQYbbLDBBhs8UYhRvqp/38h4rE80RSRFbzL/1xjj/9W9/FBErnVPM69xQZN/EPiHXQD8PvBnRcQBzwO/FmOcd+v8f1HN5y8++l0xxh8Dfgwgv/n0k2OStcEGG2ywwQYbbPC78I2vv/xq8Di7zgX4e8CnY4z/3SNv/RTwI8Df7P7/xwAxxmcf+ez/Avw/Mcb/W0R+GPj3ReRvoM/Rvwt9+vllYWvo31E6cefVQH6mdFkyd6RnK5pLPaJR2tkeZ0pCbWGRAAAgAElEQVR1BrArzaw2LWTzQDYDn6mBbDTaHXdS9Zi3GYV1NMEybQoepCMAXLAcNwMO0yFH1ZDMOj5uniY3jrJrc3PBkltHGwzjVrNoT+seL+wesZ/NARglmrH8aD57iMJRM+S16T7vfs/hen1WIol4XLQsfcrKpyxdzrTNSSQwbUuWLudgPmLeZlQuZdWm+MYopd3lIX/o215lK12xmy3omYY6JMx9ziuLp/AIR9WQcVOwalNaZ4lBCB6lzbtMYSRCay4M1/Mu+7l9xBz53GDdsjbaNg7Gkx4fOXuGXtKuzbMf1CPSznDfSGBga+qYcne5w1nbYy9d0Ea7zqcGqEPCw0r3tXYJZdryr73nFdpoKK3mgvc6s3pDxEqgCikH9RY+Cjd6LS+ODqlDwiituJUf0caEj8+eZtyU3OiNmbUFc5dzq3/SCbUNl7IZqXGEjpsxEqhDykdOn+Fab8qfu/kKvc74uw0JJ23/EfN1y2nd42A+Ium32r3ZWGIaiU5otiJF16HfDpRyNytz0c3aGcOHVE2K80kknwSqbYMrS35u6wV2yyVXyxmlbcmNo42Gs6ZP5XUKWLiMw9WAxAR+dvpe9tIZc1cwdSXbqQYcn1NqLhqWLuNqOeP2s8eYd6lJdM826+x1UHPziSt5dX6FnWxJP6kZ2PoLsu5T8XiEB/UWdUh4c7HLw2rEST1Yn8+ZcdQuIZ7TllGgMfS2V3zPn3mF0jZc7lqsD5sRry6vcNBsAXCw2uJ41WfVpGr2byA4kPMaDUKMgNH68W3SSUkE09UtUcvbtB3F7PS1zHoSE3De4ILhuOmTdLXbBIsLhiZYDuYjVk3KadVj0pY83Ttj6ZWqzY1KcI7qIbMqZ7IsmfQLnh2e8k3bDwAobcNOuljXTkCoQkodEn7n7Dpp5vn2K58nN47dRJe7W+2uwycsEY/w2cllfDBcefec6+WYVNSsPkRh3jnbpyYwbks+fXaF5z54zO3eEalxtCGhjZY2Wh7WQ5YuVzcNl3O06jMsK5rQJz812JW6YLQDHafznPDkQYat9HfTKAUfSqU9277KTnwumHsl8p4ViQRcNJw2fVY+owqpnm8YQjTkpmXuCz45vc6lYsZOsuRyNsUSqGLK0udrU/jjesC4KZk0Bbd3Tni2f4KPwlayIhW/rtnzc/ewGXF3tcO7esfkxrGfKr18p9pn7vO1AfvC5ax8ylvLHQZpva7XaVtyfz6irlJCaxBnMF0n/AuXjrjxzGtcyuYMrc71p67PpxfXaaNZn2N3z7ZZjUudOIN0nesC4cLpAyD4hGAD4gyx6eqUi4CFR2HfLphfy/jU5CoPcjV4zzpjfUNk0pZUPmHeZpyteqzqFOcsvzF6FiORlU8ZJRWpCcxdtt7WeZvTBMv3vfgpcuNITWBgK5XZdDXro17zPjffY5RV3Cgn7KUzLJGqs4SxneH/WdujjZaPHT2tJv7W84EP3luPj5HY1YKst/+k7rN0KYXVY7N06ReEXSzbjJVLmK9ymlXKk4Jv9KeVXw0e5xPN7wD+HeATIvLx7rW/it5g/iMR+feAt4B/8/dYz0+gus1PoKfNP4sx/pPHs8kbbLDBBhtssMEGjx8bH81/ScQYf5l1eNjvwp/8PT777z7yswf+wz+4Ldtggw022GCDDTb4Q0ZkHdH6TsY7MhkoihrsZpNIedTQDnQ3fW4wg1wpB4GYmXV3cP+umsGaRrvUXC5qFi1CUmt2dTqFBx+9RnGidI8EsHXEttCW2pmZ1FDtwuq5htgakonF1oJd6fK6zgsDXwkw+N5Dnu6frekYK6GjtZQeP6djU+NZ+ZRnypMvoGjPc4MHNqUNljpLmbYFr04usWgyTqd9sld6TNDtYqEmzESwS0N7qeVqMVnn4g5tRWoSpq7k9fk+96cjFsucGAy+MdgsEFtDbMGsDOIuuhvTuZDMwQ0M7SiSzHQcu11bZwGH9IL2BTD3Cl45vkU6MWtK/fz8S+dddnEB1dVANGAXhmyiptjtSOUS6azLz+7yoPmuMd9+/Q6jZEUb7Zoq3EqW63E1EuiZhrnPubPYoywnTF1BiEJpG4xEcmkZJdWavk/Fs5stuJZNWIaMNli2kiWFtOvc4UwcTUxogmXSlDxbHqsUQi0PdD0msC0rQhQ+cXyNybinhv4mEiWiYdedAX5UWUez1Uk8KiE/7Yzie2oSHlLwPaj2hEVrKU61xmYf22cmcL/rAo4GDQXo8ojP89/HH2pAIm/+wjNrU2qJUByzzjFGdBvql1b80Msfo2capl67qw2RVDyFtLQxwYpnK1nxVO9sPfY9W9MzDX1Ta2YzmvVriRy2I2ZNzsmyz8O3tzEr7dL1l1ps7vDjXI3UW2i2I5cGc3Lr1t+Zimdpc16bX+Yz48tMliWLRU4MgrGR0FiiM9hl1x19aIgJ2CW4gdZbSHWfbdcpfZ5/bVo1+o+2y4JP4eDNPQ7YY+uVBKJSxSHVOgz5RTDE9OUWaQ3+rW1Osqv8Njru6Zy1K4ZtoFwEps8YTv4V4c9c+ySWyHE7oGcbdu1iLTdoYqJ1Fy2jbI9b/RN2kiUBYcsuVY5QqKTDR4OVwNJnbOcVo3TFU8WYnq2xRLasngtVTAnRkIpjlhYYIpeyGbvJXMc3bVXCg7CVLFn6nKGtOHV9lu4GhshRps4ecaDjF5KLLnXj1ATd1ipZ8nk3poV2D5+PrS+01icnA6Z3tsmPDQ/GF/nuIYXiNGLryPKSYfpCIKZRTcNtJJ12oRGJztHlYSSbR1wpLK4KL/y513nv6G18NGu5gJ7XSmGfzxP76Zxx2+vy2gN9o7T4fjqnZxutdeM4kSFvLnZ4/eAy/iSnf0/P//ltRzJskLcLirmeu+0A/DMrtrMVg6RZU/apOI7jgNfn+xzMRozfHoEXKAIyt4jT8z1dgG0M0Rh1T7E6bslCw0R694W0M7Kvt/Uc9/m5y4rWs2ng869fxc4t+anKcZpRFzCw0mNyLntYPduSDhrcg5Jf/PgHiN36bK3/5+PuOuhh/F0rvv3253mpf8AyZMxdwcBWbHf11cSEEIWhrVh1NHlpGgpx5KZlP5nRRqvXaVQWEaLhjWKf670Je9mSp/MT5r5g6bP1csBa9pCIZ+UzTHeeDNJkPce4aJhbx7zNcN5gTfg97yO+HohsqPMNNthggw022GCDDR4LNs1AG2ywwQYbbLDBBhs8JnyFMO13DN6RN5rRwvxWhDvC3OVEA+Wxw/iILy2mCTSDFFsF2t2IyT3yVsHgIHT0gBCskFTaKq1Ud8Q2QnEMg4OIz6A88bQ97f4tTwKjtwLNyFJvG8QGelsrquWQwevaIdj2IZ9GQqq/Z/NAvW148GCbnxkP4G5JPpZ1vq4EpSYw+lk+NGF/uOCXPvUCZpJQPhB8rvSe0tVKU7lBXOdW9z9SMuw6kX0G8VtqqphjTlNsJdRXWgZ7S/7JZ96LHBQkC6E4gdVlqK+09D+fQoCtIzVzXj4VSMaG4ZHSY+kq0gw6Gr6JSNTsWeOU0k1naqJ7Ts+E9Fxy0BmMe6XEjBeSA6XgXdGZnseLrvRkpVSjGxiaSw4zNl0rcEfTZR3lk0P/QcCVwtky56NHT3H26h75iWjH6xb039b/Ta2fn72vJu837PRXHMxHHJ4MST5XYlr46b6u31Zd1nMJ4b1zirxletKn/9lsvU+m0ePm+mr0P3l/S7FVMR8s+ez/8QK+26+QdPRqqp+ZPe+JeSAZJ+x8CtqeYBsoTgPLy0I2g5CoCfHwTTj+1giXanio2xjNI5IEZZ/WhsXtViQ7VQP10NGV0Xb15XSfypNAtWtgqSb055Sxcbq87YyO01pDDqKB+NmSf9h+CDNOyE5Nl+OuyyWVZhbbFSxut+RbNe1B73dR1RK67emo1PZWRZymZGeW4VjXZxuYkOKH2rXrM81I51LF/bMtfurXbqzlLwhU76oxaWDwkRLjYeRhcUPp/+IsMrzb0AwF1zM0fTA+KhU/EnwKy+sRX0TyU7Pu2N26E2j7giuEtsdaCtIuOwPuppPfPAy0PaHZEsxK6/hcAUEWsLUenN7DSDSC60F+bvkb1eEin8DZtODHP/NtrMYF5ZupHttcj4Pr6XFLVrB4X8ULTz3kVx8+y4ODbcw4JRvr05GQ6PkSk27d3zqjzFumkxKTBORzfUyjRvcY1t30vh/oX58TIyxOe5RvpiQrPe+JoqERTr8jv7bASMR/akSygO1ZVwOPPKBJ6ki9rdTx+KWIz4X+23qsbA39BzrItlJ3DxDGL0aiF7Y/a8jHkXQRSFaBatcSTbd/qZDNOzP5ngcsozf0NdNGqh2lVYvxhWl4OofPne3y5mSH08/vkMwN+Zmer8lca63Z6uruw2ekiefnPvk+zeR2kE71OJy7P4QMqlsNkgSSuwXDezq/+wxWVyw+Tdh+U6h2df5rtz3PXznmI2/fpLnXJ52oY0czioS9huJzSuVvnel5vLglJDNDeajnCnyR9KjRWqz2wG955H5CM9Ka6h1GNex3EVcIrtSu9dU+pKeJrtNdHAdiJwtJ9MYnWQKt0B6WDO4aHZsRlMfdcM4hm8aLwIU64bWzS/zanVv4SaY57C3r2j0PElm+u6EcVqwO+wxfs9Q7+r1E3Y5kodvkC/1nXpyxcimvBsNP3P8A6UG2Xpd0cxpRP5Of6nafXxMJ3f6cS9UCZPNImkBz7cl5ivhHgTr/xs412mCDDTbYYIMNNvgGRIyPx7BdRH5ARH5HRD4uIr8pIt/5yHs/0oXbvCYiP/K1brOI3BGRT3Tr/riI/M+/12fekU80N9hggw022GCDDZ50+PBYnmj+HPBTMcYoIt8M/CPgRRHZBf4rNCAnAh8VkZ+KMX6tcYrfHWM8/moX3txobrDBBhtssMEGG/wh4HFQ5+dJih36XFj3fx/wszHGUwAR+VngTwP/2xevQ0ReQVnvBnguxjj44mW+WrxjbzTDbsOyyegdRtq+YbWf0AyFfBoxTjVSuYvY0jEYVsyeyWmGahvhC9WXtUOhOImYCKaJmDLiS8FnkE8C0QjJqrMcGap+SAKkC1gGYX+44K3tgpDkmAbKEz3W2Sxgm0gzMKTzCEtLmCZsf67ThaK6EgmqRUqqSLIUxlXG2AaiF4ojIR+j+qtCNSj5JNL2hXgmTMuCmAWMU22XzzTd6E/d+izv7h0wtNXaruR+s8M/fftlHowzYiK4vpAshPwgVUubqDqgkEHca6hHhnys2kTXU50oEWIiZFPVh/kC/K6jaVPyN1W7Z1dq/2Rr1Qe1Q92u4due4z2rOsAZ9A80zaneSTCt6rpU1xmYP50gpQNJVINmYf8TLc2WxWdq15EuI73DluPacnbWp7yvOsdiHDCfh3ok9B7ocWiGhrlE+oWmBc2qHDGw++lASISQstYA+azTmk1zGpOR3csoTlQHFUX1qiEROIbi2LG6kuL6Dc4b6h3oPehqM4PyKCARmr6hd9eyuia4kacZJoze8qx2Lc1I1rrWkAi2Ub1bKD29smF1tSQ/VU1SNoVyDvm408ye108RCKlleDcyfLPBF4bVpQSfCr1jTzMwtOfHsO8wacC4kmyi+5WuIvmZxxcGCRHTRE5eSnE9SA4yyiPYedXRjCy2CoRUNZz5WHC54PopTRohidgKsplagvmxkK4ixHNtqTAzBW4QIKoO2hWqhYwG/tPv+qfs2fnauuRhu8WvjJ/jN/wz+DpBpprokzzMNHEmgbRWTW1zucWXCdEKxKyzyhLVzDaqoYui2tJoNXmp2VLNFxGmNw0hgXxyfuwiriesOl1ZPg74THTfLYze7JKRusSx2W2DjFpcmWrKmAgSIsVpV4MDQ0j1Z8Qg4xR3kLN1AKO3PPWWKpxsHWl7nc61iszHGc++eMKiyFnUGbN2QPZmgq3V3qYLE8JWcDwuaMsEjgqSU2F4N5KsIssrhvIoUO0YPUcby2nRw+SO4l7K/ic8zdCQLHVuNM5gW9WfhleGrPYhjCJmIp1dm+p4k1q1ktHKWrsbBVw/YGuD60Fb6jl1rt8uTjp9ZwRbOKq9lGSp+m1XWvIzD1ZoS4MrdV60lcFuV7ieY0ZJ+UBIKiFZqaZ5edli2qh6whSOTweYs5SrvxFJKo/PhWagtS1BtY2uEKZ1ig8G2+icZitIF1G1fxGG9wJt3+B6Kb5UXW87EIqzSO8osLy14M8++2le/Pb7a/stI4GZL/kH/tu42yRUgwSpDcncYB5m6/kcuah7N4jEY8GVnUVcp00U26WEddpk6TnmNy29+0IzhGwmRBtJlpF2V2sGoLoc8QNPskjY+nynj71/oeMcvN2y2k9AYHnVXuiRe9B/O9I78lQ7quW2TSQ99hBhcpRy6HYo30oY3o3YWnXy4sGVQv+hJySCKzOa5zx2u2Zxs2D/o3qsmqGeP+f7P/iMY7VvOb1tOVv0mI177P9Suj5O9UjIJ7r9ElRr3fQNSR1hrvr0aLv5s7OBKo/12m+cUJz+gd5u/L4R+Zpo8X0R+c1Hfv+xLnr7S0JEfhD4G8Bl4Pu7l28Adx9Z7F732peCBb4vxviWiMy/6L1fEBEPLIH/Jsb4019pw9+xN5obbLDBBhtssMEGTzK+hqbz4xjjB7/q9cb4k8BPisgfB/468L186RCdL7cJA+DL3ZJ/d4zxWERuA/9CRF78SvT7phlogw022GCDDTbY4OuNP6BmIBH5y48051z/gq+I8ReB2yKyjz7BfPqRt58C7n+J9RVA+UUU/O/e/BjfAO4Az32l5d6RTzTFg0xSsrMuWcVA21eLl7YUJAjGRdqBJclqhkXN0iuVdp5KE1K1u/BZlxBUitKWHc3nSrNOEilOPdlMqYFoRa1jTOTBeET6dt5RRkotE6F/0BJyQ3nsqPb0ENiqo8g7+jWfeuqRJV0GfC5KIY8zpq2hfydRG4eOMk9WXbLDeapRoYk/sVJbCgn6B0s+DpS2oWcaqi6doYopddDUBmyERvfrPP0EurSIBhDoDWoW04KQ6Ped2+mcJ6rUO0qLRwtmZimOlca0nZVQFKXd0kWXWiHQ9pU69Ln+S2eekBtCqmlN6TJimoAru9SdXku9m9J7oCkYbd/q2Atki6h2P02AJiVUlnq3s9mwgnSyCdtEXGmUylwmzPKCXllTrTJ8ZVntGbWiskJMlDYzTmkbs+ho/gX0jjw+U9opWUVWe2p340tDsoTFNGeROVw/sroi9O6rBUc0ur1JHUkqwV5Z0csd1aUtXE9TfSLdunKhGQBGSBZglpblwYDcaw0gF8dqeVnpqnNrG1MLvlRqKlqte9Nqsgoo/YWISjhWCQGPrfRYqcwhrsdd0G2xDfheoDg0FCexk5xE2oHBNhHbKO0uHrKx0GwnpDOtxZCABOnskyJtKfhCyKaa6tLccDSpZbVvIUB5GomXaq4mYzyGKqRYiRiJXM5nhLbbts7+JZuodZh4dBI3YApPqA2utCyv6PZHugSpvBs7geDA1kI662yEgko7Qqbr8fW5xVVH2wm4vqbOGAfJMmhddxR6cepoRgn0HcPRiiovMK5LxXFdvc66sYpCyITVvhCToJZfTs+PbBrURsZpvcQWyiNHMk85qQfsZEtaZ8Ep5W+c7lN7bvFiQSqD9MAnEQk614jvzlUjpAuth6SKyNwS6GrIdGPouzlM9FjRjV29HwmJpvnUO2BaHYt2IJj2QnrSDiE/Ucp8dambX7LOSmh1kTzlM0jmQmtzTb0qhaQG8Tpfp3O/loXYKpLOYHdrhg+GtxuLPMjwOQzue5qR7Sh7wbiArYWs19BUhsnthGys53VIUNlFJ9PxuVBPc9rcY3K1tTsfm3pLa7ztG3wK6UwQJ7Tbnmgt0Qr1yLLVq9hJFvSMSnKaaCFaliHjbKVJWhp0LZha1nZF5/NqNHruJku1JVrP917HyBV6fJJKLYbiNCVdqGzF9zUhjCh6HklnZ+YBExGn10WXdxZmRuvftDrGrpD1OeXLQDNSa6hz+VC6jFTbRqURovUbigCNUQu6QuspXajVXT7Rk7M4bkiWBS5CqBLyscF189d5yp7PdG72uVo/+TphVickD1OyWSRavYZjdN99LhgfSecqXzmXMVmEZBq6eShSb1myeejkWY7+/ScjGQj4mh5pftlVxPh3gL9z/ruIPAe80TUDfQDIgBPgZ4D/VkR2ukX/VeC/+BKr/EHgn/1e3ysil4Fngbe+0nLvyBvNDTbYYIMNNthggycd4fF0nf8F4C+JSAusgB+OMUbgVET+OvCRbrm/dt4YdA4R+SDw97plP969XIrIX4sx/pfd7+cazQz4qzHGh19pYzY3mhtssMEGG2ywwQZfZzyurPMY448CP/pl3vv7wN//Ch8fAH8rxvhfn78gIgPgb3efv/W1bs8780ZT4NZ77vPShx+Siic1AR+FcVsybUt6tqHyKS4a2ukWAH/p+3+BrWRJIS2FaWmjZRFyPru8yqwtSIxn3ha88vAq0xct7UlBdmxpdgL0AzEI6YklWSh1ZO8X1LuO3CttpBSv0nUPP5it0zgQSMcWW8PsGSBo53S1l2iHYVDqBMBUhpArJRvlgvrzGiih3axeu6uzsXYaapedJp8g8NN3XuaneZnm01vYWlMpkqcXNPMMO7XY5rx7tKPgEqVgi7F2Jp497GNHLc//wOsAXCsnpKJdtnvpgt1kjpHIxJe8MrvBe/7kAT3TkIqjMC1j32PiekxdwcLn+ChYibw6ucSD8YjlImP8XgsmAh6cUDy02MpqyoOB5qBPNjFU+9oNOn/KELpUi0U3BvaZEpIWuuSWdgsqr536IYU2ueiStVOL61nGsyFmYUlWhnYIIROVBnT0Wz6Ja0qrOFF6dPaUJV3ELk3KEEUINrK8ZLFLSM8sy7wg72gtTd6J625IBLJJZFqluGVK+dIUJFILtM5iTSBLHd+0d0RhW14cHHDYjPjVh8/yr3/Xx+mZmlQ8PVPTxoRlyLjX7FKHhNOmz/3liLunO0wu5cxupfTuC6urEdNoQsvormP2VIJEyI4tIbXkk0gzEmZPC6axa/rTthcsj10YTK1U+mpf1t2pEmSdaiVRqT5byzrF5DxByuVKHdsuqagYe9p+wnKaYFcdLbnQlKToDf/T23+cNw73aQ96SBDKW1OsCZjDvKOX9XjCI0lSS61ZOcnoPTPjxfcfKs0cLEYC2+mKnmkYJBUAE9dj5VNSE7iVH7GbLLAExr7HsRsysLrcnWqflU/5yOFNMus5vD7CLRKyo5TiUNNh2p4wvZXR9sGME6bSI3txTnVaICEhpNC/D5NblnTeJb04nTuyU5UN+BQm77Jdt36ELvEkmwXqnYRkBW9OdzhIhrRNoglig64zOnSpYQ5iCsnC4ENOfqaJWtW2pjmFVMc4ZF1yyiyQTiz+csX8lsHnqo1J511HuKjTRHHaOQxYlZP4D0+5vj2hcilX+zPmbcbVcoaRwMA2bKdLfu3kFu/ZesD7+ndZhow2JgztikXImfuCVxdXqHzK0qV85sEV4vXI7HnLTCK+SqCySNOlMdVCNrYYB28fbxOcwZyl1Dua4DN5V6I1Z4AMqh2lYtsqgdZoes9l8NMu2SYBVwvlcaSoAtIYyD2m0a75ZqgSHwSaYZfcdRYoToTlFUgnFtPqMctmkcPxgH8cvpnDN/ZIlgJRx4obFb41FK/nFBXkY52v255KSPJppN5SiVXII/b2lG/5028SokpEcuNIjF//DOCC5d5qm/dt3eNaOtbjJZ6+qVmEnLeaPU6aAW00LFzOr711C38pcvxcRmwFs7JEiWta3TRxPSYAq5st3E1xpRBSS1tq3TRDg21U9pROOnlJ1kkqts06jaemk7ChtRRmKabS+qv2QZzKI0KX4pMuhGzWyXvmFlOr5mq1p3WaziOm0WNKhHymVLutI9nE47q0vgsK36tEyqi8xReGNrX/kjcZf0CIfGGU1pOBTwFf7JFZAX/397vCd+aN5gYbbLDBBhtssMETjict6zzGeAgcftFrDvj13+86NzeaG2ywwQYbbLDBBn8YeMJuNB8H3pE3miGFv/LMz/Nidsh1a/BEltEzC4Y2GoxEQhQaLKe+xwO3zff0Ps+WpNR4ciypWJax5c3eqwAU4lnElF/eeYGH7YjfHt/gjaN9bozm9NKG1lvuDbapxjkxiUhj2L46ZWKHNI3BLA2h7/UxeUeZE1GKuDV4D2GnhSDYoxTxQsgituoowY5ql7ld00ME7UZ0vYuOxZiDX6lBebBKmbsS+vcj6TJy+uaQMPQkBqpLnvzYUh+XpOOuk3qixtSrS9oVmp/petqeGiv3Li/4oed+i+/ov8qeXWCJjEOJR7hq5wxNwCJUMfKB8g7Pp2cYoBChEMs4OGbRsgwpTbQUxpHhubO7x2eraxy3A37u/gv4YNgpl6zajIejEWGl9C5ZQKYJ9ZUWSSLUBqk7l65uTCUqHZONagDqMgMTCUmOBMGnEHI18xav4xuDweSeKOCBKle6N5nLeozbvlKbbqcl3F5RLTLqJBIfFkirpvqDex1FU2kHZciAZUI7jF1Xvprh09GQ58b8NvOkmeNvvu8n6UnNnl2QSqCQQCGRfVOQisXHwB1X8eHha3wov48V8FHHNwBVjMw6zquKCW+7bX5p+90sXM4nz67y9t4O5bBmNc9ZhhxfJNS74MpIOtd9PnkvxCwQ0ohpTGfGr3SatErzuC0PkuD6UF/2iNPu2/NlJHb0udc69WWkdYKplUp3PZCoJuE+h9lTKhVJZwZfRNxAqPYhm0D+Zsb4Won3hsEzUxbznOXdIdFGbHculYeQTSNn7xb8tjoLLK8KyUopyB++/THe13uLbbNkETMWIedGcsbQNGx3VGQVhVlIOPF9RqZi19bY7iqwiJaia22f9RJmoaC0LVvJil9Kb9MGy8OdIdP9kuxY96Xdc5AFkl7Le68/4E/tf4rXVldYfCDn02dXOJn1qWY5ycNMXRnopL0c0mAAACAASURBVCGXHXjB9S3triM9TXAnOu75VE3Tl5d1Hjn83C6mNthKyJYqGZCgxv2u1OMZUlg86+jtrlilfcyoIbxeQuw62y0UJ3HdTZ2sQFKPHzhW17uO7JXBNNoR3TvypHPP7OmMUAReeO4+/8a1j/Ge/G0u2yW7VjAIA8mwYjAYAoFPbv0mQ+MwwENfsGtrUiIBaKNw2O/R0kmW9q/xyfl1XLA8WA05nA/IE8+izljOc3wQ6isWaQQ5LIh5wFypqLctdW2wM4utzgMPIsWxrM3N08srml6GZB55o6A8hHbv4toRrWBWQugZfC8yvyFqJm+00704i2uXi2ak3eDRRqo9dXYImdrRGIkwdMjlFvvJvhqIf64kX+p3pQudqya3VKpTHkG9pVS9eEivLfmLz32M7x58ir5pOPF9ALbtir609CQQgHFI+Vj1DM9nD3kmmdEipES2TYIncuw/RxUtVUxYxIxhWuGC5e3lFqdVj8OzIUXRkqeO1lnqJsE7i18k/Fsf+g12kwV337/L3Gf889efRwTiw4JoIslcxyikQefSYGi2Lzq67coQTSRmEUI3P9QGgtDsBYhao20ta4dH1wNXKP1td1fkRUO1yhhfF0Jj6d1Ju3Wfd8brHGVrqHaNmvV7WO0J6cDiekbdZmIELMVZQPyTcnf3teeYfyPiHXmjucEGG2ywwQYbbPBEI0J8PF3nTxQ2N5obbLDBBhtssMEGfxh4Uh6uPka8M280TeTUD/holTPODlmEnFkoGHfUg+0osDZaDJGHbsS/WD3F0K642+xRmJZtu2TqCx66LQpp2bJLAoZPLq6z8imVS6jHOQ+cQSQSnCGe5ORjg8/B7TisiWADiBDygJ1b7ErNecVrx3RMoL7k9a+aWYJpDcXJRUZsstQMZldAsxOwOw2tL9Tw26oReXamne221qzwekto+0oNFydK70QjpHMhbKuBcMjUuDkapTpDGtVsuVUqKKRKiUZjO3pCafrnLx2xkyz4rdUtnspOOXJDDppt6pBwLZuQdy3y23bJqRtwt91j5kt6pqZnaia+xywUtB29u5Us2bVzZqHkbrXL1BWcHI6ITlhuZbRVgpxmpCujZtdFJCaQTJKOltUOUFtfNO+JV0qwuQltnSCzRI2FGx1Hid24NZoRPr8Jjqj0fG20C3LR5aYv6PLitbM8GpA00CsamiYhelEKqe4oR6/5zoGu+/dyhQQhhBRxwupaRBohWSn1ZtrOtDrxvHz1AU20jP0OLVYdEKQllcAy1BiBI19w5C9ThZSfX96mjZZT3193/tedc3tu2vVrdxZ7jNKKRZOBN6xmBWacIEFpqmSp49gOI+I6M/EgJDPWZtGup+b5xoPrRcRGfC8gwWDnhnShtPh5FrStO+NpgcVTF3+x59Ouy7artZDqsiERJs9BeHpFqBNCpl2hjdHvXTYpMQjLVUZYaGY4idaCOHV0iEbUBSKJNK0aVfvOpLowLXeafXU88CVH9YCnijO2khW7iYZfGCLLkPGw3eJetc3ANhgJ7KRLTHc1SLtO3zYkrHzK/dU2lUsZL0vqWY40Xcb0QpAjpdCbm3CjN+bd+QFGIp9dXmW/XCASmaaOReEIJmpHtDPYXouvExwgjRCKQDvU+mqG2vWczkH6QBoJpSO9l9IO9XjaSk3jjYdiGTEeJAsYo7oS//+z92axlmV5etfvv9ba05nuFDcyIiMip8rKqq6qrp6FLdtytw00NMiNHzBCgBrUhhfE8AAYI2TABsuYF0tgkCzc2BJSG4vBWG1jq9RN0UDbpt20e6ghqzKzc4iIG8Mdz7iHNfDw3+dEVlKVmVXurM6s2p90deLss88e1lp7nR37+/7ft3HISK/3/KqvDD7Q+UFN0qE9q3CXdictaWfaT95DN7FAbwSe4JOzR2QS+GJzizp/xFlsyCQyk3rX73Wy3At7dN7xWnOdl9dP8UJ1SkB28oQuWUIyHLol61Dgo6UOGW+cHRI6yzwIsbHYhaNXO6gsYxa0yn7jMHNHMuqqYDqQWp0R3KqXFgVDO7daab3WYARJKheyG203DIS9jvFsw3ptCcGwsSoHccveiD6nv54jcd+TOsEuHOtbifXTkDaORak2Ab6xSLbNLo/4kRqxh3wbCgLdXiRZlQBtAzjuHF5gJfJ/rV/iKTfny5ubRAzXszl7ds2+XdMmyyJWnLT7fHnzNB8vH/Bmc0RhPNeyBbl4FkGlTU3MKEzHyWYPJ3FHm4eNY7nMWBmgE6Q1KpUpIk9lc14sHrCOOaGdcv1wQYjCuSQIQjyG5AWucmIVaYya9rulBl3YRsNI/EjdKPI5tHtCu5coHpvd/JeEPlxEjdttq7KvLPeMi5YQjV4fqZeK9fPSbq6pU++CoMEkMROqM3VRsHXElQYTEuITbuV388uHA8MTzQEDBgwYMGDAgAEfBIYnmgMGDBgwYMCAAQM+EAw3mh9RJOFP/8o/qSa+5znJJvILs6tINv0j+mRh8z01Lg/4hxXlQ0Moe+PiC3Ymr7aBdh/qo4TcrMlypc/GR2vyLJA0LprsYEmReXywnM3HPLt3waePHhARYhKWXUFMwmVTMd+U+KAV8LbOCRe5GtMC3aTPoG2Vsk69sbdpDaGxuFYr0mPem+r2nHHI2RmXJwvNzY7y+5aMAGMSIRjMsiB6Syyimg8fREzXGwS7RHNNKV0/iezduiI9LTSdo+ssjTe0wfL5s0/w5UfXCaE/3toRW6UipDbYtSE81ZKPOtqzUvtEgNZgG+mN6OmNjKH+WMN4VjPKW7pomR6syTPPYbXGR0O4aYhJmOYNi7bAB8vNyRwnkbXPOK9HGElkJhARNm2GNYnceR5dTekeFNiNVjr7sVYhb/PYN71hOEA2a4jewMOSUCZCCX6sdPk2m1oSsHRcXB6Sn5tdHrVt9E8zniGM9XVLNT2q98H22zCGzin9k6xS9CFo2/7Zr/w4jXf4YAjR0KzyJxPRxiGdkPY6jI3EiwK31D4wnR7bloYyLaxvR0bPLDiarFj7jKcmCw6qNSEZNscZXR9Un5mACFytS24fXHKtXBOTEBHaYFl2OSEZLtcVIRqujde03nH6pWuEQis8Q677dRultqDPu29VzpGteqmI5Ule9EgIFZTniWQT8U7NbG9DjEJ509N4iw+Wus64Opsgc4dbGIzRSuKIIT8XYtFX/hrIb6x5ev+KdZfTBsumySiB//PsRd682qftHDFppWc7/4S6PiRgbTFeVA6QRahtb6asQQnba0QSFGewfMlTHm1o1hkmi2RZ4Pj6FdKzYEa000rXsWgLfvPiJjfyjxMR7m/2yU3goNhQWM+sqneVp623tMGyCBWxghSEaBJtbmkBglLi4nV9O2sxJtEeWFKWVAaTC34CphV1OdjoHLGal5iNRTrA6DXYTZ/01Tb4IRaJbL+hSwVmZbGNzjcgULDLrPcVkEdO2zH/y8Mf4I3LA0I0xCi0TUZotXKYpUO8MHv+kuWygnulSnd6icp2LrYNmAbGP/aYzAYenM+wNrI/2ZDZQG49XbQ4iRpkYAO1d8ybkmdml0oHNxVNcKzbnKZz1HUGSWg/3ZG5wAxYzCtSFgHh8nsFCQJeVA3Q55qToGkypYfzRMy0P0Mhu6xwCcBxw82jOT4YRLTfQxROz6bUdUbqDNSGUOk4S3kixUTsDCnX+d2PEofPXyCSWNc53lt8Y1k0JZ978Enunu0T++2HzsDK6bgNonPtLGh/LXNYW5CErc3O3cQtVa6VDIw+pfuZFC1tsOzP1kyvNfhk9FrzekuQEjgb+ZsPP833HRxw1o44b8YcVSt8MhSZx0mkCbr+5bhis9BKdIIl5r0Rfrad57S9QqHLsuWTfu+m+tl2Hg591nq00LWO825MOC0xrZD116HpdAxK3JrE9w4LVo3kk9XzrQ8MxhtcnbC1mrWTHBI/JHd3H07D9t92fGfeaA4YMGDAgAEDBnzIkeJ7r/NRx3CjOWDAgAEDBgwY8DuB4YnmRxPiofxCRXGxzTeF8iJoVZrps5DrSMiEk4OCYGD0UNRAN+h3RJkVqseRmMHoMVx1hhvff8p+UWNIGEn4ZKi9Y+Q6DvI1RiIrX1C6jqNixdTVbEKmVYD5ii4ZZnlNPcpYdjnrLudR57DXN0pfRzXeDklIQSma2PUh2T2V56+3Sp0IiCSazir109Np6mINe9eX/NCNt3btsvYF90YzfLBsOkeMhs5bpUmDYEwi9p5eFjgcrXEm0gWLSGLT5Xz18bFW0q4zCAZaAwFM/70kKP14ldEu3ddQA5LYUVCghvAA9mHOCnjphcfE3ug4N4Fptm3nSGE8B9maZSj56uKY58ZnFMazCRmX5Wi3jpGEj5aI8OunT1OvcphE/EEkmyjNaPp22yIlIbUW6yKTcQN7G+pGq7dFEiKJrnOEzhAbhxu3+E1GXVgowq69ifK1ehuT2OucUmplwDilWAG8Nzungm4GuY2a7/z6CNMo7W4SjDvt81DRV4RDsyrwe4Hpq4ZYsDPejnZrVq+v4zcNaz9j+kPn5DYwy2qMRNa+wFeGkW1xJlAYT0yGk82M58bnOBNoomNiGyKGy65iEzKOekq9Dg4fLPb2ejdexEYS0PX0ITYSkprfrxclRB3PYhNi9b/wsbUUDzKWtyFMI7NpTWYCLxzqMSy7grXP8FPD6yfXSHmi3Ytq+hyV8qxfbDA2Qh7wSTgYb7g5nu/GwGVT8tbZAS8/uo5vv7bSVPq+SElgHIgJ8FuDejWgJqljhLaz9A4MgllbOIKX7jzkzviS3HgmtlEJhwRK0+0cGF5ZX+fVxTV+4eFLvLh3ylGxJCZDEx2z3OAkkpmAIXHRjvjqxTVsFslLz6hssRIJSeUjnbe7NgeVXCDgntrslqcohMaSvAGTaEUDAYyNdHsCJmFcpItaxa2NQS9pURcFWWfI2CN7Otf4VUZYOIpzw+ZG7OUiqiX51fu3qc8qJl/RavC8hqyXMpiud4EIcNUcYGsNNUB61wUPEtXRgV6q9OjVI+y1mh+8c5dJVrOfbbCS6KIhJsPMbXbShGUouLs+4KXJQ0a25bSbsPIFC19Qh4zHmzFGEkflmrXPaIPlqelC57wkdNESospUAHz/2nlL3WSUtxdYExFJpCRs6hzfWFISxCRm0w2l63ju4JzKdiy6knlXYk3ifDki9DS7Sg/6cZVFwjRhc5WsTIuOj+2fYSSx9pr13gTHsiloQ0W3zDFrSxSdU1IekUx1HHGaIBi6TUaK6i4AEFyA/vcjlGjgRxCuHk345Iv3OSzWXLYVhsQkU3eFmAyuH4dtdJysp5ytx3xu+RI/fufLzFzN/c0+bbTs5/Vu3Yjwhe4GbR7wbUacBtJhi5hEDIY2PPktkyySWg3HMC7hehla1zgdi7X92vn0IgejMqvoIFQR0wqdEdJ+6ml0dXpIVsfwcus+0jtb6FARTGt6BxF1TOBv86GApPde56OO78gbzQEDBgwYMGDAgA81+odH3+kYbjQHDBgwYMCAAQO+7ZCBOv+owraa0V1eRLJlxDaRmBtsl3CLDtsoZdbuF2QLiwQoLiFfJs1eztT82dWJ0YOaUDq6iWX0KPHa/WOMTTsBb2ycVvpNPPm4VerbG8TAW48PtUpwnpHKoFna3uwoRDZWqwYPPeVezZ2jC3ITKJ3HkHC9K3EbtVL28WbM4/mEl44fU1qliZUKzamDY+OznbH0qs25WIz4+yfP0HmLMZGmzinKlvVVpcfc09omwMHHzzms1oxchzOxr5LPef3x0a6qPgaDXzkkCmZlMW2fyxv7CPe+Oh7ALXlSmSlPqg7p222Xg01f/U3Or26e0/cuISb15dxplwM+rWqsSZzNx/xmvKnt37cngFilBK2NmtW7cWATz734gCrrmLjerP5tBrmGRB0cD1dTQhLGectBscFHdQSYZI1Wt4eceVvwaDmhbjPMuKU6arkxXVBav2szUOrYoLKKLzy4wfFsycefPyW3nnFvAr4JOWuf88bhAXXnmJ9OyE8yur0I04h4wdTC6ETlHkkAC2YD43vAPavOBAso5lFppVz3ny8ikqAbGWIhfOlXn9M84YOg4642pCxhRx0uD9rWgDWRV06vsVmUOjb3NYA71A4aQ+opLbt0hInn4IZWsc/yBmciudFK1MJ6qp423sSMh5tpP1ZTL4vwPU2Yc+/pPayJ7JcbnES+cPcmi01JvcxJrdX89GlHmmfkZ7anWvsxc6fjM8/ep3SeWbYBennIao8Hl7N+cEFzVWi+cu/cEDMdhJrNrIb/28zlrWRmKz8g6TiV8DaKq6/mXc5K3uCQ8/UIa9Ku/3PnqVxHaT0xCY9WE84uJsRVxt29fYxNKj1JgthIVng1UwdiEoI3HM5WzMpa21aeVAu0vVNAbgI+Gb786DqTsqHKO6ZZQ+l0nw9WU5Z1waRsyGygC5aHZzPERSazmmf2LzCSWLRaaj7Nm6/Zx4PFlNuzK26NrnAmcNGOOG9GzJuS/VLb+tFqwmpTUD8YQRLqY8iW/byYaXuFos9f91CcC/mid5yoE/kiEp0gMWE6lQJ1I0N317DKcl6dHpHbQIhKca9rdWCYjrVKfyuzWdU5L2fHGJNYrQuVuHjN0zaXGTGPvDXrSJ1FskA1abi1f8XIdeQ24CTuaOAtzpsRb13t8/zBOfv5hplT8/n79R4P15Ndm+UmsOxyfuXBHR1/mxy/zpQi7gzVW07z1l0vHVoaqk9f8NR0wWGxwZnQu2fkvHZ52Dsl5MQgpGgIndEQiSBK+zYGCVYlSqXSxSapC0G25Guejm3n3u08q3Or4ZXJsQZNnGnfJ5vAKq1vyoDZUteSCKcluMTfCJ8mBENzX0NP0lQvEJNFUi+LAeH2s6fMippZ1uzo+K28zBkdxw9XU46qFbO8Zuqandxs3pWc1SOVkUii9hn3XjvGrA3+esfR9Tml83TB7qriAUI/Vzv7NikMUPfuHZ1XqUPbOq32XzvM5kNk2D480RwwYMCAAQMGDBjwgeC7oOrcvJ+VROT3iYh9x7If/GAOacCAAQMGDBgw4Dsc22LZ9/P3Ecb7faL5t4FfFpE/klJ62C/7b4EP5c2maRP7r7QUJwu4+xA53CfujyElzNkc6oZ0dEC5bhk9OCQUYLpEPg+YTjNTYyZISLizNc4ZTDci2Zzrf6ugvAwQNTPVj8BtWnxlMT7H+IgfO5Y3He0UiqtEdRrw1ZN8YLdJmDaSXCLayMnvddS24IGbkTl9/C+SyK1SCFuD6at1SX1R8ng8wfWUD0ATeoog2ifVmJuC8LAi++IY62BzDE6gGZWUC2F8T6u+faUG8YtVSectmQtYSYQkLFYl4d6oN7XWgV7OBbvR/HXT9Sbc6oGr5vHbwvcasnXSc+0iITdqlu8TEhPRCraJvamvEF8VzWNfRbqJJeR9xngTiZkjFCXV44LFnYybjwMh0/WTU5mD6RK2TUhI2EarYk9+T0Z903O2GpNnnoueupF3lPk13nE1HxG9UE8bLtcVvjejHxctuQu03tJ4x6bO6TZOz1Pg1I6xPSW83bbpK9VDNDSXBQ+8VgxbE3e0UBctrXcs6xwfLLK0ZIsttat0l1vDwVc9zZ6lbra0uNJl0QrlRcR2ieK8JZQWP7IYnzBN1LG7Eox3jO+pPGJ8L5CcUB/pwZuu0D4Iamb8+AcKbAPXX+1w65b6uCTkkC8jdhPJ5g0xtyyedSxvZ6z3c1KCZVPs2jJ3gcwEsn7sNt5xthpT5R22b39rIk4iXbSsmrw3uTY03pJ/ucJ4OLjoDcXHIMn2wQQ67mAbouC4O98jt4EqH2MlMm9Kzs8nyOOinwuEaq1jKVuqgXy7Z1XK0Vfxuw27trWN0ri2jviRIRkhn3vyq5ZulhNKQ7SC8YniIsN4R7YeK/UL+Eqwl4EuJNybV4S9ivp3T5gGyFaJmI2pHgeKKw9oX7X7JfllSzfJWN7OmL8A9adaIsK6y3dttp0LttKMJjjWFxUcQEiGZVOQ9/Thoi7oOkuIgjWJTZ2RTgvsxrBIwuO868eh2ck8tmO47jKuzicqB0B28pLaZ/hguKwrjCTaztGsM8ZvWc2fbnXMZhtgw+5Jje00uzpeCm6TyJcJV0cICdOCbSNEiLkhj5HZGzB6aHH1AaZN5F1C9i1l0Dlm9CAnWXUAWN5yZIeCW8Ds9Y5p0YdXFAbbREZ35/hpQX7/iuWnrnHvRx0bgYd22jtQJKxEiszvzh/galNy9XjC65KYFGMmuToKzJuSRV0QomHTZYRouLoYMfqCBlPIDLJ+fOZzNGM9VzkWooERi3ml83lekdmgmff9/iQIZmPUmL8V8n58ajvpa3WqTiibI50rYqbXxuytiGmTzqlOMCHtKvmT6enkA0Pxf4woriLZOuk86tQBIFsbjBfEJ5p9y+qGztnZIpEvp2TLQDfZSh8M0QnZ0mM3gbs/NmL9XMeqzemCZW7L3bgFpbe3c+L5ckQbLFd5SeW63inEUPuMZZPvQg/azmJXRsdRFDZNtqPBt/PpdtvA11Dn6W0ODb6zSu93BjqjOffmw8NXf5BV5yLyI8DfBf65lNL/2C/7KeA/7Ff5T1NKf/mb3ObrwAI1FQH41ZTSv/Ju33m/N5ovA/8F8HkR+emU0i/x3ZAEP2DAgAEDBgwY8EHhA7rR7Fno/5y3GTmJyCHwHwE/3O/5V0Tkr6eULr7Jzf9YSun0/a78fm80U0rp50TkZeB/EJGf4btCwjpgwIABAwYMGPCRw78B/E/Aj7xt2Y8Dn0spnQOIyOeAfwL42Xd+WUR+E5VXtsCLKaXJt3og7/dGUwBSSl8Vkd8H/HfAZ7/VnX7QSFZoZ478kcXcOCblDvPgHDKHf/Mu7umbSNPCJtGNNGs1n4OvDNWyJb/oaA5LzUN1BpmvyN56AN//ApcfKwm5UD1oMOsOt2gIk4Js5ZGQMI1n9XTB+iloDhO2FSRBee4xXaSbOqUrSqMUXRMxjcW9kbP31yy2ibR7lmiF/MxjQkJ8Ilnh8nfnZGNY/+Z1JvcioUtaaWy0I7ed2U4M7imwAsVVRDzsvaaygNd/KjL+1JJZoRzkNsOaXzjg8Msd7dRiQmJ1w5KP4Zm/t8EuG+qnxmyOHfMXYPwgsf8VXe5nBe1+jq+Uvhid1IhPXH5yhC+E8iJgmohtEtlVS6gcxkfcxZrm5pRQGIrzlvNPjSgvIsXpBlvn2E0HEUzd4vcrQul2ZvtXzzqytR4DgN14pG6RuiPOKlLuICW6aYbZGCY/O2PyxoZ2X2nIbOUxjSfmlmQNyx8sKTOoryfCKyVmCeOztKveNl1ilGAUE8V1x+UnBPEwez1n7xVLqBy2CSQrZBc1KdNlZ58uKW5AMhmjzxn82FFPDdIHBow2gRFw9XxBO4VsDQcv1/jKki090mkefXEGo0lGdGpqPX8+oxvB/isdyQqm8fhKK1GjEwiCawPFVUt+Zbl8aYytE7bWbZrua9Xn+b0rUpUzuZcRM6XcYmGZfvWKZAU/K3HLFnO2QA7G2KYkvxKyz08YPYyM725oD3LamYUukZrEcqLjoT40dNegej2x/3+fsPn4MQDl/QX2aMT1TKnoUBnqZy0y1lz06jRSXHmyi4b7v3/K5A8+Ypw3O2r4YjMiPtpj9hf3MD4RnRCbSPOZjMrBjV9uEJ9Y38i5fNFw8HKkfNwQC4vdeJVdZAa37IiZoT3IsZvI4k7G+EHQ69KrC0WyAilR3J8TZhXSRaQLnH16H9sIe6/ViI8ggrSBWFpiYfGHWqEbSuWY9l4PhNwweXWhY2ScYdpAcd5iGk92uiKUh7QTi727T/EoMnlzRSidGpw7IZSWrtIK+uUdg30mstlMCJeG0QmUr7R6XpngNhFbB2zt+a2fHGOyxOgEukWB+7kjkoXMqBH2dkj4Qtgcw8iA/wdHXN09wHZKsfpRn6PeJQ0GOBSKPaV2D7/k+3ZilyPtK7OrfHZ1pB27XsKgEo6UKf0tPhEzowHbCL4QJEB55jFNQGKiOIOYGdy8RloP1hKrjPBCRraCyT2PW3XkFwF7tSEZQxwXmFVDfrUmlRmpdxMoXi45/H+UWt8c65xgehlPMno89e/KKAxM/uaeUvRWq+NdmzgIiXzeEfOC08+UjCqlkw9f9hRnDWHk+K0/lCHft2Q6WWONOjEYSWx8xqNffJq0LuFxInhop4IUcPM04dYR20WSgcuPOfJ5orhKGJ/Irzq6qcMXBtskZm96Qm7oxoKvVEoUM1RCFPU7ITd6vQtUDzZIrJg/YzGdIV96XJNwFwEiZPOG+nqFRJWRLW8pbV1eRmwdKR9tqN7qwBgIAQlJ59oukC1HjF7PcH/ngOrE97IyR8wFRMg2AURYfizDHoBfTjCvetqkc1a0EHNhttZ9ubVn/nxF9z1CqBLFWxmzn5+RLQO+UtoeUSlWtoyYTul+BNxa37tNH45RaXlJcVZj5zWLTx4yf/bDU3X+TVDn10Tk77/t/V9IKf2Fr7tNkVvAHwb+AF97o3kLeOtt7+/2y74eLPDjKaU3RWT5js/+dxEJwBr4z1JKf+PdDvx93WimlH7gbf9eAX9ERJ55P98dMGDAgAEDBgwY8A4kNAHp/eE0pfTD73PdPwf8sZRSEPma7X+9nX2jW90JcP4NPvuxlNKpiHwM+CUR+eS70e/veqMpIv/luxwEwL/5bt8fMGDAgAEDBgwY8A3w2yBCFJF/HfhX+7c/gWow/0p/k3kN+AkR8egTzB9921dvA5//OtsrgSql9M4nmV+DlNKrfXHQi8Avf6P13uuJ5tsf0/4nqIj0Q4/o4Oo5Qyj2GD1sSSI0L+1hQmJaFsRxwfqOyg0Wn/SYacejZxzuwnHwpYKYlWTrRHkRWHx8j2T3d/Tc1lQ43q4oLjNCafCFwdWR+sCSLyLtWFg/1/HC8w958NyMB1+cqYH8WcLVakwcnVa158tItx8gQTc2dGOlRWIBzaHbqYFnbgAAIABJREFU0VDRCps7nuKgpl1N2LSGfJHIVlo13I2N7ntmcU2inQl+HHFrQ3WayBfQ7Ds+88zr/NDBm7u26qLlYTvjc987YX0zo3rUVzxvtB03xzmb76n6bQv19YDpLBIrxieObmzJrzy+NEq5ZpawZ9kcCc0htFPH/mtKq3XjEl8ZinkgugmhNITc0Bzm+Kqn/aywuZ6TrV1PBY90+2OlhusDod1XmqWbZGTLjvawxNYZ9bUcX6oBd8wgXWuweeDyxTHdeMTocSBbeLpZRjstsLVSo+tbiXSjJi899SqnnmeY1mBbYXyvoTnKlc5ZeUxwdAdKNc2NRWJJtkpka4GYCDfHPeUtJAvtdQ9Z5K1/tKR6CJP7Ssl2E0MoDCFXU+vFZzrwQjspGT9MFI9rkjP4cUaoDNJXNadMqI/0O4+/v2B6N2KbQMwN7dTg6oRpIJQWyLn6WEl0ok4JIsTSYnqa3248fpIz/+wxpMTZZwztfmT2Sk4+TzT7e9g24daB5lrF5pNT7btcWD5Db9ZvKK6Ucp++vtZr7VrB5K2azVMFMYNulqgPhbPfe5PRQ09yQnttTH2Uka1CT1kKodB+Mx10YwFxbK45/AR+4tYXmNqagGBJnPsxvz6+xRf/kY+RLbTa2baWbgq+6p0XbhfUR0JzlGingm01v35zXOE2kWwRqK+XRKsuE+trluVtYXknI59DfpUoLyLd2NDsjzHdmGwVWN7KyJcRPwIidLOM9bGlPA80e5ZsreNKou5v/byOgYsmJ1SQzIzi0hOtsLxdMHt1jXQBOq+UZ29Ib0IilA4/zfps8Ug3UsmN8Yl2z+BurSnzjvlshB/nlBeO8sxjoxr9z58vsG1Bd+Sx044VpeZGZ0I+V2qxur/i4jMzjFea9OLTgru9YjUvCEWGaWB0mhg96HBrz+K5SiuP1zB/MRCtpbiwOwP2yetr1ndGVKcd7dQhCWwdWd3WfPTiSiubfWUprjztntMxetYhm0D7gmN5J7G8lVOdqktHMQ80M0u2yjA+4StDN9JrYfN8x/xjjtkrI0wHJkyY3O2wtac9OiCbt6xuV7RTIRx1+DZnc5xRnnnKiw63aLn6+AQTEvlVIGZCc7vDjVvq18fqlJF0TnTrqJ8f5lQna0JZ0s00y/wcx8ErQsiFVEY+89QDfnD/jd1cux23f/UTB3T3S9ymD3jw0OyrfIskTF6bE0tH+OSU9Q2t1M+WkZgZbN07N2wifmyxbWRxx7F6JlGdCOMTlXqYNtFOhOIi4EdGnT1yS31o8WOYnESyuSfmhmZP+2h9MydbRTbHhmRg+azuy5eW6tSyuulwG61qB5WDRSeMfuuS9U3w+x5bO5JkjB9qH1cPNqzujKmuWpqDgmSgfiqyfj6wuu2YvCGMTqPS/haqhcoq2n09lu5pj80jTZHTPLTkcyguPH5kVb4REu1MfxuyVSRbBtyyw64a5p/Y37mSRAe2zqmvVypDmH94Skx+O6rOU0p/Hvjzb1v0/G77In8J+LmU0l/ri4H+tIgc9B//48Af/zqb/MPA33qv/YrI9X5fb77beu96o/n2sncR+be/2TL4AQMGDBgwYMCAAd8A38Z73pTSuYj8KZ48ffyT28KgLUTkh4G/CJyLyD/oF1ci8idTSn+if7/VaObAf/A228uvi28mGejD81+AAQMGDBgwYMCAjzo+4DurlNK//I73PwP8zLt8ZQL82ZTSf7xdICIT4L/qv//cN3sM35ERlMlAuw+rKCSTU555NsdaBZst9gmVodkz+EKwsw0v3HjMsi04yfe5jAVuA34ESewuO8mtDPkFYGBzrIbXMcvV7N3rTm0D5ZmlPoJ81vIjR29yb7zPL61foNvL6SZGqzZHiWSTZobXjuzaim6TUR9arXC8Lfp5AvFGjdEbkKLPgH2hpbnm2PuKIV8oDb85MqyfMoSeeiRBKiLL56CbGFY3MiTCs1mDIfEb81s7k+aRbbl954y7HNFd09zc/LEjFIkHv8sQi4hpNEvXHjasy4xsntFOc9oZuLWlOVBDbPEWE6A5An+nZjnJQdzO0FkSbK45knW0M32fXxr8GB78SInxpeYj10aNtJOaYPtKaaZ2BvUzHe2lozxzZFPL6ilDvkysbj4xk29naMZ6EtbPd9gmo5s4soVlfUPw455u3UCYePZmGwoX6DpLOojUxxXrm4aLT1ZEB9nCIkmpUEae6d6GxaRivSgoT4VQCO1U6Z98qTNHyMHtNexNN5zZKcSCaA2TB5GLTyhtHjMN1K72NxS5ZzE/oN0XVjemOpatmpbnc+1XCbC5EShurFk8HCHJUh9WEKHdU5rItkaDBNZ6uH4EyToWt7RyP+Z6jRQXiW4iO6P99pmG0bRmVc+Qt4R2T0hAGFnN/g4Alm6klaCgFcpuneEroZuM6Spt2+y6xZfS54RrNWyzJ3SjjOjA1TomoukzyHszbj9W2Ue0SkFKhG6aKE3HuR/z+voII4mZq3lufMavHT5Lez2p0fXSklwizTxv/mMFsUgkE0lFZHPkAEs7E1ZPA2IwTW/cnnrz9k5p/nDcEe7nbK4Ly9pSXEKzL4RSr/MwSriVVlSnDB59v6ObgNto+7qVUFyqm0V0IKOWrPAsXnCQRWzt8FUGSeeSdjrG1SNt10KlHBJUCnDxYgUCo0cJ21qunpdddrUfJ27uLSisRySxzAPzZ0csn85310B9XU3RZeQpypb6aQh3S84/JZjOcfQbgfaoYnVDpR4kIdnInaMLVtOCh+tr2JWhPRCWt3Kiy4m5XpfZAnCJ9jiwWlrqIw1aWN2YIhGaT2v7Tt9IhOuWbpZoishZcNhax1w3ztgcCzGH8rFl/NDTHEA47liPLKGySBTyK4cvQaLVH2YDxblKLcaHa1am4spkZAtBvGC6jOgy2j0hWzp82VdkV572KWHzKMNXGb4EjJr7Z0vwhVbGS96Q556LTyfyK62mLy6F9bFBokpXlrdm+AragwCTwKp0JKc0rlkJ867g1fV1QhLqkGEkkRvPU9fmPEjCqiuxje43FLC6uZ1HDhg/8Kye66UNjyzt1OAr7XfbQHUWWd7SOaQ5AI5ral9ivOA2eq3ZRueDxR11PmknFe0UQgXLW4aYFYSsl3Elna/AUB/oNuLYMzlas8on+Iklm2tCjZ/ovOSWuUoKXjwmVBEZeVbPGNo9o/0k0ByM6UYG25S0U0M3BXOt4da1Sx7NpizTmPrYYDfgJzB/rtB52+icd3jtkirrOC9GXG2mNPuuPy+9vgDcUq9d/5zFNpbJ/YyuGtPsqxuAJH1t9vV8q7OE7f4hbzJ+myDpgzVs/xbxReCdHpk18N98qxt81whKEVmIyFxE5sBnt//eLn+P7/6MiDzqvZi2yw5F5HMi8tX+9aBf/i+IyK/3f78kIt/3jm1ZEflVEfm5b/VEBwwYMGDAgAEDPlSI8v7+vk1IKT1KKf3mO5b5lNLf+1a3+a43mimlaUpp1v+5t/17mlKavce2/xJqBPp2/PvAz6eUPg78fP8e4LeA359S+izwp4B3ekP9W8CX3sf5DBgwYMCAAQMGfCSwfar5Xn8fZbzrjeY/DFJKv8j/34PpJ4FtQdFfBv6Zft1fepsH099FS+4BEJHbwD+FZqsPGDBgwIABAwZ8ZyC9z7+PML7dGs2nUkonACmlk740/p34aeB/e9v7Pwf8e8D03TYsIv8a8O8C+7Ya4zYwepzwhbC64ahONfnBrTpCWVCdBaITzs8K1kc5802JfZxjWsjmqEVF3vduRDtaVPcmQTUqodSPk4Uk4FbQjXqt12XBvXqf+6sZaZ7hVoZsDu2B6kXYmqiaRFc7qNVSIuRbXYnqNLf2qhIhNZY65bCymFa1de3EMH2zxo9KNkeiOj5U82ZXFlOrtjI61ZK+cnGNy7biS1+4g1saYpGonl1Q5h1mpRo3CYJbqRVJc6DvJam2L6wzaAyuRvVN/flL0jaIFeSP+nOQhHSiupxO929bel2QtqNETYJJottBVIOUUN2c7RJuk9QuqE8Gkixg6wxIO81hKLShbNdbw3hI3mDyAFG1fkkglLqd7YUbKpDGML8a4fKAX2hSSNnrgJLoucWiH2gJjI34YGDlcGu1hGmnsjsHXwr5MiEeunlBV7VQW/IrPa6u0mQhcfTJQ0LbZIRoyC9Vr5b6fo+ZtokmfqiWiyyRu0AThaDNwOR+JJRmd8xEbWPb9FrNPhUlWX2NZd9m8kQjRhK6zmEb1bqGSr+XNNyDbA3ZOtKNVK+VjG67G6ulVDKaShJKIbq+vwLkV7rfrbZq+9/bfK62QzHrNcgTtYnpxsL4JGHbhOnUruWV9XXurff4wm88C8Des5c8t3+BWVtiFSGAaQWphS63vfYVJAr2wlGd6TjylV4j0cluvInX/edX2iaBtBubiGqLTavX+1bPyTZJp9LvbjWmWxuc0ePAolC9XrrKaDOn1yyqtWtmQrZKet5B7cuS2/Z5wgbVa6