03_Detecting_retinopathy_with_GNG.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Growing Neural Gas to quantify diabetic retinopathy\n",
    "\n",
    "[Chris von Csefalvay](mailto:csefalvayk@starschema.net) | [Starschema](https://starschema.net) | 27JAN2019"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#About-Growing-Neural-Gas-algorithms\" data-toc-modified-id=\"About-Growing-Neural-Gas-algorithms-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>About Growing Neural Gas algorithms</a></span></li><li><span><a href=\"#The-GNG-algorithm\" data-toc-modified-id=\"The-GNG-algorithm-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>The GNG algorithm</a></span></li><li><span><a href=\"#Counting-hard-exudate-lesions-on-fundoscopy-photos-using-GNGs\" data-toc-modified-id=\"Counting-hard-exudate-lesions-on-fundoscopy-photos-using-GNGs-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Counting hard exudate lesions on fundoscopy photos using GNGs</a></span><ul class=\"toc-item\"><li><span><a href=\"#Loading-the-image-files-and-basic-configuration\" data-toc-modified-id=\"Loading-the-image-files-and-basic-configuration-3.1\"><span class=\"toc-item-num\">3.1&nbsp;&nbsp;</span>Loading the image files and basic configuration</a></span></li><li><span><a href=\"#Loading-the-mask-and-applying-consensus-levels\" data-toc-modified-id=\"Loading-the-mask-and-applying-consensus-levels-3.2\"><span class=\"toc-item-num\">3.2&nbsp;&nbsp;</span>Loading the mask and applying consensus levels</a></span></li><li><span><a href=\"#Filtering-ROI-for-hard-exudates-based-on-hue\" data-toc-modified-id=\"Filtering-ROI-for-hard-exudates-based-on-hue-3.3\"><span class=\"toc-item-num\">3.3&nbsp;&nbsp;</span>Filtering ROI for hard exudates based on hue</a></span></li><li><span><a href=\"#Point-conversion\" data-toc-modified-id=\"Point-conversion-3.4\"><span class=\"toc-item-num\">3.4&nbsp;&nbsp;</span>Point conversion</a></span></li><li><span><a href=\"#Apply-and-train-GNG\" data-toc-modified-id=\"Apply-and-train-GNG-3.5\"><span class=\"toc-item-num\">3.5&nbsp;&nbsp;</span>Apply and train GNG</a></span><ul class=\"toc-item\"><li><span><a href=\"#Create-helper-functions-for-analysis-of-resulting-GNG\" data-toc-modified-id=\"Create-helper-functions-for-analysis-of-resulting-GNG-3.5.1\"><span class=\"toc-item-num\">3.5.1&nbsp;&nbsp;</span>Create helper functions for analysis of resulting GNG</a></span></li><li><span><a href=\"#Running-and-training-GNG\" data-toc-modified-id=\"Running-and-training-GNG-3.5.2\"><span class=\"toc-item-num\">3.5.2&nbsp;&nbsp;</span>Running and training GNG</a></span></li></ul></li><li><span><a href=\"#Calculating-lesion-area\" data-toc-modified-id=\"Calculating-lesion-area-3.6\"><span class=\"toc-item-num\">3.6&nbsp;&nbsp;</span>Calculating lesion area</a></span></li></ul></li><li><span><a href=\"#Comparing-all-subjects'-lesion-areas\" data-toc-modified-id=\"Comparing-all-subjects'-lesion-areas-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Comparing all subjects' lesion areas</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## About Growing Neural Gas algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Growing Neural Gas (GNG) is a relatively simple algorithm that, just as the other competitive model we have encountered in the previous chapter, allows for learning and representing topology. It is, thus, a form of a **topology-representing network**, and as we have seen in [Part 1](01_Funderstanding_competitive_neural_networks.ipynb), it is capable of approximating something as complex as the topology of a human face — producing something quite recognisable to a human observer while reducing information significantly. The animation to the left, for instance, demonstrates that something as complex as the human face can be approximated with significant efficiency to create an image that would be unmistakable to an observer."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](assets/gng_principle.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The GNG algorithm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{definition}\\label{def:gng_algo_1} **The Growing Neural Gas algorithm (1) – the GNG graph**\n",
    "\n",
    "Let $\\mathcal{G}$ be a Growing Neural Gas with $\\delta$ inputs. Then, there exists a corrresponding graph $G = (V, E)$. \n",
    "\n",
    "At any time $t \\ge 0$, $V(t)$ is a set of nodes $V_{1 \\cdots m}$ and $E(t)$ a set of edges $E_{1 \\cdots n}$.\n",
    "\\end{definition}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that in the above, the term 'node' is synonymous with 'neuron'."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{definition}\\label{def:gng_algo_2} **The Growing Neural Gas algorithm (cont. 2) – the weight vector of $V_i$ at $t$**\n",
    "\n",
    "$V_i \\in V$ at time $t$ is also associated with a $\\delta$-element vector $\\mathbf{v}_{i} (t)$. This is sometimes referred to as the weight vector of $V_i$ at $t$, and represents the position of $V_i$ in $\\mathbb{R}^{\\delta}$.\n",
    "\n",
    "The distance between two nodes $V_i(t)$ and $V_j(t)$ is defined as\n",
    "\n",
    "\\begin{equation}\\label{eq:dist}\n",
    "\\Delta\\Big(V_i(t), V_j(t)\\Big) = \\sqrt{ \\sum_{k = 1}^{\\delta} (\\mathbf{v}_{i}(t)_k - \\mathbf{v}_{j}(t)_k)^2}\n",
    "\\end{equation}\n",
    "\n",
    "\\end{definition}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The use of 'weight vector' may be confusing. Unlike weight vectors in error correction networks, like feed-forward backpropagation neural nets, this weight wector does not have propagated weights. Rather, the 'weights' are simply a position vector that allows the algorithm to meaningfully approximate topologies by assigning a position in $\\mathbb{R}^{\\delta}$ space to each node (neuron)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{definition}\\label{def:gng_algo_3} **The Growing Neural Gas algorithm (cont. 3) – edges**\n",
    "\n",
    "The edge between $V_i$ and $V_j$ is denoted $E_{i, j}$. \n",
    "\n",
    "The magnitude (length) of the edge $E_{i, j}$, represented by $\\Vert E_{i, j} \\Vert$, is defined as the distance between $E_i(t)$ and $E_j(t)$. Using the formula in Equation \\ref{eq:dist},\n",
    "\n",
    "\\begin{equation}\\label{eq:edge_length}\n",
    "\\Vert E_{i, j} \\Vert = \\Delta\\Big(V_i(t), V_j(t)\\Big) = \\sqrt{ \\sum_{k = 1}^{\\delta} (\\mathbf{v}_{i}(t)_k - \\mathbf{v}_{j}(t)_k)^2}\n",
    "\\end{equation}\n",
    "\n",
    "Each edge has an age. $\\psi (E_{i, j})$ denotes the age of the edge $E_{i, j}$.\n",
    "\n",
    "The edge operator $\\mathbf{E}(V_i) \\mid V_i \\in V$ returns all edges emanating from $V_i$. Formally,\n",
    "\n",
    "\\begin{equation}\\label{eq:edge_operator}\n",
    "\\mathbf{E}(V_i) = \\{ E(V_i, x) \\cup E(x, V_i) \\mid x \\in V \\land x \\neq V_i \\}\n",
    "\\end{equation}\n",
    "\n",
    "The adjacency operator $\\mathbf{A}(V_i) \\mid V_i in V$ returns all nodes connected to $V_i$. Formally, \n",
    "\n",
    "\\begin{equation}\\label{eq:adjacency_operator}\n",
    "\\mathbf{A}(V_i) = \\{ x \\mid E(V_i, x) \\in V \\lor E(x, V_i) \\in V)\n",
    "\\end{equation}\n",
    "\n",
    "\\end{definition}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With that, let us consider the GNG algorithm itself.\\cite{fritzke1995growing}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\\begin{definition}\\label{def:gng_algo} **The Growing Neural Gas algorithm (cont. 4) - algorithm**\n",
    "\n",
    "Let $\\mathcal{G}$ be a Growing Neural Gas initialised with $v_{init}$ nodes, with $v_{init} \\ge 2$ and $\\mathbf{v}_{1}(0) \\cdots \\mathbf{v}_{init}(0)$ being each randomly initialised.\n",
    "\n",
    "Let further be defined a maximum age $\\psi_{max}$, as well as a primary learning rate $\\mu$ and a secondary learning rate $\\mu^\\prime$, with $\\mu \\ge \\mu^\\prime$, and an addition rate $\\theta$, which specifies the frequency (in iterations) of inserting new neurons. A maximum number of neurons $N_{max}$ may be defined.\n",
    "\n",
    "Let $D$ be an $\\mathcal{l} \\times \\delta$ matrix, with each row containing the coordinates of a data point in $\\mathbb{R}^{\\delta}$ space.\n",
    "\n",
    "\\begin{enumerate}\n",
    "    \\item For iteration $t$, select a row $\\mathbf{d}(t)$ from $D$.\n",
    "    \\item For each node $N \\in V$, calculate the error by reference to Equation \\ref{eq:dist}, i.e. $\\Delta(\\mathbf{d}(t), \\mathbf{n}(t))$. Add this to the cumulative error $\\Delta_{cum}(N)$, so that \n",
    "    \\begin{equation}\\label{eq:delta_cum}\n",
    "    \\Delta_{cum}(N) = \\sum_{k = 1}^{t} \\Delta(\\mathbf{d}(k), \\mathbf{n}(k))\n",
    "    \\end{equation}\n",
    "    \\item Determine the best-performing neuron (BPU) $N$, defined as $$ \\underset{N \\in V}{\\operatorname{arg\\ min}}\\ \\Delta \\Big( \\mathbf{d}, N \\Big) $$ where $\\Delta$ is the distance function laid out in Equation \\ref{eq:dist}.\n",
    "    \\item Determine the second-best performing neuron $N^{\\prime}$, defined as $$ \\underset{N^\\prime \\in V \\land N^\\prime \\neq N}{\\operatorname{arg\\ min}}\\ \\Delta \\Big( \\mathbf{d}, N^{\\prime} \\Big) $$ where, again, $\\Delta$ is the distance function laid out in Equation \\ref{eq:dist}.\n",
    "    \\item For all units in $\\mathbf{E}(V_N)$, wherre $\\mathbf{E}$ is the edge operator (see Equation \\ref{eq:edge_operator}) increment the age of each edge by one.\n",
    "    \\item Move $N$ towards $\\mathbf{d}$, so that $$\\Delta(\\mathbf{n}(t), \\mathbf{n}(t + 1)) = \\mu \\Delta(\\mathbf{d}(t), \\mathbf{n}(t))$$.\n",
    "    \\item Move all nodes in $\\mathbf{A}(N)$ towards $\\mathbf{d}$, so that for any $A_p \\in \\mathbf{A}(N)$, $$\\Delta(\\mathbf{a_p}(t), \\mathbf{a_p}(t + 1)) = \\mu^\\prime \\Delta(\\mathbf{d}(t), \\mathbf{a_p}(t)) \\mid A_p \\in \\mathbf{A}(N)$$\n",
    "    \\item If the edge $E_{N, N^\\prime}$ exists, set $\\psi(E_{N, N^\\prime})$ to $0$. If the edge does not exist, create it.\n",
    "    \\item For every edge $E_{i, j}$ in $E$, remove $E_{i, j}$ if $\\psi(E_{i, j}) > \\psi_{max}$.\n",
    "    \\item For every node $N \\in V$, if $\\mathbf{E}(N) = \\{\\}$, remove $N$.\n",
    "    \\item If $t \\mod \\theta = 0$:\n",
    "        \\begin{enumerate}\n",
    "            \\item Determine the overall worst-peforming neuron $W$, defined as $$ \\underset{W \\in V}{\\operatorname{arg\\ max}}\\ \\Delta_{cum}(W) $$ where $\\Delta_{cum}$ is the cumulative distance defined in Equation \\ref{eq:delta_cum}.\n",
    "            \\item Determine the worst-performing neighbour $W^\\prime$ of $W$, defined as $$ \\underset{W^\\prime \\in \\mathbf{A}(W)}{\\operatorname{arg\\ max}}\\ \\Delta_{cum}(W^\\prime) $$ where $\\Delta_{cum}$ is the cumulative distance defined in Equation \\ref{eq:delta_cum}.\n",
    "            \\item Insert a new unit $H$ between $W$ and $W^\\prime$, so that $$\\mathbf{h} = \\frac{1}{2} (\\mathbf{w} + \\mathbf{w}^\\prime) $$ \n",
    "        \\end{enumerate}\n",
    "    \\item Iterate until a given number of iterations is reached or, if defined, if $\\Vert V \\Vert = N_{max}$ is reached.\n",
    "\\end{enumerate}\n",
    "\n",
    "\\end{definition}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While this might appear convoluted, the algorithm is actually quite simple: every time a new data point is considered, the nearest point is moved towards it by a constant multiple of the distance and its connected neurons are pulled along, albeit by somewhat less. If this sounds familiar (from SOFMs), it should! What makes this neural gas model **growing** is the ability to insert points where the model performs worst. Unlike SOFMs, which have to make do with the grid they have been given, the growing neural gas can grow and adapt to its environment and like an actual gas, approximate and fill spaces. This is what we'll be leveraging in the following use case demonstration."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Counting hard exudate lesions on fundoscopy photos using GNGs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Diabetic retinopathy, the destruction of the retinal blood vessels due to inadequate blood glucose control, is one of the leading causes of blindness in the developed world.\\cite{bunce2006leading} In diabetic retinopathy, long term inadequate blood glucose control leads to vascular damage, microaneurysms and exudates, where lipids (causing bright yellow hard exudates) or blood (resulting in pale, diffuse yellow soft exudates) have accumulated on the fundus of the eye.\\cite{fong2004retinopathy} In the following, we’ll be using GNG to quantify these abnormalities, and specifically count the number and area of lesions from a public fundoscopy dataset.\\cite{kauppi2007diaretdb1}\n",
    "\n",
    "The [DIARETDB1 data set](http://www2.it.lut.fi/project/imageret/diaretdb1/index.html) by the research group of Kauppi et al. (_op.cit._) at Lappeenranta University of Technology contains 89 digital fundoscopy images, that is, images of the fundus of the eye, of five healthy volunteers and 84 people with some degree of diabetic retinopathy. 36 of the images and all of the control images are negative for the presence of hard exudates altogether. The DIARETDB1 data set contains ROI (Region of Interest) masks, but those merely outline areas that show a particular clinical feature. Can we use Growing Neural Gas to count how many clusters of hard exudates are present in the regions of interest? You bet!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "init_cell": true
   },
   "outputs": [],
   "source": [
    "from skimage import color, io, filters, img_as_float, exposure, morphology\n",
    "from scipy.misc import toimage\n",
    "from scipy.spatial import ConvexHull\n",
    "from scipy.stats import kde\n",
    "import copy\n",
    "import datetime\n",
    "import json\n",
    "import math\n",
    "import random\n",
    "import cv2\n",
    "from matplotlib import pyplot as plt\n",
    "from matplotlib import cm\n",
    "import numpy as np\n",
    "from neupy import algorithms, environment\n",
    "import pandas as pd\n",
    "from tqdm import tqdm_notebook as tqdm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading the image files and basic configuration\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this step, we're walking through the conversion and processing steps based on the data from a single subject. Subjects are identified by zero padded identifiers appended to images, e.g. the fundoscopy file `fundoscopies/image019.png` is the fundoscopy image of subject 019. We load the image as a float first:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "SUBJECT_ID = 14"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_image = img_as_float(io.imread(f\"assets/fundoscopies/image{SUBJECT_ID:03}.png\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7efc9235a3c8>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "io.imshow(raw_image)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading the mask and applying consensus levels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The masks, according to the specification,\\cite{kauppi2007diaretdb1} contain the expert annotations as greyscale images where each expert's annotation is layered over the others, so that for $q$ experts, the intensity of a pixel with $p$-expert consensus will be $\\frac{p}{q}$ (for greyscale values $[0, 1]$). The Kauppi _et al._ dataset has four levels of expert annotation,\\cite{kauppi2007diaretdb1} and we will store the number of agreeing experts in the constant `EXPERT_CONCURRENCE_REQUIRED`.\n",