Jy8H58p50+c/uE42I1wgdDu86IrWW86ZOIeh3u9jjFJPZGNeurCpN0/KV+u120uLWeS9Thz/3LPUIwZJdG+6/vy1Cqrhl6G7QrTVhxa8jW2nfRQXWeqI/7MdBpWpBd67jYarXtRvdvvLZ1zFF7qCXUK6u6+Kt+TgyqvU4GotX9FYvIKhk6b0mdaifdUucAt4mqee+1xbbfp1/k2LVh/KBPApIneubtGI4ZpGBYz0ts0N+D6FTT3OxpP0nUeUwSGC+kK0d2ZXArbZvN04mvvPUUL8cbFG8UZAttX//pFdEbuMh1Pmi1f3XsaVvFDPxIU3lMK+SLiC+fJP9s28t07OZE7y3lXHWZxWWknRiyNXQTXYZAvtQ5YvcbZiDbaDqdWwVCadRCqREmJ57zzuCDwdQG0zyxvCM9mUuI2j/FucFvyj75qb+ugl5nJmiyXTGPmNbglxmLSUG7yimX8kQnHdi1LWg/X16NmNuIX+W4Vs9ftcRPfoddzRNLO4FupClOEmV3XUbX/84YaCffPhr6PfEd8LTy/eADe6L5rUBEfgy90fxj/ft/GniUUvqV9/puSukvpJQ+nlI6zmYH77X6gAEDBgwYMGDA7yy+C55ofrtvNB+KyE2A/vXR9gMR+SxKj/9kSumsX/x7gD/UO8//FeAPiMh//+095AEDBgwYMGDAgA8A3wU3mt9u6vyvAz8F/Jn+9X8F6HPT/2fgX0opfWW7ckrpj9O71ovIjwL/TkrpX3yvnUiA/BLKU08o1T5l/PoCs+nwh2OqkzWbp0fkVx0Scx6cz0j3KyZvCaPHCbeJ2Nb26T1Kxdg6YXyf1pGr3cXokSYTJNunGpwF6kNLdLBeWf7Oa88jJyWjU9HH+w3kbyqdkWx/jgI1OfmlkM/VdqHqb7ND1lObq0SysLptIA+M7isllIxua/FsSXRKRSarlJum6iTGj5Q+mN5VqujRrx0z746ZncPhlzvOPpWxNFNWWWJ0Ysh66mf8QOng8T2lhruRoZ1ByB35XHZ0U3GVCIVQPdYroTkQbJvI5oI/LZi9qrT13qstoTIkUXosOqU7bKspJ6GwTF4PFJcd66eUv4tum5yRNGnmOCefW9YLTW4ZPdKEiOlbUZNYvKG8jEphRlg87/ALx+ihwdZQXvbpOqJSANskTbI5FEIUutoxfdmpLY1V+5QtfbmlRSVAOito25K9NwW37ukklFrUJApop9p3dmGZZ2Pyx46YwcFXfS9DcHRjtR9JBnhVbZTG55AvEhLSztYpibYVSdNqmiPHXCZM3zCUZ2lnB1JcJbJVUluhxM66JBTbzyIhE2wbCbnRcdjouDVdYv10TswyJg96qr6G4iqSzzWtw5eGkGv/lmcCEUaPI/kyks07NtcLsmXABEe0SrWHTOUKxbn2yfhh6Om+RDOzVH0qyDYtKxSG4hz2X17R7hc0BxbTwuffeJH2rOTgi5qcdRb3+bW9CdO3BIlWac9eYpBfWkaPtK+7sbB+ml07ZqvE0ReUgm6nspPEhFwpXLeGeFJQPYbx/biTWuy/0rG4k1Fcxd5CKnD6vY5sBdd+o8FXFl8ZTVJaBUwbGN33SEi8djQjusT+y9J/N+32V1xux7+QP0rUBwa3NOSX2tezN+ITm6iksgxXp1160mpewjzD1kJ1IUzuRepDpefdCrKVXn9Xk5yzfIx7kCvN3VvCIEKzp3Y2bqNt0e4Jm5Mx47fU2iluZQ9Wjzdf6PHHDLo9oXzcj0/UOsaXkC8is9cMkqA887i1Z/58hT01TE4ibh01wcYnsoXHl2qH41YeX1aUjy3VQx1fxuv+Ric1yQh+5HCbQCgMbpWzWuZIhOJcyFZqL2YClBcRiYl63yg9GyE/dT1drclQ20S3aMF0mv7m1onUGqQI5JcqCynPE9k64UdCs6ftATB6AG5tiBYmJ4m9ryxpjiuWz1nCKqM80fGbLdWuZz4eI/1Y2Hu9pZ056n2dW/M5TyhnK4xOVPaUX3lGDwJ+4jSVLtN5YfaGJq5dPm8JRUa+6KU7lVCdJQ6+tKa+XhJyoRv1EoWgkoXJvUh12mLXHWbVEGYVxaln/tIU22o7ZOeWeDFldhcmJwEJiVAY0n1wtc4z2SoiCebPONw6sf/ymuSE+lpBKATbJEZ31yRnwAj5wpGdO5p7h8yuYHI34kdCeRF2CV0SgaTX7+qZArMWRpdCea6UvaREzARb6xxJLxWqThO2hfH9FkmJdi9DYiLkppc+eEyn5xCqDw+Z+91AnX9gN5oi8rNopuY1EbmLxlf+GeCvishPo5FF/2y/+p8AjoD/us/m9N9EePyAAQMGDBgwYMBHD8ON5reOlNI//w0++oNfZ90/CvzR99je5/k64e8DBgwYMGDAgAEfOXyXFAN9RyYDDRgwYMCAAQMGfOgx3Gh+NGEijB8EqvtL/F7Z6ygtsQD3aA4+kFcOd7HGrq8RTyr2XlHbjepRh1t7urFq5twmqW3GJtLOVBtZXXkwglt0+LFDgo6UbNGRzQ2rWwVuaWgmGaMLIV9sdXe9JjGo9lBib0EiQj5XLaTxCfGJ5DSGL1khW6gWJ7/M6GKGW6meyjaqGwLVDPlSMI0e7+ihsLKG4kr1XLq/RH6pNg+HX26wa4/baDSlu7CUZ73d0CbtIhXL80S2DGTLgG0ci2dNrxtKvTVRxG3UwiI6IS57ndwCuGd2FhxXH8vxlVqERKc6M7WqUO2f8WjM2qjY2YZsLTi6iaHZLzFdUt3gqSWfa5Slajz13FVbBEivZVurbm98/8mVrDYXGu+p9h4RtzSEa4JcOSRAcdF/tlbdJUA0W00pJNPHmS4Trk4UFx7b2j7mTgiZanCzJbiNEM4ysivV+kYn5PNAPo8k0RhD4xPdVLVg4mF8otqtfKk2Je3EkC+jfh4S09+y1AvH7PWoOs5S6EbovnO1l9nGTJpO7UfaqdDsacSp+Cf6JBMS7cw+0Zl2qieWqHqn+lDjWvUz1UYZn9RiptN27sYWuaaa5p3N0053uR17au8yv2M1dnSh1mMSn+hws2Wi3VM94vLOiJj32ucTgYcT9s4Tro6YDiZvGGKWM7mv2mNfqtZwa2EiUe1cXCM0hwZbJ6oz1fXZJtJOLdlK9a3NzOy0rflC+822sLmm/RNz2HxWdYDNnsU12obbMXr1XIHtEklUH9fsGWzjSLbEtGrBlazqffOF2iy5Jqles0k767FoBbdJO12gJGj29NhAt50vVNOJQD4X7KShzQPFr4/6aMQnY89X/bwToTqxNM2YfKPju7jS6y/aXgvd61slqK1Z2NPP3VotqJp9va6KC6Gdqj1Tsrr+5qmtBZS2PwLL0u6sn9xG9ZckHY/t2FBcBEIpdCPD6nqJq/Ucq5B0LrFqQ5UtQ2/ZZWgOc9XPN1G1jAmKS1jPHXZt+vhZbV8dT576MMM1EIPqly+/B0xrcJtEtvJ0s4zyItDMLNlax1LIBXdpCdcixaUedzfWa7vZ00hXZBsJm7CNEMc6H8fCQUy4la7vVmq/ky91vE/fgNXTqqGuDx2jBy0hzzFB5w2A0WmgOO+IWYFtElihPtJ52jaRbJHwY0uyqh8uLxKC9PONzm22jlx8ckS+SH3kq2pVJT3p++XtAiiQOEH6+duEJ78XWy1vdR61Teon15b+bumcnS8jvtBrcPlshenSbr6QmGieqiBq/YDpEqZTzfboUcTVkZhbbB1JI6vjVwAE2yVM24/nhc63kqAdS6/H7n9HnCBB52LjoZta1WLmeh1tz7+bWKpHDWFkKS4+JBmUMNxoDhgwYMCAAQMGDPjthzBQ5wMGDBgwYMCAAQM+CKQnBvXfyRhuNAcMGDBgwIABA34nMDzR/GjCdInytEWaDrsUzKYjPXiM3DiGugHnyE6uIHNM31AvSrdOjE86ikcrSInDXzwlHu9jHl9CVRLHBbap+iqxRHbZYJY1dq+C//fLJN9hv+clTO4oRg7bGOzSYjq48QunSEqkk0fIzeuEvQp7ck773LHqadqS8d0aW3ti4dTb7PEl8doeYZSRjJA/WjJ6/hh/abj26w3Z5Yb2aIStPdlrDwl3jomZwY8c7Uy71dZCdRrI5+pTV5zWzJ+dalzmVYtpPfliRHZlKB/D9K2O8uGK9mhEO7O4sSVfRLJ5iz1dED51TPnYMH0rUlx2hNyov2IhtBP1K+1G0vsSKiWwucYunjI51fEASBAwT5a7NTT7+tk2Ysx02whI1eSYoF5w+aVGfXZjwXj1QtxGNmL6bRgYnfSasD3dXrNN/ewjRZOFdqY6nuxejttAtlHdWSjUA9LWql1zHTvtUij0mGMmdEbwZbbz/JSk+sZoep/EU2gP1AMu5tDODO3U4JrUR3KqdoukOjeJ0M4c7VR1q6DrxEyQlOhmBuMTxaXQ7Bt82UevGdnpMvV9H+cW9H3T+xzK9n/Qu1c9zmQgFAkM1Eey8xqN2ZO2kii9Rqv3UPT6b4RdxOe2r3b7CqrTC7lGWm41u934SaSovuqX3Kr3QL1tVNYn20g+fePq3le01m1udZSy7c+JarLameBHqtXKL1WHuLmmGjCJZhe5mYzGKjb7Gi24HY/biDtJur2d762FJvZ0V4I00/ElUY9L4v/H3pvFyrKd932/b61VQ1d373mf+U7UvdTloIESQcmOB9qKLMVwoMROACECDMmADQTWS5A4SB4MBHlyogchsBTDhqEnK/CDYcuCrUgx4EGCzNAkRUocRfLO555hz0MPNay1vjys6t6HFO/lJcPDe31VP6Bxzq5dXV29atXq2tW/9f/6do3Sl6tMrrCbS3odJ9Q7fdua1KfVXG2bmFxeUVjupLZd7ggm8ki7ptfKL0HKdGB9lTy25Z6sy9gC+JA81+RDK/m5YLs+QxQwWfKC0asSiNkFqHWpLGRfAjW/WJXg1PWxXZ13Kw8RTS5iyHpnO0tlNtWQMkaXUFxclVhspyl/0vd+sRrwZY725QJtqzRb6Rj6kUmZiEXKRDQdxDy1T3lg2HxJMV1yeFN5SCHa9FwJilaGkJPKKfrku5qJS354F3HLiHilG1tsl3JiF4WjPElOsAnau4mpXKfrxwUTkheZX0Bx5vs+7MlPM0xr2HwplXasHraoCH5sUwnXRSqxKiFimz5Htfewq9dS7uQq+zI/bXG5wZcWP7Zklx3d1IKmHOCVc16cBUyrLK85wtTgR9JnbqbjFfryouKFelvW2aLpM03W56FC8rYzaLcA0thu+jFEzdXXvWqSxx2z9PNivx+35MoPX9+x68u9FidpjsFyz1DvpGzlbpSv92flCq/Hh95vDnl/zPtyyKk8bl/CthXaDenLFqe+nTKD+9xriXQjQW2Zcmm3Le8YhgvNgYGBgYGBgYGBx8HgaA4MDAwMDAwMDDwehgvNgYGBgYGBgYGB7zjvgjrmb4V35YVmzISL95RsiOAuG/x2he4/g2kDul0hXUS6gGk6uklyj4rzlM8WJgVhZHFVjviIbk3RzGIenlK0nuVTm+SnLebl+2jbYg4dcvsG8cEhsmyQzlOcOsqTjOphXws2RvTeQ6QsoO0wCwvLmuysRq0wfj1lE9rzJcYZ4qiAUYm5rEEVaT2oJsdrJ2VPSu0p7l8QNivi2Tlmbwv72hlZWVCOSw5/ZAc3h8nnDmA2J1vWyMaU7HJKPouoM2gQyrNAcer6TEyPuXdEUW+BbLDctYwe1tjjGXp6RvVKyXJnC7cIFH/wCnpzn+baGLcIhFnyNd3SIDH5N90IRqfJtVq5N6s68RL7mrUuuUG2UUKZHBu1vd/XO4qiuj4Z1QrdJDmdttF1jeu0XvIqIblT5Ung8gmX8uAWuq6B/qh7l3IlhWZLyR/KupZucXaVJbnKYVu9Tr1t6TaSj5Zfphq9aq4yN1d1ev3I0G4YRg+vMlJRxbTaZ0lGiIqvLN04uU2TBx7T9rWgDel9W0kZd03ELSTVp94w5JeRmKX2XNVWj07WTuHKk0114PsojQjEK2/KBNb1vlEhjJJDtfIuVzmYK282eVHpZ9O7fau2X6Nf63H5UWrz/PDqmKYnsc6gNSH1jfwyEgrhUZrN5JOVp1e13P0o5W3mFyk3MWZQnKdMzFWGpoS0nypCdejxI4Ork8O9di77zNVVO678Vvq2WrdP/75Csdpm/7uVuyZX66Qa4ikbVAKMjq/yB0MuuD7rNrrev7Xp2K2Ot6/SMSwv0vOjS97mul1X/uVIuJyXtJcFZQvjB2lH4+r8Canf+1LoNgTTCBuvBIxPzuPqvFObdr48S9tf7ho0g50vhJT7eqlkiwhB+7xNoTroHeWxwS2S4xgz6bMToZvYdeanmuQQ5zOwtZJfRmyrjO91xMLQLk1/3qX+UO8YNl5Or+kWAWLKDJWgvdctGB9ptjLcErqxxbTK9KVZny1p+nrYBndeo84QJgVqhOVuScxgdNjhLluKY1CXOoO0AbdIA0goLRKF6UsL3L1jKAq0zKjuOszpjLAbD4rAAAAgAElEQVS/gVm0XDy/RbNhsG3ELj3ueIE0LaOnSsJlcnHHd2tsE5JPWKQsy2wWqO7OMfOGbpLRbBq2/90r6M5GamdnMGGc2rMNWJ9qjWdzj2kD2czgloHoDKE0zG86VCw7nzkn5hPUCtVBpNmyfc5x7zja3lssUnuuXUu56jNqVmO2oatSnffVMuP7cfPrLo6KM1nnwdpW1xVvQs7aNV/12VWeZn55tRHb6tpxViv92KuchVSbPpunfFHtHfzoBNvG9TmBSPqsjFdjSswE064GqOT2i8LoMNBuZrxTeByzzkXko8A/B17qF/1TVf1f+9/9JPB/ABb4h6r6d77Fbf9b4Caw7Bc9VNWfeLPnvCsvNAcGBgYGBgYG3uk8Rkfzd1T1L33Na4lY4JeBHwfuAp8QkV9X1S98i9v+GVX95Ftd2XzzVQYGBgYGBgYGBr7j6Ft8fGf4CPBVVX1RVVvgHwM/9Y1WFJF/ISJfFZHPiEgrInvf7osOF5oDAwMDAwMDA99t3upFZrrQ3BORTz7y+BvfZOt/QkR+X0T+bxH5QL/sNvDaI+vc7Zd9Iyzw11T1B4F7X/e7X+0vQD8hIj/7zd7mu/Or85jyHBc3c5YfLMkvUpaW8bp2zUR7L7DP51pcs8xu2/VfDrYtCHnKADReiR/awJcpP62yUP77Y9y1fdjeRB8cYna2YLkkPHEdM6sZHVaIgqsjLGvUeySfwuUMRjlsbqSstM2S0+cLJMD1f3OJLFukyEAEdSZ5mtaApEw14yGMDHIxQ2dzXL0DW5vQpJBHrQrkYk4+20KC0D6xQ35XkJ0tNOtdpADubAF1i9ssqQ6S4+QrB8/dws7b5MIY8OMMmGI2K7qtAtsqi2sO/2eexZeSnLU+y2x0HOnGQjsRmm1ot5Xy969yDSH5a6brPcwuOW+2hfltoXoAu5+d9XWyJdUMvndM9+QeYeT6bENlfjunPPJUXzqge2IHe9EgMRKLK+9Gc8vJ+yvEJ7dn8qVTdJRhFi3qDLJoYFkTb+5y9EOb2FoozpStP5ynNqo95viceH0bWbRIlzxZHZe00y26mTA66MhPa6RukWWbwirrJmV5jkqa565z8WRBUSvj1+aYRYtczIm7G4iPyOUSvCfubtJsblKcKwTFdBE/dpg2EgqTvFbg4r0Z4pPz2G3A+G6qIR1K1m6rrZOX104k1YO/lvIWR4eR8iT0XlafLWgFIqhLfuKG7TMqu0jIDK4OKdN10SJnM3RzDMbQ7VYcv7+kuFA2vzLHni9htki/v3+IVBXx8JjY1MSP/hDHHygpTyIbL6Q2IKxOMkFOLohHx0hVcf6Tz7O4ZijOlN1/8yp6foHcus7s+V1Onrd0lbDz++fQ15o/+b4po0PP6LVz2mtTJESKU0coDW4RcHNPdILbyphfd2y80jK/lRGtrHP/kscpa5+y2erPsw6yuZLPUlu2m8Lmi57LOy712zy5kLaDvT9YIL1DF3O7dsBxBnykvVbRjS2LfcvWiw3uskWagIQAxiCHp7AxgfmScGuHi2cnFKee0e98EZ64BQ8PETHE5RKxFrm2hx6fcv6Tz9MejCgPbF+/PFK9etn3fweq2JM5dB1q7pBdCNW9BTG3jD/1AIqC+PAAs7MNbUu8mMH730MoNql3hWweyC5azKIlTEvsrEGaLvXxtqN5/jb1Vg6VwVfmymU2su5foUg16E3vOi/3BMSwzGSd0amSchVD0efIbqZs4/1PN7jzGml98vbaACGkMe2Fh2TbU5obEzZegfK4w8xqwlaFWXZI65GXX0Ou7RFff0C2u028scf0boZEpfjKQ2ha4ukZpijQpkEmY+x0Qjw4wn34Q6gRTBfAWGhbwldewN1Jn8nmC0l9k+c2aacpH9TUAVk2EAPVw44wshQnLdmrR8S9TTSz1NuW7BKq1xeIj/jtMflZS73r0L0tuu1RyvU8vKQbQSgMEsa009SW2Vzxo1Tj2xdZyiYeweKGMnlNqG+NWey7vh67RUJyiLsJjB8qiyeS76oCG68G1Ap2GVL9eFVsE9JYC7g6Y37dks0j+WUgO22wRxepLe7sYWpPGOcgwvEHK9TAzheW2NonD7wLaX5Bf66jSnNrg5Pnc7SB6mFLdrxALubgXPpsrCpYLsFa2Jgwfnoft1C2v7jAfvlVwskpdjpFQ8BsbuDvP8DubCNlCaMyjcHjERydrjOVGZXEw2PEWsJsRvyzP0Q7fWfcYxOu9O63wJGqfvgtrvt7wFOqOhORvwj8GvDcG7zcG90vnQAnb/C7n1HVT4rIPunr94+r6hffaGfeGa09MDAwMDAwMPDHje/AV+ci8jf7O4yfEZFbqnqhqjMAVf0NIOu/+r4LPPHIU+/wR+9WrnjqTX5Hv+1D4OPAB99sveFCc2BgYGBgYGDgbWBVTeybPd4MVf1lVf3B/nFPRG6IiACIyEdI13rHwCeA50TkGRHJgZ8Gfv2P7JPIjwKvquob3dFcrVcBHwJefLP13p1fnQ8MDAwMDAwMvNN5PLPO/yvgvxURT4oh+mlVVcCLyM8Dv0VyMH9FVT//6BNF5Bbwm0ArIp/pF98CfgH4uf7nXxWRJVAA/0hVP/VmOzNcaA4MDAwMDAwMfLfRxxNvpKq/BPzSG/zuN4DfeJOn58CvqerPPrpQRP5J//yPfqv786680AwFND92wcbGjFIiRpQmXL1VI0oXLCEauosKf15w/c/cpbQe19+jroOjjZZZWxCiIUZD12Q0n9mk3nbwP/7JNLElgMq1dXi1bUHihGYjye7VgRB++DZq7iAxBZS7ZcTNA8ZH/Njiyz7cdmecwr4jNDc3yM5r/N44ifA+Uh1E3NyQn7a0z95AomLagNQjwjTH9ZNhwv4W2SyiYgiFwV/bSMG2LoVFuzoSNkps57GzDtPlScp3Qhg5wsj1IcsQSoMEhx9niCpnzxqWT3bs3rxARMlMxNlAYT3ORKIKUYWzekR7Oubazz2gch1ttPhoiMh6ndWj9hlnxxvYesTrf3aybkfjQT88xgRwixTGm1+mkN5m29J9+Cb5hScUFTGTFMjdpODf6IT5rXRMtr/c0dzeSJNE5iVqhfzYontTpAmUJ4HoLNWDbt1Hup0R8cYYN+/QcYZtAt1Gjpo+TPskEguDn+RIlWHaEvFJqBcfkWWLaQKuThMgwigj5ha2kuyvOdhFA/MGc3BCu7HF7LYQPrgkyzzQIAJF5rESyV1gKkphPZdNyfHZhGd+7C658UQ1GIn4aGmjZeEzfDQsu5zZrKL57ISQG9pJCotWIfWDhaIutVU2j0n8X6SgZFsrfmzJMwMbBfGpTaSLhNIQM0O7CdkCUKW+s4lpJ5guEvefwbQRntyHmIKbN17xjL96Src/SZO8IE3WaD3x5i5GhHh9m64S2o0UtKzbU9ia0O6Nyc86spmlOoj9JIOO+tYUVyvFUU3YrMjOlmjuWF4r6CbCzquXxCLDeGinlpM/1eD/0oLCeayN2H4csCZi+3M+qKFdluxUC/bKxbqvtsHi1eAkMib93EXLg4/dojgFP3aoyfCVSQHbXmm3LMWJJ5t5Lp7KOX2fEqaexY2Cvc857DJSnNTE3NLdfoLiuMYCzf4IX8D2799Fn7qV2mpnm7gxSm3XemLuMCIUp4HJSxnjB0p10GGXHs0socrIjuZXhQNu7SARigsllA6JCptTOL/EPHkbNQasYLIMDSlQ3S0s7YbDNIGYj4i5AQp0s8QdmVTIwgpqhOMPgD61ZDyuaQBnI85EcucxouSAlcirhztsbSx4784BAAuf0wZL3Y/NISaT63A2Yf7lDaIdIXGEhDQpc/WBbLwSn30K49MkT9spanMWN/bWk4tiJtjvT+9bzW3UQHEWmN2y5DPFfe8NYm4wbUwFFzLBLlORhPzGNuOHSjcTZNGi0xF+s8RVoxT+Pi6Q27t9EQ2DbWGxbxi/blm+dx9bB2a3M0yA6rUWf3MH99J9/PfeJtpUvCFUGaEocQtPN82IDsI4J2YG06TKD2EEl8835D9xgRFwoogopQ0YUt9dfbbNHuyw98OHbOQN1yQSEXw0+GhooyX21Sa6esTl7+1hOvq+ABIs2Xw1YTZNKpMIoRBOfrjjtDPsfDqn2HG4WyPcMiJRWb5njHiwndJspYl17XaOLwskXoWwp4D2iAnQTg2Te/3z93Pq3RzTbWA6xS23U8j+ZUfMDWoN2VzJZmlM9e9/Cnd2LY2vx6fozgauKIj37hNOTnHPP4eOcgiKTKo0wagPcjc3r8NsjtuYUucGX34LU3AeN++8ykCHwN/7Bst/8dvd4LvyQnNgYGBgYGBg4J3OYwxs/7ZQ1Tlpgs/XL//db3ebw4XmwMDAwMDAwMDbwTvsQvNxMFxoDgwMDAwMDAx8t9HHU+v8nca780LTwNO7J+wUC8a2JSJrR9P096mTM2ixJvIwGK6VM8auwaIEBB8tTXTMsjatHyyHZsJJBjHr/UEDoVc9JNB7LYAIzRb4sSbvMbeYoMk/WSght8xuOfLLSCiEdgvG92B+ewRAzIVsFvHjMd3YpNBdr2SzSDs2zO+U6fVC8hZNH4DdTTJcHeimjpCl0PNu6ljcyFBJAdSLfSFaS3FsiJMRYZJRHnVIiBivuAdnEAJHH73D4jpUD4VmK8d4xZdC/T0NN6+fszua48zVGVJaj5FIVLNu59myYKdYMLIdESH0nlBUg+/XcxI5a0eczUfEfJRCm0Pyk2xD+mtPUuC7m8P5szaFsD8QEFju5wBksxTG3wIhE7Kl0uxG7NIk38cI4hU/sYTcoKYiu+xobpWIpjDpwx/IMSGnPFaqo4AvBJUsOWGFRU0KTfYjod0QzGuKu0xtryKEcUYsLNlpQ9wsUSNcvifZ3r6ygO19W0UzId4cAdfoJobFDfA3Wt53/ZDSenITMBLX/9q+33bR8sBE5k3OxDWMbLvu01GFJjommcOgLEKe+vd4jBrB1qngAALLHSEbybp9mw3L+fdG8lPDnX9d021mNBODLzN8Kcm3shaVFK7fjWH7yymgPBYGP7bJGW4i3dQhnabg99LSTg32zibRCbqZkZ+1iMnwuxW2CXBjh26jYH5TCCWUx1DfnDC/lWE6aKfJi7Wd4rdKTBtodhziodkvk384bwml4/j7+v7XbeJqJbv0nH6vcOPaOdOi7ttTv+aRm+QS+mixEtnIG3Lr121a2m7dxgYlItQh45VbHdllxul7c8pTxRfgq7SvCFw8meOWObMnleLpS25sXvBSvs9JV1CeWs7em6Em9ffmwxm2nhJGadxotp5JxRlyUJcC+W0ainAL8KMNYt4fy32h2cyxTZ7cagsxT2NEf8rhxynYWyW1V/sDFbAPmjxz8YDsrp8vCvW2wY9KulE6/qsxLz6bPGc/SuOdeXbG7e2z1G4mkFuPQRnZ5DxHhC5aLpqSvWrO1NXrdvfRrseCFSLKK7dzmma0Ds9P42r6180FPwYJQjYDPxYW14XiLLUXpPF5uSvJ675QYi40m4bldQilICH1a9ek4gW+FEYnwnLXkG86RgcdZttRP7FBLJLf7PYKfGUozgNEpZtYzp4V2i1l+pKwvFYQnVA/k1PvpOMKG+SXEbn+DH4kxEyY3xTmN0eM70WW+475dUOzB9GNk79ZGs7fW7F4b8u1/Quujy/X/W811ubG40xyORe+YL6TM8laNrL6a8aLoOmzLH3mOZyJnE53kE4wnaTPli6NryppvG03DPkFXDwDN++cMspa7t+/Q7NlKM4MMWc9Tq/74xiyS9J7tGmM7MbpMzGbKa6xhBxCLuvnmm51kZXc+mxu8CNDuJ2KKkhUQpnGqGa/TD7sNEvzDqo9VKA4D7hn9rC1p6kc0RkW1x2mU/JZ8m+LkzY59t+zR8wNsxuOept3DsMdzYGBgYGBgYGBge80wjvP0XwcDBeaAwMDAwMDAwNvB8OF5sDAwMDAwMDAwONA9N1/pSn6LnyT5e0n9Km/+9eJweBcIKoQvEEExKRsvxAMqBA6g7YGN+mwNqIqiCjaPyd2BrG9hxeEYtIyGTXkNpDZsM7gWztcokQVjuZjNkc1t8fnOIl4NVy0Zco57HJCNDibJJe793fgIvsaKVitIlGQTsCAdOC3PQhkRxnqNDlYFsQLiCJeUAumhVApYRrYv3OKsxEfkgd1fDrBvTxKTtYjhBKKU+gm9NlzEJ+bs7c5T+2nQudT9qg1kcuLEcYqMSSHRkzfRjFJYSaLGKOEvv1EFI1m/T1B9AYxinURMUrwhif2TrHmqj1Le7WTq58XPmPhM/bKBbnxtNFx0RVEFZY+S5mnKlgTObyY0JyXKU4wrmRaRbyBCAhIJ8QqIJkymta0rSPUDrlwqa0vDCFXTJcy5kKp2GXyQ2P/Z5qEPk/1EdVMTXKQ4vfN0Chknx4Ti+TC+ar3T/vMPwTq2x0be3OaLm101Tec7TMeo0mZea0jRkGjIQbB9H2TPmMvdMnHWi23WWCzWjLOWwp31Z5R5cpTJP37wvEum1XNnck5AF5TW/o+3/DRXMmDyynNlzbTa7SpLWybHDm3XLl0sLgdwSjFoe3bM/lcdgmhAjdLzrM6WDzpGe0sqcqWzXKZci29ow2Wk5e2U+bfI8cyVoHyviMUEHNNx/KJJXeun2IlEtRwthgxuywpq5amzlN/hau7CNK31SPtp6E/KJrOeen76Gr99f8VRuOWKm8RgTLrsJIyJJ1Jfq0zkcPlmFlT8MzWCbV3nDUjMhO4bAsAOm9pO4cxEe8tXdP3gTwQOkP0FoKgEaQxkCl4wS4MfjP1E3tpwSpEwdaCeOg2I6JgloYwjmimuHNLfgHtRhonMBBtX+YupGNh2/RvKBU37z0+3x8nScfKLpNH3d1qmW4tAJjPC1we1udw5gLajxurMXh13otA9PI1460YRYxSjlrqZc6oaslcYKOsKazvPXBd++C5CXg1/O6Xv4fN7TnXJjNK61OfVUPjHfM2J/QnZojCbDbC5R7fOjRC7FK/pOlP3k6QIMnbzjT12VNLzBXTSjrH5coxbG56RnsLJmXDOE8SbRctZ4sR87PR1XajpHFaIY4D+UFGN43rMb+8M2M6atbnZxsspw+nVNtLmjpbxc/253bE2EivVCL9+dwt05gFqY8aG9FoUGXd3q4IlEXHpGwoXYeuvHkE058U0o8L87Zg0WZcn16uM5KTy2zWcx5W2adHl2OahxV2YdJ5k+vVBAYDRLC14CcR3exwDwpMmz6/TJf61OrNmO6qL0aXznm77D+T8r6v+rTd1eeYrvpw/9m1GotNm7alGesxP+bQTZQX/4f//lOq+mHeRsZ7T+j7fuq/e0vrfupX3v79/XYZ7mgODAwMDAwMDLwNDLPOBwYGBgYGBgYGHgvDZKCBgYGBgYGBgYHHw3Ch+R8nEiD/91OyGVQHAT9KmZK2jiAp527lcpw9Z+gmMP1UhlumWuTFRSTkydORmFzI6GB+w7D5n52wXSxTltkjWXtGFCd9rW8EZyI7xYK9fI6RSBMdTiJ1yMj7/LPcJn/0YDTl9u1DKtdRuY6Fz9I2JNJGS+U6Zl3OrC04mVXcevohO+WSNliciVy0BfO2QERp+xruFxcjdOF4784hC5+v97HxjsWzET/P059SRjGnOdffd0BhPZkNzNpiXUd3nLXkJu1nHRyvn22yrHOMVUJrQMBmkdAaXOmJvYcVWktQoLZrF1E6AadIY4ijgHQmlcPOI/lGy/VqRm5Sm145r5GizznccDVzX3DQTNktZkxsQ6eWaVbSRcvMF/ho1j7R4cWE8c6CJ7dPqVzHST1ip1wy6/JU195Eau+og+P++QbP7JwwydpUP7x3lhYh1WMGqIPDiPLqyTbORjZGNZM+OxBSvt3KY/TRcFJXhGh4auMUfztJQ2dNua79vnpeGy0PZ9Pk6p2USG1oM8XODY0kV8x0vTNok78bxwEzt9D1TmwraT1N+YJqk1/WbAbGW+dsFvXabVu1b248mUScCViU87aksJ7tPDl33SPS6aqeOsBZW3G6rKi+/5Ct0TL1ZROog6NyHV4NE9fi1dAGy1cO93n2A4ds5PU6068OjkXv1Ho1hGjY7P3P65NLNrIGI3Hddy+vlfjGEnvXWINQbDa87wMPKG3HrEu+42VbcHN8gY+W3Ho285pX2F67aFnpCZ1Njpso7SInanLZWDqoPNoZJMja4ZVgiaPep+sM6tKxmX7vKben52wVy7XnmpnAyHQ4EyiMxxIp7S6v6Db3ZhvslAtujS/wasj68/9RmuBYdhml89yZnKe+6jw+GhY+YyNvWPgMgIumpAuWi0XJB97/gP3ikmUfJOnVUIeMqMLCZ9y/2KDtHDefO1/nXUJy83LjcY98f3fUjLlsC3bLxXr/Hs6na0/b2UjV+4jLLiO3KZs0c2HtxtatW7uZvnXYLIV6qjdIFgitRb0gWUSDQaMkRW9pWYwtkkWe3T0iN4FJVmNFcZLadGKTy1gYTxMd925vULmOvXLOyLZENXRqkr/dlkRN43EbLfNRTYiGrWJJ6Tz3Zhtcq2bMupy9coFXQ248L53vcnA85Ufe8zJRDc4EZl2xrhs+73IyE2h7V3GnXLCVL4kIC5+zUy54WEw5Op0gRpO7rslNf8+dQ8pnPTvFgpOmwkdDFyyjrLtyTKPh7GBKU2fkhafrLCLJae8ah7WRrveJNQq6cMnrd70DHySpzHV/Dtvk9fqQs/3+c7aL5boPrGrN5zakzNj+yufEVjx46UlmGyV/+tkXMBLx0fZ11O36sw7gsi7otlre/8H7a0e2DZaghmnW9O2avOTT5YhrTz7EmbjOpV12GYXzNN5hTaT2GZd1wXxeEoPQesPNm6eEaCizjswEQkznEEAXUh7ryuk2oiyb1P+9N+nzrhMoIuIiWfl1kxTeLnS4ozkwMDAwMDAwMPC4GC40BwYGBgYGBgYGvtMMge0DAwMDAwMDAwOPDYnv/ivNd+eFpkI2g8n95GGMDgPFwTz9LkbER2KV0+6O8FWOnwaKc+lzuBTTKTETRocdfmTILgP1boZbwr2DLR64DayL68zMGPpMyN4NXOWaPSiSdxejIUYheEsMgoaUJ2n6jMSs9Eyylp1ikTwYNfhoKK2ncsn3yk2gCQ7fOm5Ul5S2I7qruuKP+oEA47zldD7iYDnhYDZJPihC3WSMyg6fpYw7YyJ1Fqmylklf132VT/dwNuXwYkLmkg+jCl3n8K1FW4vMLbGMhD4vzTcuZRy2Jv0boHxoCSXEQrELwW9E7MIQQsrnSxmghq4zfFpupyy9dAgJ3mBdxNmIs4FJkfbv6HJM7gJZn5HatC7lyDUOjZLyB73B2Mj+xoxr5YzMhLVDu5E1uL7m9cqD9d7y6tk2e5PZun2taKonnEEdsrV/uDuds5nXVK5b18E2Esl6zy0iNMFRWs/d2SZnbcmt6jzVdc6Sq7Q6ris37qXDXdpZjtQG0wp0gmkFt0i1mfMziEVqmPpaaje7NCknLkJ+DjFPOX8xA7WrQD3Ly3f3U39zcZ2zZ10kyzyudxWtUeZ1Ttc4Xiu217mH1l49Z+XNem+IwfCBO/eZZMmXa/tQ0ZW77CRieid5a7xkp1gwdg1RDRGhipaN/MojhOSvPjzaY9lkhGDIskDTOPLcrzMWTebXmawiyla+oDB+3WfbaPnq6R5BDbkNyGqfg0mZuQqxsUTXv6kISu+x9etKEPCC6QxqFDczdE4xve+mEexCODue0HaWzCVPbJUJWeUd1kQyk86x0+WI85MJcu54sNNe5WO2yVeV/rhIf4xE4D23X6dyzbqetRfLJGsY25btPNUOr1zHg/mUrfGSO6NTJrYhZMmZbaLDR0sTHU2W/PDztmSrqNkvZmuXEcCZQCZhfXy3siVfOL++9iPrkOErs85TTfXMA5VriSrMuiL5zsGtc1YnRfIOQzT4MuWxds4ScrPOC1UVVElZjzFl8KpTxESMi7x0ukPnLUXuU3asCmXWMc5bumjJTDq+989SnuuLsouzkdCvC9A2GRr7TNJguLZ9yd5ozlae2mmjqIkq7BRLyr42u5HItKg5K0eUtls74iPbpTFYDZVL6866nNdOtjk8nzAqW9rO0dQZW5sLfLjKYbZZwFjFVe3a3x/Zjp1iwcLnvLrc4nA26bNHIUaDBiG2lroz0FjII8FG1Bua/nfSGgiCmxlEwY/TeLIyf20tvb+dMiVjDq8fbXE3pmLfYpTQ57YaF8hKn+6yiSYH8tQSZyUfL5/C2kjTXF0y6Nf5xXvbM/bKGVaUuc8fGROuvM/V51T5SC4qsHayfZYc0Nr6Kw+zP792ygXOxD/imkcVYn/ur1zmqELhPD4Y6i6jdRHfWUyfL9o175BLH2X46nxgYGBgYGBgYODx8MchR9N881UGBgYGBgYGBga+4+hbfHyLiMhHReQzIvJ5Efl3jyz/SRH5QxH5qoj8T9/Gdv9t//zP9I/f+mbPGe5oDgwMDAwMDAy8DTyOyUAisgX8n8BPquqrInKtX26BXwZ+HLgLfEJEfl1Vv/AtvsTPqOon3+rK78oLTYngRxCdEJ1gWiVUOX6SkR8vkLojVBlqkmOieWS5mzF+EMnmgWzmaX1GzA1uGclOltR7GaJQfXrE+IFSXATEK93YYtuI6RRfWaKDdmK4fApmIyU/E7ZeUiQoxVnAV4bipKPeTdurtw1Hf6rjC90NjNHkQV4UyV2skpu2khZdEXC55xOvPokxSuxrPn9NfWbAZIG88IRg+MqXb4FP9dLtwnDj+x7y9MYJm73LFVW4v9zks6/eWtdUTo2YMjbtYU4ryfUhQrcdKQ9NqvXbQqgMpkl1j23zSI1vB82OsvXViFphuSvkF0qzZbAtqd7yVMgvlfkNwS4t+VcmGJ/qKoeir5mdw/h+ZLlvWEaod6HbVbL7gu9r4pYh1Wue1JAtlD6qkYOPKAdGOTibYozSLLOU4+f67yo0eWEmS7Wsm9KeV80AACAASURBVM9t8fKTJS+zl3Lv+oxQ6wJd7VIengpZ4bkoSuomu/LMVHC99wopxy7LAs0y4/5slxe29gitJdYOySJiI8YlX8i5QPegws0NxUnKyizOUx8uTpV2QxgdpX/R5GzqzFGcpr7u5mCCEorkYrXT1C4SwS4FezenOlImdxsuni5oJ0Io+q9sYtJpI+BvK6IweknIZpp8TycU55F626y//2i2Ug3zL5gba18y1BYWFsZh3b4iSlZ6NAofu3j6a/rsqnZz6FKGonGp7+tZTncv9f9OobiA2XMdf/77v8jEthTWY4gcthP+4OgWv/vKe1L94iBEb3B5wHeWeJ6nDExNfrC7MGS+z8b1qbay+QY12v3YJS929XMl2CXY1lI9gJCl9jcewiiDP8xpev9tVQN7UafcWNskJ272hOKCUB5BOCgZHaTjpUbwI8hmSswFt1QW14TL5zs+9/BGqhPe1+RWFWwWcC45tatc3OVFiS09H+dpupjySK2J+GBQFdoueZOZC9TLnHu6RTlKNcRj77mK6NrhNii5C9w/2uSFzpJXbXKfgwEUsWl9MTAe1xTOc3w6QcxV7fLVGKLepD6m0ueRCoiiAqYR1IJt+rGpY50pGAoIVURfqjB5Ur5tk46VHinzCIt9wXTQbYKvlMmrsq51nQfwFVQPU7+VAPlcWe4Jx386cr4o8d4SvCVcZuAF2ejWeZdilLxILvBvv/jslafcn+fB9+6lUYxTYhDksICXJ8gYpnM4/1F4z41D/uStlzCiKVNVlJnP+VdffZ4X+zFmla8pVuGwoO5zcW0j2EqRaClOZF1jfpXp7Jbp3DUdmAjdGNwCbJvebzeBbJ7a0rTgloqv0nYu3YjRYXpdV0M0gEljcnT9+J3D/FY6d6oHgtybUpwqpRPUQXEe+1zftM3DH4aLLPCxxTOIKPUiR4PBuIBYXX+25XmaN/Hq6T4YTYOPi9gspr7R54Iiqf+YPEDVUtcZXzir1jXc15/1wnr81ZD6n/p+joBPfVEezRoW8FnKIX5HoKQ38J3nvwH+qaq+CqCqB/3yjwBfVdUXAUTkHwM/BfyRC00R+RfA88AMeD9wS1WPvp2dGb46HxgYGBgYGBh4G1j9gfTNHsCeiHzykcffeJPNvhfY7r/m/pSI/NV++W3gtUfWu9sv+0ZY4K+p6g8C977ud7/af23+CRH52W/2Ht+WO5oi8jJwCQTAq+qHReS/Bv4X4H3AR1a3ZUXkx4G/A+RAC/wtVf3Xb8d+DwwMDAwMDAx8J/gWczSPVPXDb3FdB/ww8GPACPiYiPy/XAUSPMob7cEEOHmD3/2Mqn5SRPZJX79/XFW/+GY783bx577uNuzngL8M/P2vW+8I+M9V9Z6IfBD4Ld74CnxgYGBgYGBg4J1P+t7///dmRORvAn+9//Evku5UHqnqHJiLyG8DP9Avf+KRp97hj96tXPHUm/wOAFU9FJGPAx8E3pEXml/D6mpYRL5++acf+fHzQCkihao2b7gxA802XAbL5sueVcnmmAkYQ9ysiJlFouInkWKz4eK9llAa9j4bsLOGculp9svkYe6MGD1skJAze8KmuumtobpfU5w0LG+MUAf5hSeUqf638YbuesOyytj5IrhG6caG/DJQ7+WgSnHaEV1O9iDHjx2bnzdMX+uod5PrOTqymDaiRihOal76L6b4p5fEw4LirmHvcx0o2CaCgZAL3cSixnHw4ZJQRaZfsRTnSnEeyeaew+8ZEaeyzjgLvesVL3NGr1smdxW3VOY3DPM7yX3aeDVgl4H5rYyLZwzFGZQnim0U20TyC083dYTcoALdRJjdEbBKOzVsf2lJeewoDmtOPzihPI1Ur82YPTMFVcYPDKfvE+p9qO7D7mcXfU16g20Cy2sFmy+0NDuObmRwVXIRd/7QU71ySXN9jJt1+HGGrX1y36YZVJqy014bMboHO/cj2SzgRxkmQHbhcZctd//TKYvbkTiNqWawgclXHdWBMrnbocZi64AoNNuOuz+W0+20mIcFNz6W2kANlEct7WaGBCXmhpPnHXJdscDu71SEIrlmtjGEQsjmkW5i6EbC4iMdsQqMX88oT5Wtz53T7le4WUfMDX6aUZ5dHeN6F4jJ1SpPI66OiNfkapam72uR86cdi5vJpdp4MTK+1xGezilPlfG9luyiYfb0BDXQTQ3NXkSiMH214ey5knwWyS8D1cOG2Z0CUVAxLG6Beali+0vK5PU2ZdNmYLqIOsGdt3TbBQcfGrG4HanuGTZejskxy6C4SG2WXXhCaVEnPPiIw48VP45sftliaxgdBZpdRx2ydY7hKrN0r5qz+NgeGy8rxivi4eFHhDiJ7HzGMjoKLPctszuQXSQfzS1g8sBDUNQJ2aXHNIFmv0AiHHzIEQrYfEEZHXowAlFRm3xvBNwi4ivD4qah3lO2vpw81uhSjqltlOiEkKfjbdq0vDhLLl0oIT8G00aKs9Qe7jQy+eoZ9gPbxDxDTUZ5ChsXSnka6MYG26YxxDZKOxUWNwS5EQkCB1/aJzsXJnchm0fKRhkdpv7o5oEX/3IFGx32IKeNJdnrQnGpZEtd+7y+TOfV8VOCPlvDzFJ8ZsJ0lsbP4jwyOmgxTSSWluP3b7OYwKSF7BImr3fEwmDaSH7a0uwXZOcdsUi5pQ8/nGGCcP0/tMTCEHIhuwyYLuJHlmbbEjOhq4R6z2AbGD9URoddalsn+MJgvDK5G6h3HTMMaoTlfmrra7/XYdpIs5PhR8LOl2qWezm2UyQKl69XyLlh+8uKW0bcMiJRqXdy8lk6h9zc8+JfGWOu1cgrFduf1dSmxy3LawXGK8VJS72bc/QDllBAtxWodyx7n03nwflZzmynSJ87mmqDe8CrJR6UVK8bitPkXbZTqK8po/tCfgHTex4VOPihvi++GJm8vKTZK2inho2vzPGbOXYZCCPLg48U1HtKfiFMXlPySyW/hNGRX+dAt5uO6V3PxVN5P0bDtU8F3DISc6GdWkyXMqSbzdSmEiFc65hnGdMX+nzmh55ubIhZcoo1KtllAGNZnpbs/oeM0Umkq9L4ZjtDcdh/VAs8+BObzJ6KjA4N1X3Y+cI8va9JhunS3T01kJ97FtczDv6UYd5Y7EnG9d+D8qgjFhm+MIyOUq6yHzuki3QTi1v0488y0E4dtlNUFPFpnM5Oltz7c1ssbn+jG3tvD9+JyUCq+sukST5pmyL/HPglEXGkb4N/BPhF4EvAcyLyDPA68NMkn/Nr90nkR4FXVfWN7miu1quADwH/+5ut93ZdaCrw/0hKU/77qvoP3uLz/grw6W90kdn7Cn8L2LKj8XduTwcGBgYGBgYGHgePYS6Qqn5RRH4T+APSXM9/qKqfAxCRnyd9M2yBX1HVzz/6XBG5Bfwm0IrIZ/rFt4BfAH6u//lXRWQJFMA/UtVPvdn+vF0Xmv9J/1X4NeBficiXVPW33+wJIvIB4H8D/sI3+n1/sfoPAKrrT/wxyNofGBgYGBgY+I+Zx1XrXFV/gXRx+PXLfwP4jTd5ag78mqr+7KMLReSf9M//6Le6L2/LrHNVvdf/ewD8M9KU+zdERO706/1VVX3h8e/hwMDAwMDAwMBjRIGgb+3x3eMQ+HvfYPkvfrsb/K5faIrIWESmq/+T7lB+7k3W3wL+JfA/q+rvfnf2cmBgYGBgYGDg8SL61h7fLVR1rqof/wbLv+3rr7fjq/PrwD/rJ/044P9S1d8Ukf8S+LvAPvAvReQzqvoTwM8DzwJ/W0T+dr+Nv/BIAOkfwfgkt0cH+XlHN3ZJ6lcw86R3hv0Rrg6ogO8sapSYgV16zOmc9uld8rOWUDpibsgufBKLTQrB9aUgQZHWJym7NBQP53TbI0KW4SuwecBLhlrBl4LxynLPIRGK00DMDGrAjyNaREJpMJ1SnAfqLUt20YEI81s5fjxGrRI6A1aJeXqvsTBUX3xA+9Qe5AZXR9qppTgV4swimgKpESE/rlm8NuXj86fJXhghAdptRW6k8HY1kF9EEGH8INJNzboOq2Zpog+kCQ2jQ+0lc4tbGvKzjlBa5jcyUMhm0I0N7bSXtYMSxhkSIJt5ZNkhMQXZVw875jdyuo0UvO7Oa5obkxS236Vg3dWJZjx9kDjkpy1+o0CNEEqH8RE/zvrgcoPGgLExHVsjaaLB2OKaSMgNobL48QjbAFlEvZAfuj5IOoUcL/cdpgNbGkyrtBsW8RCDIebKcs+QX6T+1m5mGK+0U0s2C0zuKfMnFVNfBaRLVGwbsZ2Qn3eoycjmcKyCuEgoU9+qb47TABMjkCZO2GXEdODHEEcRjEE8jA4a1BqKl45on9rFtJGQ53SVwbQp5FkddBNHeX/G/OY2tlXcogMfMV0qOKDOIF6wjZL9/otkt57H1n0QfJ6RX8Y0CaBWiiNDt6Esd4Ttzzf4SeqQ6oT8YA6qRFsyfS1SnhgWN6AbC9lc0zGU/hzNDaiy3E1B6d0EJAjRgIsQCqE4Ej72h98DAvndHDWg75lTjVq0D+cPuWBVyWZpwp9tlPwy0GwZUEEzMJdchXtImkRnlx3LmxW+FPLzNB5EB9lcCUX6Ozw/T5Mpmk2DCem1JILphJgppk0PXwqhEMqTkCYe9n/H2zqdgzFL/Xh03E8uOWyo93J8ZbFngfb6BNsqtk7ngelSHxRNx9DWMU2gWURCkSZL6ThgXCA7zzF96LmrleKkS+OTV0wXsMucME77rTa9R9H0GraN1Fs2TZrUVACgASQK+QyKs0go08QdP7JUDy+Jo4zpXcfxBx2y7MfEKk0SKZZXn4q29jS7aQKKWiD0RQAOl8zvVP1rKvlFR73nUtsq64lUbhkp7l3Q7U+wi45mp8RXqV0lprDykKdwcttAN7Fkl32BjPNIs5UKY9hlQKIhP3a4ZToWMQghmnUA/6r/+rFLk1JCH/AfU6EJNXmaTFgZ/MimYh7nFt0GghBzqLfSZND8GB5MNrl3dwd74XBzIZSK3qxRp3RTGB0pXSFUD1Ngv6RTnZAL41dmmO/bJGbQjYQwSpMU1QrN9RHZeZroZeuYxi+bws+jFeotIVso2UWLSk676fpJZIZsoRgvKfzepklq0ilukY6xNEpxEQmFMBOTQtSNpgD5jv+PvTeJlSxL7/t+Z7pTDO/FG/JlVlbX0GySTVKUZEGQvbM2BuiFYHhnQQtJXngjb7wwoJ022gmGV4QBL2wtLNgLG7ZlQ7C8IgxTBkVK4CB1s7u6qqtyqBzeGOOdzuDFdyMym2z2QNaQXbw/IFCVEXc495zvO/e+iP/5f1QfL9m9d0R0YtTuSw3JULzQtKeKfgq2lWtpFoZ8FXG5wU/tcH8Bt5T7ivYJFRP5VUvSBXYXaE5ku1AadACzMoRSzmV3URYoOUV7pLGtRbdR7ktGEQpNKDTFlSwIC7mMny8UySimj2pUjLgt2O2bsxjoczJsf6P4wh80B0f6v/RD3v9fkZ/H/+j7/xD4h19A00ZGRkZGRkZGvjC+yG8rvyzeGHujkZGRkZGRkZE/NyQ+l1Xnbxrjg+bIyMjIyMjIyBeMVAb66j9pfiUfNE0TmH+cyFaR+jzD1hHderS3+OMKFOS3LdFppp9o0qNSdFO7iO4iGI1dNvhZTv7oBlZrwrsPUD4x/xDcLpLfBXTr8fOC/KomFg5/VKB7MQXPlob2ccX8qaK8bCAm0IqkHPmdGK2rmJh92HLzSzO010yfRnQn+q7J8x6A6DTV855kFJMnGf4mp7iC4i7i1j3RGcLFMckMupcmkveJqbG0x4rj78m5ipc7YuGYfl+TPik5/lDMeuszy/ZBhWlg/jhQPtvhZxkqJLppwfRpj/YJu/P4omBXK6ZPE/ldIL9pKF5GSImUWezVlmiPcBtPe5LhNgbTShLl1w3RGWafNLi7hlRlTD9cQR8Ii4qjjy22iZhGjme3HmJC+Uj5vJEx00o0Ze2g0QoJ1UeM1ZCkr/LLHUlrVB8oHx2TrGPxHTHH95XBbgN258m7KPpAwHSW/LnFtKItnTyLFLc9qo8kowm5JlmFaSOzTxp25yXuuxkkMRTPVmI6vdecmibgHl2R9H2qp478BqZPOvIXO9r74vHqVi325Qrz/z6Bv/JNqkczVIDTb3tMEzBbj+4DqvMkoykuW8y2I+YOs7M4rbn/L7YoL2b9uhFzcfdyQ5zlTD9q8PMCFTPyFeS3nvxyh9o2zL/fYHYe1QdUCJTPa3TdUz07QQXF0Yc1hMDx79/gFxUoRdIKdyXO3cWiwtYF2wea8iqRlJhpZ7cNqulQ25Zwb46KieLa080yjj5KTJ62mLqXPq0spg7ouiNMMsqXiuJWsawtyiuOPpF+SErhC0f7IiO/hsUHgaRheT2hm0yYf5yYf7gjOi1xpY/opprjD3aY5Q7SnPo0Y/ppIl8F8uuOpMGuWvrjgmQN5bMd3XFOft2QrRyhAdtEsptWijpUkrM6WEiQ33QQE7t7E5JSnP3ODcRIfz7FF4bi5Q7VB8I0R4XI7t6MaKF6EfClZvK0pZ9bVIjkNx3ZUotOdtth6ozdeUW+iuS3Ae0jqo+4ZSJpRfl4RX86YbruCVlB88KRjKW8hOmncn0qgW49obAUlzWhtBRXirjMmH8iZu/lZWT+nTv8cUm0mtIn0W9q8JXDPi4oLuHst+8OS0Zj7rB3O5Iz6OUOc16hW8jvEtVLT3bXonxE7TpSYckvgQhuHVApUdwYdA/ZqsdcramswVeWmBl01zP7uCYUlvrcUmaa4jZIoYzllmy9w799hukjep3Ibmp0lxFdQVKK+SeB6ffuqN89khi97tCtH+ZFqb7Sz3O0h2wJ0yc9pvGiE3Ya3Qa601w0f0ZRPtXEq5LqGUw+7SS+L3cQAimzoBR6VXNcnLMKhuxOy9zxwqNCYnfhCC9yjj9SlFdDUYupYfleidtCtk5Mn7T4qZiNo5y0xYhG008z7A7KS5g/EmNy3Yme3dQee7vDXkPzzpFoPaeak29H8jsPCpSXeCker+jemmF6g+kibqOYPLVkq8TsgyWxdJjrDcRInFUoP7T//WPyG43pcsoXMP+kw9YB1Xry6xazaaVPTycy/5uc7U76YPq0gwjNmSO/lYIIxbqVeTmWmFa0m3Yb0NuWWGWUT3ckq5k0UjClvTchv265+osV2a2hfAmTx9vhXNVgti+aYRXEaL561hFyTbbs0LuOo7salJI8d3J/jFWGDqJTfVNQX+yK8i+Fr+SD5sjIyMjIyMjIG8340/nIyMjIyMjIyMjnw2dT6/xNZ3zQHBkZGRkZGRn5EhhXnf8M4zaR/K4nWo3d9BAjpvGYXUdyBmLE3bRMnxa0R5rJM0/1vRtoxWdTX/e4fkK6vUMZQ5g4ipc1KiSK51tU3eHPpti1aFXM7Y44yfFHGfnljvPf11x/05LfJnQXRdPSB1QQTYt7sSacTFAhsPiO6LHyqwbd9mhfoFuP2nX4swm69ZjbHbPpGdEpZt9biW7OauztDiJkdU/KLNEZzLYl2Tlup0XrCKAUZlVz8m0n2rnHK9Bg6gmmyzFtpHxeE3PRDLnrLccJ3K14bIZpTnHVMZnmorO8rUWrNWhszLKGlKi+d0M4qpj+y2dMj2Z092eiQYkR5RXZy6XoVcsctVzDpMLc7pikhG56wiRDrbbY1kv/9GHw34xktw35VSJbFqiU0Lte+qvpSLkTbWnbQ5Ghr5dUL46JFiZPBz3u9+Sa9aomZQ673kKRcxYT/dThJ4bZt25oH85xtw161xGmBW6VwEf8cYHZ9eTLAhKc/t4KUkLfbkmTHIBY5uL9WRbkty2L72qiVRTPNiStyV9uScagUiIVDnNxDs9vmT2aiibt8Vr+wo1R+i0l7I2HpgVrUZln8ryimylUjJhNQ6wy0XLmGar3mKV4oercUb5o0F1A+Yi+XgFgly16vQNn4XZJ+sZD/JHh+HsdzZlFdwHee4jatbiPLyHP8Odz1HoHvcetdxTlfebf3REmGclpsudr0vefELsePZ+iTmdkd5JLs0eK8smKMC8xq4akFCrk0qZNi2o89hrqd4948Bu30AdSYUnGYO62RHvK5DmgFcWTjej9rgrxJN31xMKRvVhD76meV8waj325IlU52keql6LjVDFibrakwqEub3Gcojovutcuoj+94vijiqSg+u41GE344CPy994hTkvcTUItN+Ac7Gru+TNp36RA+Uj2fI3LrOT59S0mLlC3ayYvpvSlYvK0RvlIsqIJNMsalTtUjKh1TTyaYJY7Ft/RZJ8uwWi4voWqpH/nDLvtUXWHu9XQB8pFhtsp3DZSfSyxiFKobUM4mZG93OCPK1SEyfNEs1DMP9zhJ47i25+CUrjrJfF8gTMK1XlICRXm6D7DNgnd9oSjEr1t0UCc5OhVDcZQPF5R3T+lvPLkL7fo5Q66nvDWCeZ2h951sNqgwgkqJBZ/qEAp7MsVcTHFrGrspSccVejh3O5JjfanZGtL9f070Qh3HZQF5naH6nJUCOgXt6ivX1A96/BFhumlekp204pfcifznt62xEkuvprP1+S3p5Q3kfz5GnW3oX//Htn3X9K9d45uo9wrFCw+0KgE+XWLXbUSs52HEFF1J7nT9xQvG+zWEa0iu+vQnSdpzfwTS3ejuPjNW2Jm0Z0nKyxJTTBdZPLREkLC3lnai4rqhac7MsRSxtNue2aPxZs2f3QHQMoMfiHjSRSP3eLTNb48orw05Hee7Fo0tKoVfTfPXqLOJ5SfrvGnFfnOc7oNZFdbVOdRzsDtEvIM3bQwkf7OrxsWHyqy2w4VwT65knvOrMJerkm5RbWe7PENcVaRVeIPPf83N2DEyzZ7KfMOvYe15I1ePCRfgmkT0SpU02N8hPWWNJ+CBlV3FJ0HrZl9UgGw+G4rc/uTF+TqbdEBVxnZc7n/9EcF+ZM7CAH/0ceon3ufNClQz6/IngOzKanMZM7xk4M/9BvB+I3myMjIyMjIyMjIZ07izXro/ZwYHzRHRkZGRkZGRr4M4viN5sjIyMjIyMjIyOfA6KP5M0pSSmo5r1r8LCdZBf3gvbipIc+GGtIwfdxQXlqym5rw0cfoX/oGar0jnh0RncFWFf7tM7LHN6RJgVv3qNWOcG+O2bSo1RbKApZrTJxi7rb48xlmF5g/0lJPedNCCKi7FabKSFaTrIEgvotH31qhr+7w794jpYR9eg1GijjrrkD9wfdIX/8axVUrNcOnuWgzlRPNV+4gBPTdlng+R92sKDMrXmxr6QO164iTnOq7l9B2xPsn6OWOZDTlpfi0EcHebAmzEuUj7maHul1LfxUW23qOPoy4yw3dxQx3vSUclaLd6YPot/7wE4w+h/mUMC0AMJuWlFnRDGUOfEC1HelkLhoeo9BNL/qZPkJMxGmO6oN4+Cmw1ztSlaF2HRmIXick/HGFvVyJJujJnejvmg7qhuMPWvqZwa5b9ItbwsNT0ZKGSCosag2pcCQrsWF34vfnVi36+Q1pMUN3HnW3Js2qgy518szjS017VlJ+6xnx3rFcm1LEwmA7L7qw3/+Aif85YmbEey8lktbE0mJ//yN4+wLyDDLH/KOapNVBzxWnBYSE3g599+kL1MP7qN5z+i+e4y+OMM9vCfcXpH/5B6hf/aZos3Y14XSBuVqDVphth960pMJBWRCOK8zjS+LFAr1uJMbqTv7//pzj310SB71p97UFSSnyj14C4B9/iv4rv4x+/ILs5Q69qdHXK9LRVMYxy1DOoYpCXDuMxr5cUbQB1QeSUbDZoQCz2pDqGsoSNa1IWov/ZN1B1xGOT0hWY24ipo+4/+/bqHffJs4L9OUKNc+ldnznUVrDZie61pjoTnLss0isMrKnS+bpCPtYtKZpWsKT5zCf448K3Kd3pGmB2fSgFJM/vIYQRIMH8Fd/hbRpCPMcXXuMc4TjCSZK3KIUsXQSpusdOINay5ygWg/OMvv+jn6eEZ0mVo5kFG4l2sFkNWrVkMoc9fFT1IN72LWMV3IG7ecS0z5KPliDenFN90tvUz3ZEspBU6sU/aIke3RDPJmiP3oCeY7tRVdYzRy6N9jbHWZjwBrSrEI9u0JvG4nNMiNOcrK7lqo0VI83JGcwy/qQv2rb4u8fY+92oGHyrCe7Ep14yjPi+RxzvZE21x3kuVzHpy8or+6g74nv3MdPHNnlHXExx9ysiUcV+uWStN1hbydkz3uSM7gnt4T37ss8cr2BSY6+XOLfuYfuAnbVUk0M+Yt68JOVeSSWDvPijnBxjFk1Ml91ntlTj/IJVXfE8yPs7Y64mJO93MDlDSyOiJMCt3Hizzh4uepdSziqMJsGfzrFvlxK7ux6rFLouhNf385D03H0gRIPz+UGdTpHvbhGX5wy+bTBfOtj1MWZ5KsXf9uUWUJeEjJF+WyHanuq5y0hN5IP58eYT55hrCX92w/gF9+HPqJe3mDfmnH0YaB4Im1KjZc82bXErz8ke3Qj3tCrFtV59CQXzfliInrys4WM73JDOJlAqlAhUT7Z4OcFZtsQz49RbY9qejAa1QW5ZRxP0HdbCg3lkwBWy3hfLaEsSM8vUUVO3O5IbUt+MiczBn+Uo7uIP5+hmx5dG9RyTdruSO88kH7sOxZ/KDpvu+5QPhJ+8R0Zj8Ki+oD/ww+w771LfrcllRmYDDObEecl+tMrmE6g6yWfMguPn6DiGcp/ts8cfybGB82RkZGRkZGRkZHPnASMGs2RkZGRkZGRkZHPGkUafzofGRkZGRkZGRn5nBgfNH82USFi2kjMrfhchkSY56IR8x7KjJhlKO9JVhOdaOPse+/Q3J+hTyqyZyu00fTv3wMgzirCLMM9uYV80EZmlvDgBPvkinj/RLwL1w1m29EdZagIxcsWYiRZixp0l2ZZE+YFqg2EKsNerV+1vemIp3NU3aN2ottLTYvtgmjSYsLcbOHZJbz/UNpWOggWjfjHxXvHxNxK/eRdh1VK/Mn6QMozqPJDA0MmLwAAIABJREFUcLs7qdMeM2lbKjL5C6tuiLMF6lxqB5tVI9pGVZFyh/aJVGQkM2hfWw+TnPjNd0laYT96hprk2HV38NtUrae/L7pM3frBY60jHM8PfpDJaNCKWDhANKPZyzuYVvQTh2u6Ycykfq1ZtYTFlFgY3N0aBcRZQXrvAaYNuGUjGtLjqfRJZgknE2JusBFSYXE3NeHBVHSFShEKi7p/Ip6CPkKegzGotsfPpSZ2rhSkhH/7jFBasssNMR/qFWeW5DT24QNS22N2HeG4HPqxBqNQZyeEaYEejmuvd6jletDpGhkrrfGnFcon9Ne/hp9l2NsdqvWiez2eEQZ9ID6id4MH7K4T302kjm6cFSStMX0gZgajFLEUjz4mBboWn9n9WPbzHLtWRKtBQTyZEUpL/v479JWDh+dyjvUGUiK8fSbxdDJBNz30QfSSvRIvy02NP58TCoOdTUiZRbc96d4xdB7V9KTjgv44xzqDebQl5lLDmMyJpu7dtwmLCtUF+ocL0UvebfH35ph1Q7xYQEh0x47gFPQ9SSswGlP39F+/wC4bYmYx1hIujgbtsBNP1qslFDm0Lf07Z+hO+ioUluLZDSaz0k/WoGIkPn9JfLiAowoRnSrRH6dEPJWc0Xdr4umcUFncqiM6TSik5jJKkRSijawy0YNvZ4R5QTKaOM3QXUQbLdrnXS/xt5igpwVEqTHvK4uxmva0AK2wiwl+npMtjglnM9FLzivcssOtoLs3xew8XByLF+O79wmVk7jykaQ13XFGft2i1w3+fCZxkhL2ZgtdR8w0MXfE0mHXom0NM4epNSE3mPWGcO8tVJWhh9r26eSYVGXoJy8JlUN3Q15ZTSoz/Cwn+9Sj5lOShpRZutOKTClp35DnyWrSYkbMNHYb6I8Liuc15mZNmhTiido3ois9mqJ3HenlFenhXPKviegu0r29QAXRm/dnE0xt8Q/ekzyuHHbn5X5hFPquIR5VohdPiVgYUpUTpjIHxUxDlPkqHheYxouOsO6IZ0dyvZOKMMnQraf/i+9jdl7iKLPYT68J9xe0xxpfKJa/MCW/Ddidx61a/MNTacvX3yIUluz+PfysxJiOcDZD+4ivDGFeYK7W+IcLzKYlzkvR8D7ZER+co3etrB9Y1RLzvXhtxipDP7+RnO08yRmSkTnC3jWiK7aaeJyTP13R3ZuSffiC8PAUvW2h6/GzXK7baMy2I54fQ4zoxTHxdIZ+nGA2hctbuCf3SvvpNd3797AvV/iHpygfZA5SCvtiib93RD8zuHUgVBZMSbvIqf71x4R37pMM2HvnhHtz9LqlP5tK/fqff0fae/+MWDn0v/o26a0T8dB954HcR8Of6RHjsyMBfw5qnesvuwEjIyMjIyMjI38eUSn9RK+f6phK/ZdKqd8dXv9GKRWUUifDZ7+mlPqOUup7Sqm//1O3V6nfGPbfH/+f/7h9vpLfaI6MjIyMjIyMvPF8Dj+dp5T+EfCPAJRSfwP4L1JKN0opA/w68B8AT4DfVkr905TSt37KU/ytlNLv/KQbj99ojoyMjIyMjIx84UiZ4Z/oBWdKqd957fWf/YQn+ZvA/zj8/18DvpdS+iil1AH/E/Af/bCdlFL/5/Ct5+8qpTql1Nmf9iq/mg+aWqHbgN71dKclzVsTzK4HIM0n9OdT0VQZIz5lWtFcFISzmdRfTeBPpzRvH6E78f9TXvzXuvdOQSnx51QKlRL+a2foXUd/VBCnOc39CXbnsdtAcpowL2nensKkpDsp6O5N0XWPihHdedp3F8SLBf3M0d2fkbSm+dqceDyjnznMN79BOKqIVlFfFGy/sSB+422a+xXdhejn+kUOStGdiBYw5Ib2NCNp8XtMmSXMMsJxSSwy+kUBRhPLDD9x9DNLd1qAj4TCEu4tCFNHmOagNX5RkQpHe1GSnKGfWkIpX4h3JzmpzOhnGWrQAfr3H6DaXs6tNd2F1JpNRoneRinaexPi8YTkNO2DmWhWc/H3C4VFxUR/PqX/+oW0v7SEo4p+lskY157+pETXrXhQns7xZzOpJd72dEeOWDja04I4LUhG0Z9W2NsdSYn+KjpNf1LSzQzZXYdqO1QUv0uUbJ/KjPrhFL8QbWeoHHrbEkrR6oo4VqObjmhFD5us7N+dT/FnE1RMhEKuO2Zm6AvxnEu5o37viPD2Gd1JTigtfpbjZxlm3eFnbmirAWNo3juhfmtKLCzRaexbD1B1SzyeQJETZgXx/onE03lFf1TQnhWwq4lWk+ZTlA/EMiM5Q/tgLrWnG09/PkX7SKgs2W0jJdL6gIpJttl2JGNkTL/5NvFrF5JyrRcPy9NK4uu4JFSWWDqar5+iUjr0DVaLRrkP+JMJtB3doiAaRT/P6H7+Ae56J7GgFCom/NkEXXf0xzlm19Mf54RT0fP545L+qKB5a4JbedEPn8xlrI5KzLKmn9pD7IV37mNutjQXE9DgZxnxZEbz9TN2vyIarm5RYFctbt2BNtJPD+ekzOLnOXzzfdEqlpaYG5p7FUkp2gczGVNn6N4Vr0SQeAiFxJgZ4lyFQD/PUW1PcvqgKfQTh7vZiVZ572l6ryJW2aAHjPiZk3wdME0kOCV1428burcXJKPpH4heVMUoWtbM4Geiz/X7fN3rYYft8tuWbpERZwW+kn7rZ47tz59ILjpNmGXEwhAmVjTwpRVP0ATpfIH99FZ030rRnuV092cyt7x7gW49fupImaG5V4o/5c4T3r0gLua051Lf2k8N7cUE3QT8VLTjfio6WNPKe2bnCVMHdSNekD5CkHkjlk5qg9+T+2N7MSW7rgFwtzXKJ2KV4S43oi1sPPXbc1RMtCcZu/s5yWj8vTnRGdqTglhkmF1PnOSYXScaei3+ua//vKn7QH9civdpSqQqx16uCZWTcS2teObWvWjyQyK/i+SrRHSK5JR4e4Lsb5ToLdtAvFhgdh3deYXuwxBXgxZ2WmK2He29iegttRoaxEGf2d6fEUtHe1YSqyEW3r0gfO3ioI2PuaWfOVJu8NPBU7SLhHmBu97Sv38PXff0pxPi+ZGc2xp009MvSlTT4edyj1F9ED/LlOh/4SHJGWKm8W+fyZzz4Ai97fCTTLTInfjYml1HNzd0R3KfUY2neLbFf+MhKohuM37tQsYc0F2gO8rQd1uS1jIeCfjVX8BcrQmnUxj8s9WbIotM/DQPmlcppb/62uu//XGHV0pVwK8B/8vw1kPg8WubPBne+2EY4D9NKf1l4NM/8tk/GR5Af1sp9Xd+XDvGn85HRkZGRkZGRr4MPl8fzb8B/GZK6Wb4t/oh2/xJj91T4OZP+OxvpZR+Ryl1jvz8/lsppW//SY34an6jOTIyMjIyMjLyhqNi/IleP/IYSv291xbnvPXaR/8Jr342B/kG82uv/ftt/vi3lXve/RGfAZBSugR+C/gLP2q78UFzZGRkZGRkZOSLJgEx/WSvH3WYlH49pfSXh9enAEqpI+DfB/731zb9beDnlVLvK6Uy5EH0n/7R4yml/j3g0WvfhP5Qhp/m/x3gox+13fjT+cjIyMjIyMjIF85Bf/l58B8D/3dKaXs4W0peKfWfA/8c0WD+dymlf/v6TsM3ov8X0Cmlfnd4+y1kFfvfHf79T5RSNZAD/0NK6V/9qIZ8JR80k1b4mcNuOppTi61FTB2tLMypzzLszJK0QreR/KqhvijwU0coNdob4tSCVoTO0M8sdmnpjzJ8pWkXJyStKGMiFEYWK0yPSQr8vCDm8kVxuzCoBP3UEDLITib0E0PeRTHHXe9gVhFcRXgwoa80tlb0k1La+84Un2vsaYWvLNldi4qOkCvak4Juqok2IziFqyPd+ZSYD4tb5paQK7a/sCBpyK8U3XGGW3uU1dJHiwkhN9idZ/ugJF9Fur0Q/6zAV5psFVA+0h1lcJKTNLRnJf1ErlH3Cd2K8bDuoyw8OXGyEOpMFjIkp+lmFmIlhtUh4acZodCykGliQUF3Lp+Hd47oJ4bprsdPHKYJYtSsoZ9a/MSgfQSnSVZRv3Mki4N8Gq6nJ+UO00VUiMRcEwpLP7PoNopBcxuIuSU6GT/TRXQf8PePD4ucQpkfxNq6T4SJPSwA608qVIiEiYMI9cMpySrc2h/isHvriH5uyW47fOWITtHen5GMwt5CKAdRfGlQEVm8UWpUb4mZxrQRf1xAgva8op8adJ8TShk/ab/Cv3OPUFrsthMjcA1hOB9JFhYkp+jfPQcN/fkEtEL5iNKQrJJFKSkRMk0/NTLudw2mDaRMxqc/lgVwdtvjz2QRl5/nmE2P8jL2KiX8LDvEgJ8MeVRaQq7p7s3wU0N+vV9UpAhvnZCMEiPtI4vJhiIAIREWFe3CYXeRkFf4iSHpgqRl4VDMNMonfCVxlS0DxVUgVhntscNZLQuhYsIfFbTHjlwp/OyY5BT+ZEI/s6iQ080Nto6ycGiIRyL03zinnxny255YOpJRNA8mRAPF1tPPc9n+/gxfGbq5LHwobsQ03lcGSkO0CrfqxXxbKbpTMXsP04KkIExkIZgKCb+o8KUh84HuWPKuO63QPqJCIinx3zNdFOP2qSHkirCoCKU5LFLRw0KJmFtZsDhcRzL6EMe+NKj7M0JuyK9q8EHiaxh3FZLEnk9DHsrCND+VhU7xoiLmmuzK050XqFCShvaHwg7HkgVEyWj6mSVpiO8d0001tg7YpqdblPhyMLQH7C4QnB4WTcq8EwqNnZeE0hCcxq5bfGngF9+S/rrtwCiS0QSjiU5jSoevDKpPtGelzAt3a1S2oHlrivIJt+5lAWJiMA1H8u4kE5P3PoICvyiIVsv8kZnDQ4IvDXbVQikL/uoHJUlDdpVkXjydiGF/qeXZwigm319BjPTHE3xp2Dw0RAf5XaI+NbjldDB2l0Ut3ZGFI0fxosGfFPjK4EuD7iJm1aLbHn9c0R9lck8aikSE9x/QnuTYrScdl0Sr8HOZ301jMcjiJVmwZrG7nvZE7iva5/hSE60jZppsKX0bnaE7m6CCDJafGtAQbS5zxankVebEeD9ve/zPvSX5MM/pJ8O8WyOG//OcZBVxktHPHOXtRuYYDaFQJKdRIcjC1IlFZ5qkFN1JQf5ii/LhcD/a/so9dJdkEeDc4TYK//YClUBrja0jujOf8VPHn4HP6UEzpfSPgX/8Q97/Z8A/+xG7ZsD/llL6O6+/qZT6n4f9//pP25av5IPmyMjIyMjIyMgbz5tXgvIS+G9+yPv/9Z/2gOOD5sjIyMjIyMjIF81eo/kGMfzU/ls/5P3f/NMec3zQHBkZGRkZGRn5wkkQ35TC658fX8kHzWREe+lnOW4b0W3ErJpBt6Zxm4BpI8nujd07sjvRbNhtQKWE3UWi06AVdhfEyHZ4PxQaU0twqChGtr406D4N24i2KFuLnsruAm4DobC4TUD3kTjJCacTYmawdSA6Tbn1oiWbOVQTUEmMmEUzJDo63ckxTRvJb+UvIbuT645OYdc9MTe4zXDOXGNq0VC6tT98Ta/bQCwM0SlU1BS3Ad3tzWwToLB1xG56YmFwWy9aKKfRXcQqBg1kIgWIhTkc023kWH0mRsIqJLK1aI3sTvRfKibcRgyF7c6TnCYahdt4urnFbQPJaPlMi6You+vE7NwNOp0uQpJ2RKvFWD+JTlM5TbSa7kg0VjHTuGWPStAtMtzKE51GdwHTBiZXG/yiEuNpxcHoWIVEmOSiybIa44N87jQpirZJt4lkHRFQPorR+taD07iNFz1kTOgASQ1axPMpqn9lWZE01BcFxXUvhulRzo1C/q01bu1JVh/6MCmJD4a2Jq0JRyXKJ5JF9LghghI9ZijE8DmUYoavmyQmy72MIymhnMa2Ulu3vZigu0hzUclY+0TMlRjPD31DFN1ePy1JVmFqT8wNdis5hpZrTkaR3/WE3EibgZQPGs3CikauFB0gIYkuTSViJpoq0ZtCtuwJmcYMbVZetlUhYRoZUwDvDHYnYxudAi1tMH06jK3diJm2qaU9+dK/Mi4PiX7usFsPAexOtgm5aOKiURgv2jHTBJRPB+NsXxrRCSoxiNed9K3bRJp7OcVVd2in7gNhIuOyH0fdB/qjTOaVSTbkeKCfW/Qq4uc5buMl5pXsY+qI7hLd3KG9zEl7nWZ/UmI2Pd35VK41iYG88lE00G2U86fBxD0k8utWcrqLJKPIVgHtJc9MLTpk0yiJCadwK4+fiT7PV6LB3M8P+/2SfaXrVkFiLO+jxGYlWuyQi2azX5SS31H0pftxMU0klAYi2J0XjXAXCZmW2FGA1oPJuUb3kWQMbiVt7mcWWwe6984lxgBfatz61VyejCK/7siMEv1qSthNB8kNmnwO+ZSMop8a7C6KqXxIot0EbB3FvN+IDpshX9Wg7ffzHN0E0a8O2lTTINrIkNi9lZMtRa+914knC/3cgZI+T1Zht1K4wm7FQF/FJO9N5Vr9YBK/15W6rfRFftOjO5kPJF4k9vb3QTdott060M8GLeiuJ9lB36hBN/HQFrvpJH68xJ7beNqHc8mN44qYGXQX6OcZ+V1PUopoldwPnUa3Q3GUkOi+thBNqxPNue6izM+FzFcS817uYaUjTXPJ7zbQLiymifT3pnJfzAatbu0Jkxy3DrjtG2K48wZ+o/l58JV80BwZGRkZGRkZeeN58zSanznjg+bIyMjIyMjIyJfB+KA5MjIyMjIyMjLy2fO5+mi+MXwlHzSjVazeG7RNTcLnlubsDF+A9qKBQYFpE77IMH1BtGrQDTH4UiZ8oXC7RMhA9xkhg5gp8eXUGrCoACFX6D6J3sSnQxtMnwiZxe0SPpdzqwQhc+gefCHb9JXCdGBrOZfpwecW2ybiXg4ToK/k3Pv9khYtoVyjIttG+tKhA0Qjn7ldxBcGc+7E07NSmE62N3066IG0l3NmW2mPbeQ8+p54y0ULpoO+VGgv2jqU6MGCe9VPphe9YcikP+szMxx70J8F6SfTSfsBSOKVZuuhPV7aU59bVJRzmk72A4Z+lfeCk7aGDHSwoq12YFroJ6/GgwQ6uMN+m/sW0yXAiZ6sn0i/tonoANzhek2XyfUMkspkJK6CU9hWzk2S8d0+sJgOfCEenCqmwc9Q2mzboe+7V+O3jwHbJJqFxK1KiA6wk/ZoL2OgIqgAvlLYnWjSkpb3g8sPYyoax0ETGtKgmYPgJJZIEG0mY1O8igW05EXSRmJ7aHO0r67RVwoVpB90lwYtIEO8uoMHYdJDHg2xHzLR0Mq1uMO59aCb1CHRT6Sfs5U9XLfP1TBGcnzTSV/qkIjmtbgy2aGfZWwhaYfpJRZMI8kUMvtqHIa8JfHH+k73EB7Y4boTIXeHOBTvVQDR9O5jLhn5t8wBFtMPefTavFCfmkMORCvjHp2Mtw4JFfZxZkhq6LvcYZvE9sIetIXRyHGjlfyLTmIjaQ7H3/dFMuLFKT6w9jA++1zS3qEChz7fzwuH/h3Oo1IiafFpVHGIveGcppcYjdZi2lfnjm6It2wYx1aOla1Fe5uUezXHtLC7Zw45u8+daIf5rVLYRuY3NaT2IaaBpBy2KWWe62U7HfKhT+T6Nw/sEJdyPSRoFiW2Hea3IV9kzpD99MNMjuUZ8naI31JyQR/pQ+wnhfR9bjFvObkvDOO6z6eQgTovZe4b5paQga0lr1QP2wtNc6xx20TIFZMXnvrEEHKLbQbNapcID9xwn8hkfUKfQDmydWJ34SR3FdSnE0ByTrTtr+53e79U1Ks5Jplhbs5fn5OnouUt9nGboYLs43YZ3WzQjncSS0lJHyado+IwbhGic4e5EfiBeEnD8KsAzQmYTrH8eiHjPfSX5O1+zItByy7xattEe6RlrAyH9si9Xvqtn/ywkt9fAgn4MeUlvwp8JR80R0ZGRkZGRkbeeMYHzZGRkZGRkZGRkc+eH1/H/KvA+KA5MjIyMjIyMvJFkyClr/43mm+ImdRnzKCp0oMew9WiudGeQ41g3YsWRHRlos2zjeiO9tsfdITDHxwqipZkf46DNqkTbYttxCvRdEneS/JeNKJfYpCF7Pdz9au/ZHSf0CEd3tuf+6ALVHvtHLidtFP3r96zTRo0V8MB1aAPNArtXx1D+3TQ05Hkv7qX63H10NYuHbbdYzrR17id6Hp0SNihrXsN3+vnUUH89exr12Nbabet06ARkpcOHHRpew3N6/1k21dtVlG0QvtxsK2Mo3hUMmhvOfSTbAu2lb60zR+/tv2+pk2HMdL9D46RbWVMdUhDX8s1va6hlRjgoENTMWH6od2JV/v1r53Hp0O/JiXn0/7VeDDoiw6xsG924qCnIsnY2FaOsY9DFV4759An+zFSaa+d4gf0XuIFK9urQbe41wnuNXm2lj43jVzfIW6T9LPp9pqtKO1p9/rSQdenXvXvIT560bBJ7AxtHbbR4dV++1zbj4v2Mi4q7ONQHWJIJQ66MNPIGKj0Kr73+a77dNB0ul0c9Gr7Ppcx3+f3Pk4OObONh/7d77fXDorGUa5xr3sz+xxo9mP7Ks9VGjSRatDpDdq3w7mHsVJBxmM/bvs4Ua/dr/Ya7n0/7ffZ560a5sik+AG9pW1e6+dh/F39Kn7lOAnb7OdDDvkHHPpzP7fu49d0HObMvT5Q9Je8ign/R/J2yJ1Dng/b7/tr337Tyf7ptf58fR7V/TCv7xIhH9Lntbh6PcfcLh7aaVuGOfy1eUIPueNfywU75JxPQ9+8Ns+91r8qvZrLSK/Gy9UJ00tMJQNum0T7GV+dz7SJbiq6w8O9ZTdoEoc5Q6XX8qkbtKXDvLTvC9LgY5qkf6VvksRlGu4pHT8wX+3nBBV/MB73c60etL19tdfzDrkXeC1vpQ37/bRPMsfEV/PpXht72G+4b+7nqv3cd7gfNENMD32+P/4PzHPx1dju8zdpiBlvDjH9ZK+fYcZvNEdGRkZGRkZGvgzGVecjIyMjIyMjIyOfOSmNi4FGRkZGRkZGRkY+H1IYa53/TGK6xOxxR7Sa8pM7/KJC73oY6u0qH+WvCKXoFwVEMG0g5Aa77ggTh123h22T1uIvuGpY/uop00c7ed8o9LYlHFXouiPmjmSl/m93nFE+WRGmhdSPTYnkDIQktbBfq8u8frdARZh+XJOcRrf+cF4Vpbax2fX4SYZpA3rbgR3q+Q71g1UbwCgIiVRYCInN+xNpax9IzqA6T5hkmG0nbQFiZugW2aEm+V4LousejDpsHzOL2TTUb8/JVj1+YjHNUO99qJNNYvDZU6D29bchOoO7qwnTHOUjKkRiZqSOeCV1nrvjTGqxD/pOs2kJ0xx7V+OPS1QfMbuO9t6EUGqKF7X4nvYBf1SgWy/1kjc9Yeowtadd5IRSM/lkQ8wsuu5J+SCqHP6I1J3US7bLhjARAZefOXQrdbKTkX5WIRJzK+MwzaTebkzothe/w8yg644wzaW/CkfMNLsL8aqc/8YHxLcvUDHKeQavyVgYdCPXHwpN+bIllAazlTrGutuL36RudcwNykf6qcNtpFbxvl1SEzqiOk8sHCTojjOyuw77/I40r0hWH8ayO59iWjl+tApfWbKl1JPXfZQ4A9QwEfrjErtq6Y8L6ns5KiayVSC73pEyi+oDKEUoLWbdEeY5ZtXSn5VDbWoREJu7He1bR7i7Gj8vICXstqc5L0FZ7C4x+2gt1zzUPvbTDHezwx+X6NZjbndsf/GU4mVNzAz2dkeYlSSn6acWt/G4xzfUv3iP/HJHd1piai914J3GvVzjF5ODhs5XDrdspH+G/iRKHW5/VGCXDf1phV13Ul/cKJrzgvLZjnwqgi93W0tflg5zuaJ754Ts2YqUO8Ikw95sIURoWsL9E3TdgpF4VJ3HH1fY2y2rXzklX3rcXUt3UlB8dE339gKz7YfjW+zVlnBUYpY1qXQyXwxtV0HqcNttP9Qyl3rWzYlj8mhHv8gpPromHlWozkNKhKMKe7mmffsYd1PTPJigQyK72uHnBdnTO/zZTOrYb4YYaXqSM/RHOdlNg584so9eEt46wbxc4R8co0IkGS31rZcKX2hmH64IU8m1fa7ouiMZg4pSH7yfZbh1Rz/LyC53xEmG8uFQS1w3Pf1xifaDDnjdEiaZeLfe7AjzEuUlbkMl+ZysxhcF2TpRXLboLqJSQnWe5Ay67gmzQubllKgvcha/9QJ/cSRzQB8P8/N+PomlozvNCU4xfdRgVjX96YTs8Q1pWkru9gF/WmHWHRhF/XCC7hL5VX2IJV85tI90R27wNVWic/RQXnagkPvTtkf5hL1a0z84wmw7QuWk7+9aQm4wjZd72s2aOKuIkwz76JJ4foRqPPXXj8luWvSuxx8X2FVDGGqFu5ua7rwie74GoH0wJ7upCZXDPV/Rfm1B/v0r2vfPMLVHdzLv1vdy6jPDxf9zKTl7t4HZRGL06TXdz12QfXJN8/PnKJ9k3IzCLhv8LCd7ucEvKpnnbze0753gbhtWvzAjGsXsUUNSkD26wd8/xixrwnFJzAzZizV+UWHuauIsR69bmWdnJfqjJ/R/4T2yZyv6+3NIie3DgoNw9UsnjT+dj4yMjIyMjIyMfA4kfuYX+vwkjA+aIyMjIyMjIyNfBqO90ZuDUurXlFLfUUp9Tyn197/s9oyMjIyMjIyM/GlJQIrpJ3r9NCiljpRS/4dS6veUUv9WKfV3X/vsbyulPhhef/unbbNS6mOl1B8opX53eP33P26fn4kHTaWUAX4d+A+BXwb+plLql//E7TtPNFKPOOYOFSJ6J5pLc7NF9QG9qtHrBhUS+YcvxH+v9iSnZft1Q7IG1fay/a4jloPu7Uhqyop2U6PrDr1pUT5i72pibsivGmKRiWfntgXAXK1RfcAtG+yLJbqPZC/WuG2kuAuybesxy1p0dOsaUsJdbtC7DlN70do5g77bEgddnK5Ff6qvV2AU5vktKgRsHQftZMI8v0XXPfauRj+/kfZcb0hWk9122Lsae7NFd9L1ImxbAAAJPElEQVQH+upO+mvTou+2qBhRrRdPwZSwm56QG9EtGYV9dodKCXO3Q+31fVE0pPmjG+mbm+1Bc+ieLQGw6w6z7agerYlOy7GuZIzszRbV9LLv8zuIkeLxHW7tZRybDmLEvVyTtCZ7dINue9xNjbneULzcMflYtEZ7zSmRQVtlMXdb6YdtR9Ja+udyhVt1uOst5nqD6gLmrh4+96KpAsxyJ/quxqPicG2bFl17iRkf0E04+NSlEERXGyMx0+hdj9k02NsG+/yO4sWW6Ucr+qklZIPuNiL6WmvQqwYVovTXuqO4rHHffozqA2bTojovbb1ak4zBXq5FL5wQvdK9+SEW9KBJczc19nojHnPLlmzZYV+uANFs6q0YCKrWk4w56BbtpqN82WJbiQO961B9EL2fj5hdL7GwadHbBrvscMtG4vB6A9ta9F03a3QXcLc10Rny25ajD3bMPlhBSkRnMFdryYGXG9Ryg9n1mE9vSLnFbgOhciifRCc8jKUKSdpqNNl1g77ZUHznOe75iuQ09q4mffpC4upqI/G4bIjZqx94VOdF06s19nqH6rxoPNvBqDYk8ptBH1caklXQiw5Qb1tQoj/Di4a4PcuJkxy6DqxFtz2q6UX7V4sW1mwauF1SXPeETPSWduuhFy2uuduCUdi7WvJ806DqVrTcqx3JGezLFWbVYJetzHF9xCxrdBOYPK0lZmMS/avVqPXuEM/7fNTLLXbnsRvRH9tNB0aj+4C73kpf3NUy9zSe/MlK+quLpOMpyWjSvEK3foiNiO4iuk8UtzKfArhP7+Tc6wa969CtxNIPzC8pobw/zGEA9nKN8hF3V2Nvd2RP74bzB7m2kFA+yPxmNHYpuUNKVJcBuw1kH77AbBrJq8ajt+1hHNxtLd6MdQTvJe+G+c3e7mRsrzfoXYvZdVS/95TqWY1Z1aA1pu7Be5JRshZAg7uSObRfFDK2TqH211z35I9v0V2gerRm9sGS2fdrimtPvvSiuUxQ/uvv456vJH9zi9l2EBKhsJIvmRnqh2uS05A5uS80PUxKWUNQZRBFG49RshYhs7LeoBvm8k2P2ragpZ1hkuFeDHl4U0Pbir9s3YOS+2Zx05MvI35REeYlTCtilR3y12x7iAFTe9yyQbced1MPuTrMq0lyIGUOu+zQNxuqZx3zD3fYq+2rOBzud6qP0n6jZF3DeivjmKQPVB9Q87no0LWS7Vct0amD//GXTkryjeZP8vrp+HvAt1JKfwn468B/pZTKlFInwD8A/l3grwH/4P9v7/5C7DjLOI5/fwolxaxabRv/IiWtFkEMEitakI1KE7ypRUVBSi2KCo16oyiCGlqFXlSKoBSilvTGSkGLhUrNEkSFXpgKJakRpK2hxobEEC8a2lqaPF6cOXiS7G52z3F2MrPfDxzOvLNz5jy7++7h2XfmfZ8kl00R+baq2tI8br3QwX25dH4d8GRVPQ2Q5BfAjcChTqOSJEmaUkuzzguYSxJgI3ASeBnYDixU1UmAJAvADuD+c0+Q5AlGg5EvAVdX1cZpg0n1YMZTkk8AO6rq8037ZuB9VbVz4pgvAF8HXgu8BjjQRawajMuBE10HoV6zD2lW9qH2vK2qrugygCSPMPodr8QG4MWJ9u6q2r3EeeeAh4BrgTngU1X1cJKvARuq6nvNcd8GXqiquxY5x1+B7VX1TJJT40QzyWHgOeA08Dzw/ap6eLnA+zKiudhaBGdlyM0PfDdAkseqautaBKZhsg9pVvYhzco+NGxVtaOlU28HHgc+BGwGFpL8kRXkUhPGI6GL2VZVJ5JsBh5Ncm1V/XupYHpxjyZwBHjrRPstwLMdxSJJknRRSHLbxOScNwG3Ar+qkSeBvzMa3VxRLpVkA3BpVZ1a7n2r6ingMHD1csf1JdHcD1yT5KoklwCfZjQsLEmStG5V1Y8nJuc8CzwDfBggySbgHcDTwG+BG5Jc1kwCuqHZd66bgEcu9L5JrgSuat5vSb24dF5VLyfZyegH8krg3qr6yzIvWfS+BWkV7EOalX1Is7IPaRp3AHuSHGR0ufwbVXUCIMkdjAbvAG4fTwwaS7IV+BlwMsnjze5Lk9xeVd9p2r9Lchq4BPhWVR1bLpheTAaSJElSu5LMA/NVtWti30bgR1X12WnO2YsRTUmSJLXuEOevdPAicM+0J3REU5IkSa3oy2SgFbNUpWZ1Tomtx7qORxe/JPcmOd4scjze97okC02pt4UpK3BonViiD+1K8s+JGcUf7TJGaRqDSjRXW6pSWsa4xJZr2Gkl9jCqsDHpm8C+qroG2Ne0paXs4fw+BHD3xIzi36xxTNLMBpVoMlGqsqpeAsalKiWpNVX1B85f3PhG4L5m+z7gY2salHpliT4k9d7QEs03A/+YaB9p9kmrUcDeJH9uSptK09hUVUcBmucrO45H/bQzyYHm0rq3X6h3hpZorqa8krSU66vqPYxuwbgtyQe7DkjSunQPoxKCW4CjwA+6DUdavaElmpaq1MyaygpU1XHgQUa3ZEirdSzJGwGa5+Mdx6OeqapjVXW6qs4AP8HPIvXQ0BJNS1VqJklelWRuvM2oRNcTy79KWtRDwC3N9i3ArzuMRT00/kelcRN+FqmHBrVg+xSlKqVzbQIeTAKjv4+fV9UFa75qfUtyPzAPXJ7kCPBd4E7ggSSfY1QL+JPdRaiL3RJ9aD7JFka3gB0GvthZgNKUXLBdkiRJrRjapXNJkiRdJEw0JUmS1AoTTUmSJLXCRFOSJEmtMNGUJElSKwa1vJGk9SHJ64F9TfMNwGngX037+ar6QCeBSZLO4vJGknotyS7gVFXd1XUskqSzeelc0qAkOdU8zyf5fZIHkvwtyZ1JPpPkT0kOJtncHHdFkl8m2d88ru/2O5Ck4TDRlDRk7wa+CrwLuBl4e1VdB/wU+HJzzA+Bu6vqvcDHm69Jkv4PvEdT0pDtr6qjAEmeAvY2+w8C25rtjwDvbMqOArw6yVxVPbemkUrSAJloShqy/0xsn5lon+F/n3+vAN5fVS+sZWCStB546VzSercX2DluJNnSYSySNCgmmpLWu68AW5McSHII+FLXAUnSULi8kSRJklrhiKYkSZJaYaIpSZKkVphoSpIkqRUmmpIkSWqFiaYkSZJaYaIpSZKkVphoSpIkqRX/BUmb2wy2h1I0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wav_data, sr = load_wav(os.path.join(train_folder, 'aaf6167e.wav'), sr=sample_rate)\n",
    "\n",
    "mel_spec = melspectrogram(wav_data, n_fft=fft_size, hop_length=hop, n_mels=mels, \n",
    "                          fmax=sample_rate//2)\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "colors = 'viridis'\n",
    "specshow(mel_spec, y_axis='mel',\n",
    "         fmax=sample_rate//2, x_axis='time', cmap=colors)\n",
    "plt.colorbar(format='%+2.0f дБ')\n",
    "plt.title('Мелспектрограмма скрипки')\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(10, 4))\n",
    "specshow(power_to_db(mel_spec, ref=np.max), y_axis='mel',\n",
    "         fmax=sample_rate//2, x_axis='time', cmap=colors)\n",
    "plt.colorbar(format='%+2.0f дБ')\n",
    "plt.title('Логарифм мелспектрограммы скрипки')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Как получились эти различия в спектрограммах не так важно. Самое главное - мы видим, что для скрипки в логарифме мелспектрограммы прослеживается некоторая закономерность гораздо лучше, чем если мы будем использовать просто мелспектрограмму. \n",
    "Именно поэтому именно логарифм мелспектрограммы мы будем использовать в качестве признаков для всех имеющихся аудио."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Извлечение признаков\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Для ускорения вычислений нам будет полезно вычислить все мелспектрограммы заранее и сохранить их. Выполнение данной ячейки займет некоторое время, так как процесс вычисления признаков дорогой. Размер полученных файлов будет приблизительно 1 гб каждый (при исходных значениях параметров)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total files: 9473\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "9473it [56:54,  2.77it/s]\n",
      "9400it [47:36,  2.75it/s]\n"
     ]
    }
   ],
   "source": [
    "path_to_csv = os.path.join('./data', 'train.csv')\n",
    "pickle_train_path = os.path.join('./data', 'train.pickle')\n",
    "pickle_test_path = os.path.join('./data', 'test.pickle')\n",
    "\n",
    "label_to_id = extract_log_mel_feats('train', path_to_csv, train_folder, pickle_train_path, sample_rate, fft_size, hop, mels)\n",
    "extract_log_mel_feats('test', '', test_folder, pickle_test_path, sample_rate, fft_size, hop, mels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Создаем модель"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Для начала импортируем необходимые классы из keras и из нашей небольшой утилиты. \n",
    "\n",
    "Сама модель сети (класс CNNModel) имеет довольно простую архитектуру - всего три слоя свертки, чередующиеся с пулингом. Конечно, используем такие модные и успешные штуки как [batch normalization](https://arxiv.org/pdf/1502.03167.pdf) и [dropout](https://arxiv.org/pdf/1207.0580.pdf)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "\n",
    "from model.batcher import Batcher\n",
    "from model.config import Config\n",
    "from model.cnn_model import CNNModel\n",
    "from model.utils import prepare_shape\n",
    "\n",
    "from keras_tqdm import TQDMNotebookCallback\n",
    "from keras.callbacks import ModelCheckpoint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Создаем класс конфига для удобной передачи параметров в класс модели: задаем входные размеры фичей, шаг обучения и число выходных классов.\n",
    "\n",
    "Далее подгружаем ранее извлеченные признаки."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = Config(shape=(64, 63, 1), learning_rate=0.001, n_classes=41)\n",
    "\n",
    "pickle_train_data = pickle.load(open(pickle_train_path, 'rb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Выделяем валидационную выборку для контроля обучения. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "validation_set_size = 1000\n",
    "\n",
    "x_val = []\n",
    "y_val = []\n",
    "\n",
    "x_tr = []\n",
    "y_tr = []\n",
    "\n",
    "for i, row in enumerate(pickle_train_data):\n",
    "    feature = prepare_shape(row['feature'], config)\n",
    "    if i < validation_set_size:\n",
    "        x_val.append(feature)\n",
    "        y_val.append(row['label_id'])\n",
    "    else:\n",
    "        x_tr.append(feature)\n",
    "        y_tr.append(row['label_id'])\n",
    "x_val = np.asarray(x_val)\n",
    "x_tr = np.asarray(x_tr)\n",
    "\n",
    "y_val = np.asarray(y_val)\n",
    "y_tr = np.asarray(y_tr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## И, наконец, начинем обучение!\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_batcher = Batcher(x_tr, y_tr, batch_size=100)\n",
    "val_batcher = Batcher(x_val, y_val)\n",
    "\n",
    "checkpoint = ModelCheckpoint('./best_weights.h5', monitor='val_loss',\n",
    "                             verbose=0, save_best_only=True)\n",
    "\n",
    "model = CNNModel(config).get_model()\n",
    "\n",
    "history = model.fit_generator(train_batcher,\n",
    "                              callbacks=[checkpoint, TQDMNotebookCallback()],\n",
    "                              validation_data=val_batcher,\n",
    "                              epochs=30,\n",
    "                              use_multiprocessing=False,\n",
    "                              workers=0,\n",
    "                              verbose=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Визуализируем процесс обучения.\n",
    "\n",
    "Воспользуемся удобными интерфейсами keras для визулизации процесса обучения. Метод *fit_generator* возвращает объект *history*, который хранит полученные в конце каждой эпохи значения ошибки и точности."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('Model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xd4lFX2wPHvSSO9kARSSAhNehWQLthRVFwbrg1X17WtZZvlt67b3NUtttXVxVXXuhbsvS2iiHQBwaB0EkINkN5zf3/cmRBCysxkJjOTnM/z5Eky877v3GHInLnlnCvGGJRSSimAEH83QCmlVODQoKCUUqqBBgWllFINNCgopZRqoEFBKaVUAw0KSimlGmhQUMoFIpIjIkZEwlw4dq6ILGrvdZTyBw0KqtMRkW0iUi0iKU1uX+14Q87xT8uUCnwaFFRntRW4yPmLiAwHovzXHKWCgwYF1Vk9C1zW6PfLgWcaHyAiCSLyjIjsE5HtIvJrEQlx3BcqIn8Tkf0isgU4o5lznxCRXSKyU0T+KCKh7jZSRDJE5C0ROSAim0Tkx43uGy8iK0SkWET2iMh9jtsjReQ5ESkUkUMislxEerr72Eo1R4OC6qyWAPEiMtjxZn0h8FyTY/4BJAB9geOxQeQKx30/BmYBo4GxwHlNzn0aqAX6O445BbjKg3b+F8gHMhyP8ScROdFx34PAg8aYeKAf8LLj9ssd7c4CkoFrgAoPHlupo2hQUJ2Zs7dwMrAB2Om8o1GguN0YU2KM2Qb8HbjUccgFwAPGmDxjzAHgz43O7QnMBG42xpQZY/YC9wNz3GmciGQBU4BbjTGVxpjVwL8btaEG6C8iKcaYUmPMkka3JwP9jTF1xpiVxphidx5bqZZoUFCd2bPAD4G5NBk6AlKACGB7o9u2A5mOnzOAvCb3OfUGwoFdjuGbQ8C/gB5uti8DOGCMKWmhDVcCxwAbHENEsxo9rw+BF0WkQET+IiLhbj62Us3SoKA6LWPMduyE8+nAa03u3o/9xN270W3ZHO5N7MIOzzS+zykPqAJSjDGJjq94Y8xQN5tYAHQXkbjm2mCM2WiMuQgbbO4F5otIjDGmxhjzO2PMEGASdpjrMpTyAg0KqrO7EjjBGFPW+EZjTB12jP5uEYkTkd7Azzg87/AycKOI9BKRJOC2RufuAj4C/i4i8SISIiL9ROR4dxpmjMkDFgN/dkwej3C093kAEblERFKNMfXAIcdpdSIyQ0SGO4bAirHBrc6dx1aqJRoUVKdmjNlsjFnRwt0/BcqALcAi4AXgScd9j2OHaNYAqzi6p3EZdvjpW+AgMB9I96CJFwE52F7D68BdxpiPHfedBqwXkVLspPMcY0wlkOZ4vGIgF1jI0ZPoSnlEdJMdpZRSTtpTUEop1UCDglJKqQYaFJRSSjXQoKCUUqpB0JXvTUlJMTk5Of5uhlJKBZWVK1fuN8aktnVc0AWFnJwcVqxoaYWhUkqp5ojI9raP0uEjpZRSjWhQUEop1UCDglJKqQZBN6fQnJqaGvLz86msrPR3UzqNyMhIevXqRXi4Ft9UqivpFEEhPz+fuLg4cnJyEBF/NyfoGWMoLCwkPz+fPn36+Ls5SqkO1CmGjyorK0lOTtaA4CUiQnJysva8lOqCOkVQADQgeJn+eyrVNXWaoNCWypo6dhVVUFevVWGVUqolXSYoVNfWs6+kisoa7+9FcujQIf75z3+6fd7pp5/OoUOH2j5QKaU6SJcJCpHhoQBUdGBQqKtr/bHee+89EhMTvd4epZTyVKdYfeSK8FAhNER80lO47bbb2Lx5M6NGjSI8PJzY2FjS09NZvXo13377LbNnzyYvL4/Kykpuuukmrr76auBwyY7S0lJmzpzJlClTWLx4MZmZmbz55ptERUV5va1KKdWaThcUfvf2er4tKG72vsqaOgwQ5eg1uGpIRjx3ndnynuz33HMP69atY/Xq1Xz22WecccYZrFu3rmE555NPPkn37t2pqKhg3LhxnHvuuSQnJx9xjY0bN/Lf//6Xxx9/nAsuuIBXX32VSy65xK12KqVUe3W6oNCaEBFq6ut9/jjjx48/Yn3/Qw89xOuvvw5AXl4eGzduPCoo9OnTh1GjRgFw7LHHsm3bNp+3Uymlmup0QaG1T/QHy6rJO1jOMT3jGuYYfCEmJqbh588++4xPPvmEr776iujoaKZPn97s+v9u3bo1/BwaGkpFRYXP2qeUUi3pMhPNcHiy2dvzCnFxcZSUlDR7X1FREUlJSURHR7NhwwaWLFni1cdWSilv6nQ9hdZ0Cw9BRKioqcOba36Sk5OZPHkyw4YNIyoqip49ezbcd9ppp/HYY48xYsQIBg4cyIQJE7z4yEop5V1iTHAlc40dO9Y03WQnNzeXwYMHu3T+xj0lhIYIfVNjfdG8TsWdf1elVGATkZXGmLFtHdelho/ADiFV1tQTbMFQKaU6QpcLClHhodTW11Or5S6UUuooXS4oREb4LrNZKaWCXdcLCuH2Kfsis1kppYJdlwsKYSEhRISGUFmtQUEppZrqckEB7GRzRY3vM5uVUirYdMmgEBURSlVtnd/2VoiNtcthCwoKOO+885o9Zvr06TRdetvUAw88QHl5ecPvWopbKdVeXTIo+Cqz2V0ZGRnMnz/f4/ObBgUtxa2Uai+fBQURiRSRZSKyRkTWi8jvmjlmrojsE5HVjq+rfNWexqK8PNl86623HrGfwm9/+1t+97vfceKJJzJmzBiGDx/Om2++edR527ZtY9iwYQBUVFQwZ84cRowYwYUXXnhE7aNrr72WsWPHMnToUO666y7AFtkrKChgxowZzJgxA7CluPfv3w/Afffdx7Bhwxg2bBgPPPBAw+MNHjyYH//4xwwdOpRTTjlFaywppY7gyzIXVcAJxphSEQkHFonI+8aYpsV/XjLG3OC1R33/Ntj9TauHhGPoV11HWIhAmAuF8dKGw8x7Wrx7zpw53HzzzVx33XUAvPzyy3zwwQfccsstxMfHs3//fiZMmMBZZ53V4t7Hjz76KNHR0axdu5a1a9cyZsyYhvvuvvtuunfvTl1dHSeeeCJr167lxhtv5L777mPBggWkpKQcca2VK1fy1FNPsXTpUowxHHfccRx//PEkJSVpiW6lVKt81lMwVqnj13DHV0BkjAlCiAjemlIYPXo0e/fupaCggDVr1pCUlER6ejp33HEHI0aM4KSTTmLnzp3s2bOnxWt8/vnnDW/OI0aMYMSIEQ33vfzyy4wZM4bRo0ezfv16vv3221bbs2jRIs455xxiYmKIjY3lBz/4AV988QWgJbqVUq3zaUE8EQkFVgL9gUeMMUubOexcEZkGfA/cYozJa+Y6VwNXA2RnZ7f+oK18om/s4KEKDpRVMzQjvsVP7+4477zzmD9/Prt372bOnDk8//zz7Nu3j5UrVxIeHk5OTk6zJbMba64dW7du5W9/+xvLly8nKSmJuXPntnmd1kp4aIlupVRrfDrRbIypM8aMAnoB40VkWJND3gZyjDEjgE+Ap1u4zjxjzFhjzNjU1FSvtC0yPJR6Y6iq9c7S1Dlz5vDiiy8yf/58zjvvPIqKiujRowfh4eEsWLCA7du3t3r+tGnTeP755wFYt24da9euBaC4uJiYmBgSEhLYs2cP77//fsM5LZXsnjZtGm+88Qbl5eWUlZXx+uuvM3XqVK88T6VU59YhpbONMYdE5DPgNGBdo9sLGx32OHBvR7QHjpxs9saGO0OHDqWkpITMzEzS09O5+OKLOfPMMxk7diyjRo1i0KBBrZ5/7bXXcsUVVzBixAhGjRrF+PHjARg5ciSjR49m6NCh9O3bl8mTJzecc/XVVzNz5kzS09NZsGBBw+1jxoxh7ty5Dde46qqrGD16tA4VKaXa5LPS2SKSCtQ4AkIU8BFwrzHmnUbHpBtjdjl+Pge41RjT6oYD7S2d7VRvDOt3FpMaF0FaQpRb53YVWjpbqc7D1dLZvuwppANPO+YVQoCXjTHviMjvgRXGmLeAG0XkLKAWOADM9WF7jhAiQrfwEM1sVkqpRnwWFIwxa4HRzdz+m0Y/3w7c7qs2tCUqPJTSqlp/PbxSSgWcTpPR7MkwWGR4KDV19dTUaW+hKd2ESKmuqVMEhcjISAoLC91+I/N2ZnNnYYyhsLCQyMhIfzdFKdXBOmT1ka/16tWL/Px89u3b59Z59fWGPUWVVO4LIy4y3EetC06RkZH06tXL381QSnWwThEUwsPD6dOnj0fn/vjPnzKuT3cenDOi7YOVUqqT6xTDR+0xJCOebwuK/d0MpZQKCBoU0uPZvK9U5xWUUgoNCgxOj6fewPd7ji4XoZRSXU2XDwpDMuIBdAhJKaXQoEBWUjSx3cL4dpcGBaWU6vJBISREGJwepz0FpZRCgwJgJ5tzdxVT761dd5RSKkhpUMBONpdV15F3sNzfTVFKKb/SoIBONgelD26HtS/7uxVKdToaFIBjesYRGiI62RwsjIEVT8H6N/zdEqU6HQ0K2Gqp/VJjtKcQLMoPQG0FlBT4uyVKdToaFByGpMdrTyFYFOc7vmtQUMrbNCg4DE6PZ1dRJQfLqv3dFNWWIkdQKN0Ltfp6KeVNGhQcnJPNudpbCHxFOx0/GCjd7demKNXZaFBwGJzuWIGkQSHwFeUd/lmHkJTyKg0KDimx3egZ300nm4NB8U6QkMM/K6W8RoNCIzrZHCSK8qHnMPtz8S7/tkWpTkaDQiNDMuLZtLeUqlrdWyGgFeVDz6EQHq3DR0p5mQaFRganx1Nbb9i4p9TfTVEtqauFkl2Q0AviM3T4SCkv06DQyBCdbA58JbvA1DcKCtpTUMqbNCg00js5huiIUJ1sDmTOnkF8L4jP1KCglJdpUGgkNEQYlBanPYVA5kxcc/YUSnZBvc4BKeUtGhSaGJJh91YwRvdWCEgNQSHTBgVTB2X7/NsmpToRDQpNDE6Pp6SylvyDFf5uimpOUT5EJkC3OIjLsLfpZLNSXqNBoQmdbA5wRfmQkGV/jncGBZ1XUMpbNCg0MSgtnhDRDXcCVnG+nWCGw981KCjlNRoUmogKD2FW9wLWbdjg76ao5hTl20lmgOhkCI3Q4SOlvMhnQUFEIkVkmYisEZH1IvK7Zo7pJiIvicgmEVkqIjm+ak+bqsth5dPw6GQeKvsFF+29jxXbDvitOaoZ1WVQcdBOMgOEhEBcuvYUlPIiX/YUqoATjDEjgVHAaSIyockxVwIHjTH9gfuBe33YnuYdyoOP74L7h8DbN4IIdb3GMyF0A/M+29jhzVGtcJbMds4pgCNXQesfKeUtPgsKxnLWiwh3fDVd53k28LTj5/nAiSIivmpTo8bB9sXw8mXw4EhY/BDkTIG578I1iwidcA2xVLD7u6Vs3FPi8+YoFzlLZjvnEgDi03X4SCkvCvPlxUUkFFgJ9AceMcYsbXJIJpAHYIypFZEiIBnY75MG1VTCuvmw9DHY/Q1EJsKkG2DcVZCYffi43lMAmBK+gXmfb+Gv54/0SXOUm5xv/s45BbArkHLfsYG+Az5PKNXZ+XSi2RhTZ4wZBfQCxovIsCaHNPdXfFTWmIhcLSIrRGTFvn0eJip994EdInrzepsBe+aD8LNcOPn3RwYEgLiekDyAsxO38sbqnewq0pyFgFCUD8jhpahgew11VVCu8z9Ba8N78MAIqNG/s0DQIauPjDGHgM+A05rclQ9kAYhIGJAAHPXXbYyZZ4wZa4wZm5qa6lkjkvtB1gS4/G24djEcOxciols+PmcyAyq/AVPPU19u8+wxlXcV7YS4NAgNP3xbvCawBb28pXBoOxzc7u+WKHy7+ihVRBIdP0cBJwFN13m+BVzu+Pk84H/GV/UlUgbARS9An2muDTPkTCWkuoSrB5TywtIdFFXU+KRZyg1FeUcOHYHmKnQGztIlh3b4tx0K8G1PIR1YICJrgeXAx8aYd0Tk9yJyluOYJ4BkEdkE/Ay4zYftcU/vyQBcmp5PaVUtzy/VTzF+V7zzyElmONxTKNGgELScvbxD+jcWCHw20WyMWQuMbub23zT6uRI431dtaJf4dOjej7QDK5g6YCpPLtrGjyb3ITI81N8t65qMsZ8oj2kyAhnbEyRUewrBTHsKAUUzmluTMxl2LObaqTnsL63i9a/bMW49/0qbD6E8U14ItZVHDx+FhNrAoEEhONXXHX7tNCgEBA0KrcmZCpVFTIzdxfDMBOZ9voW6eg+mPIp22qWwXz4Amxd4v51dQeN9FJrSbTmDV8luW/4cNCgECA0KrXHMK8j2L7nm+H5s3V/Gx9/udv86G9613+PS4a2fQpUmxLmtzaCgPYWg5Hxd43tpUAgQGhRak5AJSX1g25ecNiyN3snRPLpwi/sb8Gx4G1IGwgXP2E+0H93pm/Z2Zo234WxKt+UMXsWOoNB7IpTvt/WtlF9pUGhLzmTY/iWhGH48tS9r8g6xdKsbiVJlhbDtSxh8JmSNh4nXw8qndBjJXUV5ENoNYlKOvi8+A6pLoVLLnQcdZ0+h9yT7/VCe/9qiAA0KbcuZCpWHYO96zju2F8kxETy2cLPr53//vh0zHXym/X3G/0HyADuMpG9irivaaXtuzeWY6GY7watoJ3SLh56OYgc6hOR3GhTa4phXYNsiIsNDmTsph8++20euqzuz5b4NCdmQ7qifFB4Fs/9ph0M+1mEklzXeR6EpzWoOXs7X1VlqRnMV/E6DQlsSsyCxN2xbBMClE3sTHRHKvM+3tH1uVYkdJho868hPuA3DSP+Bzf/zTbs7m6L85ucTQHsKwcy5k15MDzs8WKTDR/6mQcEVOVNg+5dQX09idARzxmXz1poC8g+Wt37exo9tsTbn0FFjzmGkN3UYqU11NVC6u+WeQly6/a5BIfg4ewohIfa7Dh/5nQYFV+RMsTt+7csF4MqpfRDgiUVbWz8v922ISYWs446+LzwKZj9qyzN89Gvvt7kzKdkFpr7loBDWzf476/BRcKkut0mJztc1MVuDQgDQoOCKRvMKAJmJUZw1MoMXl+VxsKy6+XNqKmHjRzDwdJt125yscTDxBlj1NGz61AcN7yQadlzLbPmY+AwbPFTwcPbsNCgEFA0KrkjqbSeLHUEB4Orj+1JRU8ezS1qYGNu60C6THHxW8/c7zfg/SDkG3roRKou82OhOpCFxLavlYzRXIfg45w8aB4WyfbYHofxGg4KrHPkKOBLXBqXFM2NgKv9ZvI3y6tqjj899yy616zOt9euGRzYaRtLVSM1qbhvOpuJ0W86g05CQ6HhdE3vb7zrZ7FcaFFyVM8WOf+47vCXEDScM4EBZNX//6Psjj62rtbtJHXMqhEW0fe1eY2HST3UYqSXFO+3Wqd1iWz4mPsPO++inzODRdCe9hmWpOoTkTxoUXNVkXgHg2N5JXDIhmye/3MrK7QcPH7vjK6g40Pyqo5ZMvyMwhpHe+yWseNJ/j9+c1nIUnJyfNnVeIXgU5UNsD7tQADRXIUBoUHBVUo5dJ98oKADcNnMwGQlR3PrqWqpqHdUec9+GsEjof5Lr1z9iGMlPq5Gqy2H5E/DF/Q3DZAGhaKcLQUET2IJO02Af2xNCI7Sn4GcaFFwlctS8AkBstzD+9IPhbNpbyj8+3WTv2/AO9DsRImLce4yGYaRnYMdSLz8BFxR8bUtyFO2Anas6/vFb0tw2nE01bMupPYWg0XQnvZAQu5hAg4JfaVBwR84Uuzpi/5FzCMcfk8q5Y3rx6MLNbF7zhf3P7s7Q0REXu9WuuV/wRy802E35y+z3kDBY/1rHP35zqkpt7anWJpnB7pQH2lMIFs6d9JquKNNlqX6nQcEdDfMKXxx1152zBpMUHcHKD57GhITZSWZPRMTA1J/D1s9hy8J2NNYD+Suge1/odwJ8+2ZgDCE53+RbW44K9t8tMlGXpQaLioNQU350DzAxWyul+pkGBXd07wtxGbYUdhOJ0RH88eyhjC1fRH78sRDd3fPHOfYK+8n4f3/suDdmYyBvGfQaD0PPsUM2O1d2zGO3pmEtexs9BdDNdoJJQ+5Jk9c1MQvK9kJNRce3SQEaFNzjnFfYtqjZN+vTeh6ib8hu/l04lE1727G7WngkTPulHc7Z+FE7GuyGQzvsH2PWOEcWdjisf71jHrs1DdnMbcwpgG7LGUxa2knPmaugvQW/0aDgrpwp9s2zcNPR9+W+g0H4MmwCv5q/1rP9nJ1GX2JXPP3vj1Bf7/l1XJW/3H7vNQ6iEgNnCMm5lt1Z9K412lMIHi3tpKe5Cn6nQcFdvafY783MK5D7FtJrHNefNZlVOw7xn8XbPH+c0HCYfjvsXmuzo30tbxmER0OPofb3obMDYwipeKcNCKHhbR8bn2kXAtS2UI9KBY6iPLv8NCb1yNs1V8HvNCi4K7kfxKYdPa9wcJt9Ax98JrNHZXLCoB787cPv2FHYjgzb4efbvZ0X/Anq69rV7DblL4eMMRAaZn8PlCEkV5ajOsVnAMaW2VaBrWinfb1CmrwFxabZ/3faU/AbDQruamleYcO79vvgWYgId58zjLAQ4bbX1mI8HYIJCYUZd8D+7+CbV9rf9pbUVNiAljXu8G2BMoTk3IbTFbrZTvBobjkq2CCRqLkK/qRBwRM5U+yn0QONdl/LfdvuM9u9LwDpCVHcfvpgFm8u5MXl7Zg0G3wWpI2Az/5sN5vxhV1roL7WrjxqzN9DSA1r2V3sKcRpVnPQaJq41pjmKviVBgVPNJ1XKN0LO5YclbB20fgsJvZN5u53c9lV5OESu5AQOOFOOzz19XOet7k1eY6ktV7jjrzd30NIZfvtznUtbcPZlPYUgkNdrX2NWgr2GhT8yqWgICI3iUi8WE+IyCoROcXXjQtYKQPsnrLOeYUN7wLmqKAgItxz7nDq6g3/9/o6z4eRBpxsP8V//le7eY+35S+3K51im0z6+XsIqbiFZYstiUyA8BgNCoGudLctp9LSsGBituYq+JGrPYUfGWOKgVOAVOAK4B6ftSrQNZ1XyH3bDhv1GHLUob2TY/jFqQP534a9zF+Z7/njnXin7XKvfKqdjW/CGBsUmvYSnPyZyNZSglNLRHRZajAoaiNLXXMV/MrVoCCO76cDTxlj1jS6rWvKmWIrmu5abXdZGzTLvik1Y+6kHI7r053bXvuG17/2MDD0mWa/vvg7VJe1o+FNFOXbctNN5xOcBs703xBSW28ezdGgEPia7rjWlPP11iEkv3A1KKwUkY+wQeFDEYkDOiCjKoA55xU+utNO0ray7WZoiPDE3HGMz+nOLS+t4dmvtnn2mCfcadfhL/2XZ+c3x5m0ltVCT8E5hLT+jY4fQirKsyXIo5NdP0e35Qx8TXdca8qZq1CkQcEfXA0KVwK3AeOMMeVAOHYIqetKHQjRKXayOS4dMo9t9fDYbmE8dcU4ThrcgzvfXM8jC5rJiG5L1ngYcCp8+SBUHPKw4U3kL4ewKLtyqiVDz7Hj+/krvPOYrirKt28cLfTAmhWfbns+vs7rUJ4ryoduCRAZ3/z9cZqr4E+uBoWJwHfGmEMicgnwa6DV7cFEJEtEFohIroisF5GbmjlmuogUichqx9dv3H8KfuKcVwAYdMbRSTjNiAwP5dFLjuXsURn89cPvuOf9De5PPp/wa1tK+qtHPGh0M/KXQ8bo1jOGnUNI377hncd0VbELm+s0FZ9hJzFL9/qmTar92so9CQm1r7sGBb9wNSg8CpSLyEjgV8B24Jk2zqkFfm6MGQxMAK4XkaNnYuELY8wox9fvXW14QMiZar+7sXdCeGgI918wiouPy+axhZv59RvrqHenRlL6CBgyG5b8E8oK3WxwE7VVNkeh19jWj4tKhP4n2iGkjqjD5OROjoJTw2Y7OoQUsFzJUtdlqX7jalCoNfYj7dnAg8aYB4G41k4wxuwyxqxy/FwC5AIuLiMJEqMvgfOehD7Hu3VaSIjwx9nDuOb4fjy/dAe3vLyamjo33mxn3GFr0X95v5sNbmLXGqirtsNSbRky2w4hddQqpLoaKNntWU8B7CIAFZicw4Kt0aDgN64GhRIRuR24FHhXREKx8wouEZEcYDTQ3B6TE0VkjYi8LyJDXb1mQAiPgmHnujfm7SAi3DZzEL86bSBvri7g2udWUlnj4jh46kAYcSEse9y+cXqqcWXUtgycaQuYddQQUnEBYNp+82hKewqBrbocKg640FPoDaV7NFfBD1wNChcCVdh8hd3YT/x/deVEEYkFXgVuduQ6NLYK6G2MGQn8A2j2HUdErhaRFSKyYt++fS42OThcN70/f5g9jE837OWKp5ZTWlXr2onH32pXPS1qR28hbxkkZNuJvbY0XoXUEUNIxW7so9BYdLINXlrqIjC5+ro2rEDycAm38phLQcERCJ4HEkRkFlBpjGlrTgERCccGhOeNMUdt+muMKTbGlDp+fg8IF5GUZo6bZ4wZa4wZm5qa2vTuoHfphN7cd8FIlm07wMX/XsqhchdKP3fvY3spXz9v9zH2RP6KlpeiNqcjh5AaEtfcyFEATWALdC1trtOUltD2G1fLXFwALAPOBy4AlorIeW2cI8ATQK4x5r4WjklzHIeIjHe0p52zp8HpnNG9eOySY8ndVcyF/1rC/tKqtk8adxVUl3hWQbW4wL7BuzJ05OQcQuqIRDZ3tuFsKk6DQsByOyjovEJHc3X46P+wOQqXG2MuA8YDd7ZxzmTsHMQJjZacni4i14jINY5jzgPWicga4CFgjvG4QFDwO3lIT56aO47tB8q49IllbfcYeo2DnsNhxRPuJ5Y1FMFzYZLZqXEtJF8PIRXthKgkiIhx/1zdljNwFe/E7qSX0fpxmqvgN64GhRBjTOOF34VtnWuMWWSMEWPMiEZLTt8zxjxmjHnMcczDxpihxpiRxpgJxpjFHj6PTmNy/xTmXTqWzXtLufzJZZRUtlIuWwTG/Qh2f+N+Yln+cgjtBmnD3TvPmci208eJbEX5rldHbSo+A4p3+X8rUXW0ojyI7QlhEa0fFxJqe4kaFDqcq0HhAxH5UETmishc4F3gPd81q2ubdkwq/7x4DOsLivnRf5ZTXt3K5PPwCyAiDpb/270HyV8OGaPa/uNsqmEIycerkDxJXHOKz7Qlt8sPeLdNqv3c2TRTDRuFAAAgAElEQVQpMVuL4vmBqxPNvwTmASOAkcA8Y8ytvmxYV3fSkJ48OGc0K7cf5KqnV7S8XLVbLIy80I7zu/omWFsNBavdm09wikyAfif6fgipKM+z+QRotK+CDiEFHHcSEjVXwS9c3mTHGPOqMeZnxphbjDF+3ri3azhjRDp/O38kX20p5NrnVlJd28Kb8Ngr7SdjVzfh2f2NPd6VpLXmDJ3t2yGkqhKoLGpfTwF0sjnQOHfSc3VYMLG33XvBF3uIqBa1GhREpEREipv5KhGRpjkHygd+MKYXd88ezoLv9vHT/66itrnM555DIHsirHjStU/v+S3stOYqXw8heVIyu7H4dPtdewqBpeIg1Fa411OA4MtVMAZeugS+me/vlnikrcniOGNMfDNfccaYFkocKm/74XHZ3HXmED5cv4efv7KGuuZqJY27Cg5uhS0L2r5g/nL7aS2+jRUgLWkYQvJRIpvzTcDdbGan2J4godpTCDTuLjMO1lyFQzvsxltLHvV3SzyiezQHiSsm9+HW0wbx5uoCbn9t7dFF9AafaUt5L3+i7YvlLW+7CF5bhp9nP4nnvtW+6zTH3W04mwoJtUsafRUUDu2AxQ/bvYaV64rczFIP1lwF53LvnSuC8oOJBoUgcu30ftx44gBeXpHPb99ef2TZ7bBuMOZS+P791rvbJbvt5iWezic4DT3Hbj/6yW/txLU3FeWDhNh9KjwVn+Gbonh1tfDKXPjo/2BdcA4P+I27Wepx6RASFnxBIX+Z7amCY//24KJBIcjcctIArp7Wl2e+2s6fm+7HcOwVdjxz5dMtX6ChCF47g0JIKJz8Bztktfzx9l2rqaKd9g0hNMzza/iq1MXiB22Zj6ju8PlfdTMfdxTn27mo6KMq2TQvWPdVyFsKvSdB8gA7jBRkNCgEGRHh9pmDuGxib+Z9voW73809PMeQ1BsGnAyrnrGlp5uTt8z+YaaPaH9jBpxkM5wX/sW7OQFFeZ7PJzjFZ9rg4s0Etj3rYcGfYcjZcOaDULgJ1r3qvet3ds6S2S5sSNUg2JalVpXC7nWQPQEGz4Jti4IuX0aDQhASEX575lAun9ibfy/aytynlnGwzDGEM/ZKu4yvpW5r/gpIH2mHm7zh5D/Y5aOf/8071wPPNtdpKi4dasqgykuL5Gqr4fVrbKmPM+6DQbOgx1AbELW34JoiDxISE4IsKBSssjv/9RoPg860P3//ob9b5RYNCkEqJET43dnD+Mu5I1i65QBnPryIdTuLbE8hIbv5DOe6Gij42vOlqM1JG2Y3G1o2Dw5saf/16uvtsE97g0JDApuXhpC++BvsXguzHoCYFPtp9/hfQeHGjikQ2Bl4EuwTs4MrV6GhpthYu81tXAZseMe/bXKTBoUgd8G4LF65ZiJ19YZzH13M62t2wdi5sO0L2Pf9kQfvWWfXiXszKIDdNzo03E46t1f5fptY1+6g4Exg80KuQsHXtic04kI7JOA0+Cw72a69hbbV1dqJf3eHBZ0rkIIl5yRvGaQMhOju9oPDoDNg06dQXebvlrlMg0InMDIrkbd/OoVRWYnc8tIa/rp3PCYk3CazNZbnmGRu78qjpuLSYPJNtvTFjuY213ODq6WV29LQU9jVvuvUVsHr10JsD5h575H3hYTAtF/C/u86bke6YFW6G0y9Zz0FCI5chfp6u/Ko8d/X4DPtB7FNn/qvXW7SoNBJpMR247mrjuNHk/vwyPJiFkdMon71C3b7Q6f85Xasvb2TuM2Z9FOITbNLNdszudvexDUn53LW9g4fLfgT7MuFs/5hS3k3NWQ2pA5y9BY6YEe6YOVpsA+mXIXCTTZru3FQ6D3Z/r8JoiEkDQqdSHhoCL85cwgPzhnFo2XTCakqYvvnzx4+IH+ZHTryYE/pNkXE2GGk/OWw/qhN9lxX3M4SF05hERDTo33DDnnLYPFDMOYyO1fTHGdvYd8G7S20xtOgEEy5CnmOXnLWcYdvCw2DY2bC9x+0vCIwwGhQ6ITOHpXJHddcyRbJovjzx3hh6Q4o3QcHt3l/PqGxUT+EnsMcCW0u7BzXnKJ8CIu0Y7LtFZ/ueU+hutyuNorPhFPubv3YoedAyjHaW2iNpz3A0DB7TjAEhfxlEJlo8xMaGzzLrtDb9oV/2uUmDQqd1JDMBNJOvI7hIVt48Y03ePoVx5ad3p5PaCwkFE75g/0DXvovz65RlGc/TXqjNxOf6XlQ+PT3cGAznP0IRLZR5iskFKb9yg4z+aLsR2dQvBO6JbT9b9mcYMlVyHPMJzTNw+h3AoRHQ25wDCFpUOjEosdejAmP5u5eSynb/BU1hFEQfYxvH7TfCdD/ZLtax5OkHU/WsrfE0205ty2CpY/CuB9D3+NdO2fYD+wnRO0tNK89uSeJvQM/KFQctEOIzX3oCo+C/ifZ3KEg+L+hQaEzi0xAhp/P8IOfcEnKRjaY3pz56EoWb97v28c95Q9QXQIL72372Kbasw1nU/EZUHnoyMn2tlSVwhvXQVIfOPl3rp8XEmrnFvauD6pJxQ5TlO/5pkmJ2VCyy/MhyY7g3A63pfIxg8+0K7B8vY2tF2hQ6OzGXQW1lcQfyiVr5HSSYiK45N9Lmff55iPrJnlTj8F2cnb5v6Fws+vn1VZD6R4v9hQcb0IlbixL/fhO+6n0nMfs5Lk7hp0L3ftpb6E57eopBMG+CnnLbBHHzGObv3/AKXbCvD21kDoogU+DQmeXPqJhcjlxwCTeuH4ypw1L40/vbeCGF76mrMpH5Z+n32EnjD/+jevnlBQAxvNPlE25uy3npk9tbsfE623tGneFhtnewp5v4DvdwrxBdTlUHPB8mXGiYyVaIOcq5C21iyy6xTZ/f1Qi9Jlme5GefBirqYR502HRA+1qpis0KHQFE2+AsCjoPYnYbmE88sMx3D5zEO+v28XsR75ky75S7z9mXE+YfLP9I9i+2LVz3K233xZXt+U0Bta8CPN/ZFcRnfBrzx9z+PnQva8dOvNFT8wY+OjX8N373r+2r7R3mXGg5yrU1drKuY2XojZn0CxbCmZvrvuPsfBeu5AhbbhnbXSDBoWuYOhsuHVbwydnEeEnx/fj2SuPo7CsmrMf/pKP1u/2/uNOvN7Wfvnw/44cTik/YP+IvpkPC/9qx/CfnAkvX2bvb2+OglOcC9tyHtoBz50Lr//EBoSLXrQTg54KDYOpv7B1knzxxr3+dVj8D3jrRvfmSvzJ3R3XmorLsPsTHMrzXpu8ae+3UF3a9sq+QWcA4v4QUsFq+PJBGHUJ9D/R42a6qh0F61VQCY886qbJ/VN4+6dTuO65lVz97EpumNGfW04+htAQLyW3RUTDiXfCG9fCc+dAxSG7/0Jl0ZHHxaXbid1jTrUVXJP7e+/xIxOb7ynU19kifp/+wS5/nflXO//iTlnnloy4ED7/Cyy8x+5n7a1kwepyOxwXl27nSVY8CZNu8M61fam9PcDQMBtQArWn4NzzvK2gEJdmj9nwNky/1bVr19XAmzfYIoyn/rF97XSRBoUuLjMxipd+MpHfvrWehxdsYu3OIu6/YCTJsV4qrT1iju0RHNhih1V6jbUBoHsf+z0px755+0p85tH1j/bmwls/tdnX/U+GWfcfHrf2Bmdv4a0bbNnkgad557qL/2E/dc99105mf/kAjL3C/QnxjlaUD4j9xO+pQF6WmrfM7gue2LvtYwfNsosZDm6z//fbsugBO0d14fPNl1nxAQ0KisjwUO45dwQjsxK56831TLl3AbNHZzJ3Ug4D0+Lad/GQELi0HWUv2qtxrkJtFXxxH3zxd+gWBz943M4B+KLsx8g5dme2hffYHlB7H+NQHiy639ZaypliV7I8eardk3vyjd5ps68U59s3zbAIz6+RmA2bF3ivTd6Ut9T2AFx5jQc7gsKGd+3wamv2brA9zqHnHFmd18d0TkE1uGh8Nu/eOIWzR2Xw2qp8Tn3gcy6at4QP1u0+vLtbsHFuy5m3DP41zb5JDz0HblgOIy7wTUAAW0p86s9t2e2NH7X/ep/cBRibAwJ2dVTfGXasOdDLMntj06RAzVUo2WM/9bc1yezUva/dnKmt7Ob6OnjzeoiItUObHUiDgjrCgJ5x3HPuCJbcfiK3njaIHQfKuea5lUz7ywIeW7iZQ+XV/m6ie+IzoWwvPHGKTUz74Stw7uN2jNbXRl5kh8jevtm+eXhq+2K77efkmw6vxAGYcYfdf2KZl/fIbk3B1/C/P7pX3K1oZ/uXGSdmAybwchWc8wnu7Hk+eBbs+ApK97Z8zNLHbKLbzL9AbGr72ugmDQqqWUkxEVw7vR8Lfzmdxy45lqzuUdzz/gYm/PlTbnt1Lbm7vLTNpa+lDQMExv8Yrl8Cx5zScY8dFgEXPmuzql+62LNPufV18P6tNrhNvunI+7LGQ78TbSXXKh8sK25qxxL4z5l2WGzZPNfOMY438vauKAvUZakNe56PdP2cwWcCpuVclsLNdgHEMafB8PO80kx3aFBQrQoLDeG0YWm8ePVEPrh5KueMzuSN1TuZ+eAXXDRvid0CNJANPB1u2w6n/9XOI3S0tOE2Ozp/ue0xuJu78PVzdnnryb9vfkJ5xh1QXgjLfdxb2PoFPPsDm3+SMxUW/BlKXFjGXH7AbjLT3v0xAjkopI9qdnVfi3oOs5PSzQ0h1dfD2zfZ4cdZ9/tueLMVGhSUywalxfPnH9ihpTtOH8TGvSWc9fAi/vDOt5T6KjO6vUQgMsG/bRhyNky/Hda8AF894vp5lUW2Wmv2RFtCozm9xtoVVF8+BFUl3mlvU5v/B8+fb1dozX0PznzQbpnqSrZ6sZd20mvIVQigoFBbZYfT3K08LGJ7C1sXQmWTHveq/9gS26f84XBGfgfToKDclhgdwdXT+vHpz6Zz0fhsnvxyKyf9fSEfrNvlu3pKwW7ar+yezh/fCRs/ce2chX+xvYDT7mn9E+P0220ZCVeHdNzx/YfwwhybOzL3XdtTSO4Hk26EtS/Bti9bP79hc5129hQCcV+FXWttcPSkHP2gWVBXfeQihKJ8+Og3thzGmMu91043+SwoiEiWiCwQkVwRWS8iNzVzjIjIQyKySUTWisgYX7VHeV9CdDh3nzOcV6+dRFJMBNc8t4qrnl5B3oEgybTtSCEhdhipx1BbTmP/xtaP37/RTjaOuRQyRrV+bK9jYcCpNo+h6SfP9sh9G1682BY4vPytIyfnp/7czhO890tb5qElRV7aSQ8Cb18FTyaZnbLGQ0zq4Yq6xsA7t4CpgzMf8suwkZMvewq1wM+NMYOBCcD1IjKkyTEzgQGOr6uBR33YHuUjY7KTePuGyfz6jMF8taWQk+9fyKOfbaamTiuFHiEiBi56wY4X/3eOzfBuyYd32I1ZTnCxoOD0W21N/2Uebm7U1LpX4eXLbUC67M2jd8KLiIbT/mxLhbc2n1GUZydio72w2isx+3DJjECQt9S2KT7d/XNDQm3Zi40f22J3a1+2vYYTf2MTO/3IZ0HBGLPLGLPK8XMJkAs07UOeDTxjrCVAooh48C+s/C0sNISrpvblk58dz7QBqdz7wQbOeOgLlm/zYKOdziwx265IOrjN9hjq644+5vuP7BvE8be6vhwx81i7WmXxw0eXEXHXmhfh1avs2vtLX7cVPpszaJZd/bTgTy0vuS3eaYd9vFE+JDHb5pzUBsCyaGMcO625mJ/QnEFn2ppJ37wMH9xqexzjr/ZeGz3UIXMKIpIDjAaWNrkrE2gc+vM5OnAgIleLyAoRWbFv3z5fNVN5QUZiFPMuG8vjl42lrKqO8x/7ilvnr+VAWQD8IQeK3pPgjL/D5k+PnqytrYYPb7dj+O6+QUy/zS5/XdqOuYVVz9i9qXOmwCXzW1+xJWLX0ddUtDzp7I3ENSdnrkJxAOQqFOXZZLr2BIU+06BbvB02qi6Dsx+2PQg/83lQEJFY4FXgZmNM0wHP5gbOjpqpNMbMM8aMNcaMTU3t2EQO5ZmTh/Tk459N4yfT+jJ/VT7T/7qAxz/fQlVtM5+Mu6Jj59o3/a8ehtUvHL592Two3ASn/tn9shAZo+0S3K/+4VlvYdnjtiZU/xPhhy+7VlMppb8ts7H2Rdj+1dH3e3N71UBalprnnE8Y5/k1wiLs5jv1tbZXmDrQO21rJ58GBREJxwaE540xzRXAyQcaz0D1AjzcaV0FmuiIMG4/fTDv3zSVMb2TuPu9XE66byHvrC3QVUoAp/7Jflp8+ybIWw6l+2zd/AGneJ5kN/02GxCWPOb6OTUVtoT5e7+wQWXOC+6VD5/6c7uF6nu/OHLSua7WbpzUWYNCeLTNOWiPyTfChOuOTkz0I1+uPhLgCSDXGHNfC4e9BVzmWIU0ASgyxrixd6IKBsf0jOM/V4zn2SvHExMRxg0vfM25jy5m5faD/m6af4WGw/lP2/XoL/4Q3r0FasptsPBU+kg71v/VI61PZIMd2vnkd3DfEFjwR1ts7/ynIczNCrkRMXDan2DPOljxxOHbS3aBqW9/4ppTfKbNVdj4kQ1k/pS31M7jhLazpmj6SDthHxrunXZ5gS97CpOBS4ETRGS14+t0EblGRK5xHPMesAXYBDwOXOfD9ig/mzoglXdvnMq95w4n72AF5z66mOtfWMWOwi68hDW6u93Yp6bcLgE97hpIGdC+a06/DaqKYEkzi/mMseUqXr4cHhhhy2/3nmRzEM7/j+eVTAefZQv0/e+Ph2v6tHfHtaZCw2DCtfbf6Z8T/Vc1tboMdn/TvvmEACbB1o0fO3asWbFihb+bodqprKqWeZ9vYd7nW6irN1w+qTc3zBhAQnTgfGLqUBs/sZ+yz3nMOxnYL10CWxbCzWttHf7aKlj3ms192LXaPsaYy2DcjyHJhX0AXLF/o32zHn4+nPOo3Ufj1SvhuiU218Fbtiy0k7MHNtsNjU79U8cUOHTatgj+c4addznm1I573HYSkZXGmLFtHqdBQfnTnuJK/v7Rd7yyMp+EqHCun96f04alkdXdhxvvdAW718Fjk2H8T+yS0hVPQtk+SBkIx/3E7vfgi815Pvmt3ffhRx/aHsknd8FteRAZ793HqamEL/5mN6HpFgun/BFGXdwxSV9f/N2WH/nV1qPzNwKYBgUVVL4tKOZP7+WyaNN+ALK7RzOpXzKT+qcwqV8yKd7aCa4reelSyH0LEPuJ9rhroO90375xVpfBw+MgqjtkjYNvXoXbfTgxvHeDY6J+iS3UN+v+9g+/teWFC+1Ogjcs9+3jeJkGBRV0jDFs2lvKl5v28+XmQpZsKaSk0q5mGZQWx6R+KUzun8z4Pt2Ji+yiw0zuKNppi/AN/YGtV9RR1r8Or8yFsCi7qcx1i337ePX1sOpp+PguW5F16i9gys3uT5i7whj4Sx+bjXy2G8UNA4AGBRX0auvqWVdQzJeb9rN4835WbDtIVW09oSHCiF4JnDumFxeOyyI8VOs6BhRj4JmzbRXQAafAxa90zOOW7IEPboP1r0HKMbbXkDPFu4+xfyM8PNbWJzrWf0XrPKFBQXU6lTV1rNpxkMWbClnw3V7WFxTTNzWGW08bxClDeiJ+LCKmmtj3PTw6CcZeYfey6EgbP4Z3fgZFO+yQ0uSbbUKeN/5/fP08vHkdXLcUegxq//U6kAYF1akZY/gkdy/3vJ/L5n1ljMtJ4vbTBzMmO8nfTVNOBV/bpLYO3k4SsHMbK56Er/5pE+h6DrMJYkPPaV9OwFs3wrdvwK+2eaeeUwfSoKC6hNq6el5akcf9H29kf2kVpw9P41enDiInxQcra1Twqa2Gb16BLx+E/d9BQjZMvN6WJPdk9dUjE2yG9iXzvd9WH9OgoLqUxnkPtfX1XHxcb248cQDdYzxMxlKdS309bPzQLmHNW2JzN8ZfbZfsxiS7do2KQ3Bvb5jxazj+l75trw9oUFBd0t7iSu7/ZCMvLd9BTEQY187ox48m9yEy3P/VJ1WA2LHE9hy+e8+ukBp9id3WFBrtod3ofdF5W+EmWHSf3V+i7/QObLB3aFBQXdrGPSXc+8EGPsndS0psN6YOSOG4Pt0Z36c7fVJidFJawb7vYPFDsOYlqK9x7ZyIOPh5buslxQOUBgWlgK82F/Lskm0s23qA/aV2T4fUuG6M79OdCX26M75PMgN6xBISokGiy6o4aL8aNPq/0PDhwfE9MqHlTYcCnKtBoZ0l/pQKbBP7JTOxXzLGGDbvK2PZ1gMs21rI0q0HeHetLcibFB3OuJzuHNc3mZMH9yQ7WUtsdClRSfZLAdpTUF2UMYb8gxUs2VLIsq0HWLr1ADsOlCMCMwb24NKJvTl+QKr2IFSnocNHSrkp/2A5L6/I54WlO9hfWkVOcjSXTOjN+cdmdd3qrarT0KCglIeqa+v5YP1unlm8jRXbDxIZHsLsUZlcNjGHIRlervapVAfRoKCUF6wvKOLZr7bzxuqdVNbUM7Z3EpdNyuG0oWlEhAVXRqvq2jQoKOVFReU1vLIyj2e+2s6OA+UkRIUzoEcsvZNjyEmOpneK43tyDAlROtSkAo8GBaV8oL7esHDjPj5ct5ut+8vYXljO7uLKI45Jig4/HCySY5gxqAejsoJzGaPqPDQoKNVBKqrr2HGgnG2FZWwvLGNbYbn9vr+cgqIKjIGJfZO5bkY/pvRP0cQ55Reap6BUB4mKCGVgWhwD047Oci2tquXFZTt4/IstXPrEMoZnJnDt9H6cOjSNUF3uqgKQ9hSU6gBVtXW88fVOHlu4ha37y+iTEsNPpvXlnDGZdAvTukzK93T4SKkAVFdv+HD9bv752SbW7SymZ3w3rprSl4uOyya2m3bcle9oUFAqgBljWLRpP/9csJmvthSSEBXOnPFZjOyVSP8eseQkx+iSV+VVOqegVAATEaYOSGXqgFS+3nGQRz/bzLzPtzRUaQ4NEbK7R9MvNZb+PexXv9QY+vWIJT5Sl7wq39GgoJSfjc5OYt5lY6mormPzvlL7tbeUTftK2bS3lIXf76Wm7nCPvmd8Nyb0TWb2qEymDEghPFR7FMp7NCgoFSCiIkIZlpnAsMyEI26vrasn72AFm/baIPH9nhL+t2Evb64uIDkmglkj0jl7dCajsxJ1uatqN51TUCoIVdfW89l3NjB8kruHqtp6eidHc/aoTGaPyqBvaqy/m6gCjE40K9VFlFTW8MG63byxeieLNxdiDIzolcDZozKZNSKdnvGR/m6iCgAaFJTqgvYUV/L2mgJe/3on6wuKAegR141B6fEMSotjYM84BqXH0b9HrOZHdDEaFJTq4jbtLWHBhn3k7i7mu90lbNxbSnVtPWBXN/VJiWFQWpwNFmnxjMlOJDm2m59brXxFl6Qq1cX17xFH/x6HS2/U1tWzrbCMDbtL2LCrhA27S1idd4h3HNuSAgzNiGfKgBSmDUjl2N5JRIZrb6Kr0Z6CUl1cSWUN3+0uYcmWQr7YuJ+V2w9SW2+IDA9hfJ9kpg1IYcqAFAb2jNPVTUHM78NHIvIkMAvYa4wZ1sz904E3ga2Om14zxvy+retqUFDKt8qqalm6tZDPv9/Pok372bS3FIDUuG5M7Z/CtGNSmTGoh+4bEWQCYfjoP8DDwDOtHPOFMWaWD9uglHJTTLcwThjUkxMG9QSg4FAFizbt54uN+/ns+3289vVOwkNtRvbMYWmcPKQnidERfm618hafBQVjzOcikuOr6yulOkZGYhQXjM3igrFZ1NcbVucf4v1vdvHeN7v534a9hIUIk/qncPqwNE4Zmkb3GA0QwcyncwqOoPBOK8NHrwL5QAHwC2PM+hauczVwNUB2dvax27dv91GLlVKuMsbwzc4i3v1mF+9/s5sdB8oJDREm9k1m5vA0Th2aRoquZgoYfp9TcDQih5aDQjxQb4wpFZHTgQeNMQPauqbOKSgVeIwxrC8o5v11tgexdX8ZIQLZ3aNJT4giIzGKzMRIMhKjSHf8nJ4QRYyWC+8wAR8Umjl2GzDWGLO/teM0KCgV2IwxfLenhA/X7WHTvlIKDlVQcKiCPcWV1Dd5u0mICicjMYrUuG6Eiq0eK4A0/RlBBEJESIgOJy0+krT4SHomRDb8HB8VpqujWhEIE82tEpE0YI8xxojIeCAEKPRXe5RS3iEiDEqLZ1Ba/BG319TVs7ekqiFI7DxUwa5DlRQcqmB/WTXGGOqNwRjsFzbA2J/t9zpjOFRew4Gy6qMeNzI8xAaK+EjSEiLJSY5h1oh0BvQ8eptU1TKfBQUR+S8wHUgRkXzgLiAcwBjzGHAecK2I1AIVwBwTbEkTSimXhYeGkJkYRWZiVLuvVVVbx97iKnYXV7K7qJI9ju/O31duP8jbawp48NONDMuM5weje3HWqAyd43CBJq8ppTqlfSVVvL2mgNe+zmfdzmJCQ4Tjj0nlnNGZnDykZ5fL1g6IOQVf0KCglHLXxj0lvPb1Tt74eie7iiqJ6xbG6cPTOWdMJuNzuhMS0vnnIjQoKKVUE3X1hqVbCnl11U4+WLeLsuo6MhOjmDEolTHZSYzJTqJ3cnSnnLDWoKCUUq0or67lo/V7eHP1TpZvO0hpVS0AyTERjM5OYkzvRMZkJzGyVyJREcE/1BTwq4+UUsqfoiPCmD06k9mjM6mrN2zcW8LK7QdZtf0QX+84yCe5ewBbZnxwehxjspMYnZ3IoLR4+qXGEhHWOffG1p6CUko142BZNV/n2SCxasdB1uQdoqy6DoCwEKFvagwD0+IZ2DOWgWl2E6PMxKiAnZ/QnoJSSrVDUkzEEYUB6+oNm/eVsmF3Cd85Ni5anWeXvjrFRIRyjGPjoj4pMfRKiiYrKZqs7lEkRIUHxVyFBgWllHJBaIhwTM84jukZByMzGm4vrarl+z0lfLfbfm3YXcwH63ZzsLzmiPNju4XRKynKBoruUWQlRdMrKYqs7tFkdY8mNkBKfgRGK5RSKkjFdgtrWLnUWHFlDXkHysk/WNHwPf9gOXkHylm8eQDJwZIAAAZvSURBVD/ljqEop+4xEWQlRdGrezTZ3Q/3MLKSoslIjOqwOQwNCkop5QPxkeEMzUhgaEbCUfcZYzhQVm0DxsFy8g5UsONAOfkHy1m/s4gP1+2mtlGhqBCB9IQorpicw1VT+/q03RoUlFKqg4kIybHdSI7txsisxKPur6s37C6uJO9A+eGvgxWkxvm+TIcGBaWUCjChIdJQJ2pC3+QOfezOudBWKaWURzQoKKWUaqBBQSmlVAMNCkoppRpoUFBKKdVAg4JSSqkGGhSUUko10KCglFKqQdCVzhaRfcB2D09PAfZ7sTmBoLM9p872fKDzPafO9nyg8z2n5p5Pb2NMalsnBl1QaA8RWeFKPfFg0tmeU2d7PtD5nlNnez7Q+Z5Te56PDh8ppZRqoEFBKaVUg64WFOb5uwE+0NmeU2d7PtD5nlNnez7Q+Z6Tx8+nS80pKKWUal1X6ykopZRqhQYFpZRSDbpMUBCR00TkOxHZJCK3+bs93iAi20TkGxFZLSIr/N0ed4nIkyKyV0TWNbqtu4h8LCIbHd+TWrtGoGnhOf1WRHY6XqfVInK6P9voDhHJEpEFIpIrIutF5CbH7UH5OrXyfIL5NYoUkWUissbxnH7nuL2PiCx1vEYviUiES9frCnMKIhIKfA+cDOQDy4GLjDHf+rVh7SQi24CxxpigTLoRkWlAKfCMMWaY47a/AAeMMfc4gneSMeZWf7bTHS08p98CpcaYv/mzbZ4QkXQg3RizSkTigJXAbGAuQfg6tfJ8LiB4XyMBYowxpSISDiwCbgJ+BrxmjHlRRB4D1hhjHm3rel2lpzAe2GSM2WKMqQZeBM72c5u6PGPM58CBJjefDTzt+Plp7B9s0GjhOQUtY8wuY8wqx88lQC6QSZC+Tq08n6BlrFLHr+GOLwOcAMx33O7ya9RVgkImkNfo93yC/D+CgwE+EpGVInK1vxvjJT2NMbvA/gEDPfzcHm+5QUTWOoaXgmKopSkRyQFGA0vpBK9Tk+cDQfwaiUioiKwG9gIfA5uBQ8aYWschLr/ndZWgIM3c1hnGzSYbY8YAM4HrHUMXKvA8CvQDRgG7gL/7tznuE5FY4FXgZmNMsb/b017NPJ+gfo2MMXXGmFFAL+zIyODmDnPlWl0lKOQDWY1+7wUU+KktXmOMKXB83wu8jv3PEOz2OMZ9neO/e/3cnnYzxuxx/NHWA48TZK+TY5z6VeB5Y8xrjpuD9nVq7vkE+2vkZIw5BHwGTAASRSTMcZfL73ldJSgsBwY4ZuMjgDnAW35uU7uISIxjogwRiQFOAda1flZQeAu43PHz5cCbfmyLVzjfPB3OIYheJ8ck5hNArjHmvkZ3BeXr1NLzCfLXKFVEEh0/RwEnYedKFgDnOQ5z+TXqEquPABxLzB4AQoEnjTF3+7lJ7SIifbG9A4Aw4IVge04i8l9gOrbM7x7gLuAN4GUgG9gBnG+MCZqJ2xae03TssIQBtgE/cY7HBzoRmQJ8AXwD1DtuvgM7Dh90r1Mrz+cigvc1GoGdSA7FftB/2Rjze8d7xItAd+Br4BJjTFWb1+sqQUEppVTbusrwkVJKKRdoUFBKKdVAg4JSSqkGGhSUUko10KCglFKqgQYFpTqQiEwXkXf83Q6lWqJBQSmlVAMNCko1Q0QucdSoXy0i/3IUHCsVkb+LyCoR+VREUh3HjhKRJY5iaq87i6mJSH8R+cRR536ViPRzXD5WROaLyAYRed6RZatUQNCgoFQTIjIYuBBbcHAUUAdcDMQAqxxFCBdis5UBngFuNcaMwGbKOm9/HnjEGDMSmIQttAa2MufNwBCgLzDZ509KKReFtX2IUl3OicCxwHLHh/gobMG3euAlxzHPAa+JSAKQaIxZ6Lj9aeAVR12qTGPM6wDGmEoAx/WWGWPyHb+vBnKwG6Mo5XcaFJQ6mgBPG2NuP+JGkTubHNdajZjWhoQa15+pQ/8OVQDR4SOljvYpcJ6I9ICG/Yh7Y/9enFUnfwgsMsYUAQdFZKrj9kuBhY4a/fkiMttxjW4iEt2hz0IpD+gnFKWaMMZ8KyK/xu5qFwLUANcDZcBQEVkJFGHnHcCWJX7M8aa/BbjCcfulwL9E5PeOa5zfgU9DKY9olVSlXCQipcaYWH+3Qylf0uEjpZRSDbSnoJRSqoH2FJRSSjXQoKCUUqqBBgWllFINNCgopZRqoEFBKaVUg/8H1Ks7Bux9JjUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Формируем файл для отправки на kaggle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Считываем тестовые данные."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle_test_data = pickle.load(open(pickle_test_path, 'rb'))\n",
    "x_test = []\n",
    "\n",
    "for i, row in enumerate(pickle_test_data):\n",
    "    feature = prepare_shape(row['feature'], config)\n",
    "    x_test.append(feature)\n",
    "\n",
    "x_test = np.asarray(x_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загружаем веса обученной модели, делаем предсказание и сохраняем в файл."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_file_name = './base_submission.csv'\n",
    "\n",
    "model.load_weights('./best_weights.h5')\n",
    "config = Config(shape=(64, 63, 1), learning_rate=0.001, n_classes=41)\n",
    "\n",
    "pred_batcher = Batcher(x_test)\n",
    "\n",
    "predictions = model.predict_generator(pred_batcher,\n",
    "                                      use_multiprocessing=False,\n",
    "                                      workers=0)\n",
    "res = np.argsort(predictions)\n",
    "id_to_label = {i: label for label, i in label_to_id.items()}\n",
    "with open(pred_file_name, 'w') as f:\n",
    "    f.write('fname,label\\n')\n",
    "    for i, row in enumerate(res):\n",
    "        row = row[::-1]\n",
    "        f.write('{},{} {} {}\\n'.format(pickle_test_data[i]['fname'],\n",
    "                                       id_to_label[row[0]],\n",
    "                                       id_to_label[row[1]],\n",
    "                                       id_to_label[row[2]]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Готово! \n",
    "\n",
    "Теперь в рабочей директории появился файл с предсказаниями для тестовой выборки - его можно смело отправлять на kaggle. Модель выдает результат около 60%, сравнимый с точностью базового решения, предложенного авторами конкурса.\n",
    "\n",
    "Далее, например, попробовать улучшить модель: поиграться с архитектурой сети, входными данными - использовать другое число мел-коэффициентов или, например, другие признаки. Также стоит попробовать увеличить частоту дискретизации - изначально файлы имеют частоту дискретизации 44,1 кГц, мы специально уменьшили её до 11,025 кГц - для увеличения скорости работы, однако, для получения лучшего качества классификации не стоит делать передискретизацию до таких маленьких значений.\n",
    "\n"
   ]
  }
 ],
 "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": "3.5.5"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "00dd0366d6094ec292921424a35abcbe": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "01058314c19f45a990221fcfc68153bf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "014baea001864dcd96833c86901f2055": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "018aed93fa044868b4b99a246a3fbf4f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 26",
       "layout": "IPY_MODEL_ea2695708a844cb189e7b365c001f31d",
       "max": 85,
       "style": "IPY_MODEL_dd273d753c00420eb3f8c5e646de02f3",
       "value": 85
      }
     },
     "01931b079d6d4a75b0728a3705361bfc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "01f2e1d7011c4a9cad4b08453c53b091": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "020dac6bb7704cbd873b264fa7c467a3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "025a34b7970141afae8b174658fff9a5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "02bca7998ce049be82b68685cbcf490e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "035db4ba8b904ac98386ed83d4887050": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 16",
       "layout": "IPY_MODEL_09914f87d3b543c9885dc36672507307",
       "max": 85,
       "style": "IPY_MODEL_f8f7e3234c344d25bbf4f91a755f50f8",
       "value": 85
      }
     },
     "0375cf926e6846e18b4a254e73b55c6a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "040cb76f591640fe8be6a6186fda8860": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "04a9e1ce83b34f328ff7bb395a3ba753": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_4af30232d25343e0bd525e7be01b56e2",
       "style": "IPY_MODEL_752388bdd5bf469daefdba24cc9d9bb5",
       "value": "[loss: 2.277, acc: 0.360, val_loss: 2.341, val_acc: 0.346] : 100% 85/85 [01:53&lt;00:00,  2.52s/it]"
      }
     },
     "04b890966ed94d10af85be20bf77f32a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 8",
       "layout": "IPY_MODEL_b1ae8bc50c6f41b89c29625ed968dc61",
       "max": 85,
       "style": "IPY_MODEL_50c62bff1e934374a2323a108d2e92b3",
       "value": 85
      }
     },
     "04bd07e1a287494e974e0545c068bc7f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "04fddd423e4042ba87f50c168f637391": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0525d037ea364835a88c2ec0dcc9ae97": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_025a34b7970141afae8b174658fff9a5",
       "style": "IPY_MODEL_250815681e6847f1b88a48a78555fd3f",
       "value": "[loss: 1.648, acc: 0.530, val_loss: 1.943, val_acc: 0.486] : 100% 85/85 [01:57&lt;00:00,  2.70s/it]"
      }
     },
     "05672ba006b641cf99bb5d6147bf2bec": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "056c1fe2d4044395826bc8c38252f4e3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "05d9f4e591074fcda4003539666b3523": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "05dcabf7ab424b33ba15f1743f705856": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "068d18cbb6694ceb9a926669b814aeec": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 29",
       "layout": "IPY_MODEL_c61ad3b1a92d4c0cb1f7751e8910d736",
       "max": 85,
       "style": "IPY_MODEL_6fa034bd34104dc6b9e8eb7ab63cf8d9",
       "value": 85
      }
     },
     "06d05dc4798242ea9151c37e353fce9d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "071cf43f228745639965bf0cf846be5a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "07664b00150942269660686be5fe10f3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 11",
       "layout": "IPY_MODEL_4aad389cbd394ea4873b21d08d5cc1ef",
       "max": 85,
       "style": "IPY_MODEL_97cba60ffa394555bf4a25d80885add8",
       "value": 85
      }
     },
     "07a20e4ca88d46038d5480aba752cd3d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0818e05c3a424e4e91bc6b023a683779": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "081f7a7a0d79479fb91d7f30f376bf30": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "084fbdee7e2f4fe8a90bd37d469f9aeb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "08f255e2c3cb43ad941abd5ccf7e6560": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "08f26b1e00b74b409429fcbb783a0298": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0915628c33fb41f0977369d1ae562740": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_c8d21dd752c1452191c9644f2405ab24",
       "style": "IPY_MODEL_5d38bdf927124eb19e000d7dd0416343",
       "value": "[loss: 1.781, acc: 0.495, val_loss: 2.169, val_acc: 0.414] : 100% 85/85 [01:50&lt;00:00,  2.45s/it]"
      }
     },
     "093a5c76774743da977a3472163eb242": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "093c55fe1eac435ca23c002750384091": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "093df05c2e0a4c7d98975c4e6a4736d0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "09824284f41247509d4f61d785824235": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "09914f87d3b543c9885dc36672507307": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "099a3c309e6b40c1ae6bf33211fd1ee3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "09e54b53fa8e4c05b5a13a7ee4eebb84": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ac7f130cff2d4666a2010b2d7f49ffd8",
       "style": "IPY_MODEL_329b7a756b27422a8e7c47037719a746",
       "value": "[loss: 1.395, acc: 0.595, val_loss: 2.295, val_acc: 0.426] : 100% 85/85 [02:02&lt;00:00,  2.77s/it]"
      }
     },
     "0a1af01558c64d559c0b20948ba18d3b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0a23a79f1fea41409b87688be1eac49f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0a34f42d4b574677aa1f7fa9545b5c37": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0a88f82b474d4fd6ac62c4213d9a93c4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0aa6c9736bfb4f82b6c6f4498bf1d657": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 12",
       "layout": "IPY_MODEL_98d8bfc3f04140b98a87c960b9fc2ad0",
       "max": 85,
       "style": "IPY_MODEL_b514b0b0730e4bcb877ca61259e1c0bb",
       "value": 85
      }
     },
     "0ba56b6ec995403da8bf54c3335ee76b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0bc012afd4de449a9854450587fc7677": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0bdfcccffaed4f34a48c867a0037d724": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_2c4aa0379bb440cb86cbe6c8fc837613",
       "style": "IPY_MODEL_ea241ecb1f4e4275ad4d50d0d6cbdd30",
       "value": "[loss: 3.070, acc: 0.191, val_loss: 2.788, val_acc: 0.234] : 100% 85/85 [02:07&lt;00:00,  2.92s/it]"
      }
     },
     "0c168351b4034db1b116b8fa03e83fd4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "0d35d592827c4518a3e3b33f703222d9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "0d9bd92e5d934f50bd150c1317a1d156": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0db5df800bb44ed99faa08268f1a1181": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 36",
       "layout": "IPY_MODEL_2e3c9f4f99b04f84a2dc3f9f6bd47c09",
       "max": 85,
       "style": "IPY_MODEL_f9c59b7ae4a341fb95495ae25e2ce97a",
       "value": 85
      }
     },
     "0dcb362090284d998374a3846ab3a13e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0e1b3c7ef3ea47b6af0dd05445fe59a8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0e1f500ab9734e02bfdfe1f6ff7f2cf5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0eb0f14c032b457da307c5b344dc8a42": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0f0eeab106134961b2aca7e3ffe34200": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "0f19c0ea3e434dbdb48cf95fdab864f4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1004bb6a5c4a44d680c6810be9e75796": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "101c2ea80f9e4417a66afc1d9e69fc5f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "109b2d7f44944418814c69cd79c3b2dc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "10a7fed5ef4c4307b5bfcf4aab91f0d9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_093df05c2e0a4c7d98975c4e6a4736d0",
       "style": "IPY_MODEL_ea0ac4c5d4d5449f8231d9aaabdd875c",
       "value": "[loss: 1.174, acc: 0.647, val_loss: 2.285, val_acc: 0.450] : 100% 85/85 [01:47&lt;00:00,  2.43s/it]"
      }
     },
     "12a3e41a1dd14e089163d6e3f4710bf9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b5d1b1a5797c4a32a6ff8c23280f0d20",
       "style": "IPY_MODEL_e3dd4bca40d444c19e987e81d147437b",
       "value": "[loss: 1.121, acc: 0.662, val_loss: 2.419, val_acc: 0.440] : 100% 85/85 [02:11&lt;00:00,  3.71s/it]"
      }
     },
     "12e6a775481643ea9b39d7ac6046a003": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "134ca1b76372493f88371a814350e4e6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "135ce68b15db4b19990a33783e946775": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "13803f2cf1454bd6ab9d2a05ae575447": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1381cd3cf6a44af99293dbc35434c40a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "1382f9048fcf49baa0b54b733ef089f5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "14375834a7394a359e072ae3bb8a14f7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "14382e818345463fa4e3d003996314d8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "144228a2f653453495ade8526afcba6b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "144ab3e369064a2a908b0b2b1746ce16": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "154c9254bab6463db8af821080f5dc08": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "15ab12cd2778464ca2b75facbc0fa757": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "166def0781bb45b5bc816d4caf6961be": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "16b51918755f4ebfa6c4223b94f53dc4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "179997d11c9f4b7f98e87e13c207493e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_0818e05c3a424e4e91bc6b023a683779",
       "style": "IPY_MODEL_2b0cc150ee05457db308cb07f4cb4909",
       "value": "[loss: 1.587, acc: 0.537, val_loss: 2.190, val_acc: 0.416] : 100% 85/85 [01:55&lt;00:00,  2.45s/it]"
      }
     },
     "181db73d039343eaa6f86cae5e950d6a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "18da9f438a8d48b8a90ef3c8173122fe": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "18deeafab2ed4ead972ca9403fd0973f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1957800cd365400d9cc64dd30d7a29d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 6",
       "layout": "IPY_MODEL_b615ee544db24f49bbecd81619db68f3",
       "max": 85,
       "style": "IPY_MODEL_a4412b2c5d41423fbefdb8359e7017d6",
       "value": 85
      }
     },
     "1986c776469e42d8828841191c5febce": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1a085d86906444fbb0462458041babc6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1a21527fc86543868f04979b3accc4bc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 19",
       "layout": "IPY_MODEL_9f71294df925417486f2a84bb8834cd7",
       "max": 85,
       "style": "IPY_MODEL_0f0eeab106134961b2aca7e3ffe34200",
       "value": 85
      }
     },
     "1bc339bb051b45bab2a4483f0160fb9c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1be96ec6b7f44f868333ae67a26f8e8a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 3",
       "layout": "IPY_MODEL_0a23a79f1fea41409b87688be1eac49f",
       "max": 85,
       "style": "IPY_MODEL_3b38a4d2564648048c0acbb322f015dc",
       "value": 85
      }
     },
     "1caea7d7749c4489b4dbd83ebd1454a6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1cd58a36a61b494f83dac8c1838c332b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1d63b89b8a584ca8926013fa9a4accc8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 42",
       "layout": "IPY_MODEL_16b51918755f4ebfa6c4223b94f53dc4",
       "max": 85,
       "style": "IPY_MODEL_d711ba1d7cf2488a9b4318a500892fa8",
       "value": 85
      }
     },
     "1d731c64c565403c80f7ad9b17b8cfea": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_144ab3e369064a2a908b0b2b1746ce16",
       "style": "IPY_MODEL_54b93ad3ad6f499d83b940a34ef2aeb2",
       "value": "[loss: 1.536, acc: 0.553, val_loss: 2.240, val_acc: 0.397] : 100% 85/85 [01:51&lt;00:00,  2.50s/it]"
      }
     },
     "1e4e6ca2892b44ad9b4126859f15a52f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "1ebbbfead2724262a9220d1d55463cc1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 19",
       "layout": "IPY_MODEL_c203995a1920483db634003a99537348",
       "max": 85,
       "style": "IPY_MODEL_a0470e8f29514391a234b92758390ab3",
       "value": 85
      }
     },
     "1ecaeb35efa6411a8c615dcc296f5ed5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1efe7e8866f647628ed94601b8fb40a8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 6",
       "layout": "IPY_MODEL_212074bb5910434fa6cc7a3796911b9a",
       "max": 85,
       "style": "IPY_MODEL_6ef8e373d0d0497daf9858194bfeade5",
       "value": 85
      }
     },
     "201cd0b0a04441eab9479e5d8cffe163": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "20f10d30e3014378a5cc1dcd2849be4a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "212074bb5910434fa6cc7a3796911b9a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2145b10afb3b40af9014a92afbe054c7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "215042154bb74511916f28baef913b4a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "21cea497615242d5a16d242eb82ce6a8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "21f76b5379374028ba99fe1b95ae09b5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "239d21e464a7421b87a382aa9dd8bcd2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "24107643e350403bafcdd33d2b76ea68": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "24b442e79844455aa95f686b4dc6aea3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "250815681e6847f1b88a48a78555fd3f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2544eb4106ba40858aaa9d2933de3600": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "25728779659049c59b28c4836da88ad0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_bfdbf7f2f706457cbe4ec83bca7bc586",
       "style": "IPY_MODEL_290d8aa25df54f98be79b45c25aef78b",
       "value": "[loss: 2.578, acc: 0.288, val_loss: 2.924, val_acc: 0.233] : 100% 85/85 [01:50&lt;00:00,  2.56s/it]"
      }
     },
     "25880c90e1f14412b563b3f18194656f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "25c86704fef54a04bfa14b481a2a152b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "26059d8662834539897f0cf13c99451f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "26101d0c0d394b91ad6ff1fbf155bab0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 11",
       "layout": "IPY_MODEL_5a057e6461fb47d99c5329e3163dd603",
       "max": 85,
       "style": "IPY_MODEL_70702d11ca644f82a818eef533145a8b",
       "value": 85
      }
     },
     "271b46b0d4764329af26c15c49cce609": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2774f4afe66b4f42ada1ae4ffff7a98c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_dad746cf2db8428cb77f1bd0a82eef4a",
       "style": "IPY_MODEL_85703f52f4ca43ca8d64856c73641a8a",
       "value": "[loss: 1.849, acc: 0.479, val_loss: 2.038, val_acc: 0.463] : 100% 85/85 [01:59&lt;00:00,  2.52s/it]"
      }
     },
     "27b38fe9a1b2442d9837e42890f49c90": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_0a1af01558c64d559c0b20948ba18d3b",
       "style": "IPY_MODEL_135ce68b15db4b19990a33783e946775",
       "value": "[loss: 1.081, acc: 0.672, val_loss: 1.891, val_acc: 0.513] : 100% 85/85 [02:16&lt;00:00,  2.84s/it]"
      }
     },
     "27b85609fbb84c89941e7b6a82e1c690": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2837aef1c6604b378ce38c685ef003d0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "28983f815a284609aa0857045a84b32d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "28d89455cbc74cb995bc3f2f23a5e418": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 13",
       "layout": "IPY_MODEL_8b0cf9fffe984691ac075ede50eabf2f",
       "max": 85,
       "style": "IPY_MODEL_77b9807388704af297ff85e630071f19",
       "value": 85
      }
     },
     "290d8aa25df54f98be79b45c25aef78b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "29bbcb5f28504c1aa6e8850b1313ac25": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "29f93ccf7c474fd3bc53000361595d30": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_5d7cd41cac5e4bc0bb558743b79819a4",
       "style": "IPY_MODEL_6e43c8ce38684c4da2d481a015229988",
       "value": "[loss: 1.559, acc: 0.557, val_loss: 1.965, val_acc: 0.475] : 100% 85/85 [01:56&lt;00:00,  2.71s/it]"
      }
     },
     "2a29d8e5b4bd4a63acbddf6ccff38fac": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 2",
       "layout": "IPY_MODEL_4159bdf5c8374524b901c88152b146a0",
       "max": 85,
       "style": "IPY_MODEL_7eac89bba5914a23bf47f9fab800d210",
       "value": 85
      }
     },
     "2a3a7d6c28fc47d3823a55d4d5d23118": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 27",
       "layout": "IPY_MODEL_ab855c15ce4042d79b8e72603e090019",
       "max": 85,
       "style": "IPY_MODEL_34dd6ae0ba1b433e8618158fd94370a2",
       "value": 85
      }
     },
     "2ae9764dfb354de2a1243e829d284a22": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 17",
       "layout": "IPY_MODEL_939c172793dd4f70bc740d90b1f19783",
       "max": 85,
       "style": "IPY_MODEL_8cc98690cfaa45cf90c4aa1ea136a136",
       "value": 85
      }
     },
     "2b0cc150ee05457db308cb07f4cb4909": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2b44ba79c9fe4c1fbaa56e7e5f4d5b2e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2b9a6176c1004526bd3152e544d3f596": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2bf8d7485e934bfc956ecdc001357fc5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 21",
       "layout": "IPY_MODEL_808ffc7d5bea4161803accd3e0eb4f8c",
       "max": 85,
       "style": "IPY_MODEL_5c7b7c698a364d21bb5554efca9a04d0",
       "value": 85
      }
     },
     "2c1322f056454666babdadbefe1ea1dd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2c4aa0379bb440cb86cbe6c8fc837613": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2d0372c3380f4aa7b989b4b006b26c1c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2d3f9de353154c2593d18bde7284f47a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2d8edc78ba3a4ff4b12142adcfe71c49": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "2d9e195440ff489cb27940ee982b5626": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_81fc962a7b4e4e828377b17f17eb9cb4",
       "style": "IPY_MODEL_e5d1fd31934b4f8cbbd080d4192ba17c",
       "value": "[loss: 2.149, acc: 0.398, val_loss: 2.359, val_acc: 0.355] : 100% 85/85 [01:56&lt;00:00,  2.70s/it]"
      }
     },
     "2e07edcd82fd40f6b34f1860a0ac6ada": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2e3c9f4f99b04f84a2dc3f9f6bd47c09": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2f25cf81555e4437949af49823c0ba3b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2f53a81b513a42e7800bba66c23400bf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_7b2eec3a1e684be29b451fb87566ff6e",
       "style": "IPY_MODEL_bdaf77b79a0c4f5f8796bd76f8612203",
       "value": "[loss: 2.457, acc: 0.315, val_loss: 2.732, val_acc: 0.273] : 100% 85/85 [01:57&lt;00:00,  2.73s/it]"
      }
     },
     "2fae7445816e49959070c7fcd2cb8c1d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2fc50b09d66645688a822e540411d697": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3092cdeebc714fc792de698c8ff014d0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "30aff3681131437da217efd6ec742935": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "313911f3bc2b40e790347fecf87f472e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3149b53462634559beab7e920bd8f0f9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "31a252cb46234194b5129a47034b8807": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "31db592b35a548aea720835632ce8518": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3243d0d195f449a4a32e8986751e1dfa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 32",
       "layout": "IPY_MODEL_dbc342ac64b244ab9286fbdac5c0a74c",
       "max": 85,
       "style": "IPY_MODEL_9942705756304acea1882c2e55b58b72",
       "value": 85
      }
     },
     "329b7a756b27422a8e7c47037719a746": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "32b6a4ff74d14f24ad4ae50ccf82c20d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "330d8ef406964b40a4948bec54093218": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "331b688c2f9143b4bdee480a1ad74a2d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_46b13ef1db9645dfaf80764b262d2cb0",
       "style": "IPY_MODEL_376726004f1643bbae4e6fdb9973a53c",
       "value": "[loss: 3.728, acc: 0.112, val_loss: 2.926, val_acc: 0.185] : 100% 85/85 [02:00&lt;00:00,  2.53s/it]"
      }
     },
     "33f70b3f24654ad0b6c284e04ebd5de3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "34dc20e1544f40c09fd4f317abedf6ef": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 30",
       "layout": "IPY_MODEL_afa68b608c834f38bf0c2a9bd339db4e",
       "max": 85,
       "style": "IPY_MODEL_00dd0366d6094ec292921424a35abcbe",
       "value": 85
      }
     },
     "34dd6ae0ba1b433e8618158fd94370a2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "34e0b8fe1d054f2fb89e73e61e8d1489": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "34f6a36359314424bae3854398dfdceb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "354dc0248a514645870d88648ec4960a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "35a3694837044cee8b55515ae12ed6ad": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3604c7c282e1429ba81bb148dd65e295": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "369c0ba8aef54ccabe944e6fe102e29a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "36cf2dc7f50048ef8ff1a14c64300c58": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_752bdee4a8b74d138b513dadf499dc6c",
       "style": "IPY_MODEL_6f82fe6e8e89417a877f129f09bc05bf",
       "value": "[loss: 1.687, acc: 0.516, val_loss: 2.122, val_acc: 0.432] : 100% 85/85 [01:50&lt;00:00,  2.54s/it]"
      }
     },
     "36ff2cd1e83d40c694d0187ddb756744": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_df352299a68a4ae1acf2b0769ffb1d0f",
       "style": "IPY_MODEL_e630dd696e8d4be484395540073fb6d5",
       "value": "[loss: 1.882, acc: 0.469, val_loss: 3.266, val_acc: 0.248] : 100% 85/85 [01:53&lt;00:00,  2.57s/it]"
      }
     },
     "376726004f1643bbae4e6fdb9973a53c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "37845704705e4ec989edefe7a9e6235c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "37ccad185acd484c8b3e4b23fbeccc45": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "37e6d69ca0e94100bc0e5e66f930d6d2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "38ce6dfdfc364c7f92247bd92cf4029b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "38eea512596940f98bac99dfbf1aef98": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3901005cced24bc3969aecb9466f2b08": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9dec345b18ef4ab082d85d0f0201421f",
       "style": "IPY_MODEL_724520e857f94cfbb3fe82a2775e48f0",
       "value": "[loss: 1.070, acc: 0.679, val_loss: 2.072, val_acc: 0.498] : 100% 85/85 [01:56&lt;00:00,  2.68s/it]"
      }
     },
     "39781fd633b84db3ba5e9faed6be77ab": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 2",
       "layout": "IPY_MODEL_acb8944532fc4b6c95f8414c8c731d1c",
       "max": 85,
       "style": "IPY_MODEL_da89342fc13647ee9b08b0194e482d7f",
       "value": 85
      }
     },
     "39c66abd2d2847d8a0a9317d097ade69": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3a1b588b8dc04872acaa2d08502b167b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 15",
       "layout": "IPY_MODEL_d9a30dc6492e40cabd661d82e12c301d",
       "max": 85,
       "style": "IPY_MODEL_846c614cb6274595a670d80198b1bae6",
       "value": 85
      }
     },
     "3ac1ed727cc74cf8b27f0a584dad4032": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3b38a4d2564648048c0acbb322f015dc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3b5ecba495784eeca2015eda65563050": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3c37a33d69cb4789a26f08b97631b049": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3d04177c02c3451aa7f90f5dc74b6d97": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_521ad608969542b8b36827e141bb5c3e",
       "style": "IPY_MODEL_da4ac34f62ca452d8a3ef44b19d3c640",
       "value": "[loss: 1.307, acc: 0.622, val_loss: 2.036, val_acc: 0.457] : 100% 85/85 [02:13&lt;00:00,  2.76s/it]"
      }
     },
     "3d68e337e3e240969b3dd8d39e029467": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3e18217e917247829be26f09d70f022e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 4",
       "layout": "IPY_MODEL_2837aef1c6604b378ce38c685ef003d0",
       "max": 85,
       "style": "IPY_MODEL_88c9de3b1b6d46bea1d81cfefd18a21b",
       "value": 85
      }
     },
     "3e3996cdfc8a44e29f8f77ac5a531827": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "3e6ce02d6ad5446e8ed1aca039693b43": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "3ed60716942a4dc7b3095afbd6e030ce": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_9c20d2a4561a47969a26a61142b818dd",
        "IPY_MODEL_712121f8ad0d4542853c9ca2e8e4e3f5"
       ],
       "layout": "IPY_MODEL_ad88a0db2222488c8503734081478050"
      }
     },
     "3f39db9cab05401ab50bd257d748bf26": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_2f25cf81555e4437949af49823c0ba3b",
       "style": "IPY_MODEL_af6498c1f0d74078accce51565e25b5d",
       "value": "[loss: 1.838, acc: 0.482, val_loss: 2.267, val_acc: 0.378] : 100% 85/85 [01:50&lt;00:00,  2.56s/it]"
      }
     },
     "3f51b48deb494573acc4b766e232540f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d9c2da6e582f4fd19693a5191e235894",
       "style": "IPY_MODEL_d79e3b51f0ec44a0a1ee5834ea323e76",
       "value": "[loss: 1.490, acc: 0.571, val_loss: 2.166, val_acc: 0.423] : 100% 85/85 [02:00&lt;00:00,  2.80s/it]"
      }
     },
     "40223f0bdeac42478e12e3b61ba341eb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4136f2c7f0a64186863a3f183e867b85": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 41",
       "layout": "IPY_MODEL_dea3400b828c4f5c8293d63f7704cef2",
       "max": 85,
       "style": "IPY_MODEL_bc222e61e58a48dbaee38eaff9b1221a",
       "value": 85
      }
     },
     "4159bdf5c8374524b901c88152b146a0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "42a88d6e829e4548a40af39f9a13dd3d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f820fd8959a448daab955ec5a4c4747f",
       "style": "IPY_MODEL_59597b575e2545e2a33fcf9bef421a20",
       "value": "[loss: 2.818, acc: 0.237, val_loss: 2.626, val_acc: 0.257] : 100% 85/85 [02:18&lt;00:00,  3.35s/it]"
      }
     },
     "433ee56c69574b61b43d0afd8fc3a78b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "434ccf83cb9d431bb34abe27e40732c5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4442652502ba4431a46ce00b5d61bd62": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4488f15363f842178c44ced6d3a4b3dd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "449de225df3b4808be724f310f8a768e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ce1a42512695420ca49c6157969f01ac",
       "style": "IPY_MODEL_101c2ea80f9e4417a66afc1d9e69fc5f",
       "value": "[loss: 1.643, acc: 0.529, val_loss: 2.103, val_acc: 0.444] : 100% 85/85 [01:53&lt;00:00,  2.69s/it]"
      }
     },
     "450ffd39fe224b709ea6f60ed4a24264": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d681bacacd2543449f7b8bba4d414ba9",
       "style": "IPY_MODEL_d8a47921a5c6454d936fcf00e3b6ae7c",
       "value": "[loss: 2.774, acc: 0.236, val_loss: 2.523, val_acc: 0.307] : 100% 85/85 [01:56&lt;00:00,  2.51s/it]"
      }
     },
     "45891173831a4f1b9eba67c729d756ad": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "45b70f938bf34910a4d0499bd7bcd475": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "463ec6cd465a40e3a172cca8446e29e0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "469dbbd16c784166af2508373567312a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 7",
       "layout": "IPY_MODEL_ffd19d0be9f5436fb15f37c7f9d9da3c",
       "max": 85,
       "style": "IPY_MODEL_4862220e8d5f40a6bdfce6c43ae1bc66",
       "value": 85
      }
     },
     "46b13ef1db9645dfaf80764b262d2cb0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "472b7ff8c8144862971c3f112b9c2f92": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 5",
       "layout": "IPY_MODEL_55ae201cef74498c9b7d7c9405c14b2f",
       "max": 85,
       "style": "IPY_MODEL_2d8edc78ba3a4ff4b12142adcfe71c49",
       "value": 85
      }
     },
     "483395e120f147da94a5ed2823ff6089": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "485a28856d414777a452cb4fe50d22a7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4862220e8d5f40a6bdfce6c43ae1bc66": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "48aff79f998349c0afe9b60ac069096d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "48f14094bcdc433cb6937da8d0672a6e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 24",
       "layout": "IPY_MODEL_e187f9069322452893b57cdab4b939b8",
       "max": 85,
       "style": "IPY_MODEL_aa1e8dc7318d40c1bc3150d8c370ddfd",
       "value": 85
      }
     },
     "4a287d6f96fa4a869b064c1a92185237": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "4a4196e24b2748c9bf3b32346881d5c0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_40223f0bdeac42478e12e3b61ba341eb",
       "style": "IPY_MODEL_093a5c76774743da977a3472163eb242",
       "value": "[loss: 1.377, acc: 0.605, val_loss: 1.905, val_acc: 0.488] : 100% 85/85 [02:14&lt;00:00,  2.98s/it]"
      }
     },
     "4aad389cbd394ea4873b21d08d5cc1ef": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4aba1af8286c410b851e3d27737de2d4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "4acf6aa962de462e9577acfefa002098": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "4ad68a468edd4b45b2ad315bc62839db": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4af30232d25343e0bd525e7be01b56e2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4afbb7cf76b5466da71bb009e59583dd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_2e07edcd82fd40f6b34f1860a0ac6ada",
       "style": "IPY_MODEL_eab8d0540d634fe5b6b9bdf4ee8b8bf8",
       "value": "[loss: 1.166, acc: 0.658, val_loss: 2.307, val_acc: 0.456] : 100% 85/85 [01:46&lt;00:00,  2.40s/it]"
      }
     },
     "4b8cdee2a80042d09d3e14dac347a0eb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 0",
       "layout": "IPY_MODEL_ad894b46d9914423b0e79a187e55427b",
       "max": 85,
       "style": "IPY_MODEL_cd11262ca4fd413e862e7c26fd7789f0",
       "value": 85
      }
     },
     "4bf9fecce6df4465bf0ca272d7058cc8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f0448e9c6a694b75a0bf35e1be50f863",
       "style": "IPY_MODEL_77916ed7457f4259b9c58491061899e2",
       "value": "[loss: 1.554, acc: 0.557, val_loss: 2.308, val_acc: 0.386] : 100% 85/85 [01:54&lt;00:00,  2.48s/it]"
      }
     },
     "4d09d528dd034862bf586b69890ae924": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4dcdb47cf40246fdb82801055262214f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "4e7387666b424843a6b65f5d0e7b1795": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 11",
       "layout": "IPY_MODEL_b8860178d832433499b757847a89c778",
       "max": 85,
       "style": "IPY_MODEL_79db8c04ba004a99a524fb4655a62905",
       "value": 85
      }
     },
     "4f041daf765747c0ba9879c33d12f720": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4fc9f51a28b34758a404f10bd70a9623": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 46",
       "layout": "IPY_MODEL_ae0ed06d4a334afea5491db38b1c635d",