    "\n",
    "\\begin{definition}\\label{def:expert_concurrence_masking} **Expert concurrence masking**\n",
    "\n",
    "Let the source image $S$ be a matrix $\\mathbb{R}^{m \\times n}$ so that $R_{i, j} \\in [0, 1] \\ \\forall \\ i \\in m \\ \\land \\ j \\in n$. \n",
    "\n",
    "Then the expert concurrence function for value ranges $[0, 1]$ is defined as the function $f : (\\mathbb{Z}^{+}, \\mathbb{Z}^{+}, \\mathbb{R}^{m \\times n}) \\to \\{ 0, 1 \\}^{m \\times n}$ so that\n",
    "\n",
    "$$ f(p, q, S)_{i, j} = \n",
    "    \\begin{cases}\n",
    "     1 & \\text{for } S_{i, j} \\geq \\frac{p}{q} \\\\\n",
    "     0 & \\text{for } S_{i, j} < \\frac{p}{q} \\\\\n",
    "    \\end{cases} \n",
    "    $$\n",
    "\n",
    "\\end{definition}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "EXPERT_CONCURRENCE_REQUIRED = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_to_binary_mask(mask, **kwargs):\n",
    "    expert_concurrence_required = kwargs.get(\"experts\", EXPERT_CONCURRENCE_REQUIRED)\n",
    "    binary_mask = mask > (expert_concurrence_required - 1) * 0.25\n",
    "    \n",
    "    structuring_element = kwargs.get(\"structuring_element\", morphology.disk)\n",
    "    structuring_element_size = kwargs.get(\"structuring_element_size\", 16)\n",
    "    \n",
    "    return morphology.opening(binary_mask, structuring_element(structuring_element_size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mask_over_image(binary_mask, raw_image):\n",
    "    return np.where(binary_mask[..., None], raw_image, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_mask_process(mask, raw_img, **kwargs):\n",
    "    fig, axarr = plt.subplots(1, 3)\n",
    "    fig.set_figheight(5)\n",
    "    fig.set_figwidth(15)\n",
    "    fig.subplots_adjust(hspace=0.5)\n",
    "    cmap = kwargs.get(\"cmap\", cm.plasma)\n",
    "    experts_required = kwargs.get(\"experts\", EXPERT_CONCURRENCE_REQUIRED)\n",
    "    fig.suptitle(f\"Expert consensus mask (subject {SUBJECT_ID}, {experts_required}-expert consensus)\")\n",
    "    \n",
    "    axarr[0].imshow(mask, cmap=cmap)\n",
    "    axarr[0].set_title(\"Raw expert image\")\n",
    "    \n",
    "    axarr[1].imshow(convert_to_binary_mask(mask, **kwargs), cmap=cmap)\n",
    "    axarr[1].set_title(f'Binary {experts_required}-expert voting mask')\n",
    "    \n",
    "    axarr[2].imshow(mask_over_image(convert_to_binary_mask(mask, **kwargs), raw_img))\n",
    "    axarr[2].set_title(\"Region of interest based on mask\")\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask = img_as_float(io.imread(f\"assets/hard_exudate_masks/image{SUBJECT_ID:03}.png\"))\n",
    "binary_mask = convert_to_binary_mask(mask)\n",
    "masked_rgb = mask_over_image(binary_mask, raw_image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7efc922e5c88>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_mask_process(mask, raw_image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filtering ROI for hard exudates based on hue"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we're exploiting the fact that fatty exudates have a very prominent bright yellow colour, which can be easily narrowed down in HSV colour space."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def filter_for_hard_exudates(rgb_image, **kwargs):\n",
    "    masked_hsv = cv2.cvtColor(np.float32(rgb_image), cv2.COLOR_RGB2HSV)\n",
    "    lower_bound = kwargs.get(\"lower_bound\", np.uint8([[[0, 0, 0]]]))\n",
    "    upper_bound = kwargs.get(\"upper_bound\", np.uint8([[[24, 255, 225]]]))\n",
    "\n",
    "    structuring_element = kwargs.get(\"structuring_element\", morphology.disk)\n",
    "    structuring_element_size = kwargs.get(\"structuring_element_size\", 6)\n",
    "    \n",
    "    hsv_mask = cv2.inRange(masked_hsv, lower_bound, upper_bound)\n",
    "    hsv_mask = morphology.erosion(hsv_mask, structuring_element(structuring_element_size))\n",
    "    hsv_mask = morphology.closing(hsv_mask, structuring_element(structuring_element_size))\n",
    "    masked_hard_exudates = cv2.bitwise_or(masked_hsv, masked_hsv, mask = 255-hsv_mask)\n",
    "\n",
    "    hard_exudates = cv2.cvtColor(masked_hard_exudates, cv2.COLOR_HSV2RGB)\n",
    "    binary_hard_exudates = cv2.cvtColor(hard_exudates, cv2.COLOR_RGB2GRAY) > 0\n",
    "    return (hard_exudates, binary_hard_exudates)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_colour_process(raw_img, roi_img, hard_exudates, binary_hard_exudates, **kwargs):\n",
    "    fig, axarr = plt.subplots(1, 4)\n",
    "    fig.set_figheight(5)\n",
    "    fig.set_figwidth(15)\n",
    "    fig.subplots_adjust(hspace=0.5)\n",
    "    cmap = kwargs.get(\"cmap\", cm.plasma)\n",
    "    fig.suptitle(f\"Colour filtering hard exudates (subject {SUBJECT_ID})\")\n",
    "    \n",
    "    axarr[0].imshow(raw_img)\n",
    "    axarr[0].set_title(\"Source image\")\n",
    "\n",
    "    axarr[1].imshow(roi_img)\n",
    "    axarr[1].set_title(\"Region of Interest (RoI)\")\n",
    "    \n",
    "    axarr[2].imshow(hard_exudates, cmap=cmap)\n",
    "    axarr[2].set_title(\"Hard exudate extraction\")\n",
    "\n",
    "    axarr[3].imshow(binary_hard_exudates, cmap=cmap)\n",
    "    axarr[3].set_title(\"Binary hard exudates\")\n",
    "   \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "hard_exudates, hard_exudates_bw = filter_for_hard_exudates(masked_rgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7efc919aa668>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_colour_process(raw_image, masked_rgb, hard_exudates, hard_exudates_bw)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Point conversion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This function converts the image to a point array, which is the requisite input for a GNG. Note that for reasons of processing ease, this will be flipped by 90 degrees – this is insignificant as GNG is rotation invariant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_bitmap_to_points(bitmap, included_point_fraction = 1.0):\n",
    "    df = []\n",
    "    for (x, y), val in np.ndenumerate(bitmap):\n",
    "        if val == True and random.random() <= included_point_fraction:\n",
    "            df.append([y, -x])\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = np.array(convert_bitmap_to_points(hard_exudates_bw, included_point_fraction=0.5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of data points: 27,804\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA38AAAKVCAYAAAB/H1LfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3Xl0XPdh2Pvvb2YwBIh9I8AFBEFwpyRbErVQ1GLJdrwkjePUSW3HdtzYVdM0552mfX3N0pzaPa9tkvPOyXvty3Mi23W8xFESNYmd2JYtWxslUZSojTsIgiCIHcS+czBz7/sDi2lbshaaouT7/ZyDQ86di8HFAIczX/7u/f1CHMdIkiRJkn66pa70AUiSJEmSLj/jT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhIgc6UP4FLV1dXFmzZtutKHIUmSJElXxDPPPDMcx3H9y+33po+/TZs2cejQoSt9GJIkSZJ0RYQQul7Jfp72KUmSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAlxR/IYRfCiEcCyFEIYQ9L3L/xhDCdAjhf79o27tDCG0hhNMhhN++aHtLCOFgCKE9hPBXIYTspRybJEmSJOn7LnXk7yjwi8CjL3H/HwPfWr4RQkgDfwK8B9gFfCiEsGvp7j8E/jiO463AGPCJSzw2SZIkSdKSS4q/OI5PxHHc9mL3hRB+ATgDHLto843A6TiOz8RxnAPuBd4XQgjAXcB9S/t9EfiFSzk2SZIkSdL3XZZr/kIIpcB/AD79Q3etB7ovut2ztK0WGI/jOP9D2yVJkiRJPwGZl9shhPBdoPFF7vq9OI6/9hKf9mkWT+GcXhzU+/7Dvci+8Y/Z/lLHdDdwN8DGjRtfajdJkiRJ0pKXjb84jt/xGh73JuADIYQ/AqqAKIQwDzwDNF203wagDxgGqkIImaXRv+XtL3VM9wD3AOzZs+clI1GSJEmStOhl4++1iOP4tuW/hxA+BUzHcfz/hhAywNYQQgvQC3wQ+HAcx3EI4SHgAyxeB/irwEuNKkqSJEmSXqVLXerh/SGEHmAv8I0Qwrd/3P5Lo3q/CXwbOAH8dRzHyxPC/Afg34YQTrN4DeDnL+XYJEmSJEnfF+L4zX3W5J49e+JDhw5d6cOQJEmSpCsihPBMHMc/su76D7sss31KkiRJkt5YjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZISwPiTJEmSpAQw/iRJkiQpAYw/SZIkSUoA40+SJEmSEsD4kyRJkqQEMP4kSZIkKQGMP0mSJElKAONPkiRJkhLA+JMkSZKkBDD+JEmSJCkBjD9JkiRJSgDjT5IkSZIS4JLiL4TwSyGEYyGEKISw56Lt7wwhPBNCOLL0510X3Xf90vbTIYT/HkIIS9trQggPhBDal/6svpRjkyRJkiR936WO/B0FfhF49Ie2DwP/JI7jq4FfBb580X2fAe4Gti59vHtp+28D34vjeCvwvaXbkiRJkqSfgEuKvziOT8Rx3PYi25+L47hv6eYxoDiEsCqEsBaoiOP4QBzHMfAl4BeW9nsf8MWlv3/xou2SJEmSpEv0elzz90+B5+I4vgCsB3ouuq9naRtAQxzH/QBLf655HY5NkiRJkhIh83I7hBC+CzS+yF2/F8fx117mc3cDfwj8zPKmF9ktfrljeJHHvZvFU0fZuHHjq/10SZIkSUqcl42/OI7f8VoeOISwAfg74GNxHHcsbe4BNly02wZg+fTQwRDC2jiO+5dODx36Mcd0D3APwJ49e151PEqSJElS0lyW0z5DCFXAN4DfieP48eXtS6dzToUQbl6a5fNjwPLo4ddZnByGpT9/7KiiJEmSJOmVu9SlHt4fQugB9gLfCCF8e+mu3wS2AL8fQnh+6WP5Gr5/BXwOOA10AN9a2v4HwDtDCO3AO5duS5IkSZJ+AsLipJtvXnv27IkPHTp0pQ9DkiRJkq6IEMIzcRzvebn9Xo/ZPiVJkiRJV5jxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJcEnxF0L4pRDCsRBCFELY80P3/U4I4XQIoS2E8K6Ltr97advpEMJvX7S9JYRwMITQHkL4qxBC9lKOTZIkSZL0fZc68ncU+EXg0Ys3hhB2AR8EdgPvBv6/EEI6hJAG/gR4D7AL+NDSvgB/CPxxHMdbgTHgE5d4bJIkSZKkJZcUf3Ecn4jjuO1F7nofcG8cxxfiOO4ETgM3Ln2cjuP4TBzHOeBe4H0hhADcBdy39PlfBH7hUo5NkiRJkvR9l+uav/VA90W3e5a2vdT2WmA8juP8D22XJEmSJP0EZF5uhxDCd4HGF7nr9+I4/tpLfdqLbIt58diMf8z+L3VMdwN3A2zcuPGldpMkSZIkLXnZ+Ivj+B2v4XF7gKaLbm8A+pb+/mLbh4GqEEJmafTv4v1f7JjuAe4B2LNnz0tGoiRJkiRp0eU67fPrwAdDCKtCCC3AVuAp4Glg69LMnlkWJ4X5ehzHMfAQ8IGlz/9V4KVGFSVJkiRJr9KlLvXw/hBCD7AX+EYI4dsAcRwfA/4aOA7cD/zrOI4LS6N6vwl8GzgB/PXSvgD/Afi3IYTTLF4D+PlLOTZJkiRJ0veFxUG3N689e/bEhw4dutKHIUmSJElXRAjhmTiO97zcfpfrtE9JkiRJ0huI8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0mSJEkJYPxJkiRJUgIYf5IkSZKUAMafJEmSJCWA8SdJkiRJCWD8SZIkSVICGH+SJEmSlADGnyRJkiQlgPEnSZIkSQlg/EmSJElSAhh/kiRJkpQAxp8kSZIkJYDxJ0nS62A+lydfiMgXIuZz+St9OJKkBMpc6QOQJOmn2fRcDoB7Hj3DtsYy0iHNkb5RfuOOrZSVZK/w0UmSksT4kyTpMpmey/G7f3eY27bW0zY0zkIhZkN1MQ8cHWB7QwVv39FgAEqSXjee9ilJ0mWSjyL6xmY42jfO6GSOzXWlnB2ZoWpVmoVCgc891ukpoJKk143xJ0nSZZQOgVSA6zfVctv2GkoyGf7pjZsozWYhvtJHJ0lKEuNPkqTLpKw4y4dvbqU8u4qr1ldzrHeGbY3lDE3muG1rHb/+tlaKs16BIUl6ffiKI0nSJcoXIjLp1MpsnvlCRFlJlkw6xc++ZS35QgOZdOoH9jf6JEmvN195JEm6BNNzOQ52jnJTSw0HO8c41DVC39gcv/9PdlJXvhrgR0Lv4hCUJOn14quPJEmv0fRcjs883MGhs6M80THC9rWriaKYmvIivvJkN8NTszx4YpB8IbrShypJkvEnSdJrkS9EPNExypmRafJRxMTsAn/wzTbOjEzTNz7PxurVfH5/J/94uM8ZPSVJbwie9ilJ0muQSae4pbWGw+fGuBAVODM8zYbKEhaIWBVS3LK1mjND06ytXOVpnpKkNwRfjSRJeo3KSrJ88o4WSrNF7Gis4AM3baA8m2XX+moOd0+xtbGCkqKiK32YkiQBxp8kSZekqrSYj+9rpn1gik/9/XE+eNN63rFrDelUirLiDJ+8rcWZPSVJbwi+GkmSdImqSov5zXdsZX4hT1VpMQB3bK8HnNlTkvTGYfxJkvQTUJzN/MAIn9EnSXqj8ZVJkiRJkhLA+JMk6U0mX4hcO1CS9Kp52qckSW8S87n80vqCIwDctbOBfCFyQhlJ0iviq4UkSW8Cw1OzfPVgD7mFAgtRgUwqzXXNldz7dC+fvLWFTDrldYaSpB/LVwlJkt7ghqdm+W/fPMl739LA1oZy0iFFPop4+uwYTVUlzC/keaRt6BWfCuopo5KUTMafJL0Ky6fdLb959k20Lrf5XJ6vPtlD1eoM//d32jneP0FLXRlFITCXK/DA8X7+7JEOnu0a+7G/j/lCxHwuz/Rcjsfah5nP5Ve2S5KSwdM+JekVWI6+ex49w861FWQzKd7SVMHR3mne2lRBWXHWU+50WRRnM/zarZt4uG2Yo71jFOKY0+cn2bWuEuJAQ+UqNteX0Ts2/5KPMZ/L862jvZwamCEVAi31JTx6qsB1zZU82zXBXTsb/P2VpATwX3pJehnzuTyf29/J904O0nF+ilyhwPSFBf7LN04yPD3Lp752jAeODayMpOgHXTyytPx3n6tXpziboSSb5qr1VRSFFL1jcxAHTg5McbRngrPDs3x8X/OLTvwyn8vz7aODfOahDk4NTTKfL3DvU+d4+uwo/+lrx/j6C71Mz+euwHclSXq9OfInSS+jOJvh47c0s799hOaaUtIhRTaT5m3b1pBOBRoqi4HAk2dGuXVrnSMoF5nP5Veel3wh4tFT57mxpZovHeji1/a1UJzN+Hy9Apl0iju21wPwtu31PNExyu3b6njXVY3ML7RQXJR50fDLFyKePDNKOgW3b6snnUrxoZub6PvWLL+yt4l7D5yjQMwzXePcsX2NPwtJ+iln/EnSK5BJp0inAs31JWQzaa5rruDZrknmcnkg5rattYbMD1mOvbmFPNPzOZ7qHOPvn+9mbiFPbiFif/sw2UzK6HiFlp+jspIsd+38/nP245Z5yKRT3Ly5hsdPj/DWplpO9o1ztGcSYniua4xUSJFJBW5qqfFnIEkJYPxJ0stYHj2ZvpDji4918Wu3b+KrBydprllN28AkR3omyEeRb55/SL4Q8dy5cZ7pGubU4DS/dMN6RvfPE8URzXWrKUQxEK70Yb4pvZrfteJshju215MvRJRk01zfXMnJ3ik6h+bYtb6SbCZ9SesE5gv+7kvSm0WI4/hKH8Ml2bNnT3zo0KErfRiSfsotX6s2PjtP1epi5nN5DnWNs6e5inwUUVVafMWObXou94YcdcwXIu4/MsBz3cPsXltNtihwtHuC1oYy/vZQN9c0V/Ov7mhdOWVx+Tl+o30fbxTTc7mV52Z+IU9ZcRbgVS3yvhxqy9dcLj/eyz3nF89uu/z5y5/j6c6SdOWFEJ6J43jPy+3nyJ8kvQLLb2zrylcDi6feXck3vMvLTcwv5Pn0PxzjZ3Y3ctuW+pVje6NIp2BoMkc2PUVRKsW2xgpKsmned90G+sbmOXBmhPbBGT62dyPPdI1TiGJu31Z/SSNRP42Gp2b5L984yZqKVRDDC93jfOjGZrKZFMf7J/mXt29+RT/3V3Kq6A+bz+V5/PQIc7k8bYNTtNSW8uDJAZpry3lLUxWF6AfDUJL0xuWrqyS9Rlfqje58Ls+jp4Y53j/Blvoy1pYXUyjAPY+coWtslk///K43xNITmXSKd+5u5LatdYvHnc/zpce7aG0oY2gqx6/dtgmAdEjzVOcYuXzEsb4JClHMO3c3XvHjf6OYnsvx1YM97GmuZnDiAq0NpYQYCDFzuQJnhqfZ3z7MnTvW/MSjeeW6zVyBkwOTRHFM2+Ak+Shm59pyrmuu4KsHe9jbunhNp+EuSW9s/gstSW9QyyMpy6d1Lo/2HewcI1fIs7F6NdlMmi2N5dy6tZY9LZV86fFzPHLqPGWririppYaykuwbYkTmiY4RClHMXD5PNp3mwzdtIB9FfPVgNx++qYkXuqfIFRboGJ4iFQJ7W2uu6Km0bxTLP+8NVSU82DbAp35+N2WrsuTyER2DM1woFFhTvnj7cpx+mS9EHO4eJyaQjwrsXLu4tmAhLnDr1lqe6ZpgbeUqHmkb5jsn+inEEXduX1wz8Er/zkmSfpT/MkvSG8jF6+A91j7M8NQsv/+1I9x/tI8/faSDP324g7lcnpO9kzzSPsTmhhL+6qku/uThdr78RA9z+TzfOz7ExFyOP3v0DOMz8zzSNnRF1tXLFyIePDHIZx7u4IXucSbmcrxwbpzJCxf4/P6z/Oevn2BqfoFnuyZ4S1M5pwenGZ2cZ11lCV860PWqjnnlNNhc/gfWFfxpMjy5uIj7+Nw8f/9cD794wzrSBI71THBqcJo9zVWXJbiymTS/dOM6esbnaRuY4sTAJPvbhpnP55mez/O3h7o53DtGPoqZX1g8RfSRtqGf2p+DJL2ZOfInSW8Q4zPzPHZ6mFu31PFM1zjT83me7BijtqyImzfXsGdTFfce7OVt2+t52/Z6vndiiJN9U5QWpyGG1vpSvnNikrdtbSCbztA+NMlj7aMc6xsjVyhw5/YGpi/kViasWT49b3lk8OJw+OHJVy5+I//qAiMQxTGb68o4fX6KyrIspwdniIi5bVsdnednKEQxL3RPsr2xktPDU/RPzLF7feUr+jr5QsT0fI4DHaOkUyle6BnlqnVV7G2tecm1794MlmMW4PrmSh5rH+Hqpmqe7Bjj1OAUO9aWU1Wc5Tfu2sL+9mpu21p7Wa71LM5m+LVbN/FExyipAFevr2JPSyWBmOLM4sLzP/fW9ZRk06RCoPP8HB+7pZ6jvdM/8WORJF26N+eroiS9CVw8indxhGTSqR+Iq+URq9/92yO0DUxwpGct2xsqeehUP3Vlq3i0bYhAiuKiDNsbyhcfO4o4NTjFprrVTM7koR6ua6nkL5/qpHNkig/dvJGvHlxgx/pSvnWsl1RIMz2f5wuPdfKxvS0cPDvC23c0kM2kONI7wdXrK7l9W/3KMT50coh0Kqxse/LM6NLSDDH7ttS9oqjKpFPc2FLF4e4xTp9fPJ3zPbvWc2JgjGPdE2ytLyeTCivXBOajiI6hKT68dwN1ZatfNv6Wr338+vM9pFOB33rXVv7hhR4W8hHfOtrH5rpyPnlbyw+E7ZvhVMR8IeKRtvMc6RknX4goEBNiKEqlKMQRrfXl3H+8j8/u7+Rf3NbCO3c3XNbvq6wky+3b6ihEi+tZ7m8fpm9insfaR5jLFfja8z1cu7F2cUKfhnKO9k5z82bXDZSkNyLjT5Iug/lcnodODgGBbx/r5127G0mnAtlMmj3NVTzRMUJ6aXHt57onuLapko3Vqxkan2HLmjKymcCd2xopxBFHz42ys7GSbGbxzf/n9ndyIV+gc3iKTbWlVFesIhVSvNA1SWVpERA40TvNP7uhmfb+WTZUlfDRWzbwXNckv3JzM2UlaW7ZXEc2k+LGlmoOnxujEMU8dPI8J/on2NZQzj8e6WVL3WJoLh8nwMOnhnj89Aj7ttS+bADmCxHPd0+yY20lJ/vG2bGukj0tlZRk02xbU0nXyAxbG8sYn7/AfYf62N5QTlNtCfc9088nb235sfEwPZfjUNc4N7ZU8UzXMB+6eSMdg7PcuX0xaHdHMfu21vBExyhHe8fZ1lhG2aose5qr3jDXQf44N7VUc31zJQc6xjjcM8b2xgpyhQKd52f5yN4m2vonyOULfP6xs3zi1k0/0esjL35ulkcgF2/H5Jdm9txQVUKuUOD0+Smqy1bx0VuaON47wy2tNSvLjlw8evlmHYGVpJ82b9xXPkl6E1getbv4dr4Q8dDJ83zzaB8715fSUJHl2uYK0qkUO9au5uFTQxztmWBiLsfBzlF2ri0lk07xkX0buXHzGrLpNCf6Jzk9NEU6FXjrpjpubK2iEMWc7J/kAzesY0djJXGAkIrYWF3CrrWVFOKIVAhsrl9NOhXYt7WGQhTRMz7PwY5xjvSM82j7EEd6Jnno1CDTFxaYz+cpEHNtcwWFKOJCPs9VTeWMT1/ggzdv4JbWWqYvLLC/fZiH285z/9EBJmZzPH56eOX6upe6Ni+TTnHr1jresWsN12ysAQJ/dH878/kc2UyKhSjiaO8E/+7eF5i6kGP6wgIHzozywRvW/9hYmJ7L8ZlHTjOXyzOfz/NM5wh/83Qv0/N5zpyfphAvjoo+2bE4Mc5MLsfpwWmuWl/Gn+3vYHhqduU6yHwhYnout/Kzm57LMZ/Lr3xPr/c1hMujfp955DQHzoxwbXM5aWAhyvPIqfNsqFlFJpWiZU0pqRBWAvAncU3n8vf/WPvwyvNw/5F+7nm0g4HJGe4/1s9nH+1kbiFPDLQNTBHFMS01pRzpmaQQxRzsHF15zh44NsCfPtzB/3jw1E/1tZiS9Gbif8VJ0ms0PZfjYOcYR3rHV0aqHj11nhtbqilEEblCxF8c6Ob5c2OkQoodjRX856+fAOCO7XXsPzXMbVvr+c//cJwNVavpn5znls11lGQzbGsso61vivbBabpGZviDb7SRjyLOT84RxYEt9WWkQuBE/zRHusfZva6KE/2TjE5e4JG2Yd61ex1febKbjTUlNFUWk04FWurKeLZrhF981zr++Dun+M6xQb57fIiBiRnycUwUxRzrGWfzmjJqKop5/twEhULg3qfOck1TNdsbK8hHEW2Dk2xvLOfRU+cpRDHH+yf513duedFgWz7l8q6dawC4ubWKrxzoJl+IKEql2NpQTjoEUgR6x+b43fduX1lL8cXkCxEPtw3zdOcIrXXlHDwzxtSFPD9/7VrODM6yEEUUCpCPCpwamKKpZjVHeib47x++lkwqxdOnh8ktRKzKpJnLFShEMQ+2DfI7793Ooc4JvnNsgA3VJWSL0nzk5ia+erCHHY3l3LXz8p5aefHzdX1zJX//fDf5Qszh7glyUYG2wWly+QIPnjjPib5pDveOc82GSnavreLs8KVfX7c4Oc8QR3rHaakt5XP7z9BSX8r3Tgxx25Z67nu6lzu2riGdhvb+73+9Vak0WxvLKRQAIuZy8OcHuvjITU10