       "max": 85,
       "style": "IPY_MODEL_7fdf7437631a43ec93625b2112e1b6e7",
       "value": 85
      }
     },
     "4fe6c80d14ae4bd18f7845406738b961": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5000550ac043475c9dcba8972743f92b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "5034b1af30eb41178ca50215db697fc4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "506e0165cedc4146a7a5bc9d5a43f53b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 20",
       "layout": "IPY_MODEL_1a085d86906444fbb0462458041babc6",
       "max": 85,
       "style": "IPY_MODEL_780fe05dfdb74cb4a7de16e6a09cb761",
       "value": 85
      }
     },
     "50c62bff1e934374a2323a108d2e92b3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "515e8199957b481093a3f9781566251f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "51ccea7370c9424f8b82e4ce4e6aed0a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "521ad608969542b8b36827e141bb5c3e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "52298f588d3e47d3a5ac2aed04cc804b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 15",
       "layout": "IPY_MODEL_b07b9c0907a04e87b27fa5011dd1c56f",
       "max": 85,
       "style": "IPY_MODEL_aae2d5f30dfb494f9c3e972aa691f95b",
       "value": 85
      }
     },
     "53295a2488a9490d82cbb872e2a8ca72": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "537294c0752a49df8108199c47482eba": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "53b231e8a88f4ba8895d7942fc39ff09": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "54abc13b93cb4de0aaf9839b0cea0a91": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "54b93ad3ad6f499d83b940a34ef2aeb2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "54d08a2cdde04a8b911876d400180abc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "55ae201cef74498c9b7d7c9405c14b2f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "562b854f9eff45cba7aea9915cc75071": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "56b9d47f581d446e914d19d7710c8db5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 18",
       "layout": "IPY_MODEL_d3935d8067314d50b85e63b57a4799c7",
       "max": 85,
       "style": "IPY_MODEL_4a287d6f96fa4a869b064c1a92185237",
       "value": 85
      }
     },
     "573f8fe9ae1b4ce092069fb95944e70a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "577c2288cf3147fcadef623c9d77a27e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_5f2619ee3f854344988b0726a74e7fcd",
        "IPY_MODEL_9a0d8a3399af430d8612f4de70c320a5"
       ],
       "layout": "IPY_MODEL_bc0dcb26e85a448cb0b5c3767eb3fab3"
      }
     },
     "5807b70c780e46c2ae78131fc021c8b5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "587153d9d6fe429b936fc32d7df169ed": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "58ee01d54d7841c0903a2be070c0ac59": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "592f4b47bd124e17885104c4ece69c20": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b12d33111f684cf89406f229dd5c1153",
       "style": "IPY_MODEL_37e6d69ca0e94100bc0e5e66f930d6d2",
       "value": "[loss: 1.594, acc: 0.545, val_loss: 1.990, val_acc: 0.466] : 100% 85/85 [01:51&lt;00:00,  2.70s/it]"
      }
     },
     "59597b575e2545e2a33fcf9bef421a20": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "59d031151d0a4396bdf92fef852e7e03": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "59d83c4e333343238c1edb5d43930d1f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 17",
       "layout": "IPY_MODEL_2c1322f056454666babdadbefe1ea1dd",
       "max": 85,
       "style": "IPY_MODEL_330d8ef406964b40a4948bec54093218",
       "value": 85
      }
     },
     "5a057e6461fb47d99c5329e3163dd603": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5a3c1f8fd3024d2487ad7140a7306839": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_906aedcf5d2f4fe89d3a625d65919046",
       "style": "IPY_MODEL_34f6a36359314424bae3854398dfdceb",
       "value": "[loss: 1.645, acc: 0.531, val_loss: 1.859, val_acc: 0.505] : 100% 85/85 [01:59&lt;00:00,  2.88s/it]"
      }
     },
     "5afac633dfda4592be850a2fe83659a5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 27",
       "layout": "IPY_MODEL_54abc13b93cb4de0aaf9839b0cea0a91",
       "max": 85,
       "style": "IPY_MODEL_f7601917c047467b8af4593ddd171193",
       "value": 85
      }
     },
     "5b396f1cc40d4ea0bfac451c8813ad07": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5c142e2fd58e4bd896a08a96016b1a65": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5c7b7c698a364d21bb5554efca9a04d0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "5d38bdf927124eb19e000d7dd0416343": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "5d7cd41cac5e4bc0bb558743b79819a4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5ebdaaa3e7984a4795312ffda0f58c32": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6fa0978a03664b52a196cce58ba12619",
       "style": "IPY_MODEL_2d0372c3380f4aa7b989b4b006b26c1c",
       "value": "[loss: 2.003, acc: 0.436, val_loss: 2.382, val_acc: 0.372] : 100% 85/85 [01:57&lt;00:00,  2.47s/it]"
      }
     },
     "5f2619ee3f854344988b0726a74e7fcd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 2",
       "layout": "IPY_MODEL_dfa162d2b6064418ab2ce925a10d0bd4",
       "max": 85,
       "style": "IPY_MODEL_63b18305ab6146bcaa2b754b849ae83c",
       "value": 77
      }
     },
     "5f483b392330439b99472cc0fb32d304": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d6d5b5ae3c1b457fb14ac1f37af28dc4",
       "style": "IPY_MODEL_8639106c35164df59ddd11727af4f5f8",
       "value": "[loss: 2.121, acc: 0.408, val_loss: 2.752, val_acc: 0.292] : 100% 85/85 [01:48&lt;00:00,  2.43s/it]"
      }
     },
     "5f6926d08d474bdab0570b27ff064883": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_5034b1af30eb41178ca50215db697fc4",
       "style": "IPY_MODEL_cbcb1122cd2d4dde93c8eb788c88cb4d",
       "value": "[loss: 1.965, acc: 0.449, val_loss: 2.364, val_acc: 0.355] : 100% 85/85 [01:50&lt;00:00,  2.48s/it]"
      }
     },
     "5fab8edb39d54d6bace1cbda6c670ebe": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6a67267aeb074595b9469531ebde0666",
       "style": "IPY_MODEL_a97541280228467cb23c5f33c925fa5d",
       "value": "[loss: 1.407, acc: 0.586, val_loss: 1.818, val_acc: 0.509] : 100% 85/85 [01:51&lt;00:00,  2.99s/it]"
      }
     },
     "5faf19ca4aa548d5837ba9167d6bf26e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6057bd886d0e44148df732eecfba40a0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "60ce6d2557e0454b959df748abeeed7a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 6",
       "layout": "IPY_MODEL_e54060c112b544dbb56b309c3f5092ac",
       "max": 85,
       "style": "IPY_MODEL_354dc0248a514645870d88648ec4960a",
       "value": 85
      }
     },
     "60e3eb8827b64dadadeed868ac4074f7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6112f174dbea43e583a958211ab66a70": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "61196b7013fd4481bd5b9561d01a5a00": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 9",
       "layout": "IPY_MODEL_d9542085aad64c81a1f237232cc5a4ff",
       "max": 85,
       "style": "IPY_MODEL_bf61daf2e8104ec88cb8736f31061ce7",
       "value": 85
      }
     },
     "6209497b35524a95b0b4b4ac94cb1089": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "620e5d3b614345698fe1538c403df163": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 19",
       "layout": "IPY_MODEL_081f7a7a0d79479fb91d7f30f376bf30",
       "max": 85,
       "style": "IPY_MODEL_9e1a9bd3d28145808edafc0b3c19e380",
       "value": 85
      }
     },
     "6284f2b1e3444bd4b89a6aef90dba857": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "631d78269c804d0d8c7b8955ab3712c4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6327045cb0a0418a96dfeadab11b9670": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "63b18305ab6146bcaa2b754b849ae83c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "63c191aba0cb46ac9f5d7c882f75ee98": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "63c3a672132e4847906df563406cdf17": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ee84c90990a940b794e4adc3beb584c6",
       "style": "IPY_MODEL_feee4a77a3994b8dbfca904d4802f57f",
       "value": "[loss: 3.738, acc: 0.114, val_loss: 2.984, val_acc: 0.198] : 100% 85/85 [01:51&lt;00:00,  2.54s/it]"
      }
     },
     "63c3d01da0944ea38221cd1717c1e7a5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 4",
       "layout": "IPY_MODEL_2544eb4106ba40858aaa9d2933de3600",
       "max": 85,
       "style": "IPY_MODEL_ee9a857b63e74b9f89f23c057c895a7b",
       "value": 85
      }
     },
     "63fa3e03ab4149939e176ffceeb4012d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 13",
       "layout": "IPY_MODEL_c786eae4e6214c268a08bf50e16697a5",
       "max": 85,
       "style": "IPY_MODEL_0d35d592827c4518a3e3b33f703222d9",
       "value": 85
      }
     },
     "643cb20883754d70a7cfe6379ee23ac5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "64573e8651d749d2ae58201a06cb685d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "64fa206e5d5946d7af050ed37912b857": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "64fbc1eaea3c40089d2c988ef868805e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "650828a0b64f448c9e6747c144149f39": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 28",
       "layout": "IPY_MODEL_91583338ebb64b619397ac2c66f3ba98",
       "max": 85,
       "style": "IPY_MODEL_baf9ecfe67ad43c383650e7e1fbcd8ad",
       "value": 85
      }
     },
     "650dca87cb9b40d5aa8226621df1d5b4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_7985cd43d36448e79a66469239cfce3b",
       "style": "IPY_MODEL_ba0b67ba8f8046e289362af517c39a26",
       "value": "[loss: 1.699, acc: 0.514, val_loss: 2.335, val_acc: 0.410] : 100% 85/85 [01:51&lt;00:00,  2.48s/it]"
      }
     },
     "65b9a692297c47298d19a45a228dd959": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "66b340c4ef664ddfa809a835de70a1d9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "66cc339378af4562b25c017a51b7a582": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6704b377b472443f9e3ffa14766d746b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "6756e81290124dd88729c920ab24310f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9955336bd6214b0a912da38c408d3cba",
       "style": "IPY_MODEL_0bc012afd4de449a9854450587fc7677",
       "value": "[loss: 1.949, acc: 0.445, val_loss: 2.277, val_acc: 0.376] : 100% 85/85 [01:52&lt;00:00,  2.78s/it]"
      }
     },
     "685fbd04f1f24835a4e26109411274c7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "687062aa42134ed7acc0ca13b4d14f2b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_d6fa4614096d4890ab6230c807b4a7e2",
        "IPY_MODEL_b7ce956e049f49c5ac976ea31a2135e6"
       ],
       "layout": "IPY_MODEL_ebbaf96ffb75415cb2f0a2d4401e4ded"
      }
     },
     "68c60e31136b44f6b62d14b245ae3473": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "69a6c5efd2ca4eb0bbdb183eba09ee06": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_a30afe36a1a748528c12b56ac0459705",
       "style": "IPY_MODEL_166def0781bb45b5bc816d4caf6961be",
       "value": "[loss: 2.223, acc: 0.377, val_loss: 2.434, val_acc: 0.327] : 100% 85/85 [01:53&lt;00:00,  2.57s/it]"
      }
     },
     "6a43ae03e02547259a15cbf1ee4de052": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6a50c4670844494f8167b5a6ccbad4d3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6a51fd2ebc9b43c0b85dc95c7acc8b7b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 44",
       "layout": "IPY_MODEL_f40a8886c31f4af69fe5a0d745d43027",
       "max": 85,
       "style": "IPY_MODEL_18da9f438a8d48b8a90ef3c8173122fe",
       "value": 85
      }
     },
     "6a67267aeb074595b9469531ebde0666": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6a6875ced2e247aab4fb5e3977b24351": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_5b396f1cc40d4ea0bfac451c8813ad07",
       "style": "IPY_MODEL_cbdd797d5b604d738414bd79e848b0dc",
       "value": "[loss: 1.812, acc: 0.487, val_loss: 2.086, val_acc: 0.440] : 100% 85/85 [01:59&lt;00:00,  2.74s/it]"
      }
     },
     "6aec95823f8944899f9e9525767d9673": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_870ae49b4d3d413399b6a043fc4b5f59",
       "style": "IPY_MODEL_37ccad185acd484c8b3e4b23fbeccc45",
       "value": "[loss: 2.071, acc: 0.418, val_loss: 2.427, val_acc: 0.346] : 100% 85/85 [01:59&lt;00:00,  2.56s/it]"
      }
     },
     "6b4a53e8e07446359a36624a46c02406": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6b718152cb004ee491fbe42f354addf7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6ba0a48e9f1d43bba26619372575198c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6c105c28b48d412583da758d24ad6a39": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6ceab32829b146cbb6df801e9a0f543b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "6cf1755221d84533b5d72251dbb6f829": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6d6aaaed05904ebfa19330fea815b0b0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6dd869d503984a779311d139a9f884d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_e44d871349c2472498dd41632949d3ed",
       "style": "IPY_MODEL_3e3996cdfc8a44e29f8f77ac5a531827",
       "value": "[loss: 1.510, acc: 0.565, val_loss: 2.163, val_acc: 0.401] : 100% 85/85 [01:51&lt;00:00,  2.43s/it]"
      }
     },
     "6e43c8ce38684c4da2d481a015229988": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6eab6540d3574ab296fd98cb07afe626": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6edb0139aa434b4d8f261ece34ed2a0c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6ee124d902f04551b5945570da1a5d9b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 34",
       "layout": "IPY_MODEL_8fff6c0aac5045b081acd4070612cb11",
       "max": 85,
       "style": "IPY_MODEL_0c168351b4034db1b116b8fa03e83fd4",
       "value": 85
      }
     },
     "6ee83c215e6e4624bbd00d2acaa10c40": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6ef8e373d0d0497daf9858194bfeade5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "6f82fe6e8e89417a877f129f09bc05bf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6f8d8b3101d04faf951e48a6a83e2b60": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6fa034bd34104dc6b9e8eb7ab63cf8d9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "6fa0978a03664b52a196cce58ba12619": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6fb1bdd055354a3a9e55c7808dcf792e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 23",
       "layout": "IPY_MODEL_fdaee046070544a8966c43b24b97e06c",
       "max": 85,
       "style": "IPY_MODEL_5807b70c780e46c2ae78131fc021c8b5",
       "value": 85
      }
     },
     "70702d11ca644f82a818eef533145a8b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "70834e1c2d6241cfa5293d0d734ee5e6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 39",
       "layout": "IPY_MODEL_21f76b5379374028ba99fe1b95ae09b5",
       "max": 85,
       "style": "IPY_MODEL_ead9767ac987426bb31a08e34de4ecd4",
       "value": 85
      }
     },
     "70fac8d5e24744c59852e234534e8761": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_05dcabf7ab424b33ba15f1743f705856",
       "style": "IPY_MODEL_4aba1af8286c410b851e3d27737de2d4",
       "value": "[loss: 1.897, acc: 0.466, val_loss: 2.439, val_acc: 0.366] : 100% 85/85 [01:51&lt;00:00,  2.45s/it]"
      }
     },
     "711eb1e6b7654ad58876271ce29107b7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 35",
       "layout": "IPY_MODEL_2fae7445816e49959070c7fcd2cb8c1d",
       "max": 85,
       "style": "IPY_MODEL_fadfdb30ed5b4614a2f8310dbc9e9185",
       "value": 85
      }
     },
     "712121f8ad0d4542853c9ca2e8e4e3f5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1382f9048fcf49baa0b54b733ef089f5",
       "style": "IPY_MODEL_1004bb6a5c4a44d680c6810be9e75796",
       "value": " 10% 2/20 [03:45&lt;34:00, 113.34s/it]"
      }
     },
     "712a8f792b4649068039c559b5220762": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "71c9b6e995a2471cbbc7a965aba395a6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "71d0341fc2f54467b97eb7c954a0d6ce": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "724520e857f94cfbb3fe82a2775e48f0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "729f13b634c24796a75140aefe0342ae": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "72f6459f9e9341d78f3e1121d2fb0110": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_7d2b27f3d14d4d01b6548584f384050b",
       "style": "IPY_MODEL_8369c6bbad924a878842c04bf705bf02",
       "value": "[loss: 2.032, acc: 0.431, val_loss: 2.609, val_acc: 0.325] : 100% 85/85 [01:52&lt;00:00,  2.57s/it]"
      }
     },
     "7347595b43cc4a54a18c198c20604e12": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7374aa1738e347cb98198ae5764bcae9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 20",
       "layout": "IPY_MODEL_020dac6bb7704cbd873b264fa7c467a3",
       "max": 85,
       "style": "IPY_MODEL_1e4e6ca2892b44ad9b4126859f15a52f",
       "value": 85
      }
     },
     "737cf668c75b4dc2ae594af96fc54809": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "73dd6d820ce14b7ea778664e0fc95391": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "73f344cf14ea492f91c9fa81a695b9f6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f1b17c6145534c62ae017384a781124b",
       "style": "IPY_MODEL_f3bdd90d947f4688a2695e0ba3f1fa8f",
       "value": "[loss: 1.339, acc: 0.608, val_loss: 1.982, val_acc: 0.488] : 100% 85/85 [01:51&lt;00:00,  2.43s/it]"
      }
     },
     "749cbe4604154c29a5e2837196f54e0c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "752388bdd5bf469daefdba24cc9d9bb5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "752bdee4a8b74d138b513dadf499dc6c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "775fec72989d47a981e5c7316fc152cc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "77685d46d83b41939c1a2c44b28a568b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 47",
       "layout": "IPY_MODEL_31db592b35a548aea720835632ce8518",
       "max": 85,
       "style": "IPY_MODEL_30aff3681131437da217efd6ec742935",
       "value": 85
      }
     },
     "77916ed7457f4259b9c58491061899e2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "77b9807388704af297ff85e630071f19": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "77f12b709f124e468d01da15a6cc97ca": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 26",
       "layout": "IPY_MODEL_6edb0139aa434b4d8f261ece34ed2a0c",
       "max": 85,
       "style": "IPY_MODEL_38ce6dfdfc364c7f92247bd92cf4029b",
       "value": 85
      }
     },
     "780fe05dfdb74cb4a7de16e6a09cb761": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "794eb8d9edd8452a8b2ada05e3a30a3c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7985cd43d36448e79a66469239cfce3b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "79db8c04ba004a99a524fb4655a62905": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7a59c5d1ceb84d76b11db18d587b446e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 1",
       "layout": "IPY_MODEL_71c9b6e995a2471cbbc7a965aba395a6",
       "max": 85,
       "style": "IPY_MODEL_95db7a5890c04b96b2cc556806e4f25e",
       "value": 85
      }
     },
     "7a8276f14ec94814b9218d597c171e98": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7a898a60c5fd4d0e881125b959470ce2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7ad79baed83043159f84d00338834a99": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7b2eec3a1e684be29b451fb87566ff6e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7b57d487b5f444b4ac135a6b0ad42ebc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7c570f4745a94bc89db731641d84e664": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 24",
       "layout": "IPY_MODEL_4442652502ba4431a46ce00b5d61bd62",
       "max": 85,
       "style": "IPY_MODEL_95d9cad8204b42bd9f42a2c6ccdcebfb",
       "value": 85
      }
     },
     "7c7809581d924e589a09ee51c32e676e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 25",
       "layout": "IPY_MODEL_b92b879ba88440d4a9afa8c1333f28ef",
       "max": 85,
       "style": "IPY_MODEL_fb4c5764163a4ee1b62c0673bce04f6e",
       "value": 85
      }
     },
     "7c8f52fa8bf9489d952d7bb358b6b810": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 7",
       "layout": "IPY_MODEL_b928cede28424bf2b402f97c8ad7065e",
       "max": 85,
       "style": "IPY_MODEL_d676c51b2f004f98bdb2bcde1ff03646",
       "value": 85
      }
     },
     "7d2b27f3d14d4d01b6548584f384050b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7d43e5efd2b74a24970f2237263e12bc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_463ec6cd465a40e3a172cca8446e29e0",
       "style": "IPY_MODEL_8a8fe85a58134920b1ebe8c591a0b3ae",
       "value": "[loss: 1.604, acc: 0.545, val_loss: 2.163, val_acc: 0.414] : 100% 85/85 [01:58&lt;00:00,  2.57s/it]"
      }
     },
     "7d84347a638f4042a222582169f5b5ae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_25c86704fef54a04bfa14b481a2a152b",
       "style": "IPY_MODEL_e6b95cad31c547f390b85d140a789281",
       "value": "[loss: 1.355, acc: 0.606, val_loss: 2.386, val_acc: 0.410] : 100% 85/85 [01:58&lt;00:00,  2.56s/it]"
      }
     },
     "7dc0b85008484816832e833c9137c215": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 1",
       "layout": "IPY_MODEL_89fade78dfee42b8b98755031013be43",
       "max": 85,
       "style": "IPY_MODEL_cd0329d0256048789754a0b2ec1f504c",
       "value": 85
      }
     },
     "7eac89bba5914a23bf47f9fab800d210": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7ee8ae314ce746009a03918f74e470be": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7fdf7437631a43ec93625b2112e1b6e7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "8001f240a9c94765a0b6afbbd7f1ad06": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "808ffc7d5bea4161803accd3e0eb4f8c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8137a4d075f34e77ac426966e18edfb4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8179fa55087c49c19775f9ab05affde3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b616289f4bfe4d179829956d9cf48c6b",
       "style": "IPY_MODEL_0e1f500ab9734e02bfdfe1f6ff7f2cf5",
       "value": "100% 20/20 [38:32&lt;00:00, 112.64s/it]"
      }
     },
     "81c9bd1aa12e4d9f9666d3b34e782c4b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_38eea512596940f98bac99dfbf1aef98",
       "style": "IPY_MODEL_8599eda388a545fcafe2dd46ca6e0afb",
       "value": "[loss: 1.180, acc: 0.647, val_loss: 1.809, val_acc: 0.525] : 100% 85/85 [01:46&lt;00:00,  2.45s/it]"
      }
     },
     "81ec66f81cbc4345b8242eb43e9007aa": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "81fc962a7b4e4e828377b17f17eb9cb4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8254a2fc39ab499fa7cedd39f4db37f5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f2bf06c0592f45cbb23ce7abe1962b98",
       "style": "IPY_MODEL_d0482b5778a547b29a7289bef5a9c6bb",
       "value": "[loss: 2.617, acc: 0.273, val_loss: 2.577, val_acc: 0.299] : 100% 85/85 [01:58&lt;00:00,  2.59s/it]"
      }
     },
     "827eb4b6f0564676ad28591634b0a6f0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "82ea71c50de049ae864fe4303a6a4194": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 10",
       "layout": "IPY_MODEL_f88c34a6970c459ba405caf515b6ec55",
       "max": 85,
       "style": "IPY_MODEL_39c66abd2d2847d8a0a9317d097ade69",
       "value": 85
      }
     },
     "8326101180614b44bedcd0cef14ca93a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6c105c28b48d412583da758d24ad6a39",
       "style": "IPY_MODEL_1986c776469e42d8828841191c5febce",
       "value": "[loss: 1.262, acc: 0.630, val_loss: 2.451, val_acc: 0.411] : 100% 85/85 [01:47&lt;00:00,  2.40s/it]"
      }
     },
     "8369c6bbad924a878842c04bf705bf02": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8378e5aaa84b410aafea07f0b0317030": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_bde685064a564a1684a74f5e5e2183f7",
       "style": "IPY_MODEL_bbd4788f6154415d9b0d8112eb00bffd",
       "value": "[loss: 1.299, acc: 0.618, val_loss: 1.971, val_acc: 0.471] : 100% 85/85 [01:48&lt;00:00,  2.43s/it]"
      }
     },
     "846c614cb6274595a670d80198b1bae6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "84b2330ad3734e3d89df26c105b312a3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8558658817e143c497a856e458891338": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "85703f52f4ca43ca8d64856c73641a8a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8599eda388a545fcafe2dd46ca6e0afb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "860a5171622147ba898e076c97880cdf": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8639106c35164df59ddd11727af4f5f8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "870ae49b4d3d413399b6a043fc4b5f59": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "875a438e3e2544dd96cdea992fb34981": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "876958ccbc6344a58c58428da5a4c4ce": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8787ed10c68842ba92a072525aceba57": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "87a2da2d23654d7e9907e14b158eeb47": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "87be5f2977cd4538855e81a4f66a13d4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8824718c10654b349f2bc181116e0f7c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 22",
       "layout": "IPY_MODEL_a7ce829554dc48e69f259a322ecfac46",
       "max": 85,
       "style": "IPY_MODEL_8df3f830b6ba4ec4b246fc343c419351",
       "value": 85
      }
     },
     "88c9de3b1b6d46bea1d81cfefd18a21b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "88f84c0ba8fd44bf9be86fa56081310d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 16",
       "layout": "IPY_MODEL_860a5171622147ba898e076c97880cdf",
       "max": 85,
       "style": "IPY_MODEL_b18e6dad22d049b3ab095734e2e058ef",
       "value": 85
      }
     },
     "8978bee8462f4edfb32f543bbdb58f05": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_e51ec7c4a04d4cc8ae7a7290baeadf4d",
       "style": "IPY_MODEL_b8a672762e00416c890fc9e1051838fc",
       "value": "[loss: 1.709, acc: 0.513, val_loss: 2.100, val_acc: 0.443] : 100% 85/85 [01:48&lt;00:00,  2.43s/it]"
      }
     },
     "89e3654d3a504878a4ef2e5463b682a8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "89f0bc3276cc441aaae598ec33f911b8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_53b231e8a88f4ba8895d7942fc39ff09",
       "style": "IPY_MODEL_6eab6540d3574ab296fd98cb07afe626",
       "value": "[loss: 1.668, acc: 0.531, val_loss: 2.114, val_acc: 0.434] : 100% 85/85 [01:56&lt;00:00,  2.62s/it]"
      }
     },
     "89fade78dfee42b8b98755031013be43": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8a8fe85a58134920b1ebe8c591a0b3ae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8aa0c543c0fe4ee58f9ef125af8256ff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 3",
       "layout": "IPY_MODEL_4488f15363f842178c44ced6d3a4b3dd",
       "max": 85,
       "style": "IPY_MODEL_64fbc1eaea3c40089d2c988ef868805e",
       "value": 85
      }
     },
     "8aba82fb98a64e7e9a9d7ea9f3e85720": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 1",
       "layout": "IPY_MODEL_e4f44044a92e4ec3a4ea785a0373e86e",
       "max": 85,
       "style": "IPY_MODEL_e16d843b6b404d76acf8dc93e67be479",
       "value": 85
      }
     },
     "8ae99a9a6aa04c43b6e5a3afc9ddcce1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 38",
       "layout": "IPY_MODEL_07a20e4ca88d46038d5480aba752cd3d",
       "max": 85,
       "style": "IPY_MODEL_b8cad97b09ba4870a1fb53e912a765f7",
       "value": 85
      }
     },
     "8afd95922b384c5da2c91ed1a64f65cd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8b0cf9fffe984691ac075ede50eabf2f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8b1d12de3dde42c4a73177da1212d87a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "8b24bdf6e082428e9c97d4e945d46f18": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8cc98690cfaa45cf90c4aa1ea136a136": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "8d05db74332946d7af76facaed2591a7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ebd3e879673a49399cd79aab3b11fb85",
       "style": "IPY_MODEL_01058314c19f45a990221fcfc68153bf",
       "value": "[loss: 1.378, acc: 0.602, val_loss: 1.818, val_acc: 0.521] : 100% 85/85 [01:53&lt;00:00,  2.56s/it]"
      }
     },
     "8d17f482d9794e70aa0b240aa491b3b2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8df3f830b6ba4ec4b246fc343c419351": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "8e3a04b15a8543a5a6d25a84a8f4dcdd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8f04d42986414617b02cb6b757b68da4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 10",
       "layout": "IPY_MODEL_a4bbe3717ae44387b73742ec9a12d301",
       "max": 85,
       "style": "IPY_MODEL_794eb8d9edd8452a8b2ada05e3a30a3c",
       "value": 85
      }
     },
     "8f90cff6ce1e4239a85be39d7694f64c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8fd4f5e9880348efb2012d2e057a0fbd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_73dd6d820ce14b7ea778664e0fc95391",
       "style": "IPY_MODEL_6112f174dbea43e583a958211ab66a70",
       "value": "[loss: 2.067, acc: 0.418, val_loss: 2.368, val_acc: 0.360] : 100% 85/85 [01:53&lt;00:00,  2.82s/it]"
      }
     },
     "8fff6c0aac5045b081acd4070612cb11": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9059cbd2b7874bc795aaf144d1badff5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "906aedcf5d2f4fe89d3a625d65919046": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "91583338ebb64b619397ac2c66f3ba98": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "92144d8705af48758073f05661f84cbc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b628752baea2452bbf567eeceaa5241c",
       "style": "IPY_MODEL_c2a5266b1bba407f9181640ed83dc8dc",
       "value": "[loss: 1.083, acc: 0.672, val_loss: 2.148, val_acc: 0.456] : 100% 85/85 [01:54&lt;00:00,  2.45s/it]"
      }
     },
     "922f4afcd5fa43c4bec6ab9b65ffe5a4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "939c172793dd4f70bc740d90b1f19783": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "942f14fc8d234facb1c3659dc68275f4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "945f271796f74b26b416103d59e76c9d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "94a16b741c5e454b8d13638f46d9a258": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "94ef27d63f5d4b26816a5fb68397b77f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "95d9cad8204b42bd9f42a2c6ccdcebfb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "95db7a5890c04b96b2cc556806e4f25e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "96ae502106064281bb64bc11a1be21b1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "97294977c4d743bcb0b6791bbaedcb35": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 43",
       "layout": "IPY_MODEL_01f2e1d7011c4a9cad4b08453c53b091",
       "max": 85,
       "style": "IPY_MODEL_fe78406c8ef340469ca8ac6bdb7df9be",
       "value": 85
      }
     },
     "979002c2c38e4f4895cafadb4d32f7f6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "97cba60ffa394555bf4a25d80885add8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "98694d0ade6848f3b619a0f09ec14d43": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "98d8bfc3f04140b98a87c960b9fc2ad0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "98e5e5c9a49a49e0bb7ddb1ef98db98a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_e629b44691e54f5da9621b30c92e4734",
       "style": "IPY_MODEL_ff00122e6577464ea070c3e04f1c6345",
       "value": "[loss: 1.802, acc: 0.490, val_loss: 2.035, val_acc: 0.443] : 100% 85/85 [01:50&lt;00:00,  2.51s/it]"
      }
     },
     "9942705756304acea1882c2e55b58b72": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "9955336bd6214b0a912da38c408d3cba": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "99a424be0b2c4127bb0144c196880400": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 8",
       "layout": "IPY_MODEL_e97321dfaea346278b7adb26eaa67d7c",
       "max": 85,
       "style": "IPY_MODEL_0375cf926e6846e18b4a254e73b55c6a",
       "value": 85
      }
     },
     "9a0212feb81c43e0b8b825490cc7ab56": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_979002c2c38e4f4895cafadb4d32f7f6",
       "style": "IPY_MODEL_a1e93fd5eb3d4fbbb625b342878f96c3",
       "value": "[loss: 1.296, acc: 0.624, val_loss: 1.864, val_acc: 0.513] : 100% 85/85 [02:02&lt;00:00,  2.59s/it]"
      }
     },
     "9a0d8a3399af430d8612f4de70c320a5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ff15a00d0fe3487b94634622f6b8b490",
       "style": "IPY_MODEL_15ab12cd2778464ca2b75facbc0fa757",
       "value": "[loss: 2.821, acc: 0.231] :  91% 77/85 [01:40&lt;00:10,  1.31s/it]"
      }
     },
     "9a3bfbab951643a9af723c450a6ba2aa": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9a3ce92185b042e1943b6119e5fd0e9a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_efffcac62d764a08b8c3b3a1cdc2fa8d",
       "style": "IPY_MODEL_8137a4d075f34e77ac426966e18edfb4",
       "value": "[loss: 1.897, acc: 0.468, val_loss: 2.234, val_acc: 0.382] : 100% 85/85 [01:53&lt;00:00,  2.63s/it]"
      }
     },
     "9a436d8f2da14a39a18e591087952a2f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "9a5d5ebcf4ba48d8b4f3bacce8a6e69a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_945f271796f74b26b416103d59e76c9d",
       "style": "IPY_MODEL_be9451c7b7d6449d954da2dcac092896",
       "value": "[loss: 1.916, acc: 0.457, val_loss: 2.155, val_acc: 0.399] : 100% 85/85 [01:49&lt;00:00,  2.48s/it]"
      }
     },
     "9aa199215c1646dd821424a00852cf46": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 3",
       "layout": "IPY_MODEL_f38204caf4644cb5ba22e1759d0f2b2e",
       "max": 85,
       "style": "IPY_MODEL_c990b78ba8304f5ca2c2ee96d30e8be0",
       "value": 85
      }
     },
     "9afa188e9542468981f0e15f8883b27a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "9c20d2a4561a47969a26a61142b818dd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Training",
       "layout": "IPY_MODEL_cce9416c9f3546b38c9bdd15aa8a108b",
       "max": 20,
       "style": "IPY_MODEL_64fa206e5d5946d7af050ed37912b857",
       "value": 2
      }
     },
     "9c2ef98135944c76a2dc5f28b38eecb6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "9c641bdc096546e9bff5ca94ce66df90": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9cd0d2dadd684b54b79ae470414afe92": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9dec345b18ef4ab082d85d0f0201421f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9e069cff21934284aab88437d2f4ff05": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 18",
       "layout": "IPY_MODEL_eb65964b54e94fefbd561469b2b78a75",
       "max": 85,
       "style": "IPY_MODEL_f64f388ea8db49878daf25f247a08712",
       "value": 85
      }
     },
     "9e1a9bd3d28145808edafc0b3c19e380": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "9ed2186d63c04dddb202d08b64d2acb9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "9f71294df925417486f2a84bb8834cd7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9fae117d3c1342a9a13807bd78c14d33": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "a0470e8f29514391a234b92758390ab3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "a11bf9b59fba4a19843c8ed89053abc1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a15fea0fec9d4dc29a01c512e41a1108": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_0d9bd92e5d934f50bd150c1317a1d156",
       "style": "IPY_MODEL_7a8276f14ec94814b9218d597c171e98",
       "value": "[loss: 1.380, acc: 0.597, val_loss: 1.957, val_acc: 0.466] : 100% 85/85 [02:01&lt;00:00,  2.73s/it]"
      }
     },
     "a1e93fd5eb3d4fbbb625b342878f96c3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a298b729742944f0b3a52258d702907a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 48",
       "layout": "IPY_MODEL_51ccea7370c9424f8b82e4ce4e6aed0a",
       "max": 85,
       "style": "IPY_MODEL_6ceab32829b146cbb6df801e9a0f543b",
       "value": 85
      }
     },
     "a2e4dbfc92ea482fbe4d6dc9ae087418": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 4",
       "layout": "IPY_MODEL_5c142e2fd58e4bd896a08a96016b1a65",
       "max": 85,
       "style": "IPY_MODEL_9afa188e9542468981f0e15f8883b27a",
       "value": 85
      }
     },
     "a30afe36a1a748528c12b56ac0459705": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a349de7d2b5442b497bfbd5408223605": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a355cc0454ef44828a77b8f02b12d003": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_24b442e79844455aa95f686b4dc6aea3",
       "style": "IPY_MODEL_369c0ba8aef54ccabe944e6fe102e29a",
       "value": "[loss: 3.051, acc: 0.189, val_loss: 3.040, val_acc: 0.146] : 100% 85/85 [01:51&lt;00:00,  2.44s/it]"
      }
     },
     "a3b6bae29a784fb494743e2c8c454426": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a4412b2c5d41423fbefdb8359e7017d6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "a4bbe3717ae44387b73742ec9a12d301": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a578c42ddce347059e29c3415a510527": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f102dd6bb1bf44668f25e00092ba16a0",
       "style": "IPY_MODEL_b62ce5e701c74095b25cf3e7d346e802",
       "value": "[loss: 2.476, acc: 0.307, val_loss: 2.560, val_acc: 0.286] : 100% 85/85 [02:03&lt;00:00,  2.85s/it]"
      }
     },
     "a667710e40d34beb889986a19492e7ca": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d714cb23b87f48a48b2f0dd4ef7e0b89",
       "style": "IPY_MODEL_7ee8ae314ce746009a03918f74e470be",
       "value": "[loss: 1.468, acc: 0.574, val_loss: 2.987, val_acc: 0.330] : 100% 85/85 [02:03&lt;00:00,  3.07s/it]"
      }
     },
     "a6ad3e637f1d4c8c9c0e9597e1e061bd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9a3bfbab951643a9af723c450a6ba2aa",
       "style": "IPY_MODEL_18deeafab2ed4ead972ca9403fd0973f",
       "value": "[loss: 1.482, acc: 0.576, val_loss: 1.876, val_acc: 0.504] : 100% 85/85 [01:51&lt;00:00,  2.60s/it]"
      }
     },
     "a70d935a850c40c080546b074755c2dd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a7959f78e5d44b28a8dc5218a132ccc4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a7ce829554dc48e69f259a322ecfac46": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a7d17f4c6710453480909fc3829cce6e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_dec72468e8cf4ce5b6b63ae6bd359bae",
       "style": "IPY_MODEL_c7331230b366488aa551fa77ba05da34",
       "value": "[loss: 2.822, acc: 0.232, val_loss: 2.574, val_acc: 0.310] : 100% 85/85 [01:59&lt;00:00,  2.59s/it]"
      }
     },
     "a8558b5601ec447c87aea1e88817c5b2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b30b460096a242ec98daa8b7400bd0c2",
       "style": "IPY_MODEL_29bbcb5f28504c1aa6e8850b1313ac25",
       "value": "[loss: 2.173, acc: 0.391, val_loss: 2.878, val_acc: 0.264] : 100% 85/85 [01:49&lt;00:00,  2.51s/it]"
      }
     },
     "a8730e3e1f7c452dbfdcb08af0d4df75": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a8dcea4ad6b94f50be7262b7702c00f2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a94ffed5981a4f01bd3df5866753d600": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a97541280228467cb23c5f33c925fa5d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a9b6ca109a3f474a8725cd40e8810687": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "aa1e8dc7318d40c1bc3150d8c370ddfd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "aa31e30d78094ae397b1a5bf4b384576": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_df3127abdfd54edcaad5d587a8a9bd3d",
        "IPY_MODEL_8179fa55087c49c19775f9ab05affde3"
       ],
       "layout": "IPY_MODEL_94ef27d63f5d4b26816a5fb68397b77f"
      }
     },
     "aa6a83d9dd4740498db881ab56606dde": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "aae2d5f30dfb494f9c3e972aa691f95b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ab2baf35f41f4c9eb95c30015dc33503": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ab465563c62b4ab6b864224ad0d89b59": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ab80a5f4d94c4538a960737deca72c6c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_643cb20883754d70a7cfe6379ee23ac5",
       "style": "IPY_MODEL_ca77b799145d450d9970584fe64a4cff",
       "value": "[loss: 1.987, acc: 0.437, val_loss: 2.138, val_acc: 0.427] : 100% 85/85 [01:52&lt;00:00,  2.52s/it]"
      }
     },
     "ab855c15ce4042d79b8e72603e090019": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ac4eeab703a24da9985c3231287d2d3d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ac4f5e121c654dd19875b6831a7ef137": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_a8730e3e1f7c452dbfdcb08af0d4df75",
       "style": "IPY_MODEL_0ba56b6ec995403da8bf54c3335ee76b",
       "value": "[loss: 2.324, acc: 0.351, val_loss: 2.485, val_acc: 0.313] : 100% 85/85 [01:51&lt;00:00,  2.55s/it]"
      }
     },
     "ac67eb37964b4380964ad6e045c99c1a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ac7f130cff2d4666a2010b2d7f49ffd8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "acb8944532fc4b6c95f8414c8c731d1c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "acf66e6f87214b2abc6a9242ebb0b7b4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ad88a0db2222488c8503734081478050": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ad894b46d9914423b0e79a187e55427b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "adcbbd12baa9440797acb944af021bc3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ae0ed06d4a334afea5491db38b1c635d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "af6498c1f0d74078accce51565e25b5d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "afa68b608c834f38bf0c2a9bd339db4e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b07b9c0907a04e87b27fa5011dd1c56f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b0bb73624ae446d4a143a6a474a6a820": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b12d33111f684cf89406f229dd5c1153": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b160084b3f44424a838238045be44946": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b18e6dad22d049b3ab095734e2e058ef": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b1ae8bc50c6f41b89c29625ed968dc61": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b20f940c156c4b2b838aeaff36596b57": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 45",
       "layout": "IPY_MODEL_099a3c309e6b40c1ae6bf33211fd1ee3",
       "max": 85,
       "style": "IPY_MODEL_1381cd3cf6a44af99293dbc35434c40a",
       "value": 85
      }
     },
     "b24e403096394c7c88eb1983615cfbd3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 33",
       "layout": "IPY_MODEL_8afd95922b384c5da2c91ed1a64f65cd",
       "max": 85,
       "style": "IPY_MODEL_02bca7998ce049be82b68685cbcf490e",
       "value": 85
      }
     },
     "b30b460096a242ec98daa8b7400bd0c2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b3ae939c16d44aad80b82b47c388713f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b4d5b56d5d3a4c19a23c58a57aea6eba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1bc339bb051b45bab2a4483f0160fb9c",
       "style": "IPY_MODEL_f0af0eff6e8a42b596eb98be16952aab",
       "value": "[loss: 2.377, acc: 0.335, val_loss: 2.414, val_acc: 0.327] : 100% 85/85 [01:53&lt;00:00,  2.58s/it]"
      }
     },
     "b514b0b0730e4bcb877ca61259e1c0bb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b524448d68754c9db797af798b5a635f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b547bd4b843647ee980f65c8e34bcc59": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 14",
       "layout": "IPY_MODEL_109b2d7f44944418814c69cd79c3b2dc",
       "max": 85,
       "style": "IPY_MODEL_ab465563c62b4ab6b864224ad0d89b59",
       "value": 85
      }
     },
     "b582cb6b41e247528a91b617c16e3399": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b5ce9e25114c4ea1849942f788ff920d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b5d1b1a5797c4a32a6ff8c23280f0d20": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b615ee544db24f49bbecd81619db68f3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b616289f4bfe4d179829956d9cf48c6b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b628752baea2452bbf567eeceaa5241c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b62ce5e701c74095b25cf3e7d346e802": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b6536b9b4b4c4ff0bc3f74b343064149": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b699cd44230e4dfcb7c995a28d35abfe": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b702bdb67d1647058052fa48f830947d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b70fd210a564497d9eea0110dbd6a293": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 16",
       "layout": "IPY_MODEL_515e8199957b481093a3f9781566251f",
       "max": 85,
       "style": "IPY_MODEL_8b1d12de3dde42c4a73177da1212d87a",
       "value": 85
      }
     },
     "b72cc27b12744bff91ebf7220d0b3a41": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b75a39e157e04acc8c91423cc6a2940c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b77c12468a554904bc156dd4426bc1d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_573f8fe9ae1b4ce092069fb95944e70a",
       "style": "IPY_MODEL_68c60e31136b44f6b62d14b245ae3473",
       "value": "[loss: 1.628, acc: 0.534, val_loss: 2.165, val_acc: 0.442] : 100% 85/85 [01:52&lt;00:00,  2.50s/it]"
      }
     },
     "b7ce956e049f49c5ac976ea31a2135e6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b699cd44230e4dfcb7c995a28d35abfe",
       "style": "IPY_MODEL_201cd0b0a04441eab9479e5d8cffe163",
       "value": "100% 30/30 [56:50&lt;00:00, 112.54s/it]"
      }
     },
     "b82714d7e3734a008a160380d897df6d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b87369ec40864a0ba67d1b5314fd52b1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b880a21683494fc9884f223036729659": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_c3f4799855ac47a2b83e21fa1f27017d",
       "style": "IPY_MODEL_dc3d81b608b344fe96e09826e5c435d5",
       "value": "[loss: 1.055, acc: 0.680, val_loss: 2.249, val_acc: 0.468] : 100% 85/85 [02:04&lt;00:00,  2.70s/it]"
      }
     },
     "b8860178d832433499b757847a89c778": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b8a672762e00416c890fc9e1051838fc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b8cad97b09ba4870a1fb53e912a765f7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b8ee0b1873ef4eb695c523e5eebe9ae5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 0",
       "layout": "IPY_MODEL_b3ae939c16d44aad80b82b47c388713f",
       "max": 85,
       "style": "IPY_MODEL_4dcdb47cf40246fdb82801055262214f",
       "value": 85
      }
     },
     "b928cede28424bf2b402f97c8ad7065e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b92b879ba88440d4a9afa8c1333f28ef": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b99c38c7b8474500914cd9ac270fafb7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 37",
       "layout": "IPY_MODEL_8787ed10c68842ba92a072525aceba57",
       "max": 85,
       "style": "IPY_MODEL_89e3654d3a504878a4ef2e5463b682a8",
       "value": 85
      }
     },
     "b9db9d3bf79148a4b2228252c56e0654": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 14",
       "layout": "IPY_MODEL_8f90cff6ce1e4239a85be39d7694f64c",
       "max": 85,
       "style": "IPY_MODEL_cc2439cf363e4b70b9b2a97673ca0c44",
       "value": 85
      }
     },
     "ba0b67ba8f8046e289362af517c39a26": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "baf7e3e540d243d5beccf856212a05f1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "baf9ecfe67ad43c383650e7e1fbcd8ad": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "bb17572b2bbf48f788cc1d1ca065085b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 12",
       "layout": "IPY_MODEL_014baea001864dcd96833c86901f2055",
       "max": 85,
       "style": "IPY_MODEL_2fc50b09d66645688a822e540411d697",
       "value": 85
      }
     },
     "bb1dd5e0ac7645e3af99cb967c31708e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_04fddd423e4042ba87f50c168f637391",
       "style": "IPY_MODEL_e7b45caffdac4843a567ddc383f2c01d",
       "value": "[loss: 1.708, acc: 0.513, val_loss: 2.128, val_acc: 0.420] : 100% 85/85 [01:53&lt;00:00,  2.75s/it]"
      }
     },
     "bb72cef886134ff79560f29dc8a58b52": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_34e0b8fe1d054f2fb89e73e61e8d1489",
       "style": "IPY_MODEL_876958ccbc6344a58c58428da5a4c4ce",
       "value": "[loss: 1.447, acc: 0.587, val_loss: 2.107, val_acc: 0.448] : 100% 85/85 [02:11&lt;00:00,  4.15s/it]"
      }
     },
     "bb8bea8a753046ee8c16032324ae25a3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_eb921d464de3465d9ab88cc93039257b",
       "style": "IPY_MODEL_14375834a7394a359e072ae3bb8a14f7",
       "value": "[loss: 1.794, acc: 0.490, val_loss: 2.442, val_acc: 0.363] : 100% 85/85 [01:49&lt;00:00,  2.49s/it]"
      }
     },
     "bbd4788f6154415d9b0d8112eb00bffd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "bbf9b16cdbb84631acf7522d0d3626d8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bc0dcb26e85a448cb0b5c3767eb3fab3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bc222e61e58a48dbaee38eaff9b1221a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "bcb6eea7901744efa3650c02d4079c4f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_bd268fdaaef14f0f9c83cb9266fe7727",
       "style": "IPY_MODEL_6057bd886d0e44148df732eecfba40a0",
       "value": "[loss: 2.660, acc: 0.276, val_loss: 2.960, val_acc: 0.178] : 100% 85/85 [02:05&lt;00:00,  2.57s/it]"
      }
     },
     "bd268fdaaef14f0f9c83cb9266fe7727": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bdaf77b79a0c4f5f8796bd76f8612203": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "bdde5bd908b2406aa6cfac0888570092": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bde4597537cc4cf9909f192d93015a27": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 5",
       "layout": "IPY_MODEL_fad94aed4236459687378d4d72dd7fc8",
       "max": 85,
       "style": "IPY_MODEL_040cb76f591640fe8be6a6186fda8860",
       "value": 85
      }
     },
     "bde685064a564a1684a74f5e5e2183f7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "be635cfb8cf6467f83545586a4bdc122": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "be6874ccd93a4bed8319d38087f8c1fa": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "be9451c7b7d6449d954da2dcac092896": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "bf61daf2e8104ec88cb8736f31061ce7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "bfdbf7f2f706457cbe4ec83bca7bc586": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c0839c2ee30148e98335a7b154516c7f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "c0e0be5e06074ec798b4ea503930f251": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c1bdfb29ae744563bc3312115a470d62": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6ba0a48e9f1d43bba26619372575198c",
       "style": "IPY_MODEL_08f255e2c3cb43ad941abd5ccf7e6560",
       "value": "[loss: 3.087, acc: 0.174, val_loss: 2.767, val_acc: 0.230] : 100% 85/85 [02:06&lt;00:00,  2.97s/it]"
      }
     },
     "c203995a1920483db634003a99537348": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c2181b9abc3c4e8f97e7c79fa156c451": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c2a5266b1bba407f9181640ed83dc8dc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c2fc55b2109a45a2a495541e8260efc1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c3f4799855ac47a2b83e21fa1f27017d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c4983ba08f8b4cb885e4228bcc770fcf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "c5334219d2e84d2b897e3c3107617207": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_c91e8b2e8e464137b5d9052b71325fc6",
        "IPY_MODEL_d1c7109522814511995feec4222aff10"
       ],
       "layout": "IPY_MODEL_cd277654cff5413cb0ba5ae388eed618"
      }
     },
     "c566d314b097496c91eebf73c3215d51": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c61ad3b1a92d4c0cb1f7751e8910d736": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c62b9c9fa15f4a5aa4e8de2848d80126": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 21",
       "layout": "IPY_MODEL_434ccf83cb9d431bb34abe27e40732c5",
       "max": 85,
       "style": "IPY_MODEL_7a898a60c5fd4d0e881125b959470ce2",
       "value": 85
      }
     },
     "c633bbf6e38d486bb7d2c5a7f11319d5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c68433ffb9c640bbbde1c2d34366a46c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "c707b050dda145cdbf5a94f5f76c1bd0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 8",
       "layout": "IPY_MODEL_a11bf9b59fba4a19843c8ed89053abc1",
       "max": 85,
       "style": "IPY_MODEL_3b5ecba495784eeca2015eda65563050",
       "value": 85
      }
     },
     "c7331230b366488aa551fa77ba05da34": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c786eae4e6214c268a08bf50e16697a5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c867b32c32a04955ac81d5591bd66fe6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1caea7d7749c4489b4dbd83ebd1454a6",