nJ/h565t5H98t53xuXmO9U5zx/Y1b+gRV0n6aWf8SdJrMD2X488PdPGB69dypHuch08NQRz4h8M9PHdujN3rqmipKaV1TRlRHLF7XSVvba7gm8f62FBVQkXxKv71O1qpW72KM0NTfGTfRgC+fOAcH92+kee6Jtm+toKbWqv4yuPnVu7/T187Rj5eXNvt3bvWAYFUiLlxczVnhmb4wJ5NlBVnuHVrLbTFtA9MkieiEEeUFRexe0Ml7f0zNNeUsr2hgutaKvmv3zhBFMeUpDO8+5q1DE7k+M27ttB1fg5SMT/3lg1ULk0m0lS9miiOeeD4AO/evY5cocCZ4amXHdVZfsNfV76aX39b6w9sf9dVjSu3X8npgYUoorI0SzoNN22uZu7CFv7x+QFaakvpHJ4CIB1SfGzfRh5vH4UoXvxaqRTXbKzm6vVVbG5Yzf0vDDKTX2Amt8Dn958lFQL1FUVsb6ygEEc82zVBc81qcotl87opK87y3qvWL4Z13zhdI9O01lVwx7Y1pFOBzuEZrlpfwap0mpJsmp3rfjKzfObyBeJCTMf5aXaurSRXyLOusphsUWDLmgoKcYF0SLG5oRRYfI5vaq3i4JkxjvVNAnDtxmo+vreKXUVMAAAgAElEQVSZspIsH7m5iT9/vIuasixffPwcRekU1zdXuYSHJF1B/vebJL1K+ULEoa5xNtWs5nD3FB+4cR1nzs8yPV/gzNA0+Sji5tYqYuD00DS947PkCgX+4kA3cRSTSgUmZvP8b195ju+eGCACvnygmydOj/L0mWG++MQ5Xjg3xndPDnCwYwKA585NUJzJsLFmNZmQBmDf1hrSqUDP+Dx/+WQPFwoFukZnSKcCBzrGaB+cYl1tMYfPTXCyf4qd60sZmrqwEjNnzk/zwrkJUiGQTaX5hRvW8cCxQe7cWce/+8vnefbcKC/0jPGPz/cyn1+AEFMUAlvWlDIyNc9crsCDbUNLE8G8csXZzMpHJp36gdsv97xPz+c4OTBJiOFwzzj/9RtttJ+fJJcvQIgZn8nRUrd4SmRxJkM6laK6YhVffqKbh08N0T85z9jcBX79i4eoKcswMLG4fELv2Bz5KKJ3bI65hTwPnhxkYnaBkwOTfOfYINPzuVf3S3IJMukU79zdwLuvbuTDe5voGZ5jQ00x//B8L4+cGloM9UyGXesq2dtaQzbz6l/Kl2N9+bTd+Vye00NTfGRfE1etr+StzeU80naeKI45NTjFXC7Pd44PcqR3nO+eGORbR/u5/2gfX3i8i28f6ScfRaRD4MaW6pVZR8uKs2xZU8bQxAWiOGZTbSlfOnBu5TRbSdLrz5E/SXqVMukUN2+u4aGTQxzuHudI7wS/estGnu2a5Pbt9bxlQw2ZVIrBqQvcuKmKnvFZ0qlAAJpqVpMiUJJNsXdLDUWpDNsbK2gfmCSdCuxpruNjt2zkUOcEmy6Uks0ENq8p50TfFIVCoHd8jju3NdI1PL04A+fSNVcfunkD9x7ogQB7W2s40DEKMfSNzHN1UyU7Giv420N9NFaUkA4pilIpNteXkQ4pmqpXs6OxgjODs+xaV0nX+Tlu21bP3Xds5unOcTbVlHFuZJ7W+jIIgeKiDNdtrKVkVYrm6tVkwuVf0Hv5Wrhnz40QgObaxcC7pbWWR9rPs6mmlNu31XGqf5KyVUUU4hnyUUQ2k+IdOxopW1XEDS1VFAqQTqW4Y0cddeXF/M57d/DUmTFu2VLDoc5JesbnaB+aIleIODsyy+a6MuI45sCZEW7bUr8SrPO5/A98z6/2+3+5CWeW79tQXc5X7r6ZslVZKlYVAzHpVIrC0sQrBzpGX/FjLu+TLywuCL848dAox/vGFyeUyUccOjvOyb4pDveOc+uWekqyaXL5iDPnp2muXs3H9m3kYMcE9x/rZUP1arbUl0EMV6+vJlfI81TnGLdvq1853fdt2+uBQDq1+LxvXZPhUNf4T3wxeknSK2P8SdJrkEmnFt/MNpRzvH+c4kyG65orONw9xvJ72t98eytff7aft21bw74ttaRDiuN9ExTimEIUU5RK03F+mqaaErauLScdUmxpLOO5rkmO9E7wzNlhfmVvM22DU1y9vpq3Npfz9Rci0mlori/lue5RMiFNPoo40j3F7g2VK6Nw2Uyaj+xrIh/F3Pd0H9e3VJJOBZ7rXoyF5rrVnBudZX1VCd1jM7ylqYprmso52jNBIYrJphdHF+dyBQ50nueu7YunZi5EBbLpNFdtqCKbSXNdc+3KDI+Xy/Io1Vuaynmua5QP7d3AM50TtA9Mkc2kaKkpZde6SgBCKnBVUznfOtbHgTMj5BbgwbYB3rlzLV964hwzCzl2rKniRM8kxZkiXuge52jPONl0mlu31vJc9zCZVIr11cV86Ob1/PXBXgrx4rV3J/qmuGp9Jbe01vJnj57h6vVVpFMBiF/RtWwXT8ry+Olhrm+uorgo8yNrLP6wuvLV5AvR0u/V4tc72j9OikA6BBYoLAb/mRHeuWvtSz7WfG5xAXaA6Qs5nugYIVcosLF2NTe0VPF89yi5fMTWhnK+c7wfYti9rpJzo7PsWLs4YdHJ/lnevrN+MT5DTPvgNOkQeGtzOf/tm21srF48TTadCuzdXMszXRPMF3K8e/u6le9zPpdnej5HWXHWAJSk15nxJ0mv0eIbfzjSM8GDbQOc6p+hEMdMz+f5j393lP6JWa5pquKhk1OcHZlh25oKtjSUcXpwmvahKWLg569by+/fd4SrmqoJxPSMz/Ku3WtpqS3l4ZP9bF9bwd883cOOteV85YlznB2a5kjPBP0Tc3QMTXHL1jpO9E1yzYYq9m2p5c8f66JtcIqNNSX8n18/tzgSuLmeP/7Oaa5rqubRk4Pk4wLnJxf4P96znfsO9jI+nWN+Ic9fPNnDxtoSspkUURTz2f2d9I7OkStEpFMpjvaPcbRngk/c1sLJ/tmVtdwu5xv4xRG/IeZyBU72T9IxMs29T/ZwIc6zra6SM0MzbG+sAODz+89y6OwI+TimEEVcu7GSJ06P0jE4zdt3RqyrXM2fP95Hc1U56VSgta6crtEZ3nP1OrKZFPP5PMd7J3jX1eu492AXuxqq2dZYwbG+MXYsfQ2A+Xyec6PT/MrNG6gqKebRU+dXjvXFnovlmS4/t78TArTUlnK8f5y/e7aHptrVXLO+ijt3rKE4m/mxj/Hdk4OsryyhqaaEZ8+Ocl1zDesrV/PVg51sqirnG0d62bu5lqrS4h95nOXRvkIUcU1TOX90fx97W2o5NTTF8+dGmM8XeKZzlN6xeX77Z7eTTqWWlgKp4tattZQVZ3/gWB5qG2RD9WriOCYi5vmuKSBmU10JR3om6B6b5XD3BBuqS/jMQ6cpFGLqykq4vrmKzz3aydNnR/jwzc387DXrDUBJeh2lP/WpT13pY7gk99xzz6fuvvvuK30YkhImlQpsrFnNtsZy3rmrgdb6cp7vnmBzXRkd56fYva6SE73j3L6tkXQIbKorpX1omoGpWXavraJ0VZqjfROUZdOEVIqdjZWcn77AxNwCm2vLKcmm2bG2grlcge0Nldy+rY65XMzVG6ppqi3h4/s2UVtawvuuW0ff+BzrKkvpn7jA+akLS8cwzepVaUqL08wtRPyLOzYxMpOjf3Ke6tWr6BiYZG1VCf/kreu4a1cDYzMR3WMzHOuf4pdvWM/8Qsz4zAIfv30T58cv8N63NHC8Z4q9rfVcv6mWjbWryRalSS0F8OUSxTFnzs/QNjBF65oyttSXs297LV890EVRJkUMNFaWkM2kqSopIpUKbFtTwbnRWUqzWbKZNM11q6ldXUIqFVOcSbG+ZjU711VSlE4xMn2Bh9uG2LKmjNHpPAtRRGtdBbvWVtE/Mc97r2ng+XMTNFQUc3pomlVFKYanFhidu8DpwVm2NKzmoZPn2bymhBe6J6ldvXgMsxcWyBalmZ7L8aePdDB1YYHKkizv2F3Po23DzOYXqCguIkVgYHKOhUJMQ/kqHj89zLrK4pXfsWXZojS3ttZy7cYqHjg+yMDYLLduq6dvfJ69rXVUlBRRlArctLmOKI753skB1lYUk0oFonhxNLimNEPv+AXGZvO8fXc9z5wd50I+T//YLEWZNIU45o5tazjQMUZFSZqbW6v50oFzPH9ucWbR4mxm5fHODM3wwZs3EEWBoal53nftWgoF6BmdJ4pjVhUFitJpGipKOD87SzqVprI4y9DUBYqzKXrHZ0mHNG/ZUEm2KH1Zf4ckKQk+/elP93/qU5+65+X2u2z/3RZCeGsI4ckQwvMhhEMhhBuXtocQwn8PIZwOIRwOIVx30ef8agihfenjVy/XsUnST8LyqFdVaTFlJVk+vq+ZbCZFz9gcsDgyuCoLH9nXRPfIHPP5PI+1DTM5f4HHTw+zpmIVpwZmISztm0qztrKY00NTHOkd54kzw3zzcD/H+sf47P5OHjgxQMfwFA+eGGI+X+CxjvN88fEujnRPrJzuubm+jJJsmhAvPmYqpMhHEc+fm+Bk/2K8bKkvZ2J2gSM9U3zhsS7++mAPcwsLFKVSrKnIcqhzgpMDk1woFPhfT/eyEMc81zVFIY7pHp15XRdIX14m4pO3tVCSzXByYJL/8cBpiGK2N5ZRFBYXvr+ldXHym6JUipJsmrfvaCS9NLHO/9zfyfDsPO3908Qh8OCJQU4OTPLIqSE+cOMGUsScHppmLldgYPIChQgqVxexa10l+Siib3Kem1uraaop4W+f6WFiNs+JngnWVma592AvY3MX+MNvnqKpdhWf23+G+4/08/tfO8rw1Cz720c4fX6KF7rHuf9YH1890M3MwgKPnxoGoLW+nN7xeeYW8nzvxBDPdo3x6KnzPHBs4AcWSQcoK8lSXJRhVSrN9S11FGeKiKOYp8+O0j44xZbGcva3D/PNI318YX8nf/LQaR44NsADx/t58MQQn3+8k1y+wFXry/jMgx188Kb1XNtUyzXN1ZRmMrxr91oqS1Yxv7DA3zx1jifPjNJSW8r66pKV9ftgccKej9/azH1P95LLRxzuHuehEyN87bleNtSUsKW+ghN9k2yuL10cQUyniaKY9sEpcoUC3cNztNSV8bFbNnKwc9QJYCTpdXQ5T/v8I+DTcRx/K4Tw3qXbbwPeA2xd+rgJ+AxwUwihBvhPwB4gBp4JIXw9juOxy3iMkvQjLn7DvRw6r2QijWe6xgG4c3sDZ85Ps7upmlMDMxQXZWhdU0Y2k2ZXYxVdIzPcub2BtsFJNtcvLvr+lg1VfGRfE3/wrVPsaq2kpCjDR2/ZyFMdEzxwsp8NVSXctWMNJUVFFKKI470z3LF1ce28ohC4Y3sdBzrGeKR9iH//7m0Uonhxoo6haQrEEAd2r6tgR2M5N7VW8VjHEL+yt4mj3VPk8hFnz0/TUl/KY6eHyeUjQhxTlEpRiGNSIXBDSyXLk4283qfpZdIpykqy3L6tjnQq8InbNvF05wS3ba3l4cww3z7ex/Wbqjg9tLj+XCGOODsyw861lZRkU+xYX07f6Dxb15TTNjjBHdsaODM8zV07GmgsK+Gf3diycgrvuspiCBGQ5rrmCr56sIfbt9STSaU4OzzLtU3VvG1nDedGpqleXcxNrbX8zl8f5ufesoG/PdRHFMW8tbmC473ji6dZFmBj9Wpa15TRPTq7copq79jiCHBJNsPaylWL6w2eGKCpppRta0v5fx5o54aWKl7onmTflrqVayqLsxl+4+1bVmac/Y27tgCLp6J+fn8nkCKKIypK02xtKKMQxZwYmOCjezdy+NwYJwYm2NNSSUN5MZmliWPGZgv8y/e0crh7iuuaF68NvWpdDTdvruazj3Zy+NwYv3TjRp48M7oyUUtxUYYoirlrZz0l2Qw3tFTSMTRJ39g8m+pXU12WpSSbplCAolSanY2VlGTT3LljDbduWQzyTCrFkZ4JjvVNcvftmy/rdaOSpEWX81/aGFi+SKIS6Fv6+/uAL8VxHANPhhCqQghrWQzDB+I4HgUIITwAvBv4y8t4jJL0A5YXbi9EEXMLeU4NTrGtoZy372hYeXP6UvFTiGJO9k2wdW05m9eU0jYwRRTHzOUKdAzN8NFbmgDoPjDDja1VPHCyn60NpTRVLS5qXlVSzL955xbue6qXhSjiSO8YV6+vXplRs2xVEbdvqwfg0VPDtA9NERGzvbGcTCrFDS2VPHtuhC8fOEeKQCoEPrqviee6pkinArl8gVODi+vgralYxd881Us+isjHBa7dUEs2k+K3fmYrJ3qn2dJYzu3b6ni8fZSjfWM82zVBNpNm35baK3aNVnE2wx3b65eWQigmk05x69Ya7j/ex32H+vjI3iae7pygEEVsayhnb2s1BzpGefeudZQUZSjEEX0T80uL1S9OSrK/fZi2wUlSYXE21rMjM6RCikxqebbNYk4MTJAtWrz//Teu51jvNNvXVnBjSzWPtY9wdVMVlaUZyorLyBUK3Pd0HwtxzJGexWU05goLi7O9pgL7ttZQnMnQNjBJSTbDtc3l/OORHrY0lLG2qoQdjRX8zVO9VJUW8XTnOMd6F5f6uHhCmeXfw+UQyxcinugYpXN4mrfvbCSbTnN2acmPXKFA5/AMBzvGaaor5sz5Wb7wWBddozN84bFzpFOLExOVZbMcPjfGkb5xfuWmJg53L84mW5RK8XNvXU/ZqqKVazyXpZaeo2uby3mhe5LWhjLu3FHPkx1jtNaWs2dTFc91TfK27WsoyS6e2pkvRBQXZViIFv/8+L7m1+m3R5IElzf+/g3w7RDC/8Xi6aW3LG1fD3RftF/P0raX2i5Jr4v5XJ7/+fhZNteVUogj7j/WTy4fcfDMCKcGp1ZGL15sZsdMOsUtrbU8f26Mtv4pQgpCDIUQc6xvnN6xeT799eOkQ4q1VcU83TnK+Yl5Tg/OUIgjjvVOQhy4/2g/DZWriOKYoz0T/PNbN1GSzXBTS/XKMgMAt7TWcPjcGJ3D05wbmaFtYJJMKkP36CypVOCuHfWcGZrlua7Jpf1r2d8+QhTHHO4d5+DpUT62r4UzQ9PsPzXMP9/XwvNdk/zxt9upK89yoneSQgFODU1ytHtxkpeybPaKj84sf//Lf1aVFvMH779mZXQ2nZrkRN9i4BJijvdM0j0xx53b6ylbleVdu9YylytwtGecXesqyBUCfeNzvGNnI9e3VPLv/+oFtjaUEseBQgRtA9M8cnKI1royLhQifve+w/zTPRt47NQIEQUePHGeDdWrKRSgbWCS3ok5futntvKXTy6+nDVUZPnsI91sWVNBCni6c5y9rTVLcRlzqHOCQiHm4RPnSadTfOK2Kk70T3CyZ5JP3NpC++A0N7XUvGhwX/xcXNdcwRceW+Ds8Bwfunk93znRRxRHnBmcJV+IONw7xpGecXZvqCCbKmJzTSkfvWUDT7SP8R//11F++YYmIFDIx9z7VC8bqxdng928ppRH24d5167GH1naYntjBZ/d38nz58b42Wsa+fKBLkIqpmNwljiO+fIT3eSjiK6Rad6xu4G2/inaBqf4wJ51vNA9xvjcPPc+1UtUiEmlA5+8tWVljUBJ0uVxSa/iIYTvAo0vctfvAW8HfiuO4/8VQvhl4PPAO1icp/qHxT9m+4t93buBuwE2btz4Go5ckr5vei63ElbrqrKkUymIoLGymExYPI3u/dev5x+fG2DLmrKXfJxMOkU2k2bLmjJubq3m8fZRTvRP0DeZ47Ytazg3NsPGmtX0j82RL0BlaRHv37Oe+57upmhpiYJ/ONJLKixeq5cOUJz5/mjXxcpKsvz6na083DbM0d4xQgh89JYNHOwYJ1coUJzJEAPTFxY4c356ZTHwHWvLyaYzfOjmJs4MzlK8toizo9MUZ9JkM2nu2tFANpPil29s4hvPDZBZGvk53D3FyYGpN+TpeRcHw10713D7tjryhYj97cNsaSyjpb6McyOzfPCmxej6yoFu3ndtEynSdI7MsG9LHSXZNFXFq/jF6zZSlErxF0928rFbNvPPb91IPipQtqqIbQ1lPNN5nlScZk1FllRI8a7d6/j/2bvT+Lbu68D7v7sABIiVJEiC+75qsyzJ2iVLtmM7q5O6buLaSZomedI2M820TZ8u+bROptNpO80z02nTtJk2cZI6jh3HS1bbsWXJ2iVq5b4TBAEQBLEDBAjc5XkBkWOncSI7SWPH9/tGogiC9wIX0P/gnP8529ucXPVn6Klz0Oaxc8WXYjaaoaXSRiCRp7/JhdUkcnN3LVCazdftdbC7o5ILviRv39xAaRaegNtq4caWKjY2VOB12vjowet7vD2Ocj5///b1+29x27HIJt67uzT/sb3WBoBZkuj1OlF1jUvzSSRRxFkuYTWZ+NihjtIohqLCv53x88G9LdgtZt7S733ZBw/wfwfRH+qtJl9UsJhkLLLp2oiQFZqrrNjMZgqqyqW5ZUaDaUYCSe66sYHBhRQbG9zIosimBjcdtVauzCc5Prm83vXUYDAYDD8fP9U7rK7rt77S9wRB+Arwu9e+/AbwL9f+vgA0veSmjZRKQhcolX6+9N+PvsLv/QLwBYDt27f/yADRYDAYfpy1ZhqKqvHAt0e4ucfDDc0uvnh8jg/tb0UUREYWkrxzSxOnZiL872enuLmnholwmlv7a39kJsZilvnw/jbOzpbKJKeXMvTVueiosTMTyaLoGqenl9nf7WFsMcXwfJy/fXqc9EqRv7pnM08OhJBF2NJYwc091SiahttmecVzsFvN3NpfgyQKjAaTXJovzb7zJ1c43ONF0OH50TCJ7CqnZ9zsbq/iK6fmafFY+daFIN11DgAyK0XOTMcxy9K1od46/3x0ln2dVUiCxLHJJf7krT2/kP1+r9ZLZ8mNBlPMLGdoqiyn3WPnr78/gddlob3Kji+aRRQF3rO9nv/57ASHe2t5fizMiekI/+nWTsSzOtvbXFzypVlMrVJQSvtA03mFXEFl0J9Alkp7OL92ZoFur53xYJqZWJqGinJiqVUOdtYSTuX4L7f18E9HZijqGrf11TIVziCIvCyj+9LjP9xX87JzuV52i5lj4xFGQyl66lylEtSBII1VVqwmE1ZZprfOhappvDC+RKO79LiUyRK7OyrWg+i163jtuF4pG7d2fGsjKlzlZnIFhbyq8OBJH4d6a+mpdXBjq4cPH2gjr6g8fiGESZRZiOf40gkf795ez//7jau4y81IQukz4Ns2eF/315nBYDC8UQm6/vOJnQRBGAV+S9f1o4Ig3AL8ja7r2wRBeBvwceCtlBq+/G9d12+61vDlArDW/fMisG1tD+Ar2b59uz4wMPBzOQeDwfDLqTQ7LsJgIMHHDnawnM3x6LkA3bUOYtlVvK5ycsUigwsJbmiuuLZ3Kc3ujgoePOnjw/t/fHnaWlD54sRyqeFGMIGi6aCDRmmAeqvHxrNjIT71tj4G/Rlu21C7/nM/nGW5nvNZa0zz0vvIFxROTce4scWJu7y0Ry6TK3BqOsbQQoK++tLg7oKicWv/y0tZM/kCl/0pdrVXoqgadqv5P7TL589CJlfg6HiEgqJhNUtsbXHylZPzyKJIs8eCRTYjiTC4kKC/3s1wMIGu62xscDMUSPDhA22cmY5zZSGO1STTUW1nOJhgc2Mll/zLbG3yrP/8h/a3cnIyzrOjAfZ31vD8eJieaiftNTaOjkfY01HFfHSFzhoHc8tZWqvLOdDtYSiQYVd75c8s25XJFTg+GaWgKsxGVqhxmvju1UX+269sZDRQygCfm02QKyiAwPRSim6vkzs2vXw4/Gt5rvMFhedGwowEkuTUIh/Y28bj54Noul4aEi/B1mYXl3wpLvmiCKKISRSZXE5xa6+X6aUMN7ZUcrjvR3+4YjAYDIZXJgjCBV3Xt/+k2/08ays+AvydIAgykOdamSbwPUqB3xSwAvwGgK7rMUEQ/itw/trtPvOTAj+DwWB4tdY6eW5rcXHJFyORy/PwmYVSeZo/xlAgyWfevZHvXkowH8siCAKqpmMvK7XZb/PYODsbY2+nZz0o+mFrGZG1DM6h3uofeZu3bKjFbjXjddp+6mHpL/3ZtWOyW80c7qv5d99bK4384TK+l3LbLOzrMv/Y27zeWcwykijy/HiI5opydrS50dF5z456/tcPpri1txZJlOivd7G9zclUOE2zx0pB0SjqGv/n+Oy1gF3jnpsaGPSn6a1zsr3NyTMjQba3OVE0nacu+zk5FUUSRW7pqcMsi7RX2fjwgTYUTeO50TBWk0xbtY25aJZ3bqvj734wyfRSlm6vnRfGw9zWX/dTP75rHUABJsJp3reziYfPLHBrXy3fOB+goCiouoaq6ZjlUgOWvnoXh3pr1rOla/dzajrKge7q63r+1z5wSOTyTITTNFfZODKxyDfOB+jzOrmhxcnXTvvxxVcYWkjRXetAF2Bzg5sbWpycnFxmd2cl/ugKN7VVvOGuM4PBYHgj+bkFf7qunwC2/Yh/14HfeYWf+SLwxZ/XMRkMhjc3RdU4MblMrlgEXWQ2luHLJ/3MLqdorbKj6Tq2MpHPH5nh47d2MPl0Cl90hXBqlc+8awP5osKR8UUOdtfw3Ogiz4+G+fN3bMBuMb/sdwD/LpOztqB96eiItSDtp1nsrmUxC4qKqulMRTJ8eF/bj+1Mer2B5ht9ES5LIvu7qhheSKDpOi+MRRiYi9FV4+RAl4ebe0pB+ReOzTAVyXD3TfV89dQ8/tgKoUSWWlc59W4LJyYiiIgIwKX5OKqmo2gaXz7pZy6aZmoxxVCVDbMoI+ga7bV2FlOr5BWFq/7MtU6rOi9MhGlw23jiQhCv00JHtZ2xUJpAIsfu9qrSvrlrJZSv5bGXJZFd7ZVkVgt8fzjImZkos8sZzs/m2dDkYsiXQNNhOV1gT2clFeVWQEdRNfIFhb9/fhINHQmBuVipEZFZkikoKmZZfNnIiTWlTOMywwtJzvqi3NBUwWQkRTyzSkd/aX/so+cCqLrO/k5PqeuqpnN5Pk5nrZ2//PYYl+ajHBlborXKxunp2How+ka//gwGg+H1yNhVbTAY3jRkSWR7i5svnpzj3p2NFBSVnR1uvnRijvfuauKxswG6alx0VjvwRXIc7vFilkUkUUAWRY5PLuO2ycwtr9BaZaPSZuL0dIyipmASS1mmC75lzCaJD+xu4Yo/Ta6gMLqYpK/OiaZrjAbTbG+tYmdb5U/d2XAt0CwoGqPBFEVNI68WOTqxhCSImGXxZ/J73sjWGuMcn4wyuBAHTWU4lCCYyGGWJG7uqWZjo5tcQeGyL4U/nuXjt3bxuSOTfPKOXuxmme4aF9YykUIRLvuiyBIc7Krh5l4Ppybj1LstfGBPC7IocHY6gVkWuam1kofO+vnA7hbaa7sYC2QBgT6vE0kUGAolQNDZcG2Q/InJKDPLGe7d2cQVf4otTU48jvJXda6KqpVmC2o6e9qrCMYL/Mnbezk3nWBPVwUPHp9jY2MFXXU2/uyJIf76VzdzxZfin49P82s7GpmLZmiutHHXjnq+cc7P7vYq8orCV0/P013j5OTU8nqn20Q2j6JpfOXUPKuqyvt2NjEbT/NrOxt5ciDIxkYXh70a/xEAACAASURBVPuqOT0do6GijNnICkcnlmirstPusaNrGmZJ5Hdv7+Jvn1E50FWDWRYZDSZRdQ1JEI29fwaDwfBzID3wwAO/6GP4qXzhC1944KMf/ehPvqHBYDAAoiiQziv4ojmeHwshiwLPjYSJZRR6vE68Liv+eI7dnRU8OjBPZ60dEPDHc1yaj3FyYpnffUsXZ6diXPTFaKws5/GLC4iCwC39Hr56Yg5XuYmJxSwmqVR+N7WUoqDAhbkYM5EM9+5u5uvnA/R7HZhN0ms6D0XVOD0dpdpuJpDIccemWjJ5jacuLWCSJJIrRaxmkWMTy2xpdL2pF9Fmk0Sbx8bWZjdum5Uqu5mB2SiyJKIDmxrtnJqKYbNIDMxGiWWLlJeJTIdzDMzFubwQB02gzm1BEAVcVhMPnpwlmlWIZVeZi61wZirOC+NLzEaz1LstPD7gJ1MoMhnO8M8vTNNb70BEoNphoagp/GB4kVi2yLturOOZ4UXaq204ykwcn4zii2b58kkft/RVv6q9gKIo0FhhxWM38fXzC3zkQAtDC1meHg5SJpmJZPJ47BaOji8xGkiQzClc9ieIpPMsxPPYzRKdNU4eOe9nOpzGYpb52lk/kVQeVQenVabdYyexkufjD13k7EwCWYYr83Eq7WZ0VWB6aYWmShsnJpfY2+3hyOgSR0bD7OyoxLe8wk2tVZgkkYVkDlWFIyMRvC4Lo4tpfMtZNja6mAxnuDgf52B39Wt+fRgMBsObzac//enQAw888IWfdDsj82cwGN5UZEksDS+finK4x8vBHg+6JmI1S0iiyJ6OSra3ufj62QBep5Xtre71zMfmxgpEBHyRHO/d3cjkcopQMs9bN9VT7bBgkSU2N1dgliTu2lbHZ5+Z4EBXNXOx0oD3y74UBUXDY7PSWlnOqenoa25uIUsiO1oreHEiwhV/aZ/X4T4Pm1sc+CJ5tjQ5+MpJH331b+7Ab81ame0dm0rTiQ731mAxlUosv3JmnntuqsdjL+fGVjePnllgMZXnQ3e2c2E2xUIiy317mrDIMhOLaWxmMx/c2469bC2jqvO+Xc08fj5IU5WV/d1VPDu6SEeVnY4aOwVNJRDL01xVzmQ4TVHTqHZa+M+3dTIayKBrOqoK87Es9+1uRtE0Hjzhw2J69f9Fy5KI22bhgXf0r2d8v3VVQNVV/PEcndVOyiSJf3j/diZDWTbUacwspWmvdjK1lMJqlritz4uq6bjKzfzxW3s4O51A1eC50UUkUcAsSdy9rZk9XRWcnU4STOSYDmf5zf1tWOTSMau6yuMDQdqrbRRUFZMoIUmlcRYIOp0eB+/f28zAbApVV1FVgYlwEn8sR1eNA4E3ftmxwWAwvB4Z76wGg+FNx3Ktxf50JMOx8WWml9NsanIgiQLPj4X56ul5mqusSMCpqRgzy1nGFlPs66rivr3NgMBIoDTHrbHCyneuBMjkFb58ah4dna5aB1d8KYqqytZWN6Kgc36uVI43uZQis1pgainDUCC5vt9q7c9MrnDd51FaHAsIOuSKCv/y4iz/69lJtjQ5sJeZ0XTY2uI0FtEvsbaXzG2zYDHL2K1m7r2pkccuhEis5Hnk3AKdtQ4OdddyeT6J1SxyoLOWgbkEp6djdNTYkUSBYCIP6OzqcLOQyDG0kOKd2+o4NrHE0bFlZFHgvbua8EVWmAplaKq0cmY2RnuNjf46N62VNp66uEiuoNJQYUUSBZqryjk/m+DLJ+cJJHPrZb0/yXrDlWx+vWkLQCKb5+tnFzjcU8OezkoUTWUqkmZVVXnsvJ/BYKlE9a4dDVjNIkVNLZVcigJ2i8yGBhuXfCnGFlMgaNQ4y1DV0rU2Hk4yMJdAknQOdXvRdJ0XJ5Y5NR3l+bEwRyeWyBUVZiNZ5mNZppYy3NLjRRIFJhczqHppwP3a12ZZoK/eSV+dC6tZXm9IYzAYDIafLSPzZzAY3pRkSQQdxhZTTEVS/O3Tk/zBHV385Xd9aBp01zpY1VQmw2nqXaVZez8YXuKxi/Pcv7uVXe2VfPtqAB3QdQ1Jgs4aG8+PLdGna4yFUkQSOb5xbgGv01oqb3NbuOKL0Vfv4L07G7jqz5BZLfBvp+fp8ToZDaWYi2b4zLs2YreY10c2vJJ8QeGmNjeqpgM67dV2zs5GODkV5WB3NbPRDH/1vQk+9fbeV71/7M3EbjHTXePg5GSMIyMhlF4NEZFQMsfv3d7F3z49gabptFbZMMsSH9zbAghMLqXZ1VGBrsPQQpLvXg0SjGd5QdXQ0ZkIZfnYoQ4UrQ27xcwdG71c8CVRNZ1urwNJEBlbTOOP5dB0CCdXua2/DpMocnt/3XWVfOYLCienoiRzqzw24GdrSwW/sbeV//bdMWocZQjAZDiDJIjE0qvc0lvLRDjDFV+MCkcZBUXlc0cmeN9NrVz1J4DS/tZ2j52//M44ta4yFhI58orK1fnYtcY3Olf9pfElZknGF81Q1FQuzcf5le2N+CI5osk8b+mr48Y2F5/8+iXwODDLIgg6XbVOxheTTITTvH9PM+OhFACzkRX6610c6PZwqPfVlbwaDAaD4foY76wGg+FNyWKW+e3DnWRWC/z5t4Y50FWNLAoomkanx8m2VjePnvfzjs0N+GJZ2j12piJpUDW2t7qxl5m5tdfL3q5KXhh1Yi8zsaejip3tFVz1Z9CBKpeV/nonVpOJzU12ZFHk80d1hgNpxhezdNc4+PKpeVRVY3dHJVtbnHz62yMcm4hgliQmlzJ8aG/rj5z7l8jm+fOnhmmsLKej2obVJLOnq4IfjIWYWMxwS28tN/fUIAkSXz7t43du7jIW069gbSyHompIosi+rkpOT8cxCQKjgQytFTZ6653c3FO9PtT8tg216yM8uqudvHdXAw8e9xFL5znUU0swnmdPR9XLmu24bRYO9pjJ5At85jujHO6pZUO9C13XEBForSznlr5qbt9Ye13dLhVV4+RUlEy+yFwky7tvbMReZkYWS41+gvE8HTUOppdSbGhyIJ4WEJGQRZFNzRWUSRLdtU5Ap9JexrtvbMJqkhlfTDGznKHObaG/zkWf18VwKIEI9HhtOC0W7trWwNNXFrl3dxOnJuMADAZjhOIF7tnZyFw8iyTBFV+KSqeF+/c0M+hPMxpK013rYFNjBZIo4C638LFDHeszKo0unwaDwfDzZawEDAbDm9ZaMNTiLgdB51+Pz5HMFHnPnQ14nTY+/e4NjAUyKJqKL5qlx+ugs9rBJV8KSPHs6CKqrvH8+CJ3bqzj6HhkvWPj1fk4LZXl7GyvAODR80HaPTb88RVaKm0oqoaqa7RV2RhZjHNsIgKAiMBIKEW/10W7x8ap6RiqrrK/s3o9CCw1e4lR4yyjtdLGc6OLNFSUMnuHe2qxl5nJFAo8cdHP229oYDaSve4SwjertaDjjk2lDpNrwZ0siT9y9MBLv/7YoQ4sZplPvKWHfFH5sVnbtZLTP3t7H7Ioomgaqq5ysLuaTKG4PuohX1B+7LiONQVFY2QxSb/XdW2khMrnj05z1Z/gf/zaZtyWMvJKgScHgjjtJgSxdB0Igk5fnQtV1winCuSKRU5MLXPHhnp6vE6KmoKIhCQKqLqGKAhsbS2No9jcZOe/PHyF3noHpybjPDca4kB3DUvpAr+5v52RQIZ9HVVIgsiVQBwQuDCXQNM1WqtsSKKw/ti+9HE0gj6DwWD4+TOCP4PB8KYnSxJTixkWEjn6G938w3PT7Gqr4olLfjY2uginVjnY7WQmnCWnFXGYy7h7Rz3fGdTJrWrMLqXJFWpYTGZp89ixl5np8jp4diTEX3x7DE3XuLXfy8YmB/92epXuPi9TS1