       "style": "IPY_MODEL_6284f2b1e3444bd4b89a6aef90dba857",
       "value": "[loss: 1.084, acc: 0.670, val_loss: 2.585, val_acc: 0.397] : 100% 85/85 [02:03&lt;00:00,  2.85s/it]"
      }
     },
     "c8ab81c8cb614143bfa35135e5822687": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 0",
       "layout": "IPY_MODEL_9c641bdc096546e9bff5ca94ce66df90",
       "max": 85,
       "style": "IPY_MODEL_35a3694837044cee8b55515ae12ed6ad",
       "value": 85
      }
     },
     "c8d21dd752c1452191c9644f2405ab24": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c8da7164ff67434abcca8b6344494837": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c8fc253cc05c4b8c9099dddb43cf7552": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_712a8f792b4649068039c559b5220762",
       "style": "IPY_MODEL_b72cc27b12744bff91ebf7220d0b3a41",
       "value": "[loss: 2.505, acc: 0.306, val_loss: 2.748, val_acc: 0.220] : 100% 85/85 [01:52&lt;00:00,  2.54s/it]"
      }
     },
     "c91e8b2e8e464137b5d9052b71325fc6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "Training",
       "layout": "IPY_MODEL_7b57d487b5f444b4ac135a6b0ad42ebc",
       "max": 50,
       "style": "IPY_MODEL_215042154bb74511916f28baef913b4a",
       "value": 50
      }
     },
     "c94c14c32abf49f0b355cb652dc571c9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c990b78ba8304f5ca2c2ee96d30e8be0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ca29e199721f49dba80db4a626f760c9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 13",
       "layout": "IPY_MODEL_b5ce9e25114c4ea1849942f788ff920d",
       "max": 85,
       "style": "IPY_MODEL_7347595b43cc4a54a18c198c20604e12",
       "value": 85
      }
     },
     "ca57cebe465449e3a78ec73350aa634f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ca70959c9c0a4b8081fd86b22daef6d5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_e4fda969d56d436fb418772e3a7ab951",
       "style": "IPY_MODEL_45891173831a4f1b9eba67c729d756ad",
       "value": "[loss: 1.453, acc: 0.580, val_loss: 1.877, val_acc: 0.489] : 100% 85/85 [01:49&lt;00:00,  2.41s/it]"
      }
     },
     "ca77b799145d450d9970584fe64a4cff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "cab548d5080f4e969cdc5813f39568a2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_be6874ccd93a4bed8319d38087f8c1fa",
       "style": "IPY_MODEL_58ee01d54d7841c0903a2be070c0ac59",
       "value": "[loss: 1.209, acc: 0.638, val_loss: 1.932, val_acc: 0.503] : 100% 85/85 [01:47&lt;00:00,  2.42s/it]"
      }
     },
     "cab647fd5f3846548baf5105c64e4cd5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_2b9a6176c1004526bd3152e544d3f596",
       "style": "IPY_MODEL_25880c90e1f14412b563b3f18194656f",
       "value": "[loss: 1.322, acc: 0.610, val_loss: 2.178, val_acc: 0.445] : 100% 85/85 [01:48&lt;00:00,  2.49s/it]"
      }
     },
     "cb8b30c6650d43868798697a0d0c0ed1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_bdde5bd908b2406aa6cfac0888570092",
       "style": "IPY_MODEL_a7959f78e5d44b28a8dc5218a132ccc4",
       "value": "[loss: 1.747, acc: 0.505, val_loss: 2.121, val_acc: 0.425] : 100% 85/85 [01:52&lt;00:00,  2.45s/it]"
      }
     },
     "cbaf0c7a1bfc4e79b7b89b4125dc22e8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ac67eb37964b4380964ad6e045c99c1a",
       "style": "IPY_MODEL_f9747ba20f1744d3845ff9d00dc83342",
       "value": "[loss: 1.253, acc: 0.629, val_loss: 2.054, val_acc: 0.467] : 100% 85/85 [01:47&lt;00:00,  2.41s/it]"
      }
     },
     "cbcb1122cd2d4dde93c8eb788c88cb4d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "cbdd797d5b604d738414bd79e848b0dc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "cbe01cec6d5e4d39a65ae20a8ba5eb9a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cc2439cf363e4b70b9b2a97673ca0c44": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "cc5b5c7329734282b70c04341de2b3c7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cc6f198c58584793bf2c1198e63b0626": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d2e7372addae417890604eb5302416c3",
       "style": "IPY_MODEL_96ae502106064281bb64bc11a1be21b1",
       "value": "[loss: 1.216, acc: 0.636, val_loss: 2.197, val_acc: 0.438] : 100% 85/85 [01:47&lt;00:00,  2.41s/it]"
      }
     },
     "cc755292ef8541a698a1a69f9c369ddd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cce9416c9f3546b38c9bdd15aa8a108b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cd0329d0256048789754a0b2ec1f504c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "cd11262ca4fd413e862e7c26fd7789f0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "cd277654cff5413cb0ba5ae388eed618": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ce1012850e904af9bb235b372c88c85b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ce1a42512695420ca49c6157969f01ac": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cf211d101e544349b9118cbffdec8d52": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "cf273bc748cb4fcfa48fc0821db03123": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 9",
       "layout": "IPY_MODEL_7ad79baed83043159f84d00338834a99",
       "max": 85,
       "style": "IPY_MODEL_12e6a775481643ea9b39d7ac6046a003",
       "value": 85
      }
     },
     "d0482b5778a547b29a7289bef5a9c6bb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d0513700d31a4ea3935acfe7261f0b70": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 7",
       "layout": "IPY_MODEL_81ec66f81cbc4345b8242eb43e9007aa",
       "max": 85,
       "style": "IPY_MODEL_ef6336997ad14c01af6bab9d9db8bf5b",
       "value": 85
      }
     },
     "d1305d2dd231404bafcbb25decd55ce2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d1c7109522814511995feec4222aff10": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_c633bbf6e38d486bb7d2c5a7f11319d5",
       "style": "IPY_MODEL_6b718152cb004ee491fbe42f354addf7",
       "value": "100% 50/50 [1:37:54&lt;00:00, 121.04s/it]"
      }
     },
     "d24717259ee34b1e965aa51e77ebb5be": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 2",
       "layout": "IPY_MODEL_d773a83a5ef548cc8a821106b286bc9d",
       "max": 85,
       "style": "IPY_MODEL_37845704705e4ec989edefe7a9e6235c",
       "value": 85
      }
     },
     "d2e7372addae417890604eb5302416c3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d3935d8067314d50b85e63b57a4799c7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d471706ce0f5484db89150eb520a458f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_fc9402b5173c452fa246602403c6a058",
       "style": "IPY_MODEL_f216bda7de4e4fec93f3963b00587f02",
       "value": "[loss: 2.095, acc: 0.406, val_loss: 2.381, val_acc: 0.349] : 100% 85/85 [01:54&lt;00:00,  2.50s/it]"
      }
     },
     "d4fdea74cd4b4a7a83ae9760fec7f30a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d64a828ddcf6457d86014f5ad0337a6e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d676c51b2f004f98bdb2bcde1ff03646": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "d681bacacd2543449f7b8bba4d414ba9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d6c33b96891b48d2a8a2cd0c5d261508": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 15",
       "layout": "IPY_MODEL_3c37a33d69cb4789a26f08b97631b049",
       "max": 85,
       "style": "IPY_MODEL_c0839c2ee30148e98335a7b154516c7f",
       "value": 85
      }
     },
     "d6c577ea8cae4a218b69b25605737b83": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d6d5b5ae3c1b457fb14ac1f37af28dc4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d6fa4614096d4890ab6230c807b4a7e2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "Training",
       "layout": "IPY_MODEL_05d9f4e591074fcda4003539666b3523",
       "max": 30,
       "style": "IPY_MODEL_b87369ec40864a0ba67d1b5314fd52b1",
       "value": 30
      }
     },
     "d711ba1d7cf2488a9b4318a500892fa8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "d714cb23b87f48a48b2f0dd4ef7e0b89": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d773a83a5ef548cc8a821106b286bc9d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d79e3b51f0ec44a0a1ee5834ea323e76": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d7fd1d85a6054306b78d030b4602d63a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 25",
       "layout": "IPY_MODEL_b75a39e157e04acc8c91423cc6a2940c",
       "max": 85,
       "style": "IPY_MODEL_6704b377b472443f9e3ffa14766d746b",
       "value": 85
      }
     },
     "d8a47921a5c6454d936fcf00e3b6ae7c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d8c498d975114687bfcf6f36dd3a9d26": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 5",
       "layout": "IPY_MODEL_ac4eeab703a24da9985c3231287d2d3d",
       "max": 85,
       "style": "IPY_MODEL_ed0f475c381e4c2eb24d364c76c74a77",
       "value": 85
      }
     },
     "d8eeeae335514b6bb97d5d3fee1035b2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 18",
       "layout": "IPY_MODEL_8b24bdf6e082428e9c97d4e945d46f18",
       "max": 85,
       "style": "IPY_MODEL_64573e8651d749d2ae58201a06cb685d",
       "value": 85
      }
     },
     "d9542085aad64c81a1f237232cc5a4ff": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d985ede72f464866bf97a63f5295d681": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d9a30dc6492e40cabd661d82e12c301d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d9c2da6e582f4fd19693a5191e235894": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d9e8eaec591042cf9ff44df1516243ba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 17",
       "layout": "IPY_MODEL_537294c0752a49df8108199c47482eba",
       "max": 85,
       "style": "IPY_MODEL_20f10d30e3014378a5cc1dcd2849be4a",
       "value": 85
      }
     },
     "da02c5c6cb3843279a6a2aa50409b73e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_8e3a04b15a8543a5a6d25a84a8f4dcdd",
       "style": "IPY_MODEL_d1305d2dd231404bafcbb25decd55ce2",
       "value": "[loss: 1.135, acc: 0.663, val_loss: 1.856, val_acc: 0.514] : 100% 85/85 [01:46&lt;00:00,  2.40s/it]"
      }
     },
     "da4ac34f62ca452d8a3ef44b19d3c640": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "da5c0485f2554b61acdc8bacc2854971": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "da89342fc13647ee9b08b0194e482d7f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "dac89d1ec735479daf870e887494b0da": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "dad746cf2db8428cb77f1bd0a82eef4a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "db719a59b0384fb58bae4e7da46c5ff5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "db7a62c2b4434f059274c5b1aaefc042": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dbc342ac64b244ab9286fbdac5c0a74c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dc3d81b608b344fe96e09826e5c435d5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "dccfb14e804d4c599b3b6faa827a3895": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dce04254c2df45f9a52d9a40024dff9d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d985ede72f464866bf97a63f5295d681",
       "style": "IPY_MODEL_aa6a83d9dd4740498db881ab56606dde",
       "value": "[loss: 3.725, acc: 0.117, val_loss: 2.912, val_acc: 0.183] : 100% 85/85 [01:54&lt;00:00,  2.65s/it]"
      }
     },
     "dd058d7b028d4c3ab9084c3dd28ad6e3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "dd273d753c00420eb3f8c5e646de02f3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ddfe77a031a14044addf4c1d268f9446": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_4ad68a468edd4b45b2ad315bc62839db",
       "style": "IPY_MODEL_c2181b9abc3c4e8f97e7c79fa156c451",
       "value": "[loss: 2.222, acc: 0.380, val_loss: 2.172, val_acc: 0.397] : 100% 85/85 [01:59&lt;00:00,  2.63s/it]"
      }
     },
     "de26fcd6a8f24d83b2712a8f462be318": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_53295a2488a9490d82cbb872e2a8ca72",
       "style": "IPY_MODEL_9ed2186d63c04dddb202d08b64d2acb9",
       "value": "[loss: 1.467, acc: 0.580, val_loss: 2.552, val_acc: 0.336] : 100% 85/85 [02:10&lt;00:00,  2.81s/it]"
      }
     },
     "dea3400b828c4f5c8293d63f7704cef2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dec72468e8cf4ce5b6b63ae6bd359bae": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "df3127abdfd54edcaad5d587a8a9bd3d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "Training",
       "layout": "IPY_MODEL_59d031151d0a4396bdf92fef852e7e03",
       "max": 20,
       "style": "IPY_MODEL_6209497b35524a95b0b4b4ac94cb1089",
       "value": 20
      }
     },
     "df352299a68a4ae1acf2b0769ffb1d0f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dfa162d2b6064418ab2ce925a10d0bd4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e16d843b6b404d76acf8dc93e67be479": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "e187f9069322452893b57cdab4b939b8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e1893c4a4fd54969b66c2af2f89c91af": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 1",
       "layout": "IPY_MODEL_e873b5b0662f48e4a838ab5cbf5a5bda",
       "max": 85,
       "style": "IPY_MODEL_c68433ffb9c640bbbde1c2d34366a46c",
       "value": 85
      }
     },
     "e1a5750e7ab54a819b2f515df8c17a3f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e3dd4bca40d444c19e987e81d147437b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e44d871349c2472498dd41632949d3ed": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e4f44044a92e4ec3a4ea785a0373e86e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e4fda969d56d436fb418772e3a7ab951": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e51ec7c4a04d4cc8ae7a7290baeadf4d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e54060c112b544dbb56b309c3f5092ac": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e5d1fd31934b4f8cbbd080d4192ba17c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e5e3dd221d4345f2b1897719ff273542": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_271b46b0d4764329af26c15c49cce609",
       "style": "IPY_MODEL_737cf668c75b4dc2ae594af96fc54809",
       "value": "[loss: 1.228, acc: 0.633, val_loss: 2.133, val_acc: 0.479] : 100% 85/85 [01:47&lt;00:00,  2.41s/it]"
      }
     },
     "e5f19b026b5847ca9cc41727dc25cbe4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6d6aaaed05904ebfa19330fea815b0b0",
       "style": "IPY_MODEL_fc4f76aa7edb4db796b9e2e4749af377",
       "value": "[loss: 1.148, acc: 0.651, val_loss: 2.048, val_acc: 0.472] : 100% 85/85 [01:48&lt;00:00,  2.41s/it]"
      }
     },
     "e629b44691e54f5da9621b30c92e4734": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e630dd696e8d4be484395540073fb6d5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e6b95cad31c547f390b85d140a789281": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e6c84c2e6b104537bdc4e9cac319f23c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e7ab0947421841efa0327a00a5e846bd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_922f4afcd5fa43c4bec6ab9b65ffe5a4",
       "style": "IPY_MODEL_685fbd04f1f24835a4e26109411274c7",
       "value": "[loss: 3.741, acc: 0.117, val_loss: 3.053, val_acc: 0.185] : 100% 85/85 [02:13&lt;00:00,  2.97s/it]"
      }
     },
     "e7b45caffdac4843a567ddc383f2c01d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e873b5b0662f48e4a838ab5cbf5a5bda": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e97321dfaea346278b7adb26eaa67d7c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e9b70841b448414da94dffc1d28a9ef7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ea0ac4c5d4d5449f8231d9aaabdd875c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "ea241ecb1f4e4275ad4d50d0d6cbdd30": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "ea260bf295c94f45b48bb2033dc8defb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 0",
       "layout": "IPY_MODEL_0a34f42d4b574677aa1f7fa9545b5c37",
       "max": 85,
       "style": "IPY_MODEL_3e6ce02d6ad5446e8ed1aca039693b43",
       "value": 85
      }
     },
     "ea2695708a844cb189e7b365c001f31d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "eaafbd107aa04b61bccb80ca610e0041": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_05672ba006b641cf99bb5d6147bf2bec",
       "style": "IPY_MODEL_33f70b3f24654ad0b6c284e04ebd5de3",
       "value": "[loss: 1.764, acc: 0.495, val_loss: 2.378, val_acc: 0.370] : 100% 85/85 [01:51&lt;00:00,  2.55s/it]"
      }
     },
     "eab8d0540d634fe5b6b9bdf4ee8b8bf8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "ead9767ac987426bb31a08e34de4ecd4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "eb1ee01678f44b22bce962ee9a011e95": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 28",
       "layout": "IPY_MODEL_f03720db172145889462c000397981e0",
       "max": 85,
       "style": "IPY_MODEL_9fae117d3c1342a9a13807bd78c14d33",
       "value": 85
      }
     },
     "eb65964b54e94fefbd561469b2b78a75": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "eb921d464de3465d9ab88cc93039257b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ebbaf96ffb75415cb2f0a2d4401e4ded": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ebd3e879673a49399cd79aab3b11fb85": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ed0f475c381e4c2eb24d364c76c74a77": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ed687afe3df840d597d94f01ae8de371": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ee5a60f89c3245e389fe4a4f49159356": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ee84c90990a940b794e4adc3beb584c6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ee9a857b63e74b9f89f23c057c895a7b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "ef275875c1c6450fa0b36aa3f22a1c84": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ef3c774c08cb4496810ad148d822ceb4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_827eb4b6f0564676ad28591634b0a6f0",
       "style": "IPY_MODEL_a349de7d2b5442b497bfbd5408223605",
       "value": "[loss: 1.467, acc: 0.574, val_loss: 1.958, val_acc: 0.479] : 100% 85/85 [01:59&lt;00:00,  2.48s/it]"
      }
     },
     "ef4f00a7fb4f4895a4f837827eb38766": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_87be5f2977cd4538855e81a4f66a13d4",
       "style": "IPY_MODEL_587153d9d6fe429b936fc32d7df169ed",
       "value": "[loss: 1.269, acc: 0.629, val_loss: 1.893, val_acc: 0.501] : 100% 85/85 [01:47&lt;00:00,  2.45s/it]"
      }
     },
     "ef6336997ad14c01af6bab9d9db8bf5b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "efb06571a19e493ebf4f8d26d162cce0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 23",
       "layout": "IPY_MODEL_6a50c4670844494f8167b5a6ccbad4d3",
       "max": 85,
       "style": "IPY_MODEL_c4983ba08f8b4cb885e4228bcc770fcf",
       "value": 85
      }
     },
     "efffcac62d764a08b8c3b3a1cdc2fa8d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f03720db172145889462c000397981e0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f0448e9c6a694b75a0bf35e1be50f863": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f07aada8d81a4c77bcbe47649dca7645": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ed687afe3df840d597d94f01ae8de371",
       "style": "IPY_MODEL_dac89d1ec735479daf870e887494b0da",
       "value": "[loss: 1.525, acc: 0.559, val_loss: 2.090, val_acc: 0.452] : 100% 85/85 [01:57&lt;00:00,  2.45s/it]"
      }
     },
     "f0af0eff6e8a42b596eb98be16952aab": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f0bbc9547cf94dabb54ba6c7b1a198e5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_45b70f938bf34910a4d0499bd7bcd475",
       "style": "IPY_MODEL_26059d8662834539897f0cf13c99451f",
       "value": "[loss: 1.727, acc: 0.509, val_loss: 2.358, val_acc: 0.385] : 100% 85/85 [01:50&lt;00:00,  2.59s/it]"
      }
     },
     "f102dd6bb1bf44668f25e00092ba16a0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f1b17c6145534c62ae017384a781124b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f1fcdecdd62c4d3d84f99f4fb3df560f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 12",
       "layout": "IPY_MODEL_5faf19ca4aa548d5837ba9167d6bf26e",
       "max": 85,
       "style": "IPY_MODEL_cf211d101e544349b9118cbffdec8d52",
       "value": 85
      }
     },
     "f216bda7de4e4fec93f3963b00587f02": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f2a9edbc6c8043a1865ac7d31c6e47f9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 14",
       "layout": "IPY_MODEL_c0e0be5e06074ec798b4ea503930f251",
       "max": 85,
       "style": "IPY_MODEL_9a436d8f2da14a39a18e591087952a2f",
       "value": 85
      }
     },
     "f2bf06c0592f45cbb23ce7abe1962b98": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f2f7627313314476bea82690cbfa9107": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f38204caf4644cb5ba22e1759d0f2b2e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f399ff33d7a347ae8f219d22d5b3a3b2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f3b2eed3c328463e81af94d244751d41": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 9",
       "layout": "IPY_MODEL_729f13b634c24796a75140aefe0342ae",
       "max": 85,
       "style": "IPY_MODEL_dd058d7b028d4c3ab9084c3dd28ad6e3",
       "value": 85
      }
     },
     "f3bdd90d947f4688a2695e0ba3f1fa8f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f40a8886c31f4af69fe5a0d745d43027": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f497c86fed65400c934d6b6ccb4fdd32": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f4e0b9a72f034212a7a5d1239082db21": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 29",
       "layout": "IPY_MODEL_2b44ba79c9fe4c1fbaa56e7e5f4d5b2e",
       "max": 85,
       "style": "IPY_MODEL_4acf6aa962de462e9577acfefa002098",
       "value": 85
      }
     },
     "f57310a94ff94ae594246c0d5612422e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "f5ffaba77af24a7785a04a4233d7bb2a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f64f388ea8db49878daf25f247a08712": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "f6752a63939b44afb57745f3d6b48238": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_056c1fe2d4044395826bc8c38252f4e3",
       "style": "IPY_MODEL_1cd58a36a61b494f83dac8c1838c332b",
       "value": "[loss: 1.582, acc: 0.543, val_loss: 1.910, val_acc: 0.484] : 100% 85/85 [01:53&lt;00:00,  2.52s/it]"
      }
     },
     "f7484c326a4b467d8b306f622bb5028a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f7601917c047467b8af4593ddd171193": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "f7ac3bcd06104fc4a77f18201ae58148": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_3604c7c282e1429ba81bb148dd65e295",
       "style": "IPY_MODEL_9c2ef98135944c76a2dc5f28b38eecb6",
       "value": "[loss: 3.038, acc: 0.188, val_loss: 2.683, val_acc: 0.252] : 100% 85/85 [01:50&lt;00:00,  2.50s/it]"
      }
     },
     "f820fd8959a448daab955ec5a4c4747f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f88c34a6970c459ba405caf515b6ec55": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f892523a0ea74748a57c573fe05d6ca7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f8f7e3234c344d25bbf4f91a755f50f8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "f9747ba20f1744d3845ff9d00dc83342": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f981b2a4b06d48fb9271d29794fff7a7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 22",
       "layout": "IPY_MODEL_9cd0d2dadd684b54b79ae470414afe92",
       "max": 85,
       "style": "IPY_MODEL_acf66e6f87214b2abc6a9242ebb0b7b4",
       "value": 85
      }
     },
     "f9c59b7ae4a341fb95495ae25e2ce97a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "f9dcc3d86bd44c15a8b75ad169234de0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_54d08a2cdde04a8b911876d400180abc",
       "style": "IPY_MODEL_b702bdb67d1647058052fa48f830947d",
       "value": "[loss: 1.361, acc: 0.598, val_loss: 2.247, val_acc: 0.417] : 100% 85/85 [02:03&lt;00:00,  2.78s/it]"
      }
     },
     "fa5783b72dd04d5cad7e582b0c799c84": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fad94aed4236459687378d4d72dd7fc8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fadfdb30ed5b4614a2f8310dbc9e9185": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "fb17e85b01d9443ebfb0222389a0e910": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f497c86fed65400c934d6b6ccb4fdd32",
       "style": "IPY_MODEL_0eb0f14c032b457da307c5b344dc8a42",
       "value": "[loss: 1.106, acc: 0.671, val_loss: 1.858, val_acc: 0.533] : 100% 85/85 [02:07&lt;00:00,  2.85s/it]"
      }
     },
     "fb4c5764163a4ee1b62c0673bce04f6e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "fc4f76aa7edb4db796b9e2e4749af377": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "fc5c5d7f6a774f7c84e8b6f4b9bbbacb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 40",
       "layout": "IPY_MODEL_da5c0485f2554b61acdc8bacc2854971",
       "max": 85,
       "style": "IPY_MODEL_144228a2f653453495ade8526afcba6b",
       "value": 85
      }
     },
     "fc9402b5173c452fa246602403c6a058": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fd13ac6a0943491a8cb90552d3b37dd2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fd1c692010e942f19b4d02745330d79c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_942f14fc8d234facb1c3659dc68275f4",
       "style": "IPY_MODEL_5000550ac043475c9dcba8972743f92b",
       "value": "[loss: 2.330, acc: 0.347, val_loss: 2.730, val_acc: 0.285] : 100% 85/85 [01:58&lt;00:00,  2.92s/it]"
      }
     },
     "fd5186e4deab4de8880bdca9bdad47d8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 10",
       "layout": "IPY_MODEL_e1a5750e7ab54a819b2f515df8c17a3f",
       "max": 85,
       "style": "IPY_MODEL_f57310a94ff94ae594246c0d5612422e",
       "value": 85
      }
     },
     "fd7ae4b9d51844559ef9a8e7560be03e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 31",
       "layout": "IPY_MODEL_a3b6bae29a784fb494743e2c8c454426",
       "max": 85,
       "style": "IPY_MODEL_6327045cb0a0418a96dfeadab11b9670",
       "value": 85
      }
     },
     "fdaee046070544a8966c43b24b97e06c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fdd4e3efef904b7db7dcc721ddd75804": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "Epoch 49",
       "layout": "IPY_MODEL_2145b10afb3b40af9014a92afbe054c7",
       "max": 85,
       "style": "IPY_MODEL_b0bb73624ae446d4a143a6a474a6a820",
       "value": 85
      }
     },
     "fe78406c8ef340469ca8ac6bdb7df9be": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "feee4a77a3994b8dbfca904d4802f57f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "ff00122e6577464ea070c3e04f1c6345": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "ff15a00d0fe3487b94634622f6b8b490": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ffd19d0be9f5436fb15f37c7f9d9da3c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}