mOjC5R7TTz4lgYfyxHu8fOJ+/s4bGzASbCaTpqbKi6xrMjIYYCSTbWuznUWxocPxxI4o9n6a51UFA1ArEc3dUKJ2eiHOqpZmC2dHt/NAfCL/gBfgN5aSByvUHJWpBmMcvXFbCtBe9DgSQFVeWKP04qX+Dhs34+tK+VuUiO+dgKb9lYi9Vk4mBP9b+7v7V9frmCwlVfnM5qOzOL2fXvb2p0MbiQIpNX+OLxOd6/u43J5RST4RWKqsrJySh9tZXMLGfQdB0NHVXXuKHFwUOnFxiYW8bjKKOxopxAPIcoCOzr8jATWWF7qxubVSQQzyGLIsvJHOPhFAVF5dx0guFQghMTy7x/dxuBRI5Gt4WhQIqLvmU8Dgs9tS72d/37mYEGg8Fg+Pkz3nkNBsOb2lr5p6JqnJqOcVObm2PjUUZCCVw2ExvqXWyoKzXAaKyEJy7N89f3bGHQn6bJXY7dKvHh/d1YywTu7m7godPzHOzxYJZFPv2uDSiaxn/9zijDwSTTkSx9DS52d1ZiLzPR5ilHEkR8sSx7O6oJJfOMBjJsaHSTzK1ybCLC/q4q6l3ldNe4GA2lKKgqt/TW0lJlY3I5hSQKlMkSezursFtk/ujObgZmk4yFUmxuqlgfZP5mnvX3epMvKFz2xwABTdexWySC8VXu2dGI02KhvVpiRSlQVFUkQSSTL+C2lfadrgV9L4wtoWo687EVbt/kxWqSUTSVLY2VFBSNoqaALjC1lKG5upyDfVXMxdL0eEsNa7prXBzqq8I6LZEprjIVzqDpcNEXp9frRNVUumod2MtMdNcW2d9dxdfOLHDvrkYGZpOIgkBblZ379zRzrMrGXDRLNJVndDGJSRTZ01WJ3WKiuaKczlobJtHEVCRJa5Wde25q4OxsjJ1tleuZ7LWM5w8HhC/Ngv6kPbAGg8Fg+MmMGguDwfCmt9b18XBfDXaLuTT24dri1izJzEWz9DfYWYit4LaXcXE2yXAgSVHXmFzMIIgqPxgO86UTczw3Eua7V0M8M7LImZkYdrOZm7tqWUytUu+0MrKQ4m++N0FqdZUnLvoZWUyQTq9yanoZTdcpqCo72lycm4tzU2sl00srBJM59nZV0FHt4KEzczw7ssgLE2F0HfZ0VnJLTx1npuMMLiQYmEswsVgaJ2CWxfVzM7w+5AsKR8cjXJmP0+KxIosSVpOJOpeFZ4fCdNeVMxFOcXxsmaeHFhnwRfmzJ4dJZPMksnmOjIZ5biTMt68EGAsmecumGh457yeeyzPoT7KxycGRsTAPHp+jvbYckyDS5XFwbjrBVX+C8cUMI8EUxyeXePCEj2RulSNDS6wqGvWuMqbCK6iahopeGrtQZ+PJSwtkCiq+WJZTUzEGAwlEQaStys5Dp/08dXmBJrcNSRTp87rpq3NhlmQkEVRd5/hElA1NdgLRHA2uch49F+DcbJTPH5vmH56b5FNPDvHdqwH+8YWpl3W7zRcUjowu8UffvMLTgyE+d3TyVXXDNRgMBsO/ZwR/BoPBcM1aqd+Bbg+bGivY1OBme5sTRdV47HyQLq+D2/vqmFvO0uIpJ5jI0+IpZz6a55N3drO1qYomj4VgIs8nbutkKpzl+OQyM8sZWirKsVtk7t7RzB++tRtnmYWtTaXf8b7dHdy5sZ6eWhdHxpaQRZE/eWsP4dQq7R47HVUOLLLMzg4XIjo3trpprbAhCgIDcwnMssAtvbX017uRBJF7dzeWGsB0VBn7qF5H1hq0AGxsdLMYX2VTg4tP3tnDcrrA5kY3o4EsGxvc/NahHt66sQFREFDRODIa4c+fGubSfBxVA1ESaPJYeXZoiZvaK7CZTVQ4zLgtZj71jl7u293GM4NL9NY7+dD+VuwWmft3d/CRA63017loqiynvcaGy1rGnZvr6fe6iGWLvGd7PTNLGQbnEzRVlPOtiyE6vXaeuhhiX0c1M0ul0tIGtxVfLEtXjZNKexl2i8y21ioO93mQRJFgMsfGJgdWWWZ3RxV2s0RbrZ1QKkerpxxJENhY7+K+vU0c6imVMk8sJTk+FVkffXJmJkZ3XTmxbIGNTQ6ml9IcHY8Y+1cNBoPhp2CsCgwGg+GHyJKIJAqMhVJ89ZQfURC4e0c9VrMMQqlNv1mS0PXSIjRXLDK4kOKGFgcmSaKjxsbQQoo2j42tLU5MosD79zYzGU7z3Ogij5wLUFBUuuuczC2vUNQUJFHAahbxOi2cmy11T9RUnUBihXt2NnBqOsolX4obmz1MhLLcu6eJxgoLqiowGIiX2uijc9kf45IvRY/XYZTIvU5JokC5ycQ9OxsxyxLfOL9At9fJfXuakcTSIPQ9XW4O9niwyhK/c7gTq1niYHcNG+pd3NJXzR/c3k0oscpdN9YRzSgUFI0ap5lj48t86aSPE9NLvHOrF7Mscn42AcDGJjvnZxNMhkuz/kqBnE6P14kvlmV7awVzkTz37W1iS7ObQ30eOqrtRJIF2qpsSBJ01Tror3NBacgJN3W4aK0sfU/VNWSx9Nqpc1p45NwCtU4Lj1/086UTPg501tDrdZYGvQObmhx8/WyAq8EY+aJCLLVKtlDk5FQURdXI5Is8eSHIu29sxCKLRJN5rizEyeSN7J/BYDC8VkbwZzAYDD9kLfu3uamCjQ1ueuucPD4QIpXP840BPwVVBUFH0yjNTJuPMzAX56HTCyTSeTRd43tXQ3xvKMCXT85T1HQsskxXrYPUSpF33VjH9FKa6aUVah0Wvnh8lsFggk1NDkyiSEHR+bfT8/TWO7lvdzOPngswtFAaiP3Rm9sBgYdO+zkxvkwmX2RwIcHWFicgcHUhSUHRjCHZr0OyJHKwp5pDvTW0Vpfz5KVFuuvKmY9lyRWLfPaZSTJ5hcGFBJ94+Ao/GF4iW1T4va9fYcAXZTycZDiY5PmxMJ/9/gT5osolX5yCovLsSIhnry5y3hdlLpJhPJjg4TN+cgWV4UCS5Wyejz04wEV/jBZPOaOBFI0V5aiazvOjYcbDSR46NU9jVRmXfGmGFhKcmYkhiQI6OkVN4dFzPkYWk6X7XEiQV1QePrPAfHyF1GqRE5PLJPKrQCkQnI9mmYmlcJebKagaT11e4GowxmigNFvz4bMLVNvLOD4WYSKcxW4zMX2t7FRRNWaWs7R5HPhjOQZmU2xsriCUyPPFE3MvG2ZvMBgMhutnfCxsMBgMP4LFLHO4r2b963xB4exsnPfuaMVqligoKqIIv767iY5qGwuxHPftaaLX62BvVyUTixlmohmaqiz4lnOcm41zS18NuaLCTHiF1mobUMoCHeypoafGycNnFhCAG1oc17IjpaBRFkXev6cZu8W8PpPuQLeH39zfir3MTKW9DI+9nDs2lbOvqxK7pbTHzyj5fP2RJZF8QeHUdIyP39LOty4tUuMow2qW2NlWid0i8aH9rRSOqvhjWbY2VdJV7cQsi4wsxtF0UDWdFbVIR3UN0+EsDZVWRARqXWYcZTL3vqWbB755lTZPaazCqqriW16hyVPOxnoXB7urGV9M4YtlaPc4EERocdtoqLTyxIUA9+9uRtU6uLnbQ15RuLIQwyRJbGh001ltx7e8wh2b61lOFbn7pga+emqOYCzPB3aXxjxk8kWW0gX+821dfPtCkP3dHpwWMxI6FlGmyWNlJpamtaocq0lmf0819+9p5r9/b5QN9S4OdJfGmnz0QDsAL05IgM7OtmpubHFyxZ82rm2DwWB4jYx3T4PBYHgFL51DZreaOdhTzdu21HG4r4ZdHRXE06s8dMrP8akIq6paarYSTnNyMga6TiyV59hYBF8sQ0FVySsK374SZDmb57mRMM8OLzISTNFV68BqlhF0naYqC599ZpKCojK5lEGWRD56sB23zfKy0QMWs4zHUb4+cHztONduZyyOX7/sVjOfelsv/ugq9+yoJ5pVyBVUvnU5QLZQKmmMZFa5d3cTuzoqWIivMB5Oc3kujj9WyrKN+pMIkkZXrZNQIgfAZDBLQdX4/tUwfU1ufNEsqqazmMzTU+vkLf11mCUJRdMIJfMUNA1fNMv+Tg9mSWYxkWdqKc2XT85TUBVOTUf5Py/Ocmk+wXAwSSCxwkQ4TU5V+PblANlikYfP+rk6n0DRNMwmgcxqkRNTESptMt+5vEhWUXnolI+8omKRZd67uwl/NE80lWdqKUOuoBJI5Dg7EyWRLrC9zbV+7a6NzjjcV8PhvloO99XgLrf8yNEXBoPBYLg+RubPYDAYrtNLF5z2MjM7WqvprXeyUXMxvZQBXUDQdaYjGdqrnczGsxzsrsYXzSEJIrIo8o7NDQTjORoqrABous6JqWX++K09qJrOWCjJLb213NpfC3Bd+/aMhfAbj9tmYXdHKZP7F+/aCJQyelPhFURi1DjKrnVuzQDwvl2NKJrChroK5paz3L65npu7S5np58aC3L+nhYXECpsb3WxrdfPfvzdGc4WJvV2VSKJIQVV4cTxCY0U5BUWjqaIcHWipsvHiVIS97R5mo6VSUEXTODIW5o/e2svQQpINDU421LkRBQFREOjw2BmciyGLIv1eN4Km01pt5/mxJTo8dn77lg7+8cgUHznQwKA/Q7/XhdUkU9QzDC6kuPumBqajadqq7ExF0kQSOSYWs1TYzQzMJrGaM2xrca9nutfGQQCcmFxme4t7/QMQg8FgMLw6xjunwWAwvAYWs8zHb+1aD7wUVUOWRG7uqSZTKPC10wsIuoAvmqOgKRRUlX95cZaZaIY/vLOH0UCWbGEVkyiTKyoMzCVQdR1V19nXVWUsbH/JrV0va3++MLbEixMR6iosjISSzMWy9HnddNc6yKwWeeTsAsMLKX7r5k4O91XzuRcm+fKpeXpqHTS5bciigHatAdGgP029y8qGehdnZmIUFJ35WI5d7VW4rGVMLaXprnUyHk5ilkUaXBYkidKohjoHuiYSSOS4PJ/kPTvq+ZPHBtnU4Cpl9yQJq0nmvbvakUSBvV0VgM5IKEG9q4z3bK/nyQsh6pxWBv1pVE3HbBIoKCqzyxlA4Ko/SeO1bqFtHhuz0QybGyrY3OAGILlS5FNPDHLHpjpu7q4plZ76U2xrcZMrKPzLiVlWiwq/c6jLGGNiMBgMr5LxcbHBYDC8Rhaz/LLSUEXVODoe4bELIe7d3UhntYPmChunJ6MUNZX2agdLyRUeObtAJl/kkXN+LvnjPD4wz/euhhgNpvDFspybTRjt7H+JKarGsfElMrlCaW7f6CIFRUPRNMoEiebKcubCGUYXk2RWi3zrcoBWj5XtLVVY5NKHAoP+BG2eciRRRNV1/vX4LCP+FAVFK5V6pvL0Nth5ZniRI+OL3Lmllu8PLbKrw819u5uQRIFAIo+qa4iCgG95hRqnmbmlHDPLGRpdFiZCaUYDGe7e1owoiJyYiFBQNaYjGSRR5Mh4mH89PsdQMIE/nmMuVrq22zw2+uqdqJrOUCDB94aCDIeSJFeKtFeXA9BR7UAAZpeziALs7HCtd9gdD6cIJlYYXEjx90cm+Z2HLnFqOsK/nphD1XTuurGOy744z4+F18dCGAwGg+H6GB8tGwwGw08pkytwajpKJq/wwkSY37+9e32R7io38cE9HdhMJra2OHl+vJxeb6kzp6aqdNXYMUnwm/vbkEWRr5ycZ09HpVHK+UtKUTXyBYVzs8tkVouMBFPMRjPc1luHKMK9e5oY9Gc40FtDn9fN3HKaOzd7qSi3sm1HJaemo2xuclDhKGNPZyUWWeayP4YsCnzkYAe39dciiyJDC3GeuhjiQGcNE0tJZsIrvOuGemRR5Isn5lhVVOpcFsySRE+dgy3NLv7w0Sv8zT3tTIRWyKwWONDt4aKvlB28ocXJMyM2bmhys7O9ArvZzN6uCk5OxpgMp/m927v57NPjaLqOhspkKEtnrR1V14kkc9zeV8/9e5oZDWRQtCxHJ5Z4S78XgNnlNGenE4yEEvTXu9jS7GI2mub+PU1c9qVYKRaRRZHGijImwmkkUWRDo4uxUKqUiTTL7OvyGK8Zg8FguA5G8GcwGAyvkaKWWtL/84sz5IsK/tgKiq7x2e9P4HVbCCXzdNY4CCZzLMRWALhzQz239teSWS3w8FmJuegKi8lVLvlSmGWJjY1uo+Tzl5SiapyYXCa5UuTY+BIFVUPTdQbn4+xrryGeLnBuJo7VJCOLElazRL27nCcu+3nbxgb+x9PDtNfYOT9nR1V1vnrKT3etA398BUXVuOxLEMkU+Y29LQiiSGe1HVXTCSRydFQXOD0TRZZgVVHRdI0ySaagaLwwESZbVIinVxnyp9nY5OBPvzmDWSqNC3l6OMRYMEVzhR1Vhb/8zjgNFaWy0omlFKqqMRZM01Jpo9vrYGIxy+xyBh1oqijn2GhpDuGj5/301LgBnXh6FXQBVddIZgtkVotcnouh6BpXFxLMLWY4PxdDVyXOTEXZ113DiakYt/fXsaPNxaA/hkhpJuKuduPDEoPBYLhexgrDYDAYXoO10r1tLW6KioooCLRW2eiosWOWZCbDKfZ1erBbZO7d1cjAbBJV07FeKxX1OMr5p/fvwGKSUVTtZQGfsZD95SRLIvu6PABsaXHw2LkALVU2wulV3rKpmoVElvlojvfubODp4SB9DTYeO5fAYSkFgk6LxIHOWubjWZqqyumudWCWRW7tq2VPZyUPHp+ju9bOyckY/vgKXTUOppbSKJrO7HKWfR3VTIezaLpOmSRz354mBmaT1DjK8EVWaKm1sb3NxaPng9y1tYF9XVXIoshUOMPdN9Xz0Ol5JFGkzm1B0HUAumucjISSFIo6giBgliS6auwomkK7x87Mcoab+7xIosDATJSuGjtbGqvo87oxyyLDwTS9DaWGMFtaKtEBURTY21uDRTKxtd3JkQkX/8/BNgZmk+zrquTcbAKzSeaDe1vWm8IYDAaD4foY75gGg8HwGuQLChfnYxwbj+JPrNBf76Svzs1MJMO2Nie9dS5enFwis1rkSyd9DAeTjIZSbGiwrd+H3WLGYpaxW80v2zto+OW19vyOBjIUNQ1XuYk/urMXiyxjliTu292ELIrUOCx843yAxopyykwyW9tc1LitWMtEQMAfzaLqGrmiwvHJZfKKhixJoAuML6ZodFtB0Vhp3AAAIABJREFU0GmrttFUUY4kCJhNAl21TkyiSFOVhUu+FIOBBP74Co2VFkySjKLptFSWM7GU5iun51E0jQ/ua2EkkGU+ngN0erwO2mrsmGURSRTxxTIMhxJouk5B0ZhcSuOLZpmKpJEEgV/Z0YjVLOOymxEFEUkUsJpLWUVF0xhZSDK6mKSp0oosivTUOrHKMgVV5dJ8kj+4VkY9FkpxYmoZgLt31GO3mNez7waDwWC4PsYqw2AwGF4lRdU4PrnMfGwFDYUWdzkHu6tRdZWB2Sh/9d1xuurKiaXz5BWVi3NRWiptKJrGX31vgqcHQ+vNPvIF5Rd9Oob/YIqqMRpMsZjK01Jt5U+/OcTnXphiLpblxYllHnhqBF8sS15R8MWyeJ0Wvnk+QJPbhlkulXMmMquMhlJMhjPsaHXzP5+ZJKcqIOjIgkhbdTkvjC1xZCxMMJGjs9bOWCiNJAoUNY0nLgTIFRUUTSOWynN8IkpBUfmr74wxFEwwtJDknVu9fOX0PMcmItzU5qbJbUXVNZ4dXuSJiwtsbnKwq8ONrkMwmUfRdOaWs7RX29aDuMZKK5946BLZ4iqCDpIgomoaqq6Vypzr3WxscqHp8NCZeVRNA0FH1TWGAkmevLTAH35jkM+9MMX4UpKxYJpkbpU/fPQq370S5J+OTXNkNGwEgAaDwXCdjLJPg8FgeBXWMg2qpnOopwaTWCrjVDSNucgKvQ1OzJLEkxeCbGisQBIATUc26fTXuWjzKEwtZShqCs+PRADY01G13jnU8PqyNooBWO8s+VqytC+9H4tZ5rcPd5JZLfBvp+d522YvNrOZ7loVgDq3BU0vBXG9XhcjoTiKVurKiV7Kmm1qrmBjfQUjoQQzyxn2dVZzZCLM5ga4Z1cDj18I8ft3dDPkz1BQFfZ0VvL8WJhf29l47Yh09nRWMhdZ4Z1bm3BZzQwFE/TVOTncW0MiX8BtMdPoLufhs3NsaXbhi2fp9jpo99h517Z6rvrTgEBzZTkf2NvKQ6fnuXd3IwNzCZorypEksIoyjZ5ybmiu4ImBBTY2OXjk7AIaOh/Z38aZ6TjLmSJ/+rYO7t/TjEUW+drpBQqqhlWWaaws51C3E1XXuDwf5e6bGhgNZLmh0c3erkrMslQKYl/y+BoMBoPhlRnvlAaDwXCdFFXj6aEgzwyHeHooyGQ4A+h0e51c8ado9dgZC6ZAAF3XEQWYiazgsJmZDucYCSU5NR3l7Vu9zC7lUNHpb7DzxZNz61nAtQDDyGT84mVyBU5MLpPJFcjkCvzjC1P84WOX+fsjkzw9FLyu52itu+eJyeWXPbeyJGIvM1NQNE5MRhkKJXh+NMyzw4uAjigIlEkS29qcSAiIAmi6XsqKabCcKbClxYGu65wYX2ZrW2lUwlAgyaPnAjRWWHh8IEhBVXhhPMKpyTjh2Aqf/f4EBUUlmMxzaT5Jq8fO00MhtrQ4WUzmmQpneG50kT/+xiCfemKIgqLispkYmIsRTeUpqiqarvO556cZWkjS32BjJJgmr2hcno9zajLOaDCNCoyF0hQUDXSdpy4EcdrMXJxN4otmueyL8eJ4FACv08LZ6QT/3zOTPHRmgaZKGyOBJO/aVs9ScpXJSIqjE0vEMwUG5mIA9NS5uOJPky2s8vBZHy+MLRmvGYPBYLgORubPYDAYrlMmX+ChU3P0N7qoc1vpqrUjiaWOgzvbKskrCs+OBumtdWI1mRgLJmi/1u3z13c3ctmXQtNLLfh/fXcjD57wMRLIUOcqQ9V0XpxY5spCjI31FVjNErvaK43On78gmVyBfzk+Q2eNg88fnaav3kFTRTmTkRQNFWU8PRRiX6cHt83yI39+LRA5Nh4hVyyCLvDMcAir6f8+nwe6q9nUWAFAj9eFoIOOTl+di5mlLPfsbOCx80EQBDqr7RydiLCxHsyywL7Oar51aZHeOie9XheNLhufelsvXzu9QHetk4KqUFQ1JEGkyW1lT1cFRyZs3NrrZWeHm/Fwkt3tVQDMLKVxW8zc2udF1TXmoznetbUBSRDxx1ZorCjHJMhsaqpgPpqnr85FX52LW/pqkCWRd2ypZzKUZUtjBWaTQHetk6FgDEXTmIpkSGeLvOuOer57cRGrWeIzd20gU1B48kKI5spyAokV+rwuGt1Wfn1XI7IocmQixFggw6GeWibCKVqrbHjdZcwu5eirMzMdyXDvrka+djrOjU2V7DdGPRgMBsN1Md4pDQaD4TpZTDK9DQ5kUUQUhGuZP2G96caZmRjoOlORDNvbnGxodLOnq4J9nVUM+tPr96NrpU6J8/EsLdVWHr/gJ68U2dxkZy6aBXQ2Ntg5ObVsZDN+ARRV4+xsHBDY3uZiLpbh0nwcXzRLjaOMyXCGSDLH6enYj3x+1kY65AsKuYLCd64GueSP8dCZOeK5PAVF48J8jMxqgf1dVXTW2tnZ4UIQBEyihCQKqLrOoD9NW3U5PV4nVpNMvasMsyxSUDQC8Rzv3OplPpqjqCm8OBHhqj9DX72TG1ocPD+2hKJpFNRSJ1pZFPnU23rZ2eHmSyd8XJiLcWx8mdMzUVqqyzk3G2colGB6Kcvbt3rxRVeYi2Z5z456FpN58moRqyzTXetAEgWmIxkUTeO5kSUevzhPXinS7XUiCSIIGv5YDlGA9+1q5B1bG3lmMMw7t9chiSJnZmI8ctZPW3VpSL12rXMowFdP+Tk1vYzXWcZkOI2GiiyIdNc6iKeLdNeW9hmuKgoDcwn66p389uFO7Fbzf+AVYjAYDG9cRvBnMBjeUNYW24qqkcjm1/++Vl738yybzBcVJhaztHlsbGqowCQKbGqy84ORMEOBJOilNviBeI4vnfAx4IvymW+N8sXjc1ycj1FQVYLJHO01Ns5OJ5kKpXj07AI2i8zkYpZTUzFEQWBTk4Mvn5rnynzCCP5+AWRJ5GBPNR+7uQOLLON1WljOFLhrez0jgSShxCrvvrGFQ701L8s2rV13iqqRyRc5PhWhoOgkMnlaPTbc5WaOj0eJr+R5bijIA0+N8J0rIb5wdIovHJtjPr5CZ40DSRDRrnXyHAulGQklKGoK4XSBTU0OZpczNLjLeXwgSI3TxJdOzJHMFcgWVikoGg+d9rOqKCiazosTEbJFhQeeGubPnhrmL741xkwkja6V9q1enk/yzQt+WqqtDPoT5FWFv3t2kqlImoKqcsWXYrWocHw8Snt1qcOnqmugw+npOAVFJZEtUFA1joyFS2WegIZOQVP52+9N8NTlAG6LmT97fIj4So4r/jgDszFUFaA0IkKSoKfOga7pXPEnGV5IUeeycmIqRmuNFavJRENFOQAnJ2Nc9MUYXEgiiQIWs2y8TgwGg+E6GfVEBoPhDWFtUX1yapmdbZUcHV/m4XNz/PWvbuLyfBJJEBkKJFE0jW0tVUgibGly4nGU/8yOwW2z8Nlf28xjAyFu7a9BEgUuzCZRdY3795Ra9P9gNMh/urWTpy4ECaZW+ZO396JoOlfmk6gqNLrLMUsS+7qqGA56aa+2Y5UlWjw25qM59nZ4GPSn0VWNnnoXUGo0YpR//sdTVI0HT/pYiK/wyTt7cVtM1Lis3NxVSyCxsn6btefn2HiEzGoBVdN5bnSRpkob/V4Xd21tIZhY4bZ+LwBTS1lcFplP3N7FsD/Njs4KtjRWIAV1zLKEqmnMx3NoemmfXyC+Qo/XyS09tYwESoHf2FICsyhjM5vZ2+UBXeDhsz42NbpLB6/rWGSJ37+jm0fPLrC7w8O3Ly3w6V/ZxNX5JJPhNAd6qjDLEgVNYTKU5Z7tLRzuqyavKHz6WyOIgkBB0RBFgYM91QiiRq4A08tpfnVHIxOhLIf7qgG4sc3F0fEICDqSUDoHsyjziTtaefiMD7tFZnODi5nICpIg4rLLSBJIgkBbhR1N19B1gf4GN4OBOJua3NitEp+4rZNHzvrpqpEwCSKT4TSt1VYq7WX01zu5qa2CxEqeK/4U21rcr1iGazAYDIYSI/NnMBhe90oD1SM8NxJmYC7KqekokgjdXhtfPeXna6d9JHOl7MOV+Tj9DTYyq0V+9+FL69nBnxWvy857b2rgij+NqmmMhJLoGlzypbjoS1LntPL9q2H6610c6Kph0J/mkbML/GA4zHOjiwCMhlIAyKLIqall7tpRTyiV5+1bvZybi5MrKmxodFNQVZ4bCfOFF6fJ5Ao/0/MwvLJ8QeHIaJhT01H+f/buPLy1uzz0/XetJcuSNcujLNvyPO0xe8ie984ICaEQQpqGEEIYSmlvn6f39LaHFtpToD1tT9t7enp7uDxlhpAJQoAQyESy53nenkdZlmXJtubZ0hruH9p2Ay2FXhISYH2eZz/b2xos/bS89u9d7+/3vnduacS/nOHvnxvniyfmuHODlzs3NdHXZAfg+eFF/vw7wyTzRbLFMo+enuOF4TCqBpqiMbmcZiGep7W2hsBKjrlono66GqoMEtfmUzw3vMjZ6QT9XgvLmRKKqiKJAi1OM1WCCAKoGkxEMiiaSklWeWU8womxFVprzUiiiFESMRpE7OYqREFAEgQkSaC/ycHV+RSBeI6aapHBFgffubjIy2NLLCQLnJqJUpIrffYuB+PMLme4GEhiMhjwuS10N1iZXE6TzJQoqyqPn5nnxdEwM8tZ/u4HE1yeT3B6JsH0coZ/eG4K9fpy5o2tVoIrOWRV5ZvnFhgKJplZyfLgPh/mKgP3727FVFW1Pt5tdTW8NLLCN87Ps7HVWnnfwAvDER4/E+RyIM7MSo67dzbT73Ews5yn2WUG4F+O+vk/Hr3EGX+U//adEf33RKfT6X4K/VKyTqd708sWK9mU0cUUC8kCHzpQyYg9OxTiLYNu3re3lacvhBlosiNR2SslCSJbvC5MVa/taa5Yknns7AK+WjNlRaFKEOmss64XnFBUjT1dbmRV5S+fHeOWviYkQeBP7upjKJhlU6uVp84vIqsqfU12VE3j2xcX6W2w8/zQMr93SyfPX1vmjs0O/vypITa3uehusHJqJqa3hPgFkBWVY5NRrgWTlX13oRTv2dXBUrqIhMD+HjcAE5E0OzucTIazuK0Gzs8lANjS5mRLSy27u5xcCqTZ3Grlgj/N9HKGzkYrRknCaJB4YHcntw40IAkSQ6E4gZUC+7vqK/v9VI1qSeLBva18/WQQr9uMqmkcmVjh43f1AQKpwiqL8SKWJgOSILKry8mLo2EGmhwomkooVWBXl4Ovn5ynyVGNpgksZVY52GND0KCl1sR0JM9As517t7cTiFZeXyWrvkIoVUADREFgW3stlqoqtrQ5+Z1DXciqytdOVLKMe7pcFMplBpudFMpl/Ct5jJKBWwaaePhAO5cDGQY9Dvb1uLkUSFUCynCOW3obQRMYDaeYi+e4ua8RgLFQlh6PlZFQivZaCx880E621ArAn397lHdsbkZWVcYWUmzyuNjQbEfQNB7Y28r/eG78jThkdDqd7peKPoPQ6XRvatlCib94ZoRBr4VNLS5u62/ivD/JmZkEiqqBJuA0VfPQ3jaMBpGuRhslRUESBQa9jtflNfU02BhdzPD0hRC/sd3DphYnslrZc3RtIcGJ6SiXAim8dhPbOyoZoiuBNIqq8uiZIF6nma+fDjIWSTLgcaKqKof6arl3u4fnh5a5e1sTz16J0NtspbvBystjES7PJ/jskRleGg2vL4FN5oo/stdRbxj/2un32NnQ7KCj1sqlYIKOhhr6PQ4OTyxxZGKFfLnMyak47XUWwqkiqXyZqaUMolAJEE0GA4qq8tiZBcpqmbu3e5iL5lE0lRs7nEgSFGUZo0GkShLp9Vg4MrXM1YUEo+EUqqZhMhjoabITzazS22ij3m7kgj8Fgko4UaSsKigKlT14QMv1PXGzKzmaHZXlj6oGC4kCkiDRaDcyFy3QUmvirD9Ba62ZnR1OrCaJzkYrRyZWyJZKTC5l+N2bu9juq2WT18n9uyvBVyS9iqyqXAtmkVHZ2eHkxFScJ69n7OZiOe7f5WVnh4MtbW4uB9KUFBlFUzkzk+DyfBxFU8kWZcaWUoyGk3Q3WtE0DUFUGV1KMBJKoagaC9cb05+cjvGxb1zjm+dD/OldfVhNBra0uHj3dh9mY2X59KZWJ1ZjFV21dv3CiE6n0/0U0ic/+ck3+jX8XD73uc998iMf+cgb/TJ0Ot3rJF1c5Zvn5lnJlKmzGamSJCYiGazVBqZWMpRkjSvzSZKFMi+NLnF6ZoWl9CptbgszyxlKikabuwZRFH7u1yIrKqdmYuxsd7JaVmlzW3jm6iKxfIEXhpbZ6rPzyKk5BFHgtsF6hhbSDIcyyKpKqlDGZjLy4kiIaqPEu7Y38/LYMndvb+aFkQhGg8SxyRg+dw3JvIIkwvHJFXobrVwJJPndW7uZXMyymC6wpcXB0Ykof/2DMaolie9dW+CiP8GVhRQbPDaMVdJrMPK/nlRNwx/NUVIUqiQRUdRIF0qMhjM0Oox88dgcoqQxFEiymC7Q6rIwHc2ymCgy0GxnLJLBIMFLI0s4aqpYSOQ5PLbM5fkkgqgxGs5QVlWePBvgnD9Bq8vClWCChXiReH4Vt9mIigoIaJqApmm0uSyIIrwyGgFRoMluwmWpJp4rsZQpEojl8UfzBGI5ZFWjrKoEE3nCyRIeh5nhUIrf2t3Ck6fncVmNTISz7Gx3c2o6xsxKjnC6gMtczUw0i9lg5OTMMt84v8DD+30cG4/yyOkAoihSXQ3T4QKCpPLdS4u4a0xoqAwHEzjNZl4aCbOYKvGDoQgtLjOTkQzn/HHOzK4gSgKBWB5Z0bg4H2d2KYOGwLt3elmIFQgli7w0HOEtGxsxGw2ML6XprrchIhJK59ngcfC5ozOAhstiZGo5QyxbYlVRmQhnUTR4x9ZmakxVP+0j1ul0ul9Jn/rUp8Kf/OQnP/fT7qdfItPpdG9asqIyHMry7u2+Sk89QWRnh4OH9/kwGiRAAzRkRWFmJcsf3dmL21rpV2Y2SmxscbGv+7Xt/1UoyZyeiTMUSrCUKfIHt3eznC5xoLcOq7GKd25tBQ0ePR2krCoMeOxs8jp5aG8btwzU4bZWI6sq3zofos5qZCyUpdlhZke7k0GPA0mCkqzij+awmQ1UiQZMZpFnr0RocddweS7Gsall9nS52Oi1saHVylAwRWudiQd2tXB8euVHqp/q/vMUVePwxAqFsoyiwFJmlWaHGZPBwM19DWxrreMGnwuf20IgnqPDbWF/Tx1L6VX2ddUhIrGqKKAJmA0GPnXPRg50NzAVzrK7vY75aIE7NjbT4bZwsM/NnRubef++Nt65pZUH97VRJYhoqAx6rUxGMkwspTBKEn3NdkSESpVMQaOvyY4kCLS6zXQ3WPHVWhj0ODGIIoIA79jmYSVb5NP3bMRpqmKbr5ZNzS5aXGaCyTxelxmf28LsSobxpRTNzkqmurPOitdtoq7GxEdv7uL+G328Z3cLiazC3Ts8LMZX2dHpYjaaRRIFNrU52dPjZFOrg4EmOy1OM5Io8OGDHdy5sRmXtZr37m6ju87Kw/t9vGVDE+/b28X/+dYenr6wiAaYqiR+56Ye7trkZWubAwGBiUgaSYLb+z2YqwxYzSKd9Raml7IsJPP0e+zc3NfIR2/u4lBfg97uQafT6X4GevCn0+nedF7dqmGHz0kwnueHo0tcCsb59DNjHJ2IAhoiIgMeO71NDqpECYMoEIoXubHLidlo4GBv3WtSJfPVwdRIKEWhLLOYLNLqNuM0GRGFSlbxC8f8HJ6MoGoasqqykCyw1WdnZDHF188EKcoyXXU2qgQRWVWJpFcZ8FqJpFd57MwCm1utTEayKKqGpmmkc2UKZZnpcI53bvNgNko4rNVMhHNkS2Wuzif41vlF7trk5fRMgiMTyzxyco5kvshLIxE+c2RqvfWF7mdjkERu39DIn93Vz0K8iNEgcucGL1taXCwkinz05i7eurGRXV0NfPSmLja3uDBKEguxPO21Fk5MRynJGqFkkZKiIKsq//vlGWaiGba2OjFXVz77F0bC+OqsnPcnGV1M8cjpeVKFVb5xdoGCKnPZH+exs/O0ui0sJAv0e62cm45RUhRGwyleGI5QVuVKoIfA7EqWKlHkUF8tgx4nAgLfuhAiX5L5Xy9M8ZffG6OlzozZKNHb6GAxWaS73sZcNMvMUoa5WI6rcwkeOTVPWdXI5MsUZRlZVRkNJ/nWxRAHuxuYDOdZVWQmFzOUFJmiLDMUTPLl4wGGFiqVbzvqrLxyvc/g9g47gZUcl/wpVODCXJLjU1EKJZl/emmaQlmmr7ESxF6YS/DccJi/e26SciXCRVFgdiXL0GKC4Eqe2ZU8g81O3r6phdsGGzEZDfo+WJ1Op/tP0Au+6HS6N4W1AKVYkjnrT6xXPdzXXccHD7TzpeNzdDXYkKhUDnxgdwvPj4aRBJFAIs/D+3w4LSYe+e1d1NlqaLJbfq4J4Y+X8C8pMqWyhj+epbPBws19jRyfXuH2wUZu62/CaJBocZs5P1v5nqaKhJIFzs0kmYtnuam3nkfPLKxXiUTQqBJEJsM5/viOXq4E0lwOZEADSYQNzU4MgsDbNjdhrjJQV1PNzHKYdreF3kYbF+eSuGwmOussuK3V/MmGXi74U2xurZT6v7aQZDFVILta4mowzaG+Bn2C/DMySCJ1tho+cqjz3/TxW7uYcKivfj1QPNBTy6mZONt8dg5PRtjWYWcskiQQrRRUaa+1spDIc/eOZp69HOa3D3ZweMyG1WSgpCiUVRVV0Xj6YpB7trUxF1Nx2KrpcNswG6Xr+/9Eujx2NjW7AAjEc8ws53jvnjaeOB2kx2PDKFVem9EgcPtg0/V/a2xugR6PhU9+e4T7b6zslfM5zZirRe7f08p8Ks8f3NbDty+E6G20M7GUxmWt5uR0jKlIlsuBBL+5w4fZKLK3y42iVi5uGESJuZUCm1tdbPI6qRIFZldyANRZjXz5RICOejO1FiPbOhzMRbPs7nSzu9PNV47P4XWYMIgiCBofOdTJS6NLBOMF9nbWcXgqwmBzpbl9Z4OFuZU8//y+Hfhc1vXPRD+edTqd7j9PP3PqdLo3XLZQ4ujECi+NRPjs0WmyqyVKskp2tcyxyRUMoshHDnXy1o2NbPPVstHrxGk28Y7NLbx1YxMfOdi53t9rra/fzzMxLJZknh8K8/FvXyOazVMoyVwLphiPpHFbDcxF8+ztcXGgpw5ZVZlcyqCoGv5olv5WB6IgMhvN4nVWytErqsrsSh5ZUSldX6KKJrCp1UWhLPPY6SBDoQSSCPft8mI0iJiNBnwNNZz3p5iL5Tg9E6erwQpoIGiMLqZpdpoIJvJs9zm4FswwHk5TVlVOTsWRBIHfuamTS4EU5+Zievbv/4e1jNLan1dnkV8dgJiMBowGEYMo4nXU8OTZBSLpIu31ZkpljeMzlUbr//D8OIVyJZt2eHKJbFFmKpIlki7SVW/BWmPAYNCuf84wG83S77WgaRoX55J8/G39OGqMIGh4nCZUDa7Mp1AASRAplMt85WSAS8E4E5E0oCGJAoWyzHcuLuK0VLGrq1Khs6PByvhihu9cCnOwq4Hnri6homI0SPQ12vG5a5AEkft3t7CtzYUgqoyF0xgkkRt8NmI5mc6GGnoabFRJIof66nj4gI/NLS56Gq0gQElRmI8V+at7NzOzVCBTLl3PgCuI15eF3nujl6cuBEkWSxweX6LZacJg0BCAfd213NjhZCFepLXWzOeO+JEVdf3z0Ol0Ot1/nn721Ol0b6hiSeYLJ/wUSgqFssx8LM9wKMXoYooXRiKcD8T47NEZzvoTyIrKLQMN3DLQgMloWP/7tWyAXmkkH2MolCQUy/LF43OML6YRgFa3iXPTcertlazG42fm+OyRGeYTOfq9FoKJAsPzKcbCaVRNA6Gyf0wSRQY9DnoabByZXKG9roZAPM/ODgfj4Qyz0SyLySK9Hgt///wkhbLMoNfC107Oky2WEag8D4LG8EKKUlljIZEnnCjSVV8JFCRRxFdr4cp8guHFJLlymf/y+BUW03mOji+RXdX7n72etrbaOedPoqHR77HxsTv7mI8WOTyxRJO9GlXTmAynkVU4MhZjJpxmeiXDw/t9fOzOPhbiRQLLWV4aXWI0nCKZXaUoK/yv56dYThV5bijMk+cX2OC18PLYEqPBFHD9uAAUFQ5PLHPnlkYWEwUCsTyKqpFdLfOtC0FkVeVATx2PnVngs0dmePrSAr66Gu7d0UwkXeDtNzQRThUpyiWmljMEY3mGQkmeOLPAKgpHxqKsygrFssxTF8LsbHdyfCLGWCSFP5bjlbEV/vr745RklelIlqH5JADttRZ+cDWCr97E6ckYbrORf3h+ioKscGY2zjfPLWI1Gzg7G2MqkuaH4xFeHFkill7lzEySK8E0D+1tw20xcdtA02v6u67T6XS/jvSzqE6ne+OpoGgK87ECv39bN9+7FOb+3S189WQANCjJMn2eGr52Zp4P7mtfnwC+Hlf/DZLIvu5askUZfyyLKMDdO5sZD2UBgX099SzE86ga2GoMGESRPZ1unrkU5ua+BsYjaTZ6HSiKwItjIdA0mpzV7Opycmo6DprG3u5KKwAACbhzYzPbOxwMBbPkSmVmlnPc2t/IZ953A06TkVfGlzEaRPZ21VOSVe7c2MSBPjdfORFA0VQ+f3yW9+1p4+xMGU1V6aqvFMcJxLNYjRJ7umsry+tgfSmr7meztt9z7VgrlmSsZuN6JlVWKvv3/Ct52mrN+KNZREHEKEk8sKeFx04v0OwyMbGcxmky0FlnIZIu8t69HdTbTJz3pxheTHDfrhYKcplBj5PJpQyb25xs8brYfsDBicnY9f6AIiaDgbdvbqHHU4PTVMV5f5IqUWRXl4PvDwd5+kKIW/s9TC6lkESB2UiOAa+NgSYHL08scWt/A0ZJ4j27W3n6/CKzK0HKqlwpQGSrZmY5T2edFQkLujEgAAAgAElEQVSB/mYHJVnh5YklDvTUE4hnAbh3h4f//v1xPE4T3Q0WgvE8JUVB0TS2d9g51FdL95gNSRQwGkQ8DhPX5lM0u02Yq8X13xNFga6GGjRVxVJl5EMHeliI5+lpsiOJArcNNqxn+Q71VYq56Bk/nU6n+/noMwCdTveGMhkNfPhgB6dmYnTUWfnMyzMoqsp5f5xQIk+jw8TIQpInq4x8aL/vFxK4GCQRs1Giu9ZGWVP5s28NcUNbLf0eKzWGKhQ0+hptlV5k9XZemQzT5qrBKEmEU0U2elUkUbq+zFPl9FQUUZVYyazSYK/m0TMLaIqGjIKIiKKpPHV+kRZ3NdmCzDu2eTg+FeX71xbxusxcDSZ4zy4f2VKJb18KYjZKoAkEk3lWZZWhhTiqCu11ZlIFmanlSr+5dLbMSCjDcCjN54/5+cB+H9+4sMhDu9v0Ihk/xVrQd2wyylAoyaDHjqKpHB5f4Y/v6GEklLu+DxSePDfPu7a3gCaSzJVp6anh2kICSRR4YE8Ln35mjLKq4LKbCCeL3Hujl398cYo/fVsfp6bjXJyNUSwrjIbS9DU6CCbyaJqGJCQYDie5GkiwqcWFQRKYWMrgdZj5s2/NckObi/4mOwPNdgyiSCpXJkCW9+1tY3Yls15sKJIu8f59Dl6eiKAoMB3LYjSItNdZGI+kCMTztLmshOJ5fLUWgok8Hz7UgUEUOTKxUqneKYE/luPzx/wMNNvw2k0MNDsZj6SQVZVALE+dtYq/e26CW/ubmFpOUyVKNLtMfOfSArYaA8lsZXn3n9zVx1QkzeRSFkmA7iYrRoNEvd3A27d4/t2LO/qxqtPpdK8NPfjT6XRvuMqeKYm9XbWYjRKbW61cC2Zpr7Uy2Oxke1sdN/XV/cJKuRskkVsGGjjYW4esqByfirGny8WVYJr79zgZDeXY5rMzsZRiT4+TI1NhNnld7Ohw8OJYmMlIjp5GK5l8mYEBBwNNDqzVVWz12Xnq/CL37/JywZ/mxbEwB7rrmFrK0t9kx2gQefe2Np69HKGjvgZREvjAfh/HJ2wYJYkL/hQbvA52d7o5709ya18TE0tp3FYTPQ2VTEtXk43NXheSKCIJAl31NqoNIgMeB1eDae7Z1sSpmRgAtww06pPqf4esqDw/vIgkCqCJlGSZkqxiNhrY11XHE+dCdNRa8EdzdDdYuXdHG0ZDZRw3tzqZT+QIJvJs8DoAUNFor7Uiqyo9jTaGF9KsygqnZ2LMRQv8xg0tOMxGNnvdIKh01lopyGVEofKcW1tdCKKAqql01NmYW8nT12ylv8nOzEqWrgYLJoOB/3HfFp46G+JKIMPkSppHTwfprLcwn8hxdT5Fk72yR/ShvW3IqspjpxcoqyqZbBlfVw2LqQIDTU7MRglZVTk9E2c8nAZgdiWPqql0NViQhMr+xn09biRRxB/L8u6dXv7xxUma7CYUTaVKlLhnZzNPX1jkbZubsVZXMRZJMdDkYCiYYUOLkz1dbmRV5YlzIR7e26RfkNDpdLpfAP0sq9Pp3nAGSWR/TyW4O9hbx9RSgb1dbra313LbYAN3bGr6hffwWivkYTUbuX1DI06Lia2tdv7ni5WCNMcmo1yci3NuJkmLs4bdXS5GQln2dtbR77FxsLeOwRYH87E8U9E0k5EMV+ZTeF0mRkJZbhts4ONv68NoEAnEciiaiqJqOGqM9DXZ2Nddy8ff1odBFDk+vUKqUGJkMUUoWeT41Apj4TQIGoMeB931NhAqxT2662zs63Gzv8fNgMdBKFmgt8nG/h43JVnjyycDnA/EGA6l9D6AP0EyX+Rrp/wML6TZ1GolmCgwHkmxzWfHajLQ5qpBkuDenc0YDRKKVimGUpJVjJLEoMdBs7OaUlnjkVNBJEHgXdu9jIUqgVRJVlE1jfFwlvZaCydnoiiqxhafDf9Kjgf2tBLNlOlrtLO11cUHD7bTXV/JNEuiwF03NDEeSrOh1Uar28yLo0t85vA0j50J0N9sZ0eHnXimiKxWXs+B7nrQhEoRmroaTk7H+JsfTNLqrsFmrOIT79zAcnqVZkc1CJV9gp/4zhBXggk6GywMNtt5/75WDKKEucrADT47CnDen2Srz0YiU+JaIE2dvQpREJhayuJ1V/PNcyEyqyWevRairMmIQmUZ6HAoRUlWOedPMBzK8vBeH1azUQ/8dDqd7hdAP9PqdLo3hbWJn8loYE9XLVazkUN99W+KbMDaz7eajNzS18jcSp6xcIZ0rkShpDCfyHNmNk4iX+SRU35+MBzm8PgKwwtpmuxmRoNp7tnZzGgozTfPz5MtVnrvPXo2yEgoRVlTGYukK3u9WisFXP7ye2N8+ntjfP7YHLlSiRNTUTrrLbS7LMws5WmrNfHSWAQEjTa3haOTy5QUBYAvHp/jS8fnmFzKcPf2yhLSE1MxsqslLsxGCcUKtLrNXAgkf22qgK4t43z1Xr2fdL/z/iSoGu/Z3cJoKIeqqWhqJdgplBReGV/i2ash/v65SRL5Ak9fCJIvl5leqiy33eKzMzyfYmIpzYDHwa0DjfzgSgS7xcCuLidz0TxoGqIgcGOXk1i6yKVgjD/55jUKZRmTQcLrrPTkU1T46skAhyeWKcsqw6EU//jCJKtlmX98fpIj48t4HSZ8tTUcm4yuHwMi0FFfw2g4xbcuBlFU8DlrkESB8XCGBpsRq8lAs7OGv3l2nFxZZj6e54WRMM8NL3J1Lk5RKTMRyfLy2DJQaUGyqdXGY2cWCMbzjCykrrccqaZQUjg1EUPVQFEUjk1FkVUNgyjishqZXc4TSRfZ6qv09BsNp1BUjR0+p74HVafT6X6B9DOuTqd703mz9vEySCJ3bGritsEGZEVlY7MTSRQQr1f1tBiN/NaNPlbSRVRNRVBVqqthY4sTq9FAd4ONyZUUklR5vlJZRRAEWpwmRCqVEZ+6GObe7R52djj5wjE/7XU1zCeyHOqrx2E28sBeNxf9lWba0VSRVL7Mxfk4ezrrkEQBX52ZYKzIPTubefzMPMMLaXZ1uAGYXs6BpvEHb+3h+1ci3LnR86Yb49fSqwO+41NRJFFA0VQOdNdz1h9nV4cbk9GwXsQFKp/xzg4n3x+uwWmqZm9XNVfm43Q32FBUjbImc/ugB9DY1+PmUiDNO29oZSGeo8djQ1Hg6nyK/hYbW1td7OxwcnI6hq/Wgj+exSCK3Hejly+fLNPfZMdqrGK7rxZREHjH1mZcNWZOTsUJpQo8uLeNK4E0kiBy20ATI+E4Ax47IIAA/+WOXqbCeSRR4GBvHTf11zESynHheg/IUHyV7no7l+aibOuwoyJjlAxIgoAqVAK5p84vctdmD0aDSDCRo9lpRhQEWt01bGl2V5a+omIQRRrsJoaCGXobbTy0tw2DKHJ6Jk6rqwazUaK9wcIH9vsYCmbpWS1x+2AjJ6ZivH+fD6uximypzPRSgft2efnbH0ww6HFwfCqK0SCyq8P9C8/u63Q63a8jPfjT6XS6/4RX9xi7a4sHWVGvT5A1SorC8yNh3nVDC8FYgYFWJ7PLeRYSeb58co6SrJHKlUETyJZKLKby7Ouq58jkMh21NowGkXu3e/iHF6b5/Vs7CaUKiAj4XBbMRolsUeZvvz8BAhzorcVRU8U5f5wbO9z4o1nmYjlimRJb2pw8eTbEhbkYJVVhdCHDb+300V1v5dLcCtfmU8wn8m/sQL7Okrki5/xJhhcSlFWVxWSRQ70NHJ5coiSrjIczXA4k6Gqs4fD4Cv/tNwaps9WQzBUZDmW5Y7ASGBfLMiICsytZVhWF45PLfPhgJ4vJErcaGgAIJnMMNDspyQqvTCxRKMsEY3ke3NPB107Nc342ymCLg2aHmc8f8zMXyzK7nKNKELGaDPzeLd0UZZm/fW6SP3prA0+cDeF1mPnGuRAAG7xObvDZ+ObFABuanWxpcVIlCPzg6tJ69VuDJNJktOI0mzg5HePurW0c6Kmt9BScivDIqXlOTa/w+Yd3cIPPxie/O8Kjp4N0NdTw8vgKLc4akplVOmqtSIJAtSgxs5Klva6GhWSRF4aWOT6xQqfbxkp2FUVTMVcZKJQUFlMFuhstVBsMDC2kKZQ0vnN5gZ2dLl4YCfP9oRAtrhpW0iVuG2zk4txqpRUKMBJK0dlg4Usn5/jgvnY9ANTpdLrXmaBdPwH/stqxY4d24cKFN/pl6HS6X2PZQolTMzG2+Rz8+XeHuGNDM6WyxpGpJW7tb8QoGdjf4+bEVJzhUIIH97bx1PlFMqUSG5qcvDQe5hN3DfDU+RD372rh1FSChWSBFqeZvT0uLvgrZftHQikKisyGJhczKxkURaG11sIZf4wmu+n6hFpjc4uLXZ0u/uKZEW7rb2I8kmZzixOjZOBCYIUdvjr2dLlxWkxv9NC9LrKFEn/2nWFu7mvAaBApySr7etwYRJEvHPXT3+xgKJQAQFZVhubj7Oxq4D27Wvi/X5jiY3f2cjlQ2Z83sZShr9HOgZ5aAF4eW+FQXy2mqkp/yWyhxJdOzPHA7ha+fjpIk93ETDSDoMGurjq2tNopygpPnVuko97CVCSDikZ3g523bGj4kWXN2UIJq9n4b/Zirt3+/FAEs1FiX3ft+m3/3pLJtSWta49bC4SzRRlHTRXZ1RKPnfKztb2WvkY7L42H+did/Tx6OsD797WvFzQ6708xEkqSKZWoqTIysZRkY7OLe3Y0808vTXNTbwMzy1lKioLRIPHA7hacZhOHx5dJ5cu4rUbi2RJHp5doc1kY9DjY1eXkkVNBSorM7xzq4kvH51A0lUGPE0eNkf09db/S2WidTqd7vQiCcFHTtB0/7X565k+n0+l+DrKicmomzuhiim0+Bx0uK4qqEYjlaLBXY5QkymoZgygiibCqKDx+doF2t4UjU2kGmhQ0TUNWNUpllb96dhyf28J7drfwxJkF5mI5VhWFD+zzUZI1Xhxb5AP7HJiNEulikXCyxI0dLuzV1UxE0pRVFf9Knn3dtVRf7yW4mCpiMea470Yvz40somiVZXxrQUK2WMJq+skZl1+myfjaeyqrCpIocIPPzt98f2K9Gmd/s53dXU6GQnH6Gp1MLad5+9ZWrNVVPHo6SHa1xKmpBFPLafo9dvoabRzoqV0PyowGka+dDnDfTi9NRitWs5GPHOrEIIn0NdkZCSWpliQe3u/DZDDwlVMBBjw2PnywA5PRgLzxXwOzHx/XtazXT9oDd8empvXH/kd+/HanxcQtAw3r43NyOsb79nWzv8fN6ZkEoiBwwR/n2nyCR0WJfo+Nq8EMB3pqKckKj5+d42/v28xTZzUe2tt2/bY6bh1o4K0bmyiWZU5Mxbgwl2R3p5uRUIrZaIbbBjwEE3kO9TTgMBtRVHjszAKyqhGM55HVylgE4wW2trrZ3en+pTrWdDqd7peRHvzpdDrdz0kSBfqaHJyfSwACs8s5ml3VPHM5RpvDyuPnA4gYUFSVU1MrHOxrxFwt4nWY0TSBeHqV//n8FAd66ilrCr1NVoaClYba79jm4RPfvMqXBY3BRheqpvHkuUW8DhPPXF3kT98+wB8/eYUPH+pksNnJ9HKGh/f5cFpMfPodGzjnT9JTZ+OB3S2MhHLc2tcImsBnj8yw0eugJKs8fi7Ag3t8GCUJSRQBjZKsXm/SLbG11b4eHK5Nzl/d+PzNQlZUXhlbYm9XLT5nDQCXA2m8TjObWm08fmYBQdNQNJVrwSTdDTZkVeOZqwts97mRVYhnS4xF0iwk84BAv8fGWX+Cfd21nJyOoqgqHoeJP3ziGv/PA1ups9WsB2s399dX9hWqGsOhLPt76vjwgY4fCfR+njF7LR5baZhev/71zf311++hsd1XR1+THf9KjsHmKkxGA/t63Dw/uojJIIIgYBBFCiWZo5MrWKurONhbz8VAiqsLCa7OJxgJpfG5LczEMihqZSn0WX+cT9zVz9Vgho1eJ4NeC//1yatcmEsy2OzEIIoc6KnTC7/odDrdL4B+ptXpdLqfw9pkei2j0uexI4kik5E0A14bi+kCzmqRfT0urNVGjAaJm/rqkFWVl0aX2NDs5Ddv9LGn281QMEskvcrGZmH9+UeCGezWakRNZFeXA0mEHR0Ovn4qSI/HSiCap7W2BkuVkZv7G7htsGF9Er2W8bmxw8nFQKVSZWUZpMZcPEtvkxUQqLUZ2dHuZDycZ6O3Um3066eDdDdYGfRa+dQzo9zUX4+5ysDB3noMksiJqeh6puanBSWvDhR/UtC49v1XV+F89b9/lsCnWJL53rUQg14roiiiqJVtDYqm8ejpIKqmscnrZH9PLYWSgtkoMdDkYDGV4z272xgL5agSBR7c28ZFfwqAmeUsAx4JWVEpyRqTSxnaay3UmATOzyW4fdD0I5Vq1zJsa6/5zRYgw4+Opclo4PYNjQDc3N+wPuZr97kUSCOJImdnEwSTeU7PJDAaJO7Y0MzB3ko13l0dLq7OJ9h6ven81FKGFmcNgVgOs8HAJ+7qpM5Ww6E+0/rP/3/ft2P9gsKrj1mdTqfTvb70s61Op9P9nNYm+bs6XHzlVICH9/rY3+PmKycCdDda2dLiwlljWq8WujbBvmNDMyVZ5ejkCvPxAh/a3052tYTZKHFjh4uSrFxv5VBDT4ONr56aR9IqmcZVRWEilOED+zqYDDv/w8zJ6Zk4AJNLlUqNRoNIs8MEWqXv2p0bvFwLZuj1WPjU98Zoc9fQ7DAxsphiLJzGXiMxHq60MQC4scO13uvQaBDZ4XP+SKGOtQqbBkmkWJK5EEiyw+fEIIkcnlji5r7G9dvXxuLkdJTtPicnpqMYJWm9KufxqSiKpnJT778GJj+p/YesqqykCnz1ZAAVjfFImp4GG6ChapXXXlZljk7EePpSkC1tLjY2O7ipp4lnry5x73YPl+ajPH52AU3VGPTaGfA4KMql69VCRbzuakqySjZfXg8uf/xY+GXz41nJV7+HWwYa2NtVqYp6a3/jjxxj6y1Qri99PetPsKvDhfn6ffZ2udf7Zf748756v+kv45jpdDrdLys9+NPpdLrXiNVs5MMHOtYnux+9uesnTqzXltwdm4zS7rLQ3WChKMt8+3KIe7e3cHQiysvjS9TbqhkNpeiqsxCM57mlv4EDPXUoqoaEwGQ4x8aWn9wrrViSeWlsiY+/ra/yGFHgBp+NZ4cWMIgZNrc42Nhq5alzIb5zeQFZVWm2m/nu5SDb2tx46008cznGu7dX9sUpKnzpxBzTy2netqmFGzucfOGEnw/v71jfF3dsMspQKEl3vZXpaIY2p4V/OTZLR62FJ87PkS6WODUVp63WQl9jpUH9SCjNmdkoQ/MJ3rWzlSMTy+RKZY6MRVnKFLkaTGI0SMzH87x1sImb+ur/TWVIq8nIu7f7mFnJEIznaXXVMLWcAQT6Gm1ML2V4aXSZm3sbcVmNiIiMLqYRBfjQgQ5OTscYXkjxGzc0sxAtVBq2e6388ZNX2Oar4107mvn9R0bx1Vm5b2cntw00/coHLgZJXB/n/6gS51pfToMkrmc/f9XHRqfT6X4Z6WdmnU6new29Oghby1D9pEnw2jLBDx/qIBAtcGEuybu2tmCUKmX293bVsqujjr/5zc0sJlbxOExMLWWQVRWz0cC29loO9tZzy0DDf/gz7tjgwWQwMBnJcCUY5/J8igabkf4mG+niKh/75jVaa020uWrw1dZgNVXhshl5cF8bkWSJu68XRLm5v4GdHQ7u2dGMrKkgVJZkTi2nOTUTI5krcnh8GUXVyJfLDIeTFEsKL09EKJZlSrKKXJaZXc6zv6eO9+5uYXwxxUgotV78o6/FjqKqRBI5Zpby/NGdvbz7hlbMBonf3OmlzlbFyGKKfz48STJX/JH3KisqwUSernobrS4zG70O+pscVIkiVlMV99zoRRA0DAaN2/qb+O2D7QjAfKKArKrMLOd4ywYPkUQJTRDY3GpjNJRjwGvnfXtbWYit8qH93fS9qlKn7l+9+sKGHvjpdDrdm5N+dtbpdLo3kEEScVpMPLS3jemlHMFEpWl3V72Nc3MJNrdamQznEAWBgSYHoVSRc9cLkNwy0PATl0C++vnX7tfbZGMpU6KnycrpmTjZ1TJHJqMoZQWToYreJjvVkoHtHXaqDQZMBglV1ZiP55iIZEgWivz+o5f462fHmI1kKJQUirKMJIr0eix85WSA50fC9HhqGJ5PEIzl6W1wIAgaKiqz0QyiKNJZX8PxyRVOTcdRgc56K9WiBAiMBlP8cHSJcLJIi8vMU+dCnJyN0uq28o2zC5yciNJgNXFkbJnPHpkhWyiRLZT+9Q1rMLmSZjFZvF6wRqCn0cag18JnfjhDg62a0cUs37kcIluSMQiVJbAX/ClWZZUXRsLcs7MZCXj0zALxbIlT03Eu+FPs7nTzru0t/N4t3Xo/Op1Op9P9UtL7/Ol0Ot2bxFp/t/W+b8USV4JpFFVjV4cLk9FAsST/1IDvJ5EVdf3x378axmgQKMkqE+E0D+1v44mzIQplmU1eJ4qq8daNTRweXwE09nS5AfjM4Wna3VYOT0XoqrNhNhrwOsyE0wUe2NXKsYkYZqPE1WCCAY+TfT0uDo9F8UezdDbUMB3J8uA+H18/Nc/a/z7VBomH9rYBcHgsitEgMRKK85GbujjvT1EoyThqjGzwWvjS8Tk+cqiTF4YrAWJnvZXj0yv85Ts3YjIaSOaLfO3kPO11Fm4ZqOf0TJzvXgnS1WCn1VnDQjxHj8fGljYHz1yO0OIyMb2UZVOLkxt8dr52IsAHD3Zw/nrBF6NBZIPXQp21Rs9m6XQ6ne5NS+/zp9PpdL9kfnwZodNiYn/Pj7ZY+HkyTq/ev/XWjY2cmY1zoMeJ0SAxHs7z4J5WvnJ8jvFwer24itEgoqgq5/xJrgYThBJ5VE2jo9bKYLOdXZ0uLgfS9JhsFGWFVyYi3D7gYeP1APLLJwIEYlk8TjNTSzlkNB4/u4CqaWhoDHgcqJrKmZkEI4spZmMZ/vztg4DGpUAKRdWYWk7z/n0+LgfSBJMFjk1GOTUb46beemZXstRZq8iWSpz1JyjJCl0NVsYjKRw1VWxqtfHF40X29dQTTFQqo05GspQVhc46C5IoYhArY3thLokCfO3UPG11Jt6+qUVfwqjT6XS6Xyl68KfT6XRvYq9X4GEyGtjfUweA2Sitt224ob2WXR2u9SqNaz3h1loutLpqODGzQovTzFQky8xyjrKi0u+x8dfPjrOUzKFoKmgC/mieAY+drgYLRsnA8EKchUSOW/oamVzKEk4VkRUYXUyxudVJUSlzbS7Ol48HWMkUuXWwkVJZ42IgDgj0NFoRgLKioKmVPoSdDRZeHA7zV98b45b+pvXm7AuJAu/dY+GiP008u8qxyRXa3BaOT63QYK/m80fDfPTmLm4b9LKnywXAF47OomoqTQ4zXzw+x029DdTZal6X8dfpdDqd7o2gX87U6XS6X2GyopLMFSmW5PVln2vfWwss919vE7HWs9BqNv5Ief61P9t8dvyxDLcPNHH/7lY+fLCDD+5vBzR2dbrwOE247JUWEmPhFA/v87G/p5b5WJ7dXU56m5zE0quMRzKEk0Vu7m3EXCVx5+YmtrS4qDEYectmD799qJ2b+htAEwjG8/Q1W+lptCIJIh5HNcFYkVp7FVNLWRRVw+s243XVoKga84k829udeB0mnjwXQlFV2huttNda2Ox1crC3jvfva2dvTy0mQxXFssznj89yeiZOQVEIxPOYjRIf3NuJtVrf1/ezWNtz+eoeja/+WqfT6XRvHnrwp9PpdL9i1oK8YknmpZElfu/rF/nnV6Z5fijMP78yybcvLvC7X7/IQiLD0Ynl9b58P/544FXPE+FLx+c4Or5MNFPko1+9WCm4UipzZT7O0cko4WQRgyCSLcocm4iSLZU550+iqnDenwRBxWk1YhAl2tw1HOhzs6oqfOP8Aj0eCxICK5kyp6YSHJ5YZnIpwzu2ezg/lWAolGR4McViqshvbPNwbiZBvlzmyMQy84k8fY0O/NEsJVnh4lwSQRTprreCoGGSDAw0OSgpCl8+7udvvj/O8EKKkqLwyvgyF/1xOhtrCCULxFJFRhczzMVyHJuMUizJ68HN2pj8+Hj9OkvmivzFM8Mkc0VOTEXXLy6cmIrqY6TT6XRvQvqyT51Op/sVIisqRydWGFlM0Vlfw+5ONyXZx64uJxf8afzRHMWygsko8MjpeaoEkVyphN1kYnenm8PjywBMLGV4cHcrj51doLO+sk/uoX1teF1mllKr7OmupUoQuRJIM+h1MLeS57bBRhQFQsk8/3D/FqaXCkiiwAcPtHNiKsbUUg6DINHbYCMQzXI5kMZmNPLX797EC0PLdDdZAZAkKllBo0RdTTV7e+uQRAHQEBFwmAz8zsFejAaBfq+VP3vqGjd2OZBEOOdfoSgrSAhIosD4Ypb2WitmowFJhP/90Hacpiq+cMzP9nYnj54OUmMSefpiiFZXDV6nmS0tLkqySklWODy+zIsjET7x9n6eOB/iwV2tXAykKCkyB7rr14vw/DpW/5QVlYuBFI02EwZRpCQrnJyOsa+7lpKsrAd/+p5JnU6ne/PQz8g6nU73K6RYkjnUV899O5t5fjjCF4/PcS0c57EzC5QUGY/DBEAqW0JWVFpcNRweX2Gj10p2tcSzQyFKikKprHB+LkGL00yhpKCqGmdnkpybi3PfLi/b22p5cF/r9YIpErKqYDSIBKJZmhwmPn/Uz0avlX3dtRRlmenlLP1Ndny1FowGkR6PjQM9ddy93cPf/2CCFqcZSRToaLAwHs4wsZymJKscnYiymCoAAgMeOzf1NfDEuRCKpjAeTjMUTGK3GLkyn2J7hwNBg8lIhtZ6Ezs6HFSJAvftamFPl4tsUeaffzhNslgmkMjzyKl5ZFXFIEmIiAw0OTAbDPR7rbw4ushYOM2mVhseRyW46W2wcVw8iRsAACAASURBVM6fIFVY5btXQnzm8DTPDy/y8W9fI5rJr2e9Xm+yopItlNb/Xvu5v+hMm0ES2dXhQkMjWyqhqBqKqpJdLXF5Ps4PxyJ6BlCn0+neZPTMn06n0/2KyBZKfPrZMf7wLV1cC2aQFRWf28LTl+a5e7uXiUiG5fQquztruRaIAbCQyLO70815f4rhxSSxdJEtbQ6mwhlyq2VeGVshmlnlnu0tADTZqzk3k+Sl8TATSxn6Gu0omsJKuszWNgejCykWYnk8tmouBVIUyjKPnw1w38429vW4mAgnGY+kWUwWrheFyTHotbG3x8VfPjuOpqg0Ok0sp1dRNQ00EK9fp5xeyqKh4XNbOTK5TIvTwlQ4h6ZpjITSlMoaNksV/miW4xPLTC9nUVWBP3riCts7agnG8pQUhX96YQpZVRAF6K63UyWKPLCnhaFghoIs8/fPjTOzlOVATwPfOBcimMhzeibOgZ46jkysMBHJEk0VaXdb2Nhi56snZvmXI37MRomNXge3DDS+btmuYknm8PgKL41F2N9dy7HJKF6nmQGPHUmC2wc9v9BMm6yqXJiLMRfL0eI2s9Hj5isnAlwOJJAQ+ejNXXrmT6fT6d5E9OBPp9PpfkVYzUb+6x09PHE+xNu3NGKsktjd4+LwZIRAtIAoCPxfd/QyFsqyvb2eXo8NRdU4PL7EWzd6eN+eVrrqrEyG86yqClPLOQ72NDCxnCQQKyAJIGsa08sZREHg/XvbKMoKz42E+OM7+hgKVpq4P7zfh+F6C4UjE1FsNQYkUeC8P4kmCHzoQDtFWWEinOfenc08enoegyjS4jTT5q5hJVPivXt8PHMxzKoi84d39DIVzlOSFa4tJpiP5/C5LHzwQDuyqvKp743SWW9hejmDIAi011rxus1s8boplGWG5+Pcs93LVDhHoSxzoLeWC/4UG1ttPHUuxKqq8OS5EP8fe/cd3ch9Hnr/OwVDgEQjCRJg722btL1XaVVcZVuWZcUtur6KXzs3iZM37eb6Ok6xr9Nz8vomdmTHlotkSZYtxaqrtr0X7S657CTYQBAgiUqAgynvH1hSK3lXWsmSrWjnc47OwRkOgJkBBc6zz/N7ntriQsZjcwQ8dsrdBXgcCu9dUUF0bp7B8ByaYdA7maDBV4hAKa1+Ny8FE7iKbCCYtPpdZHLa2/b5arrBvt4Ikgi/f1Mz50aT3NgRIK3O89LYLKF4ZrEUdWFUx6XejiBMFkXKPA4q3A5Gp9Msq/TymS11fGJTDXZZ/oXxJRaLxWL59bK+lS0Wi+VdxFtop8nn5GenQuxoKefCeIqmUifNARd94TQ/OxmiudxJk9/JcGSOjKYRjs9xfjxObziFYMKdG6p55NQIAIZhEkmo7F5SyuoGD3/7VA9/eEsbZ4IJAL7+ZA9TsxlODcUZjKSY13WODM7QP5XOZwUNA0WSwBRQNZPhmST7emZ4oXeSGzv8fP9QlFPD07SUuxmdnWN8NsPv3dzCU+emuGN9FV9/ood/eqaXGzrKMQ2RzpEYt6+rQRREjg/FUHUdEQGHzUaz38mJ4Sh1zU4ePzfGF3Y0czqYoL3Gw789P8j6hlIeOzOGJIhcCMd56nyIMrfCVFJla7OL4HQaw4BQbJ5oMoNLUUirKt85MMyOtnJ002Aup3FwIMqWllIGp1LkTJNkWoUyWFbj4n8+fI7trWV4i+xv+Wer6Qbdk0nqSgp58Ng4umHy0XVV/OlDg7gcNmyySEpVOT0aZ3lVvkRVN0x0M/8ZbGste8uDMadD4au3LUczDL78aCddE4nF3yNBFFhe5Xlb3tdisVgsb471bWyxWCzvIrIksrO9DEkUyORyALRVeFBkkaWVbjY0FXOwb4YD/VG+eFMLLwUTBGfSdAQ8SKJAz2SCk8MxRAG2tZbjLLBxXXUJkghnggkCbgdnggn6p1IA3NgeQDfAoYg0ljl5tnuSjoAn32kTE0kUuKmjAkXOdwGdTc6TM/LHtb6xmJ5QklK3nU0t+Vl7veEEPz05QUfATcBdxJ+9r51v7x+mdzJNi9/Jitr8fs90hqgrdSKLIjctqWBHW35mYedEDKdDosRZgCyK6AbE0jm+eFMrA+EMt6+pRZJgdHqOqmIHBZLEjlYPofg8d26o5txoipaKQv5pTx93rK/mwaNj1PoK+eSmWpyKjcfPhtjeUs5wNM2yai9rGzy8UOYkOJ3m1FCcElfB4tD4t5pdkbl7cz1Hh2apKXEQjM5xfjTJh1fXsLO9DM0wePhkiNtXV/Ddg0E03cAwTYZn0tSVFAGwrbXsLR1cn2/6EmNzs49bl1UB5mLAKQkiqp5vArO9rcwq/7RYLJZ3AOub2GKxWN5l7IrMugYvL/REyGo5ukL5DNmFUIJ9vVH6wkl8ToVHTkyQMzTARLEJAORMA10Hv9dOcHoOSRTY0OQFoHMijoEJwG2rKxiMpMmneAxUXUcSBVRdpyecIGdoqJpBdyhBztTQDRidSfPe66ooUhR2tpXjVBSWVHqpKXZwqH+G4ek09b4iBBN00yCVVTkdTDIRy6AaOoNTaXKGkV9zl8hSW1LIsiovkihwdGiGWHaezrEEug7VxQ6ODMwCJttb/AyE57iuxoViE9B1QDBZWuGlLeDGWWDjrvXVPHIyREbVePRUiC1NPs6PJhERqC0u5P6jo6TUHLppsLWtBM3QWVnn4vhQjMFoirrSIkZm0vz5+5e8rZ0/nQ6F1XUeDg/O0FheRNdEnAP9UY4PxTk8ME2lp4DO8TTN5U4+s6WOuzbWUFtcSFN5Ian5HC/0hNnbE1kcW/HL0nSDc2MxUvMqqp5/za5QnGc6Q3SF4lyYSJJ5i97LYrFYLL88K/izWCyWdyGnXeHmJRWIyJwbi6PrYJomz3SGqCm1o0gSNcWF9IVTzCZV2itd9E0meSk4y5mxWbrG4qiGjqoZ3HdoBFXLZ5FGp+c4PTbDPz7Tz4fXVNAdSvHDI0M8eW6CdC5HIq2S1TX2dE3RFYqT0TQeOj6GqunUlBTxny9NcHp0hue7wuzvi6KbOqMzGZ7tCuN32Tk8GKXOV8jT5yf59oFhdMPkhvYADknm/asquDAep6GsCI9L4cjQDMtrnPSGk6SyGg8eHafIIQEwFstwbjzOufEYA5EU8bkc3zs0wuNnJ7gQjiMgIIgGfZNJzo7NYpdlWspd9E0lmc9p7O0L0xtOcufGakRBYDia5vuHRphNzjMzl+PF7jDf3j/MqdEZgjNpgtNp2gJuvIVvfbnnq3mL7PzFB5bynuWVrK4v5SsfWMrKOhff3DvIs11hmvwOOififHv/MF9/ooe+aII9nWHuPzLM2dEEmVyOg/3Tb1kXTkOHb+4b5N9e7Cer5QABwzRZUuGh1e9iMJq2On5aLBbLO4RV9mmxWCzvQrIksnupH4DtbaU47QrR1By/+4NT9IZTIMDwdJrmMjd9Uwl+cmKcT2+upaLYzthshj993xKe74yQM7V8+ahg0uZ3Y5gGhilQ7rLhVGy0B1wMROLUlBQxOTvPR9fUE4ymcRTLLKvyohsmNlG8OKcPSl0KrX4nogBrG7wcGZilotiOiEA4kWVLcxlbW0t5vifMR9ZU0huaI53Ld/585MQ4BXYRuyzR4nPTGnBxbjRFJpdDkUU+sLqCYw9FUHWdCk8Bbf58MFd/MSv3sXVV/OiQTluFh+ayIuyyQmO5yLPdIVKqyso6Fxcm4tSVFjIez3DXhmpOBxNMxrN86f0d2GWZPZ0unjgzycaWUlrL3fRHEmxrKWNHuw+fs/BXVtq4kF28tJzyu3ev5ehAnIeOjROcTvOe5ZV8ZG0Vf/tUN9uayxmeTvGx9VUMhDNv2XHYFZl7djTyYk+U8ViGHW3l2OUYoUSGzS0lnB9PcXdH+Wuu+dN0Y7FBzcK5XPrYYrFYLG8dK/izWCyWd6mFm+eF5iNeh53V9T4MTAanU9yytAJMgeRcjsFIkv09M/z01CiuQhvPXpjkkxvqebE7ymRsDgMTWZDQDINQfB5JEDg8MIMkCrT7vdy9tR67Ld9lMqtqHB2aZX1DMUeHZllWVUzvZAIE2NVawVAkxZmRGC9ciDIUTdE1GuP6ulI+sLqCbzw3AKbAdDzLj4+OUV/m4D/2DXPXhjqGZlL0TaTILdMRhHyTF1XTeWk0RqvfzWOnQrRUutjXO8VgJE1HwItgwmA0SXvAw7nRFAgCOUPjp6fGuWNtLZIoMhXL8JXHupAEkV3tfoan0zQUO7HLMrphUldciCyKnAzG6YvE0Q2YjKnYpATD0RQnhmc41D/NVz+8/Fc+7P3SAMnnLMRpT9NR4cE0TdY0ePjewRGisQzD03MEp+f452f6+fIHluC0K29ZcOV0KNyyPMCONt/Fx/bFx1ta7Fd8n4XZhPt6I2xqKuXo0Aybm33Iksjenim2t5VbAaDFYrG8xazgz2KxWK4RdkXmd3a3Estk+f37zyCJ+QDqI6vrGZlJI4kimAZbmst46qUxnHaZP3lPG/cdGEYQREAgmsxxU0cFuzrKsNtkjgzOcM/2xlcEPU6HspiR2t5WBsCNS8rRdIMTwRhb20ron04yPJ3mro01NJe7kESBn5+ZZFOjD4cisaq+lPaAG4CaMgfhhMq6ujI+vakeX6Gdb7zQx8amYjTD4InOcRRZzJexRhJsbyujrrSI3Uv8HLDN0B2KI4kCumEgAjZRZmmVG4dNZmNTCRcmYjT5XQQjc6xr8rKrI3/Mhwdm6A4laPQ7+cGRUW5bVcG9++Isq/ZSU1LIsopiMKGhFFZUF//aO1ouXO+sqtEfTnI6mEAAAiVFfHpzLQ8fm+ATm2relk6ksiQu/g68+vHlaLrB3p4IGVXj6a4QaVWlPzyHbpisqvNwamSG9Q0lv/Jg2mKxWN7trODPYrFYriF2RSagOPm/n1yD066QVTW+c2CY5dVeVta5efqCg+mkyrLaYgam5pAEEVmSWFLpYUtLKVlNozs0t5g52tLiu+wN/kJWByCWyeJ12LErMhsaS9B0A7tNpqnMxcPHJ5jXdJZW5juEXgglWFblpTXgwqFIqJrB+5ZVs73Nh2YYnB9PYa8QOTMywwu9TkRTwjRB16E/kiSn6QSnMyhSPkspidDid6HIErph0BJwI4kikwkVEMhqGoos47DJNJa7+OL9Z1hVX0KLv4iB8BySIKBIMmpOpzeU5sOra9jaWsqPDo+yvsnDUCRFW4WLne3vjCzVQuC1MFx9Z3s+6F7Y9usOUC+VUTVUXcfnUhgIz6HqGpmcxncPBDk6GKUj4OGW5a8cWr9QDppVtcWupVaJqMVisVy9d85fAYvFYrH8yixkf5wOhXu2NwJwsH+atjIPd22s5oEj4zSUFTEwlUJEQBIFjg3F6J1K8pmNdYs325e76c6qGv/yXB8CMG/oPPnSBLcur+K6Gi+KLLKttYyv3rYcuyKTyqp890AQRRZZ11DMYCTNyjoXX3+yhz+4uZWHj0+Q0TR0w2R/f4QdreWcGMrRWumieyKFYZrMJrPkTB3DNBFFgRZ/ETbRxn2HRjAujh349OY6zo6mGIgkuGtDNd0TMZbXOPnbp/rY0FCKJAqsanDxr89nSc9rPHhsjJW1xXx6cy2yKNI/lWRFjYuvPd5zMbA0OTua5LPbGhaHqr+TLAR5lx7XOynwy6oaT3eGyGoasZTK1+64jn/e08d11SK3r6vi+NAUqq6/Yj2gphscGZzh+ho33zkwzLIqD5uaStnfH2FnW3596zvpHC0Wi+WdyPqWtFgslmvcwg3z9rYyNjeXYlfkxczRQvZuIYjY1up73RtsTTcYj2fY2VbG+sZiNN2go8KdH/tg6Oimwc42P7Ik4i2yL76XLInctaEauyxTUmSjJ5TmttUVfP3JHmyCyPqGErJ6jsFQhvMjcVbVl9LsL0I3DIYjcwBIosjA1BxN5YW0+l0sq3Hxf564wA+PjPEbG6o5OzrLof4ZMobGmWCCzU0+esJxbKKEbphsaS9neWUpBbJAW8DFj46MkTMMVlQVY5dlqjx2dNMgZxqomvGODPx+1RYCtFRGXbwer5eNsysyN3ZUcGYsignYZZFyl0LO0OgNzbGspoS+cIp9BRGWVDnpCc2hajoAB/pm6IskaCov4slzkzx0MoiqGQSnM3xqY+3bUtZqsVgs7xbX9l8si8VisSySJXExsFu4ibcr8uLjS3/+WpwOhT+5tRVngcLpYJJoah4Ek8byIoLTaTpHE7zQE14MLBdeP5bO8pc/v8CT50Mc7I1SXWrnkRMTCMDt66rY3xfh2/uGqPAWIAkCNd4iXuyJMjqboaXcxdp6H398azvtAQ8/OT7GS2Mx7j8yynRintriQk4HE8zrGk+fD7H3QoSXxmLohslELEudrxDdgGgqh24YTCVUJEGkvrSIidgcK+tcHBuaRQcuhBK8FJylcyL+ls3L+69I0w2yqsbenikm4ym+9Og5nr8QJpbOcqAv+rrXRpFFJuPzzCbmuXffMHu7w+zpnKK1opACSUJCIJrO8t/+4zjhZJrz43F0w0TVtXypr2Gyt3cK0zBYXe+lrqSQ+w6PXNOficVisbweK/NnsVgslrdUVtV48Pg4n9pYhyyKZFSN4WiamuJCBFHgzo35geo724xXlPUdG4phmAYbm0t4rifMY6dD1JcUoekGvkI7X/ngUu7dO0RxocKyGi+hRIaAu4DO0RiSBKvq3HztiW7+9D3t6IaBQ5HJ5HIIgNOe7wxqEyVuXlrJueIZOipcjExn2NZaxnB0jo4KNze2B5BEgZ1tfhRZRBJFdndUcnY0iaoZgMmyymKaypw4C2ycCMauuO7xl72GdkVeXNsGV26e8uuQyqgcHZpBN+D0yAynRmbxu+ysqHHx3YPD1JYWcrA/yubmy2eKF0aRrG3wcO++IVr9HkZjKba1lvHIiQlM4M6N1TxweIwiWWAoModNzJ9/XziNTRLZ1V7OtlYfL1yIciaYYCiaH27/TrpOFovF8k7ztn5DCoLwPwRB6BEEoVMQhL+5ZPufCoLQf/FnN1+y/ZaL2/oFQfiTt/PYLBaLxfLmLazBWsj+vFpG1TnQN81z3WF6wnGqigsYjKbRdQNZFGj0FZHNaXzjhX4eOzPGkcEZ1jV42dnmJ+Au4n+9t4Mmn5PgdBqA/f0RZFGktcKFbphMJuZp8TtZXuXl+rqSxTmCfpcdWRTpm0qwvMbJ2GyW1oCLlXUuBqZSmMCqBg/RVA5FkrhzfRUOm4xp5tcGSqJI50ScvqkkAJuaStjQlJ9XCAIjs3NkNZUD/VFUXWdNnfctHWCeVTVi6SzfeLGPyXiKb+0b5PGz4zx/IfyOyGhpukEqo/LN/QOomsmquvzsx/qSIkxMjg7OkpzP8fyFKVJZjYP9UWLp7GVfS5ZEnAUKhYqNTS1eihQFRZKoLy1CEgS6xlPcubGalQ2ldFS4aSwvZEtLKStri/nd3S3IosiRgVn29U3RORHjttUVOBR58Rgv/d1ceHxpIyKLxWK5Fr1tmT9BEHYCHwRWmKY5LwhC+cXtS4A7gaVAJfCsIAitF5/2DWA3MAYcFwThMdM0u96uY7RYLBbLG7NwA31oYPpiQAQ94SSf3dKwmHHJ5jTGYnMYpsnIdJqpZJbRmQwbGnycGo7w7f1BQrEMmZzGwFSSE4NR/vbO69jbG+Hhk2Osqvfy90/3Yeomuzr8IJicG41zZiTGmZFZllS5MEwDSZDomYwjCNA5nuTcWJyJeIZYVuX5rkkQBH5jQw1ff6KHgUiK1nIPXaEYj5yYIODOrwt74Og4mm5gItAXStFe6UYSBOZ1Dd00eO7CFHsuhKgrcfLRdVU8fh4GIxnSOZUXuyNgCkwksovn/8tknbKqxr+9OMC8ZnBiOIqmmTSWFfF89xSVHge6YbKz/bUHpv8yXm/I+qXjGQYjaZp9Lg72T3MmOMO8bjARm0MzTMKJefyeAiRRIDWf4yuPdfHlDyy57Fo8uyJzz/ZG7IrMH92Sb/BjmPkSYd2AB46McX40jmGIHBmMYBNlslqO3/7+Bba2lFMgCWxu9jEcTfOPz/SzvaWMF7qnePp8iJ3tfoIzae5aX8Px4VkujCdYVu0FYFtrmdUcxmKxXJPezm++/wf4P6ZpzgOYpjl1cfsHgQcubh8SBKEfWHfxZ/2maQ4CCILwwMV9reDPYrFYfs003SCVVfnm3gFEQWROU1EkmbZyNxk1x3MXpugNJzFMgxU1Xm5sD7C5pYR9vVF0HSRRYDCS4kOra3HYbOSMHMPTaf7oPW0cH4jRNZ6ibzLFrcv89IbSBNwFLKssBsGgbzKNCSyp8FBbUsj4TIbqYoHNLcUMRVNUFRchCSKDkRTbWsqwyyL15U5ayp3YZYkar4OOgIcNTfn9b19byYNHx9jSUsqOtjI0w0AWRY4OzbK6Lr/f//rZOZZVeekKJcjkNOp9hTx0bJymEhfvX13Bl34S5Q8/1MGznRE+sDKAphuvWeZ4NddX0w2ay50sq3Gh6jla/PnS0i/e1MJLwQRnx2ZRZGlxhuJb9bkulJa+0BNebMTz1PkJdrS+PL5i4Zwyar7zasCtIEmAKbGk2kOBJJG7eB1/76YWHjk+DoIJhoDPZeNkMM72tssPll8ob/3Z6UnuXF+FLIrcu3+Qz25tZG2DhzpfIbuX+PmNuRp8hQX8+74hNjWXsqzKA0D3ZILWgIt6XxFjsQx3rK3k3OgskihQ4bHz7f1DGCbYRJEVNS6+f2gU3TDZvTRglYhaLJZrztsZ/LUCWwVB+GsgC/y/pmkeB6qAI5fsN3ZxG8Doq7avfxuPz2KxWCxXQdMN9nROcmZ0ljPBGd57XRX3HZqgvqwITIGhaAoQ0AyDUCxLU3kRw1NzIJg81z1FtddBe4UTzTAITs9RV1LE4b5pNrSU8uOj40wmMvzxra2cCc7y5NlJvvrR5TzVOcn6Jg8PHB2jxe8GTAx0jgxF2dhYyn+engDgzvVV/OXPu6ktLqSl3M1zPZP0T6WpKymiZzLJcHQOURBBMDk+FKO21EHXeJp5Xefw4DSYIj3hBJ/ZVMd1NS5+cHSU21YGGI6kWFrt5uzYLCOR/PBxmyTyvpUBfnpyAkPXefTUBA2lTv7koXNcV+NFFAQyOY0bOwJXHQAuBF+HBmY4OzpLfyTJ4+cgHJtDFCQ6Am7+4ak+ytw2JhPz/NZ2z1sa+O3pDPNU5wSbmkt49NQEumGypt7LN57v4/TILEWKDRP4ws7m/P4XJvEVFXB2dJallV4k0WQmrfFbO2r5vR+EqC928bMTEwxPp2kud9E/lWQqqbK67rWPW5ZEmnxF/OjIGJUeB4+eGkM3wCaInBiZAVPkp6dH+Oi6akzDxC7LKLKIqpkMz6TRDINwYp4/urWVzvE0siSRUXX29kXQdIOakkJa/C5ODMUZnk4BvGNmM1osFsuv0i8V/AmC8CwQuMyP/uziaxcDG4C1wIOCIDQCwmX2N7n8+kPzCu97D3APQG1t7Rs/cIvFYrFctayq0TOZxDBNltV42NbuY3//FI0+Fx/fUM3fP91Lc5mLoWiKMrfCcCRDzjTBFKjy2AGTgfAcumkiYuIptPGbm5sZmUmzpNLNpzfXEPA4+fyuZvb3TVPtdfIXH1iKXZFZUulF1XS6Q/k1eAG3A7tNZlmNh+NDcQAM06C9wsXm5hIGp5LomLRXFDMwleDuLfUAHB6Y5dzYLOPxDP/jxmZ+emaOJZUeFElAyxl8c+8gBbJEQ1khXnsBm1vK6R5PUSDJbGotxWmXub21lH94pp+NjSWk1BIkQcBRIHJ9tZeOCg9doQRPnZtEEsSrKs9cCKovhPJjCxrLi8gZBrXFRTxyahhRyGdM/Z4CbKJIXbGMLL51wYqmG0iiQLlLYTSa5QPXV6FIEnZZ5u7NzYzH5rhtdQWPHM+XxtoVmVuWVqJqOhOJORRZBFNgV5ufYCRDY4Wb1XUlbGgq5r4DIzgUietrS7inwfu64xdkSWRri4+eUBKHIrGs1st/21oPwNgTGTa2eBmYimMTZJZczPhtbCrh2FCMaq+dVr8LmyDSE0qzudnHdTX5MR3bW8uQBJHhaBpJAkWSuKHDjyJJVuBnsViuSb/UN59pmjeaprnsMv89Sj5z94iZdwwwAN/F7TWXvEw1MPEa2y/3vt8yTXONaZprysrKfplTsFgsFstr0HSDE8EYd66vwiaKBGfS/PDwCLWlRbT53VwYT3FjewBPoUJDWRE9E0luX1eFYRog5DM0Syq83LmhGpsoopkGaxu8lDgLEAUBEHj4ZIisquF0KOxemi87dDryJYLbWn04FJkVNcV8ZksdBaLE4FQ6n83DZGNTMS2lbhRJ4vx4is9ub2BlbQmrG9x0VHqQRZED/VEkUaCtwsXq2mJ+cnycaDwLpsDWFh+3r69CEgRuX1tJdyhBSlWZSmbJajma/U4iCZXO8QRnRuJsbCwBU+DCWJxmvxNJEGkOFLG9zcf1NcVUeO2omsG+3sjrNmjJqhqdYzHi8/N0h1L0TCYRBQFFFvF5HHxyUy2KLCEIJp/aXEuz38mhgem3pGmJphscH55lVZ0bmyhRXeJgKJoikZ3nvsMjSJLJ7WsreeTEBOmcxqGBaQB2L/Vz87IAH7q+li3NPgYjaRyKxM72cv7uI9dzy/IKfK5CPn9DM7uXBtjVUX7Vc/ecDoXP7Wziho5yOsq9nB1N0R2a43+/r4Oe0ByiKKLq+Vl/3aEEJ4MxWisKGY9lCUbnWF7jZXNzvvPqhVCauzZUMz6bxWm3cdfGahRJYlWdm5HpDP1TKavxi8ViuSa9nWWfPwN2AS9ebOiiAFHgMeBHgiD8A/mGLy3AMfIZwRZBEBqAcfJNYe56G4/PYrFYLFdw6Y2xbph0jqdo8Ll4+vwEwsUCDlmQkEWBpvIi1jV4OdA3Q6lbQTNMDvRGWVbl5VOba/nB4VEGplKous7Bvigrq33saPMhiQKbm0uBaDtozgAAIABJREFUl9eUvTobY1dktre9/I98K+tL8uWZh0aRRAGnXeGeHY2vyLJlVJ0/+PFLrK4tIZ1Tefj4GH/1keX8+FiUAz0RPrGxHs0oBuDFngjfPzSE12njcN8sz1+YpMlfhM9ZwN6eKNvbffmSUr+TrrEEI7Nz+D0FFLvsrGss5nsHg+zvmUKSBLY2l/H4uXFEIc5C4cquDv9lM0z5pjkzzBsmh3qj/ObWRiamszSWO+mbSqLrBmdG4mTmTfZ2R6gvLeLR0+Osqfexqs7DS6MJtre9+bJFWRJZW1/Mvt4Ig9EU/ZEU08kMoViWL9zYzF882nUx2wvnx2KsqCpefJ4siezqyL/3PdsbL9vo5s02U1l43ud3Nb/iWl1X4+Lk8DR7eyNsbSlFFASWVjn5+pO9CMBnttThLFAWn7+xKf971VzuZFWdmx8dGSOTywHwmc112G2y1fDFYrFck97Ob77vAN8RBOE8oAKfNk3TBDoFQXiQfCMXDfiCaZo6gCAIvw08DUjAd0zT7Hwbj89isVgsl5FVNV7onkKRRba3lXNdjQtZFNEMg2d7XPzuja184/l+msuc6KbBA8eCGBjYJRs7Wv30hdJ8enMDNy+twK7IfG5H0+Jrf25HE95CO7IkXnXjkkv3WXjO53Y2IUsiWVXjRDDGhsaSxXmBumFQ5rZzx/pqHjk+gdMu89DxMUQEPr2pEaddRjNMesNJPr6hmpMjLhRJYm2Tl4GpADZBJhTPsK2ljL7JNAYmiizS7HdR6ytkZHqO+pIijg3OIgrw0XU1DE1l2NoMFZ4CmstcPNcTYnm1d7FRjt32cnC7cD6KLPLpzbUMROLYZYlGfxGKJNJYVsTp4DQ5XUcS859BkVLAihovH99QzQ8Oj5LRtMUs15tlV2Q2NZVyfjROfVkR/eEEd22qpT+c4SsfWkpfKM3aBi9Pny9EkUVSWRWnXXnFZ/J2BVDZnMbJYIyMarCnK5TPNAfc3LWxhgePjVNT6sCpKPzZe9uRRRGnXWFPV4idbf7Ff7hIqSpPdU6gmwZ3rKvkPw4Mk1F1TgZjrG8oeVuO22KxWN7p3rbgzzRNFfjEFX7218BfX2b7E8ATb9cxWSwWi+W1pTIq+/umeeL8BDcvqSSamuP3f3yWkkKFG5aUIQsij58JU+5SGIwmMQ3Qcjp7OqeoLnYQjmep8DqYSs7zvhWVwCsDhMsFC5cbK3AllwYdWVXj3gNDNJc5eaEnjCRIqJrO3r4pfm93C8ORLHdtrObf96kIJggCOBSJ1HyOA70RtrSW8/1DI5wfTbCk2s0/7+lnc3MJqxs83H9kiL6pFJOJLKYBreUe9vVPoekGuzsqyeRyPN89xbymcWJ4hlV1+c6mZ4KzNPncxJIqqmbwbNcUPzwyxPW1xdhEgY4q92JHzQ2NJTzbFWZsOsO58RjRpEp1cSFLq9x8fH0jkgjP9YbZ1JjPkk4m5jk3miLgtvPomVGyOe2Xvp5Oh8Lnb8hn2Z4+n5+tp2om/7JnlICngLSqct+hITa1+HjyXIgblpTjcRSwps6L06Fc1Xu8UbF0lr/4+QW2t/pQpHx3022tPr57IIgii2RyGg+fGKV/KkWBLLGsysOKGhc/PBwkrebY15MPnE0TJmNpnjofQs2ZnAnOAgJtfjffPRzks1sarOyfxWK55ljfehaLxWIB8oHfvQeGaCgtIuAuQDcMzo4m+epHlnF+NIUiC3zp/R0AfO2JHna1lbK+KT83LWfm2/xvb/XjUCRAuOKNdVbVyOY0zowmWFPn5dDAdH6eH79Y9vl6NM0gNZ9jYjZDa8CNqutUuO389MQEHZUujg7k6ByL86E11dglhe7JOCBQXerg+upiJDHfpdQhy/zmlir+8Zle7DaZ919fQziR4TduruOxkyEcisQf3NzK1564QFZXefT0OB9eVYskCrzYF6Yj4EG6uA5xQ0sxkgQOm8zaBi9d417u3lrPvt4oPaEUkiAB+RmGNy7xoxv5oHRVnZtTwUS+HLajNB8gNhXzo6Nj3LW+GkkUWNvg4QeHR1lZU4LdJr9iXaH94oDzwwPTrK0vvurAxq7IpDIqe/si7O4IsKEp32RnOJrGbpPZ3OLjuppiOsdiDEXSfGJjKd/cP8AXdrS8pcHTQufTM6MJtreUIQkCqmawtzfCqnovwdk0n9hUAwg0leXHYEhifp/u0Bx/9ZFlDIQz3NAeoG8yQXVJIY+cGKa6uBCHIrO6rpRmv4uxWIZPbay1Aj+LxXJNsr75LBaLxbI4biA9r6KbDsZnM/lUGXAyOMPyKi+qZvLAsXGay5xsac5no86OJmmtcC12Aw1Op1la5UGRpcu+Tyqj8q19g/RPJbh5WQXPdYfZ0zXJihoXneOpNzwnb17XefTUGB9eU4Oq6zzfPcX2lnIGo6n8fEHJZEmVh8HIHMsrC5AEgfevquBLP4mwusGDXZboDiVor3DjK7RT7ipAzZkcGAyzqy3AEy+FuWtjNQ8cHUfVdWYTWQam0hhGvqGNp1Dhize1EoxkafI7ePy8nR8fHWMomqSutAjdNNAx0QyD/X0R/vjWNpyKwr37B9nUVIrTofDe6yoALq6lsy8+BvAW2mkPuDkVjCOJQr6pzdaG/GeW0/juwSAZTcNhk/nMpjpkUWR5lYsjgzNsaCy56mvpdCj8zo1NOBWZ/zgQBODjG6qRRYE9XZNcV+uhbzLFneur0QyDE4PTxNbX4OXl411o2vNGMo8LFjqfDkTT3L66ghNDuXwX1DInpU6ZHx4eIZvTOB1Momo6+/qm2L3Ez4amMr69f4jf2FjLj4+P8emN+bV/65s8OBUFRRbZ1VGG3SZzfZ2L/nCGm5b637aspcVisbzTWcGfxWKxXOM03eDI4AypbI5zo3Gaylz4PQVgmgSn54inVdoCLhw2mYZSJ50TcYaiSRp8Llr8TvrCKSRBwCaKLK3ysrO97LJNQPIB5jQ1xYUcGYig5kx6pxJMzs7xHweCyKKIbphXbJJyOQWiyAdXVeNQJM6PxdFN82I2T2cgkqK+tIiJeAYRkY+vr6F3MsHDx8fQdZPvHQyiSBLmxedkNY1zozFqSwoZmEyxqw1ayp04FYW2gBvd1FlVV8onNtfxHdPk8GCULc1lfPmnQ3xoVSX/+PQYNlHkY+ur+bunezFMk4yqczo4y5IKLxVuB2dHk6xrKMYw4OjQ7C+se7zcea9r8HJsKN+Z01mQ74K6tyfCubFZqkvsPNcdZ3dHgH/fN0xwOn/OLQEnRwZn2NLy+usCNd0gmprj898/xcamUsLxeXxuma8/2Uulx0FJkY0fHx1jIJrk+4fGAJMiez5IFAUAgY5KJy9054Pb7tDcVb3vq4+hbyrFB1YGePDYBKqmY5jQPRnn3EgcV6GNZCbH8honxwZihGNz7OmcYl6FZ7smSas6FyZiqDmTtoCTn5wY47aV1RwajKLYBNScyU/PjPOPH1thBX4Wi+WaZgV/FovFco27dP3ZsmoPfZEkXWMJPrymmla/G1UzUCQJVdfRTZPGsiKGplPU+xwMR+awiSJ3b63HbpMvG/RB/ub+YP80p0dmaK9ws7TajSSBLEosr/WyvMqL9CZm2ImiyOhMhla/C1EQ2NVWjiQKTMTn+cNb6jg/mqTKW4iByU9OTPCx9VX86PAoK+qKKRAlGsuL0I2XR8p6nAoORWJjSymKLCKJIvv7okC+66koCBwfmmE6neNP39OO12HHWaCws72MW5bm1zg6HQp//v4l3HdwBEyBUncB65u86Ga+EcnhgRlaAy42N5deMUDKqtrFjqDTnByZARP+8+wE71keYGebn/UNxaiagW6YmKbJusbi/GNDpyXgJBjNcPeW1w+iNd3g+QthAD61oQFPoY2OKid//3QvG+p9jMfmGJmd44Z2P0PTST65qRq7LPMvz/cCAqquE4plaSzLd0j9/uFROgLuN/w5ypJIo6+Ix05P8qlNtWhG/lodH4rTWOZkIJIiq+c4ORyjfypJqdvB5mYfik2gzleIJMKSKjdg4LDZ+MjqGpx2G5/b2UhfKM3wdIq//egyfK7CN3xsFovF8m5iTTi1WCwWC5CfnWYKoIgy7VUuhqNZnu8O8+ipUU6PTfPUuRD3HxuhN5zCME3GZ+e5e2s9n9/VjLfIjl2RrxhsyJLI6joPE/EsGVWncyxBXzhNY1khM2mNLS2l7Gwvu2K56OXYFZl7tjdyXU3x4vP6Iyl0w8QwDe4/Msb58RijM3NMzGbQNYOTwzFMBBRJ5M6N1fRMJnnweJCzYzH29UYBgZ7JFF1jCXonU6yqc3MhlEA3TAam0oDA0NQcuzsC+JyF2BWZ3Uv92BUZp0NZzCp5C+10VHoYnZmjocTJ0YEYPzk5xpIqJz2TSfqnUlc8r6yq8S/P93Lv/iFmM1lOB2doLndS6bGj5kwO9k+zvy+KIousaXATT+f43sERHjk1Rlulm5uXVnDP9sarynBpukHXRIKMqjM6OwcI9IbS1HgKmYhneN/KCrpHE/RMJoklVU4MxdEMg+5QitoSB4okUeG1MxhJL75ebzh5xRl6rzVbT5HzAaBmGPzlf17g2/uH0A2DiVgWzTA50jfD+fE4bQEPVcUODg9O0x9OUVXqIBybZzyWAWBtgwdPoUJ8Lsfv/PA0L43Ngglex9XNG7RYLJZ3MyvzZ7FYLJY8Eybj83zxplr+7/OD/Pdt1ZwJJjk9Os2KKi8rKvMNTTa3lPCDQ6N8ZkvdVQ/wBnDalcXs2Lp6H5/ZWsdLo0luWVqJ054vZ3yj5YJOh8KujnI03SCjajzXM8nyKg/bm/0EZ9K0lLvRTRObKHLH+moeOzVJY1kh+/oinByOIQDFToWPrK3i56cnaSl1cfv6KgpEaXF2nGGYbGkpZUtLKbIoLmY3F47zcscrSyK7l/rZ2f7yjEJFlgi4ixZHX1xpPZ6mG4zHsvz2DU08cnwcb5GNtY3FBCNzDERS3LWhmgeOjvHZrfkA718/uSbf/CWnLV7H15PKqDgd+X2by51IoogAbGwqxmlXWFHj4keHx+gLpbn1ukp+e1cLKTVH13gKZ4HCVz60lJ+fnqQ94KFvMk5juYvhaJrWgOsVAfyl6/9SGZXTo/ErNqNRNYPeyQQIJj6XAgiL1/1A3ww2UWBlbSnLa5w81zPJ7o4A65u8/PXj3fzuzS380558NvL4UIy1DV6ODMSo9Tn4re2Nr5gBaLFYLNcy65vQYrFYLNgVmc9ub+BfXxxgMDzH79zYRNd4iuvrXOy5EAJTQJEFVN3AW2jnczub3vDN9EJABLCzveziAPdXNjh5M3PrFub7DUfTVHsdLKt283dP9eJ3F6DIEiYGDWUuHjsd4o51VXSOp9nWUoYkiDSVOxmcTvHTkxP85uY6UqrGT05MUF3q4KXRBEurnAxOJ3m+e4qRmQxtfhe7lwauej7hpfvtXuq/YlnspeyKzI3tAX5+epKcYYAp8NCxcVornCiSzNnRFIZhvtwU5mIAfrWfRyyd5cuPdfKVDyxFFkV6w0nu2lDNiZEIB/qjbGn28eDxCerLCukKxbFJ+WDu/iNj+XWMOY29vVF2tZWzo60MhyIhiQL3bPcvdhw9MjjDmjovR4dmuK7GjV2WuffAEHeuq2Jfb4QVNS4CHufiMWm6QedEjPaABzVnMhHPUu11cHhgBlXXUSSJpVUe1jZ4+NHRMTY1lnJDRzkpVWU+p9EXSlPuUVhS4eHsWIxTo9N0+L2IAthla6C7xWKxLLC+DS0Wi+UasFBu91qBhyyKhJPzpLIaf/NEDwAvjRWxqy2AJOWbl/zk1CjbW8veUMbvFe/xqiDvlxlS/urX7aj0cCIY5f4jo2RyOcIJaK8q4v97thvTFFhRVczZ0SQralz89eNj1Hgc6JhggmmY7OuZ5oFjw3iKbBwbzPHhNdV87+As4XiWnskkbX435ydibG3xvammIW9kluEtywNkVR/7+6YpECXaKzxIYj5ozuby4x0WRiO8kcAmq2rs7Y0Qnp1jb08UBJPecJw9nTaePDPB2EwWXYdGXxEAY7E5qr0ONMNgZDZNjbcQh03mf7+vYzHLuKuj/BfOT9UM9vdFOTMa41t7+7l9dS094TjfPqAyMp3h759J88P/vn5xDV6+k+csjT4Xg9EkNV4HkiiSUQ0eOjHCh1fV5Ju3SDLlLoWfnR5HkWT2XJggOJWmrszB15+M8MkNDZyfiLGvZ4qq4kKGp9JkNQ0nVpMXi8ViASv4s1gslnctTTcWy+729kQAk+1t+RLJywUMdkXmT25tBWBPd4hdbX4cNpllNS5+djJEW8DNR1bV4LS/826kZUlkbYOH+w4N0FrhwiZJ7Govx1dYwMZWHwWSBILJ2dEYqq7TUFzEHRuqODeaAkzu3lrP8aE4XqeN//neJZwaSjAYSdLiL0IzdJZUeAAYnp5jf1/0qrN/v8z5OB3KYuloal7lvkMjrG3w8oPDI1SX2PnWvkFEUeATG2rwFtpJZdXXDMo13WBfb4ShyBy3ra4BwaRrIslscp6RmQy7l1ewrKKYgUiKjgo3W1t8gED/VBK7LPP+FdVsair5hbWdr3Ud2vxuRmZSIJjUeQsRBZGdbeXsFaewyy//DsqiiM9jRxIFJASWVnuRRIGVdW4Gp5LcujzArcsDANy7f4jrqz1savEyOJVke4ufhhInX9jZiq+wgFW1pTT5XDjtMjvbAjgL3nm/rxaLxfLrYjV8sVgs16yFbNilg7LfDRaCvsfPjvOtfQNEU3PoRr5cL5qay29Lzi3ul8qopDIqezrDfPXxHr59YBhV15EEkXPjMf70oXNUlxSwsi4/IPu1mnb8OnkL7dy5roEim42q4nwzkgN900wlVZrLXeiGybye/6w/vK6Sh09MsLzGyUQ8iyyKrKxzkZzTeCmYQDd1ACRBxMDIByWiQOXFAOVXZaGk9bsHgvSE4xzsmyGtarzYE6XeV0htcSF//lgnD54I8vkfnGIyfuVGMpBfd3jb6gqGoil6J1Msr/Lw5Q8v57rqYuyyjCKLNJU5kUQBuyJzy/IAn9vRtLi2cmGd4OtZuEYORWRXW4DR6TlaK9y0Bdw4C2zcsqTyFf8A4XQo/NUHl+EpVFhRW8KWllIkUeTB4xMsrfa+oqHOZ7c2sKreR+d4mvZKN5PxLC/2RDg8FOXf9w+hajoTsTkQTFbVX7mjqsVisVyLrMyfxWK55iwEPft6I6xrKOa+w0Hu2db0lpci/jpkVY2D/VHaKor48bEg77u+ir95qo+NjSXs7Ynw+LkQlZ4C/vLn3exqL0MSJJ7qHKfB56KhtIhSp4xmGASn0hjLdJZXFaMaOoPhOUamx2j1u96x10eWRG5e5kcSBZ7umkA38vP7Ap4CAJ7tChNJZjBMgWfOhzFMk5NDccDkQN8MiizwwZXV6Ab89NQot6+pJWdoHOyN0lLmxmGTmUrOs66h+Fd2DRbGPWQ0jXhaxUBHFF4OQnUdJmNz7O2ZwlEg8MDRcT634/LrMWVJ5PoaN19+tJOJ2TR3rG0gk9P4/fvP8KnN9YzOzCEJArIksqzKu/icN/r/hSyJbG4u5WD/NKqmMx7LUO9zMhxJY5jm4j6vHgbvLbKzvS0/JP748CzbWn1sa/X9wjrJhUB08RotMS42Cyrl2FAMMOmo9KLI4muO07BYLJZrkRX8WSyWa0pW1djXGyGT03jqfIiMqqNpBtmcxsn+GLphsq217L9kIJhVNZ7tCtMdSnBqZJbmQBFuu40/uqWF08EEu9oD9IbjtAZcZFSdvskUjeX51vofXFXBQ8fG6B5PcuuKSiYCGeySjevrXDxzIcSOVi8Omw1FfmdfD7sic+OS8sVMXUbVuDCeYEWVl/csr0I3TByKxLoGL7IoXmw64+NkME5G1Tg8OM321jI+tKqGba0+jg/F+OTGBiZiWVr9rsXOpK/26jWVWVVbfHylMturJ+AqUPir21fgKyygbzJNzsyPnqgpcXBdbTHtfi9D0RSf2Vz3mu/lLbLzpfd38N39QZx2Cd2QqCtzUKTINJe5Xnde49XKX9d8p9OtLT5OBGPctPSVcwevFKDKksjGpisHbZpuLJa4XnptvUX2VwSFC69nsVgslpdZwZ/FYrlmvDxofBZVN+gNxdm9JEBrwM2xoXzgd3ZsZrHkEcFk95KK/xI3kFlV49kLk+zpCrO9rQxdF7h33wQ2USZn6AyE0rRXetBNk86JOOdG41xX60XXYSYxz4+PjlIgSXzptqU8ez7CzR2VIJg8fHyCHS3leBzKYhblnX49FmbvLdje5vuFgO1KGScQ6J6IE4ylGZ6eQzBAlARa/W4civQLmaSFLPKRwRkA1tTlM2b3Hhiize8CwaQvnObuzfWvOQfxShYaqqyqc/O1J3rY2ebHBApEkTvWVfH1J3rQTZMVVSKKLGG3vf6fdZ+rkN++sWUx+6bIIttafbD06juGXu2xQz5T90ZHeLx630ub2+TnTQb5+4+toHM8xeZm3+Jxv9N/Ny0Wi+XXzQr+LBbLNUOWRNY3FPPSyCx1pYWcDSqsayzmTDABmKysc/Hzc2NoRr40LRTPsLW57E11dvxVWihh7Q2l2NXuv5jZKgZgKJLk6fNhGkqdrGnw8GTnBNXeIq6v8bCiuhgQuL6uBFkUqS5x8M0Xh/jiTc2cH0uAKdAWcC+Wz/1Xapf/6uDuavZf6LJ545LyXyhJvFwmeCGLDLCpqRTNMPjXvf20+l1kVA3dNJCQaPQVLQ5l395W/qYCQG+hnZuXVrCpqWQx++orcvCe5VUAi/MEr/YzujRY2tVx+WNauAaXrvG8XLnm1Z7Dm7WQ0d5zIcyX3tcOQHuFk6MDMfrCSTI5jR2t5W8quLZYLJZrzX+dv+QWi8XyFrArMh2VHlRdw+1UODYQY8+FELva/ZwOJih3KYCJYYJmGGjGO7O5yavphkmTv4il1S66x1McG5rFUSBw16YavvbEBdorXMiiSENxEXdurOHkcIwNTcW8NJqkPeBBEgV6JhMk5rPcf2SU4ekUjT5XfkD2VQ4Ofze42symphs82xWmdzJJzjDQDRNV1zkxGGVO1RiLzdHmdyNJGrphMjKVoS3gelOB08JxLQRp/z97dxrd1n3de/97BoAACZAgCc7zTEqybMm05sGS7dhpBruu2zqJk7ium9umt7cv2t7b1TlNmvau3ufpuJ52JWnqOk7SOnlSJ05cT5FtzbIpS7IkzhMIzgMIgAABHp7hvqDIyImdyNZASdyftbhEgSBxeEgC+GH//3vvaQyufJ171he953EP7/S1f1w0meZ0OM5tFdm8PjC7sn9yR13+yuy+5TENV9Nytf7McITUosGhnim6x+cJR+ZRVRUceO6tMTrG5qgvzGRfU9H7HkMihBBrwdp4NBdCiAuWqhgO6UWT8PQ8iYVFusfivNw5QWx+kY6RORwHUAAUXh+YvW67Wy4zLZvzIzGee2uUT33pOM++NUIkYfDi+QlOh+JUBLLAUfi3o0OYtsPXjg3x9eOD/MuhQZpKMnmtexJNVagKehmdnqcs4CUyt4Bp25wMxVb727suRefTvNo9xYN3lAJLc+1wFNZX5KA4MDCR4OxIlP86O8bL7eM8vHWpQnewe+qyfp/ShskrnZN86WA/BzomiM6neeJYiLRhXrHf02gyzee+187UXIonjoQ4GYpwOjzLqaEIh3pmaBuM8NvfPE00mb4it/fT6JrK7VU5jMcXwFHoHk/SUOhDuRD8Wkqyqcz1Uprj4Z9e6eNPnjlPImVc9eMSQogblVT+hBBrwvI4h5fbJzg/GsN0bHY3FbKnOcih3knKAx5yMl185LYy3LrKwFSSilwvO35K44nrha6pbKrK46EtZfz5s+fYU1/IwMwcRdluNlT4OTsyC4pDU5Efw7LAUVgwFwGFkwMxinMyaK3JoW0gxq6mIvavK2B4dp5NlXnSLfEdJFIGTx0P8+v7augem2coksR0bHBgPGawp74QC5tbynNAcRienQegbWgavzvjbQ2FLpVp2RzomMCyHTrG4iQNg9Siyb8eDlFXuDSQ/bWuyfe1rPTHb+dkKEbApzMym+ahO0r5xrEwi7aNDdxRk8OmKj9tA4FL2l94JQSyPHz+/g0c65vFsm221eXSPRFHVZaa+qAoeN0a+5sKeXxv7XW/TFsIIVaTPKILIW56acPkSwf7eeH8GC+1j2PaNqOzKVQF/uPEMKUBD17dRUuZj++dGuaVzklqgv6lasMNYLm1fsdIgkRykd7pOKGZeQZnkvzv57p4czDCudEYhmVxoGOS82NRzofjlOd66BmfIzQzz78eDtE+GmMokuRrR4cYn1tg+4WB3uJHlkYvROieiPOPL/exscLHXc3FuBQVFCjOzqBveo72kTi2YxOOzDM4meQHp8d54ew4D9xe8rZzeqnVOtOyaR+NY9kOv7SljPbhOCnDYnAmwcvt4yQWDM6PxkmkL6/qpWsqt1b46RlL8MDtJbSPJGBpCyxv9k/zT6/28Vc/6MKyl5rcXKuqeCDLw+6GfPqnkxzvm8XCob7Qj6aohGbn6Z9McmtlHoFMWfIphBA/jYQ/IcSa0FycjVvT+L0PNvI7H2jmc/evp6U4gGFZtI/Eqcj3cC48hz9L5971JexryacqkHnDVL08bp3dDUFaqwvw6Dp3NRdTnZ/FfetLeWBzBS5Fxa2rlAc8rC8JEPBnMBFf4OHt5VTnZ+FSVR68o4xEymRdaTZ//tH1snfqHSzvufvIxnL2NBTwVniOwekk9UU+cGBizqA6z0eeP4Ntdfk0Fvh5dFctcwsWv31XM+UB39KogpRB2jB5rWvykpZP6ppKTTCL7ok5PLrGpso88rIy+L0PNqGoCrqqUhvM4qkT4ZUq9/sV9Gfytx+7jcGpNMsNgT6xvXJpP6Sqsq+5CJ/HxbbavGv69+HzunlsZzUA54ai9E0l2FaXS1NBNo/vqeGe4e7YAAAgAElEQVSe9cU3zN+rEEKsFnlJVwhxU1seGL2xwse/Hg7x/Plx/vjDzRzpneHFjjHKAl5ystwMTqeoDkIsucj6Cj9nwnOgKKt9+O+Jz+vms/vrOdoXwTAtphKL3Frl529f6qUk28O22jz6J+fZ05TP4HSSj28r5/xIElVRqC3MIjSV5sHbK5Y6e16jJX03oqWgnc8TR0M80lLBcvHLq+vc1VRM/+QcoHB6KIauaRT4vdx/W9lKxe/5s2P817lR7moupmsizndPD/MHH2om6Mtcad7yTqMOeibmsByHNwaiNJX62VyVw5uhOKoCCWMRTVV4dPtPn/N3qYL+THY1eFZu+3h/hE/uqOfOptUdq+DzurnvlhJ2NSz9jnrcOr++752H2gshhPhJcm8phLip6ZrKpoocvnJogOHZeYpzPBzpifDC+XFm59LsayogQ9OoL/TTMzmHbdv848t9/P7PNXJ+5MZrduLzulcGXe9qyOfsyBx/9KFmzoTnCGR6eGxXNbqm0ljs5/xIglsrsjkzNEv/ZJKHt+XyV8+Fea1ritpgFp/ZW/e21v43woy/a8XndfP47hp0TcXr1thWm8e+5qW9fIm0wROHQ9zZWMjdLcVvO2+JlMG5kSido1GKczy4VJX5RZMvfr+LfS1BDndH2NdSwN0txT8RaHRNpblwad/my+0T9E4m+dVd1aQMi79+vhtYGqh+pVw83mLXha97Pfz8dU19W1Vagp8QQlw6xXGc1T6Gy9La2uq0tbWt9mEIIa5zacMksWDw1NEwTSV+QGFTlZ/zI0kMc2nQ9e1VORzrm2V7XS6BLM9lt9C/Hlw8q820bF7uGAdHoWMsTmORH69bY12Zj6dPjFBXlMW5kSgbSvMwLBOvW+P8SJzQTIKqoI9NFblsqcmV5aA/5p3GN0ST6ZXfofSiic/jJm2YtIWiRBILfP9cmPtvrWBrbS5fPjTAxtI8vG6VxpIsnjk5RlOJj3vWlfzEXEGAI70zpAwLt66uBP3lj12vzU7e74gLIYQQl0ZRlJOO47T+rOvd2M9qhBDiEnncP1oidnFFI+j70ayy5blpyx+/0YMf/KhSszSXbpInjw1wa2Uutg3tYzFcisrZkRg1wSy6xuYYiab4xHY//+vpM2yszAVHYdGxsWyLRNric8+28/n7N1y3IWM1vFPwe/JYiF+6o4ynjoR5YyjCQ5srGImleHRHFYd7ZkimLAzT5lD3DIe7pviVndWcHIjxzJtjVOZl0jORZF/T2wPT8u/j3qaCn7jt6+HncXHzl4urxYmUQVsoSmtVQAaxCyHEKrvxn9kIIcR78E5L6X7a/28Gy4OyDcsk359Bda6PQ30T1OT7eXh7OWfDc2iqSk1BFqHZJADBbA8bSnJZOh0OuqKiaQ6F/oyb8hxdKYmUwRNHQswZBl8/Pkx5Xibd03G6J+O41KWGKXevK8IwHXom44BCRdALwMud4+xtDJKT6eYD64ve9cWH6/H8m5bNa11TWLYNKJwZjrC5Mo/bqwI8eWyI8lwPXzk8wLqSnJVh9UIIIa49ufcVQog1wLIdBqfn2VtfxNBsElCoLcji222jJA2DLTUBwjMpSnO8PHNyjD0NQe7dUMS2ulxUBdaV5oCjMB5PXfdD71eLadm0haKUBbyMRlOU52YQiiQpy10Kd0lzkaN9MwD4PDoZusYntpeT6XZzciCO7TgMzaTZWpN7XVTyLsVy59JE2iB1YenpurIshiLzNJVkcTIUoyQng8GpeUpy3IBDIm3IIHYhhFglUvkTQoib3PJ4Ast2ODs8iwKU53rwulwU+Gz+5eAA2+ry2VAewDAtNlT4+YNvn2VrXT7ffn2Et4aifHJHFV1j89zTsjSnbjkAXtwQZq3TNZXWqgBfeq0fy3Lon0piOwoT0TT5fjcHzk1wW3kAXVvaq7encalz5hcfuAWPW+eulqWGMTfKcmPTsnn+7BjPnRtdmQV4d0sxXz8eJmWYfPPEMNV5WRzum2JrbR7/+eYID2wq54+/ex7HtvnCz98i+0eFEOIak0drIYRYA5bHE+iqQnW+j9FomtTiIhOxBbbV59M1Ns+WmgDd43HaBiPkZLr45vEwVcFM7t9cTvtIglsr/GiqQtoween8OAc6Jpbebx9baTiy1quCPq+bx3ZX01yYwy1luWyuzOPe9aVsrS7gI7eVc09LycpeuOWQ5/O60TUVn9d9wwQ/WB48H8NyHHY15FOZl4mmLf8OOFTmeRicSbK/qYQst4uWsmx8GS72NRZhOjavD0TX/O+LEEJcazfOo4wQQojL4nHrtJQGeGt4FtNaWgb6yR0VPHU0zNaaXEzbxrCWmpD81t0N/OMPewnPzgNQlefj1NAs4dl5UosmP+ycpC7oY2OFn68dC2HZDne3FHOkd5qtNXnA9dGEZDUEsjx89q76laro8f4I22rzuLOpYKVqalr2StC7uIp6o9FVlfKAl+FIGgcHy14qAc4mDLonknh1jVur/Dx9YphzQ1Fur8hnR0OAV3sm2FITuCG/ZyGEuJFJ+BNCiDVC11T2NS91itRQWF8WwKPrhGaTJAyDp98YxQF21RXwgzNj1Ob5eGRnBadCc2yq8tM2EMO0bTRFpTzg4ePbyvG53fxyaxVuTSW9aHJqMELbYISR6Dx/fv+GNbus7+JB6LsagpiWvbLfDxQ6x2N8ansVHpfOkd4ZLNtmR13+SmC+EZbSetw6j+2u5quHBqkuyKR/Isng9DyNhdkMzMyhqwq1hVn8vy/0sLshyFB0nm11uZwbSfDB9aX4PGvzxQEhhFhN1/cjixBCiCtqeZzFw9vLuWd9EWnTxLRt2gajPHh7CS5Fwe1SsG14ePtSuLNsm68dDZNcNAjPJlm0TUIz8/zLoUG+dLAfGxPDtDnWFwGgochHetHkWF9kZTnoWvdy+wTnhqNYtsOOujxKcjw8eWyIlzvGiaUWOBmK8JXDAyRSBqZl89zZYdKGiWnZK/9ej3weNxvKA+R4M/j1fXU8tqsar1sjQ9NZV5LDnsYgDjb3rCvir35+I0F/Jrsagm8bqSKEEOLakcqfEEKsEWnD5GD3FBsr/Py3J0/ydx/fxD+81Itp2ZwdjtExOofpLC3bG5hO8MXvd1IT9GPYJuGZeXom40QTaUxTwXQswrNJ9jUV8lL7BBW5WdxSFqC5NIfbqrL5jxODpM1FjvTOsLepYE0/0U+kDX7YOcHu+gLcukY0vcB3T4/yhQfX8503RumdiqOoCnc1F3G0b4bGkiz++r+6WDDAm6FyfjROa1X+dXkel5vXLDvcM41b1/jghjK8bg2PrlOXn43Hpb+tGiqEEGJ1SPgTQog1IJEyONoX4dm3RthY0cyWmlzaR2LsbSyidyJGbVEWLtXFwHSC26sDnBuJkaFqPLSllG8eH+ID60ronohT2VSAW1dxqRqluV5cqo5hWqiKgmHa9E3PAQq+LB2XKg8xpmVzOhznzsYivG6VWyv8fOPEMA9uKifgWWoO81L7BNvr83i9f5b20Tgpw2JvSz5ul8KpoVneGprl0zsqV/tbeVcXh7ldDcGf+Niv76u7oRrZCCHEzUxefhNCiJtcNJnmyeNDrCvLwrRtXuueon04zgtnJ9BUqCn0caQ3gqbCQ3eU8s0TwwxHUzy0tYxvvTHC6dAsm2tyADjSF6FrIs7ehiKydBeLjsn8gs0vby1ncDqJbUNzWRYjMylaa3Kuy2rVtaRrKttq8/C6NQzT5o2BKLXBLAA+//0Onj87wZdf6+XLB/v5dluYD95axLnRKG39UZILi4zMprFth7aBGId7pq/b5Z/LljuZLr8BEvyEEOI6snYfkYUQYg1IGyZPnQjz0duKOB2KgwPh6TQf3lTKB9aXoqkK4UiK37m3Abeu8fVjQximheI4nBmK0ViYTX6OhzNDMXRVZXddAZqiMB5P8/Ht5UzEFvniQ7dQ7M+ksTgb27E5PxznzsZiAl7Pmg5+y5bPwdmRKG9d2Pc3HE3xe/c14NZVAl4dj67z0VvL+cHpcTI0jb/7xGaKsrO4q7mI7XWF3NlUwK6GoJxPIYQQl0UeRYQQ4iq5Hpqd6JpKXdDHN48P0z0xx+6GIA+0lvLcmTFsTCzbYcE0aR9JsK4si4GZBBvKctjXVEzfZJK9Tfn8wc81oykqDUV++mfiNJX4eXx3DUFfJrVBH987NcbB7mks22Y0lqZnPMGtlbkSVC7QNZUtNQGGo/OAgltX+dT2SnxuN8PRFJ9/6FZ+4846fB6dusIsagt8fPfUCE0lmYzF0zy+t2ZlFqAQQghxOWQthhBCXEHLy/LShskTx0I8srWCQJZn1Wa56ZrK7oZ8eibmuHdjIb/91Jv0Tc6T5dEwLeifihOaSdJUlMPXjg4xMJlgfXk23zs5hu04HO6Z4blzo+iqyo66fF7rmERTNO5qLgLA59F5bGcNuqZyvD/C5+/fgMelv23ZnwCPS6chmM3D28p4K5zgcO80XpeLQr+bP3+mnQc2lfNK1ziVeVlYjs0rHZMsmja3lufJSAQhhBBXjIQ/IYR4F4mUga6peNw6iZSxMqB7OdRcHG6Ww92BjgkALNuhyO/myWNDfGp7Ja8PzKKpCrdXBd737Lv3O/vN53Xz2K5qjvZFqC7yUVeUyUg0Sf9Ukqq8LMLRedy6yi1lAeoL/ISm0mwoD7CjLg/Ttnm5c4Lf/2AjAa8HX4abXQ15nArH2F6X/7aliLIs8d153DqP7Kjg3EiCWGqB17on+aMPt/CNYxHu27gUpBUVGov9bK4O0D+TxKVquHU5n0IIIa4cxbnQ1vtG1dra6rS1ta32YQghbiKmZZNIG3zu2Xaq8rN4eEsZf/18DzvrggzNzrOuJAdNZSXIJVIGp8Ixbinz88SRQSrzsuiZnGMkmuLOxkI0Dfomk9QV+DnQNc5fPHDLyjDvH5c2zJUGGRdXEQHaQlFaqwJ43EuVtfcaBk3LJjqf5qmjYQzLoqk4h1Akyce3luPLcHOkd2kI+c76/LdV7hIp4yeGj98IQ8ivJ2nD5KtHBnlwczF/8f1OxmPz/M3HNvHkkRAnQxFKApnsqM9jPGJg4xCaTfKnH1lHIFP2TQohhPjZFEU56ThO68+8noQ/IYT4EdOyea1rCsu2iaUM9jUX8MZAlJRhMTiTpK7Ax10thRzqmeHF9jF+594Gvt02yiPbKjgTniORXmRgOkl9oQ/LXrp/fblznLtbiri9OsBfPtfJXzxwy9uqfz++VPTxXTUAHOyewrLhxfYxinM8NBT4GZhJsq4km33NhbzSNcG+pqWq0XvpqLgcJpdD3I+HTQkbV8dysJ+em+eJQyHWl+dwOhwlFEnw+z/XzPdOjfOp7ZV4XEsV5nd7gUAIIYT4cZca/mTZpxBCXMS0bCzbIWVY/P9tYRTF4UhvhN31BVQHM+kcj3FXSyHb63J5PTTF14+FCUWSHOmJoKkKKA6PbK/Ao+u82D6OW9MozvagKRpfOxrCsh2O9UW4Z33xym2+dH586XNRKM3JIL1ocjIUo300xiPbK7mlwsc3joXpmZxj3jQ4N+JwS4Wfp46FSBkWfVMJfm13LT7PpTUFuTgoXnx9CX1X1/J5D/oz+e93N1y4VCFDUwlmelhfmnPJP0MhhBDi/ZDwJ4QQF5iWzfH+CAC31+TwfLsLl6bx2/fU88zJMeYWDM6Go/ywY5JF2+Ro9zS/srMO07bpHI9jWBZnw1E6x+aoDfr4l8O9fGpnDZNzC2yo8PFixxhlAc/KbR3pnebWimyeOzfG3c1FWLbDv78Romcywa/uqiaRXuSp42EWFk0UFMpzvTx5ZIxdTUtjG35hcwWaonByIIJpOWypyWdrTZ5UjG4Ay0HwnvVF7GsuwOPW1/xMRCGEEFefPMoIIcQFywO5Ab79xiilOV66x5N8641hHt5aRpbLxfqKHCzboW9ynm0N+dzZkk+GqtJQ5CND02gpz6a5OBu3S2F7fZAst86Oujy6x5L8j7vrydB0NFUhsWBwJhzlSM8shmViOTaaquDz6ti2w/G+WfqmEjzUWortONQWZpLjzWBXUwENBX46xuLYWOxtCvKR28pwqSpNJVk8eXzouhgxIS7NckOh5feFEEKIq0keaYQQ4iIet87+lkIe3VlFjsfDx7aVMRZbwOd286mdlUQSi/ROzGHaDt2jc5zon0VVVbwunfrCbLpG4izaJp1jCUai87w1HOVbbcOcGJjm71/sJTw7T2NJFk8dD2OYFovOIrYDPRMJDMtiPmXSVOwHlKVKZN8Mbw5GONAxhWGZuFSNw71TJAyDfzk4wA/OjPPs6RFSlsm320b5+Jby97T/TwghhBBrhzR8EUKId7HcoOPibpfL4x8A0osmPo97pfNlIm3wz6/2saEsgGU7PN8+yo66fH5wepSPbatmb2MBpm3z72+MUBHI5NxoBFXRGJiJc9/6UrwuF4ZpcVdLIUf7Irw1FMFCIW0tsq4ogM+jY1gWOCqaCrH5RXIyXZwZnuXW8jzcusr+lkKpIAkhhBBrjDR8EUKIy7RcQbt4D93F71+8XM+0bE6GYri1pdlsW2py6RmfY2bO5G8/vomgL3MllD2+q4b0oskr3RPsrA0wGpvHrelYto1bXxqxsKUmwLqyLF7vi/J8+yiZupuWkmw6x+JsLM/FMG2OD06zr6mQLLeLO5uCKyMghBBCCCHeiVT+hBDiCjEte6UKqGvqyt67d1uGuTw4fvl6Xzk4wIJl0lCYzYGuCabiaTaU53AuHOXvPrEJgL98rpPfu6+Js+E5zoaj3FaVy466fGnyIoQQQqxhMudPCCFuMNFkmq8eGgQF6gr8aKrCtroATxwK8fjeGp48NsRc2iArw0VTUTaaqrCnMSh7/IQQQog1TpZ9CiHEDSaQ5eGz++uBt3d+3FSdh8/j5rGd1ZwYmGVrTa50iBRCCCHEeybhTwghriPvVMVbnv/m87plFpwQQggh3jd5BiGEENe5i8OeBD8hhBBCvF/yLEIIIYQQQggh1gAJf0IIIa4p07JJGyamZa/2oQghhBBriuz5E0IIccUsj61YHndhWjaJtIHHpa+MvzjaF+HU0AwbKwLcs66EtGHKqAohhBDiGpDwJ4QQ4rKkDROPWyeaTPPEkRCLtkVTsY89DYUc7pnh68cHua0yl/WlOZwfi9FQkE3b4AwAmypz+Ovnu/njD68jkOUhkTIkCAohhBBXiSz7FEII8b6kDZNEyuAfDnQzPTfPk8eGKMxxkV60eOpoiH96tQ/LdlhX5qe5OJvUosnpwQgbq/wEszOoLsikbSBG0JfBsb4I47EEf/jMWRIpY7W/NSGEEOKmJJU/IYQQ71naMPnKoQEK/DovnxujviiLkmwv3zo5yPqybNaXZ+NSVSzHZiqxiOXY9E8laS7z0zk6R3nAx6GuacpyvZi2w7mRGGfCs0zGUqRNEx9S/RNCCCGuNKn8CSGEeM88bp2Ht5QRmk5RFszEpbhw6wr5/gx0VcOtaTy8rZy8LA//875GvC4XtuMwETewbIcNpTncs66EpuJsJuML/OKWMjZX5fOXv7iRM+G4NIMRQgghrgIJf0IIId6ztGHyZiiO5Ti4NZVtdQF8Hhd/+KEWWopzCEXm+eaJYTaU+Xj69REMy2RdSQ6VuZkAdI7HAdAUlcJsN/9xIkxqcZFvvT7C2eGYhD8hhBDiKpBln0IIId4T07I53h8BYF1JDsOz8wAYps3Xjw2hoGDbDo5lkzYteqbiLJgm60pz0BWVRcsiNJMAwHYgFEni1nQaCrPRFHh0ZxUetzw8CSGEEFeaPLoKIYR4T3RNZVdDEFiqAHaNxTjeH8EwbYajKX7n3kaePjFMTdDHN48Ps2Ca2I6DW9NoKM6mfWyWirwsPr2zkraBOEOzCfY1FeLzuNhQnovPI/v9hBBCiKtBln0KIYR4z5bn+C29aXSMJuifSFKRk0nHyFJVLzw7T2mOl9BkElVR2VSVjaYqjETTaCicH0mwqyGPhvxscrxu9jQG2d9SiK7JQ5MQQghxNUjlTwghxPvmcet8dn/9T1zu1jVur8rhcM8MBX4PH9tWwbdPjlEX9PGhW8rY3ZCPx700+P2zd9WvBEkhhBBCXD2K4zirfQyXpbW11Wlra1vtwxBCCHER07LRNRXTskmkDQJZHtKGuRLwJOgJIYQQV46iKCcdx2n9WdeTyp8QQogr7uKQF8jyAEgTFyGEEGKVyUuvQgghhBBCCLEGSPgTQgghhBBCiDVAwp8QQgghhBBCrAES/oQQQgghhBBiDZDwJ4QQQgghhBBrgIQ/IYQQQgghhFgDJPwJIYQQQgghxBog4U8IIYQQQggh1gAJf0IIIYQQQgixBkj4E0IIIYQQQog1QMKfEEIIIYQQQqwBVz38KYryu4qiOIqiBC/8X1EU5e8VRelVFOUtRVE2X3TdTyuK0nPh7dNX+9iEEEIIIYQQYq3Qr+YXVxSlArgHGLro4g8CDRfetgL/BGxVFCUP+FOgFXCAk4qifM9xnNmreYxCCCGEEEIIsRZc7crf3wD/k6Uwt+x+4ElnyXEgoChKCXAv8JLjOJELge8l4L6rfHxCCCGEEEIIsSZctfCnKMpHgRHHcc782IfKgPBF/x++cNm7XS6EEEIIIYQQ4jJd1rJPRVFeBorf4UN/CPwB8IF3+rR3uMz5KZe/0+1+BvgMQGVl5SUdqxBCCCGEEEKsZZcV/hzHufudLlcU5RagBjijKApAOfCmoihbWKroVVx09XJg9MLld/7Y5a++y+1+CfgSQGtr6zsGRCGEEEIIIYQQP3JVln06jnPWcZxCx3GqHcepZinYbXYcZxz4HvCpC10/twExx3HGgBeADyiKkqsoSi5LVcMXrsbxCSGEEEIIIcRac1W7fb6L54CfA3qBeeBXABzHiSiK8nngjQvX+3PHcSKrcHxCCCGEEEIIcdO5JuHvQvVv+X0H+M13ud5Xga9ei2MSQgghhBBCiLXkqg95F0IIIYQQQgix+iT8CSGEEEIIIcQaIOFPCCGEEEIIIdYACX9CCCGEEFeYadmrfQhCCPETVqPbpxBCCCHEDe/igJc2THRNXbm8LRRlQ5kPX4YbXVNXPiaEEKtJwp8QQgghxHsUTaY53DuNW9NILCxyoHOSqjwfpm0BDk1FOfzm17vZXJnHpspcttTk4vO4JQQKIVaV3AMJIYQQQrwH0WSaP/3ueZ443E9s3uRA5ySGafLRzcUMRpK8ORhhc00OBdkemouzMUyHP3rmHAc6JmU5qBBiVUn4E0IIIYS4RKZlczIU5bfuqcelqexuymV/UzGqqvLMyVHKA15y/W4CHjd/9pF13NVSSGtNNjNzKTZW+Fb78IUQa5ws+xRCCCGEuESmZXN2OIZh2vxiaxUeXcfrVqnOy2J9aQ6GaTMSTXGgY4qeyTiqqtBSnMO6shzaBmJ43Ql21gfxuOUpmBDi2pPKnxBCCCHEJdI1labibAzLomMixpcPDWCYNm5NZVtdLm5dBcehYzxG/3QCy3RILZqMxRY4PxojZVgc74/I8k8hxKqQ8CeEEEII8R4YpkX7aJw3B2YwTAu3rlId9PFmKAZAfTCbxiI/qrp0ef9UkorcTNaXBvC6dbbV5knjFyHEqpB7HiGEEEKIS5Q2TA50TVCZm4XjODQW+dlUlc0r3ZPEUgaaqvDAHaXkeDMozslgYCaBAjQV+7l7XSF7mwpkyacQYtVI+BNCCCGEuES6plKVm4VbV4mnFumenONY/wzpxUVe656iLTTDf3/qFCW5bjqGowDUF2ZzuHcG07Kl4ieEWFXy0pMQQgghxCXyuHU+u78e07IZmErQVJJN30SCpoJsHtlZia6qjMRSvPDWBJsr8/nMnXX4PG4+sL4Qn9e92ocvhFjjFMdxVvsYLktra6vT1ta22ochhBBCiDUmbZjomopp2ZiWzalwjO11+SuXA7LEUwhxTSiKctJxnNafdT25RxJCCCGEeB+Wg91y0Ntel4+uqVLhE0Jct2ThuRBCCCHEFSD7+YQQ1zu5lxJCCCGEEEKINUDCnxBCCCGEEEKsARL+hBBCCCGEEGINkPAnhBBCCCGEEGuAhD8hhBBCCCGEWAMk/AkhhBBCCCHEGiDhTwghhBBCCCHWABnyLoQQQohVkzbMt83Hk1l5Qghx9Uj4E0IIIcQ1Z1o2acPkq0cGaS72o6kq4LCzPojHLU9PhBDiapB7VyGEEEJcU9FkmtcHZgEoz/Wwp7GAxILBGwNRjvTOsLUmF5/XvcpHKYQQNx9ZWyGEEEKIayaRMvjcs+2cGpolljL4zpsjTCdTfO7Zds6NRkmkTZ44FiJtmFfktkzLXnlLGyamZV+B70IIIW5MUvkTQgghxDXjcet8YF0Jlm3j1lX+7IF16KqC7Th8cnslb4XneHR71WUt/TQtm+h8mi8+18Xuxjw8mguAc6NRNpTlsK+pCF1Tb6j9haZl31DHK4S4Pkn4E0IIIcT7thxKLjWc6JrKvuYCjvTOkEib/Ml3zpGT6UZVQVdVuicS7GksuKzjef7cKP1T89xeGeBAxxSVuT4ai/wMTCcxLRvLdvC6XNd8eely1dG07J8ItxdXJJfP5fL5NC2bwz3TbKvNe9vnXW4glEApxNqjOI6z2sdwWVpbW522trbVPgwhhBBizVheRglwsHuKLTW5vD4wy57GgpUw8bNCxfLnR+fTvBmKs6UmQCDLQ9owL6vqF02m+dyz5/nNu+r5/qlxKvO97G8uJG2aPHl4CAsHcFhXEqB7Ms5v7K3H49bfMZBdKctLTk8MzGLZDp3jMR7bWbMS8EzL5kjvDIZpYdkOvVMJ1pVks6UmF5/HjWnZvNw+gdetsbepcOVrHuyeYkdd/rse97v9DJZ/fge7p1Z+ZhIChbixKYpy0nGc1p91Pan8CSGEEOKSmZbN82fH6JqYoyY/ixc7xnhjcIbw7DxpcxGP7sZyLO5ZV5sUJ+EAACAASURBVPJTA8Xyx4L+TPa3eFb+f7nLPU+GYgR9GfznyVEqcr30TSbRlGle7BhlX1Mhjq3yas8EVcFMTg7M8GKBD01V6J6Y47f2N17RALgcsl7pnFxacloaYFNVNinD5IedE3SOz7GxLICmKsRSBqHpeQCqgpnE0wv80X+e5b4NpVi2w8udE9zdXLQS2iwbvnt6iJOhCLdV5GI5Nm5NwzCXltNqqsLO+iDwo3O6vI/yYPcUibTJga7xlesvB00JgULc3CT8CSGEEOKSRefTvNA+RnG2l/NjUWbiKUpyPEzGUpwfmaOpKJtDvZNsr80nkOW5pK95pQKHrqlsrcnl1OAMC6bF4d4ZKgNeDNOhZzROaU4mLkWlODuD8MwCDSV+usbnCM3MMzOXIr279opVwUzL5kDHBKBwfiTGwEySmmAmX3i2E9O2cXCIzBk0FPgB+M83h7l/cykezc0L50cxbZvZxAKG6dA/GafAr9M3lSCxYHBuJMaDraX866EFVFRqCzI50jPDjvogR/pmuHddCZrq8ErnJF0Tczy+qwaArxzqp77Ix/mROIPTCRzFwbLh9NAs339rlA9vLGNPo4zaEOJmJn/dQgghhLgkacPkGyeGCfrdtJT46ZlI8AutVbTW5jIaS9NUlI3XrbG1Jp/T4Ti7Gq59Jcnj1tlYmYdlO1i2za6GfAB6JmM8truaE31R2seiVOV7+XbbFB/fWkNlXiav9oxj2jbH+mbYXpd/2Xvp0obJuZEYj+2qZmOFjyePDKGiUZ2XRXm+F01R6ZyIAuB1a3xoYykjkQUebM2nayyT+qJsDMviA+sLeVVXee7cMHc35/HGQIz+6QRnQnE2V+fz8LYKvtM2Sr7fxcB0gtIcD7sa8jjWF8GwLFLGUoXPsVVMc2m/Y21BFgpQX+hnZ0Mu4GA5DhsrfLwxOHvZ378Q4volf9lCCCGEuCQet84v3VFK1+gcOAqO4/BK1yR/80I3lm2haZAybJ49M8KGMt+qBAhdU9nfUsi+5gK8bp2ToRjH+iKMzy3QNhilZ3yO8GwKHJXYvMHGqmyO9E4zNDnP8f4Id1TnXnbwO9Axwavdk/RPJzjQMcU3jg+zaNsMReZ54I5SjvbP0Ds9x2g0zcudY6QWTY4PzJBatPg//9UNioJh2nzv9Ahp02RXQx6KAm6XwqYqP5UBLwCKotA9Nk91fhZj0QVM28Glqpi2TcdYnI6RGAPTczz31jjPnx+lPD+TnrEEh3unqQp66Zmc46uHBjk3NsvAdIJ/f32ETRU5EvyEuIlJ5U8IIYQQlyzoy+ST22vZ3ZCPW9cYjqXYVRfkUO8UmqLi9Wh8bEsNgcxLW/J5NSwv3dzb9KOuobsblpYz7qo3eOJwiDtb8jjcHyCYmcEDmyrYWOEj6Mu87OBjWjbto3Eq872U5GTQMzlHc3EOfZNxHrqjlG+3jbK7Psi+5gLSpsW/Hwuzoz4Pt6bRWpPDN44N8+AdpTxzcoymEh9nwnFurwpQEcjCsh3+/cQIKAoALlVhY4WPE31RHGWpW2pDkZ/j/RFaSnJILBioqkpZIJODfZOMRtM8vrcG07Z56lgY5cIxa6hkuDQe2VZxTbufCiGuPen2KYQQQoj35OLxDmnDRNdUDnZPs6cxeMndPlfTckfRRMrA53Vf0ZEHiZTBi+3j9EwkMGwLt6rxq7ureaVzCq9b40w4SjiSpCboR1UUTNvGresY5iK3lAdIGRa903P8YmsFf/NiN3/84RbeCif47plhPri+hNuqsjnWG8GX4cKyoWMshmFZLNomm8qDaKrC8+dH+d37mvg/z3exvTbIWDTNom3zG3fWrezDXG7+ssy0bAl+QtzApNunEEIIIa6KiwPecmDY31J4XQe+iy03NFk+9it13GnD5CuHB+gej1GU42EsmqYqP4sfdk7yr4f62VlfCIqD7Tg0FPpBcfC6dG6p8POXz3UCKv3Tcc4NzZJK20xE5zkVirO9Lo+3hmZJGRZf/H4n/RNz/MZdDeysz+fsyCyD0wn6pxJsLMtja00eTx3r42jfDOPReY71z7C/qYihyDwe14+e9klTFyHWJvnLF0IIIcRlu1GC35V08dD2ZY9sq+ArB00cFCryMmkq9jM4Nc8nt9fgdWt0jMWxARuLAx2T3Lu+hPaRJOUBL7+yq5JDXREsx6GpKJupZJrtdXl4XDq1RVn0Tya5e10RRkMhO+vzOdg9jUtVubOxmJm5fjRVQVdVNlbkEZqap7U6yKO7qgj6Mq/qHEMhxI1D7gWEEEIIId6D5aWur3VNkTJMOsbjmJaF1+2iqSh7Ze9d+1gMr0unrtBH53gc23FQHIglDebTDlPxNDgK68qy+P7ZEdoGo/RPJ9AVlfBsksrcTEzb5uWOcV7tnKIoJwOAw71TdE3GONIzza/uqcGxFOJpm0Ta5N+ODqEqCg5QV+jj/EiCvU2ZEvyEEIB0+xRCCCGEuGRpw+RLB/tIpA1ShknnWJz0okkoMs8j2yrY3ZBPQ7GPnQ15TMwtsKkqG69bx7QtXKrKY3uq2ViRy1g8TdC/tP/uP14f4c6GQnbVB9lYnktt0EdjsR/HgSePDtE5Ose2ulwyNA0chfJcL7qqUhnMxKVqDM/OUxnMYiiyNEuwpSSApigXwqlF2jDfVp0UQqxdEv6EEEIIId4D03R4rXuK86MxqoKZqIrKVCyFadu82D7ON06EMG2bbTV5vBWeY2OFj8k5g6pgJqbtMB5P8+mdFfzlgxvZWhfAMC26JmIc65/hjpocdEUBR2EkluKjm0uoCmZytC9CeW4mQ5F56gp9NBb5qQv68WW4eXhbOXUFfkzHoW8yiVtX+NTOSkoCGVi2w1cO9XOgY0ICoBBCln0KIYQQQlwqXVOpK/TRPZ6gZyqGpijUFHh5rXORA52THO6ZYdFY5EDnFE8cHmBfczGW7bC9Np+D3VO83DFOJL5A20AMy7H5YcckC6aJA+iqDo6KqqpoqoJlO/w/z3cBsGjbhCJJfuGOMv7upR4cBypyM9lclc1Tx8KEZpLoqsq960sBONob4dxwjKYSP/WF2WiqvN4vhJDwJ4QQQogrYLmqpGsq0WQaj0tfGQdxM+03My2bzvEYD9xexueemeGB1lL6J1I8sr2W8VmDPQ0FvNLt4FI0agqz+OSOCk6H4oxG0/zOfY08fXyY0lwvt1Vl8/TrI9zZWIhlOwxMJ/i1PdX4PG72NS/NJ9RUlc7RGFXBLPqm5lCA758a5951JWyry+VMeI5ApodHd1bx1YMOjSV+7mwq4NWuKQan57n/tgqGZ9KsK81gZ33+mmzKI4R4u5vn3lgIIYQQq8K0bA50TKCpKrdW+PkfXz/FrZW5rC/NoXcqweO7am6aGXKJBYMDHePYONy3vpTvnhqlJs/PG6FZ9tQX4PPoNOT7cesq1Xk+TofidIzFMSyLjpEENQU+nm4b5BvHwliOjabB4NQ8Ds5PjGK475Zi7mwKcmJglpxMNzvq8lYG2C8NsfegayqBLA+fvat+JWz3TyVpLPLjdet8+NYSPG5dgp8QApA9f0IIIYS4TGnD5NxIDMt20FWV2ypyaSjyoakKhmFytG/mptlv5tF1djUWUpPn50RoBmPRRlOh0O9iJJrijpoAj++twa1ruBQFFIfq/CzGYmks20FTFYLZHj6+vYIMTUdTVD6+oxyXqpJeNDnQMcGXDvYTTaZX5ijubSpgf0shPq97JfjB28drLAc8j1vnM3true+WEvY2Fax8jhBCgIQ/IYQQQlwG07I5MTDLomWzpSbAyVCM+iI/QzPz3FLhx8LBsp3VPswrwrRsjvbNMDKbYmAmQWmOh3UlOVi2w3h8gQ/eWsRTx8Mc64tgORa1hX4s2yE0k6Qs4KW1Jof+qQSVuVkEPBlUF2TSPTFH20CMBcta+jzboSQngyePDTE9Nw+wUtFLGyaHe6ZJpAwSKeNdA/VyEJTQJ4T4cXKvIIQQQoj3TddUbq/KYWJuAdO2MUxrKeDkZfHN4yOcGYpgOTdH1U/XVLbU5OLSFeoLsnGpKigO/VMJ0osmf/9SDynDJGVYtA/HeLlzjM6xOLUFfjRF4VRoadafqiikTZPXuqaozs/CsCyGZ1MYloVhWbzaPUWhL4O/+q9upufmSRsm//xqL690TZAyLP751T7+8JlzHOiYvGkqqkKIa0PCnxBCCCEuSyDLw598uIVzIwkAVEXB59H5tT3VfHpnA3e3FN80VSifx83dzSV43SqLtoWmKjzQWopL06jIzaS+0EfXZJyaAj9lAS8Nhdm4dQVdVbmjJsC60hxsx8a0bcoCXizHZnBqnsILA9xxVAxrkQw3/O699TzdNkp60cQwbXAUNlX5qQpmck9LEVtqAjfNeRVCXBtyjyGEEEKIyxbI8rCrIcg964v57P569rcUEcjycM/6opuu22fPxBxpc5G3wlHOhKP87Qs92I7NQ1vKGZhKcmZwls6JGOFIiq6J2NK+v2AWZ8JzpBZNzgzN8m9HhzAtm8O903xoUzGdI3HOjcRILS4SGk/yw/YpfG43n9pWCUA4miS1aPKFZzv58sE+0uYiTxwNkTbMVT4jQogbiYQ/IYQQQvxMl7K88OJOlO/UlORmYFo24eg8LlXnodZKfnV3Nfubi8BR+N6bYzSV+PnCL97Cpop8ynI9uBQNw7Q51DvF+rIsRmbT/MLtVfy3PbVsrMhlf1MRoak0999Wga6q+DJc/ObdLXzh5zfgceu0haK8PhClJs/P/uZC/uz+ddzdXMydTYWoKKt9OoQQN5ib6x5ZCCGEEFfEckUpbZgkUgYHOiZIGyZpwySaTK80IFm+3nITkkTK4KX2MRIpY+Xym4nHrXPf+lI0VeWVrgm+diyMYVlomsKHbiumc2yOv3mhm8SCweScwUNby/C6dO5qLsKj69QX+ghHkgCcH40udUQ1LXpmYmgouHWVezcsVU11TWVXQ5D9LYV8Zm8tZ0fm8GW4ub0mn0Cmh8/srb2pqqpCiKtP7jGEEEII8TbTc/M83TbKL7WW8uSREIu2w/DsPEnDoHs8yanwLL+wqYLBmSSqqvDLW8r46+e7cWyFgmwXZ4ZmOTcc52Pbyvnbl/r4s4+su2nm/Omayj3ri0ikDV7sGKM6P5OD3VNYlsM/vNRLQbabieg8A9NJSvwZfKdtlEXLprk4myeOhKjM97JoO0TTBqdDszQUZmPZDke7p/nyo3dQnJ31tmrp8vs+r5vtdfkX5vsVSDdPIcT7ojjOjd1+ubW11Wlra1vtwxBCCCFuComUwR9/9xy/fU893WPznAxN01SUA8BQZJ7KvEzOjkbw6C7sC88hvG6d4uwMttXn8fTxYTqnYnxwfSljsQWq8jK5q6Xwphw0nkgZeNw6acPEtG2O9c1imBYdYzEe3VXF+ZEk68uy+MaxYeqKsrAs6J2cIzSbpDI3i7S5yO2VQdy6Rsqw+NCtJTfdORJCXBuKopx0HKf1Z11P7mGEEEKINc607JW3tGmSWjT5TtsoiQWDgZkkPZNzuHWVR3dWoWkwFktTV5hJU1E2TcV+Hmot5Zk3h/nm8TDFAQ/xhIGmKvzSHaVYjs2XDvbzUvvYTTeWYHmAus/rJpDlYV9zAV63jlvTVoLf+ZEkdYU+esYT9E0lWFeaw+/d10RzcQ6Tc4toqsqexqAEPyHENSHLPoUQQog1zLRsDvdMk7qwd+/cSAxNUXiwtZRz4QSxxALVzYV0T8yxrS6X9tE40/E0XeNJhiIJYvOLbKzMITvTtVT96p8Gx6ZjbI7usTgnQxE2VuZyPhxjX1MRuqaSNsybcq+ax62zv6WQPY1B0osmX/hBJ3c2FuJ1azQU+dHUpQYtf/9yH3sbCqjL97G7If+mPBdCiOuT3NsIIYQQa5iuqWwo8/HkkRCNxdm0lGRTlJ3BM2+O8ci2Ck6FA3h0N7aV4kjvDC5F5dbKXD6xvYKnTwxTnOPhP9tGURWF5uJsNlXkkUibeN1LXS4HZpM0FvsIT6VIL5qkF02eOBri8V016Jp60wWf5b14HrfOn31k3cqy0K8eHqS5xM+exgLubCrA49YxLfum+/6FENc3WV8ghBBCrGHRZJqnjoVZsGw0VSFlWHzzRIjkwiIJw6RjdA4Um9oiH6GZFA9tLWM8vsC3Xh9lwbJ4fTBCXYGPirxMwpEUt1Vlc6RvivaxGKGZJOWBTEJTKVK2xZcPDfClV/vomojxw84J/vm1vuuyG+jKEtgL1dDl9y9eHnspLl4W+pm9texvWZp5uHy5BD8hxLUm4U8IIYRYo9KGyVPHw1Tme3GpKnfUBCjI9vBPn76dzZX5PHNylIdur2RrbS5DM/N8fGs5HSMJygOZfGxbGVkuF5W5mexpysera9QV+DkdilMa8HJbRR4PbyvHq2nUF/oZjaaoK8iiviib8oAXTVExF5fm311PewHThsnhnmleOj/OVw4PMD03z0vnx/n/Xunl+XOjvHR+fGXsxXs57pux4Y0Q4sYjLzkJIYQQa5THrfPojioO9Uxj2zZvhmJsrsrhTDjOHTVLHT5nU/+3vXuPkbM67zj+fWbG41177V1f8N3YxjdsoAmw4RKn4U5ImwCqEpU0VWhIRaskykWtEmirpP0jUqNe0lSJIrmE3AVBNElRlNYhhHAVxCbhEgPG9hqz6+v6sma99np3Zk7/mFmzGDvYrNc76/f7kVae98yxfUZ6dGZ/73ve8/by/cfbKadq0FnbsY9Nu/fz9Cv7+MiKM1m9qYtCLsfCaRPI5+C5LV2UKLNs9njuXb2VvlTmggXNPLBuG/2VMht29LCl6yA3r5gIwPodPVyxtDLiwWggzD2+cTd/MHcCz7Z3c+P5M/jWo5uppERfpcxzHfvIRZCPoK9cppjPc+Wy6SM+dkk6XoY/SZIyrKFYIJ/LUchX79H7/hPt9PWXebq9ixkTx/KNB9dz1TnTKRbyrN7UxbwzxvFkWyfPtO+lXEm8uK2bZzu66Nh7gMuWTmXz7h427Ohm+95DXHLWZB5/egdBjpkTG3nwhU4qKZHLBT/89VbOm93MLSvmj/jyx/0H+1j5cBv95Qqbd/fw46dhTksjT23eTUfXQS5fMo1frtvJnJZGls+aCCnHS9u66a+UefeSMwx/kkYNw58kSRlWyOcO71A5sBPn4xv3UK4kDvb3c+25M1gyfSIPre/k3FllioUczU1FKsCL27oppQrLpjczd0oDExsa+If3L+PRdXsoFnKs39nNOXOaWT6zmXwuKOb2sXx2M+9aPIWGMYW6eFB5b1+JX63r5FCpxPKZk1g8bQI/W7uFSkpUgGuXzeSypVN5cfs+CrkcxXyB3lIff/KOWfzbqpfqasmqJL0ZT1VJkpRxA5uPDGxOcuWyaVy6cBIPrttJAl7ZdYBypUI+cpBy5PM5cgSLp01ka9dBntmyl+8++jJ7Dx7krie28OjGTvJ5iETtWXbB5UvP4MIFU7h6+XRaxjfUxT1wpXKFh1/axaq122jb3UOxEFy0sIXunkP0Vco8+8peiMT+vn7Wduzj7JkTOG9uE//10Ca+99grIzp2SXorvPInSZJep5DP0TCmwNzmcWzuOsCnrj6Trz2wnrfPm8hz7d2cOWkchVyOdy5uARJtO/dz6eKpNBQKLJleJFKFFYumsGLRFB5bv4d8Lijkc1y2tL6WSBby1QesL5k5js/98GmWzW7iufZuls9pYWy+QPP4MfSXy9zz5Baam8Zy8VmTmNHcxJ23tLJ2Sw8XzmumqbE40h9Dko5bpJRGegxD0tramtasWTPSw5Ak6bTT1dPLyl9tJJfLsa/vEJMaGjhzSgPrtu+nQqIhXz2H/MGLZnPXE+3VZ/3NnEBvf4ltXX30lyv0lko0FgqcP28Sly2dVlfhb+AB93t6evnB421csGAquQg6ug7y2WsX82+r1jF/chMfvnQu33lsMxMaitz67rMOP6Ovnj6LpGyLiKdSSq1v1s9ZS5IkHVVTQ5G3z5vCvClNrN64m8kTxvDQul207d7Plj0HmT9lPKlS4a4n23l5937KlQrPdnTx499s4cYLZ7Jk+gRe2Poqi6dPZMWiqXUXlgr5HK3zWni58wBTJjaSi2DpjAnMbW7gxa3dzGpuJB/BU5tepXN/H3928ZzDm9PU22eRpOPhsk9JknRUA5vBlMoVypUKW/cd5NO1K2KzmhtpahjD0pnNrNv+KoV8jrNnTqRtRw/nzWlh446DvOfcGVy2dCpNDcW6DUsDD2B/ZP1uIAFBsVCgXEns7unntvcu4MVtB/jH9y+nZXzDSA9XkobE8CdJko6pkM9RKlfY/movlUpi/bYDLJg0nkI+z6ULJ/F0+6t8/IpFQPWxEb19JZ7ctJcVi6bQUCyM+GMcjkdTY5Frzpl++PiKs6v3Jl519nSaGotMbRpXt+FVkk6E9/xJkqQ31dtXAl4Lg8Ax733zfjhJOrWO956/+j8dJ0mSRtzgK3iDg93RQp7BT5Lqk7OzJEmSJGWAV/4kSdIpUypXDi8bHbyEdODYq4aSNHwMf5Ik6ZTYf7CPR9bv4tmOvYzJ51k0fRwbtvdQTgGRuODMyYc3ipEknXyeXpMkScOmt69EqVyhq6eXOx7ZxNPte9m85wBzJzfStvMgS2Y0s3DaOCqVxIXzmnmibc/rrgZKkk6eYTu1FhH/Arwf6AM2Ah9NKXXV3rsd+BhQBj6VUlpVa78O+CqQB+5IKf3zcI1PkiQNj4Hw1ttXYuXDbYev8C2cNgGofsk3FvP8xYp5lCoVvv3oZtp2dVOqVChXRvcu5JJUz4ZzXcX9wO0ppVJEfBm4Hfh8RCwHbgLOAWYBv4iIJbW/83XgGqADWB0R96WUnh/GMUqSpJOoVK7w0LpOAPpKZdp27ae/nIhUoVjIUa4kls5s4aXt3ZQriU2dB5jeUuTnz+/nVy91MqnRB6lL0nAZtmWfKaWfp5RKtcMngDm11zcAd6eUDqWUNgEbgItqPxtSSm0ppT7g7lpfSZI0ihzs7+dtcyfwjgUtXLNsBn912QIWzZjApQsns2HnfsqpzPwp49nUeYAZzWPZ3V3iQxfNZ+veQ1w4r9lNXyRpmJyq2fUW4H9rr2cD7YPe66i1HatdkiSNEvt7+/je45v4wk+e5ws/Wcuq57exY38vKx/aSG+pxIKpTTzw4g42dvZw44UzeWzjLm68cCZrt3Vz/fkzeG5Lt/f8SdIwGVL4i4hfRMTvjvJzw6A+fw+UgB8MNB3ln0q/p/1o/++tEbEmItZ0dnYO5SNIkqSTqGFMgYvnn8EXrj+b95wzi0pK3PPrdsYXq1/zjcU881rGsXDaOJ55ZR/buw7QVCzwmWsW8tNndnD+XK/8SdJwGdI9fymlq3/f+xFxM/A+4KqU0kCQ6wDmDuo2B9hae32s9iP/35XASoDW1lbvDJckqU40FAt8/KpFNBQLXLWsyLptXRyqVGhuGsual7t416Kp7D1wkLadPcybOo5IiYfX7WbT7h6Wz2z2MQ+SNIyG7dRabefOzwPXp5QODHrrPuCmiBgbEQuAxcCvgdXA4ohYEBFFqpvC3Ddc45MkScNjIMA1FAucN3cyY3I5enr6WbvlVX75QifffuxlDpUrNI4ZQ+v8MygWcuQj+MPFU7zqJ0nDaDhPr30NGAvcHxEAT6SU/jqltDYi7gGep7oc9BMppTJARHwSWEV1F+g7U0prh3F8kiRpGBXyOa45ZzpXnH0Gvf0lGsYUKORzvHvpFJrGFinkc1y9fBqFfI5SueJVP0kaZvHaaszRqbW1Na1Zs2akhyFJkiRJIyIinkoptb5ZP9dWSJIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZYDhT5IkSZIywPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAXUX/iLiuohYFxEbIuK2kR6PJEmSJJ0O6ir8RUQe+DrwXmA58KGIWD6yo5IkSZKk0a+uwh9wEbAhpdSWUuoD7gZuGOExSZIkSdKoV2/hbzbQPui4o9YmSZIkSRqCwkgP4AhxlLb0hk4RtwK31g4PRcTvhnVUOt1NBXaN9CA0qllDOhmsIw2VNaShsoZGr3nH06newl8HMHfQ8Rxg65GdUkorgZUAEbEmpdR6aoan05E1pKGyhnQyWEcaKmtIQ2UNnf7qbdnnamBxRCyIiCJwE3DfCI9JkiRJkka9urryl1IqRcQngVVAHrgzpbR2hIclSZIkSaNeXYU/gJTSz4CfncBfWTlcY1FmWEMaKmtIJ4N1TNZXegAABKNJREFUpKGyhjRU1tBpLlJ6w34qkiRJkqTTTL3d8ydJkiRJGgajNvxFxHURsS4iNkTEbSM9HtWniJgbEQ9GxAsRsTYiPl1rnxwR90fE+tqfk2rtERH/WaurZyPigpH9BKonEZGPiN9GxE9rxwsi4slaHf2wtlEVETG2dryh9v78kRy36kNEtETEvRHxYm1OutS5SCciIj5b+y77XUTcFRENzkN6MxFxZ0TsHPxotLcy90TEzbX+6yPi5pH4LBq6URn+IiIPfB14L7Ac+FBELB/ZUalOlYC/SSktAy4BPlGrlduAB1JKi4EHasdQranFtZ9bgW+c+iGrjn0aeGHQ8ZeBr9TqaC/wsVr7x4C9KaVFwFdq/aSvAv+XUjobeBvVWnIu0nGJiNnAp4DWlNK5VDfGuwnnIb25bwPXHdF2QnNPREwGvghcDFwEfHEgMGp0GZXhj2rRbUgptaWU+oC7gRtGeEyqQymlbSml39Red1P9ZWs21Xr5Tq3bd4Aba69vAL6bqp4AWiJi5iketupQRMwB/hi4o3YcwJXAvbUuR9bRQH3dC1xV66+MioiJwLuBbwKklPpSSl04F+nEFIDGiCgA44BtOA/pTaSUHgb2HNF8onPPe4D7U0p7Ukp7gft5Y6DUKDBaw99soH3QcUetTTqm2pKX84EngekppW1QDYjAtFo3a0vH8h/A54BK7XgK0JVSKtWOB9fK4Tqqvb+v1l/ZdRbQCXyrtnT4jogYj3ORjlNKaQvwr8ArVEPfPuApnIf01pzo3OOcdJoYreHvaGeu3LZUxxQRTcB/A59JKb36+7oepc3ayriIeB+wM6X01ODmo3RNx/GesqkAXAB8I6V0PtDDa8usjsYa0uvUltjdACwAZgHjqS7RO5LzkIbiWHVjPZ0mRmv46wDmDjqeA2wdobGozkXEGKrB7wcppR/VmncMLKGq/bmz1m5t6WhWANdHxMtUl5lfSfVKYEtt+RW8vlYO11Ht/WbeuORG2dIBdKSUnqwd30s1DDoX6XhdDWxKKXWmlPqBHwHvxHlIb82Jzj3OSaeJ0Rr+VgOLaztcFane8HzfCI9Jdah2f8M3gRdSSv8+6K37gIGdqm4G/mdQ+0dqu11dAuwbWBah7Eop3Z5SmpNSmk91vvllSunDwIPAB2rdjqyjgfr6QK2/Z0gzLKW0HWiPiKW1pquA53Eu0vF7BbgkIsbVvtsGash5SG/Fic49q4BrI2JS7Sr0tbU2jTKj9iHvEfFHVM+854E7U0pfGuEhqQ5FxLuAR4DneO1erb+jet/fPcCZVL9QP5hS2lP7Qv0a1ZuYDwAfTSmtOeUDV92KiMuBv00pvS8izqJ6JXAy8Fvgz1NKhyKiAfge1XtM9wA3pZTaRmrMqg8R8XaqGwYVgTbgo1RPwjoX6bhExD8Bf0p1J+vfAn9J9b4r5yEdU0TcBVwOTAV2UN218yec4NwTEbdQ/R0K4EsppW+dys+hk2PUhj9JkiRJ0vEbrcs+JUmSJEknwPAnSZIkSRlg+JMkSZKkDDD8SZIkSVIGGP4kSZIkKQMMf5IkSZKUAYY/SZIkScoAw58kSZIkZcD/Ax5rqjFpxW4cAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7efc917de0f0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scale_factor = 1\n",
    "data = scale_factor * np.array(df)\n",
    "print(\"Number of data points: {:,}\".format(len(data)))\n",
    "\n",
    "plt.figure(figsize=(15, 11.52))\n",
    "plt.xlim(0, hard_exudates.shape[0])\n",
    "plt.ylim(0, -hard_exudates.shape[1])\n",
    "plt.scatter(*(np.array(data).T) / scale_factor, s=0.01, alpha=1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Apply and train GNG"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Create helper functions for analysis of resulting GNG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def extract_subgraphs(graph):\n",
    "    subgraphs = []\n",
    "    edges_per_node = copy.deepcopy(graph.edges_per_node)\n",
    "    \n",
    "    while edges_per_node:\n",
    "        nodes_left = list(edges_per_node.keys())\n",
    "        nodes_to_check = [nodes_left[0]]\n",
    "        subgraph = []\n",
    "        \n",
    "        while nodes_to_check:\n",
    "            node = nodes_to_check.pop()\n",
    "            subgraph.append(node)\n",
    "\n",
    "            if node in edges_per_node:\n",
    "                nodes_to_check.extend(edges_per_node[node])\n",
    "                del edges_per_node[node]\n",
    "            \n",
    "        subgraphs.append(subgraph)\n",
    "        \n",
    "    return subgraphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "class MarginClamp:\n",
    "    def __init__(self, x_px = 1500, y_px = 1152, tolerance = 0.05, x_shift = 50, y_shift = 15):\n",
    "        self.x_px = x_px\n",
    "        self.y_px = y_px\n",
    "        self.tolerance = tolerance\n",
    "        self.x_shift = x_shift\n",
    "        self.y_shift = y_shift\n",
    "        \n",
    "    def clamp(self, x, y):\n",
    "        if x < self.tolerance * self.x_px:\n",
    "            x = self.tolerance * self.x_px\n",
    "        elif x > (1 - self.tolerance) * self.x_px - self.x_shift:\n",
    "            x = (1 - self.tolerance) * self.x_px - self.x_shift\n",
    "\n",
    "        if y < self.tolerance * self.y_px:\n",
    "            y = self.tolerance * self.y_px\n",
    "        elif y > (1 - self.tolerance) * self.y_px - self.y_shift:\n",
    "            y = (1 - self.tolerance) * self.y_px - self.y_shift\n",
    "        \n",
    "        return (x, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_ideal_label_position(subgraph, x_px = 1500, y_px = 1152, preference = \"centroid\", label_length_offset=90, label_height_offset = 25):\n",
    "    \"\"\"\n",
    "    @param preference: preferred location – options are \"centroid\", \"middle\", \"corner\"\n",
    "    @type preference: str\n",
    "    \"\"\"\n",
    "    \n",
    "    assert preference in [\"centroid\", \"middle\", \"corner\"]\n",
    "    \n",
    "    clamp = MarginClamp(x_px = x_px, y_px = y_px, tolerance = 0.05, x_shift = label_length_offset, y_shift = label_height_offset)\n",
    "    \n",
    "    x_coords = [node.weight[0][0] for node in subgraph]\n",
    "    y_coords = [-1 * node.weight[0][1] for node in subgraph]\n",
    "    centroid = [np.mean(x_coords), np.mean(y_coords)]\n",
    "    \n",
    "    centroid_h = centroid[0] < x_px/2  # True => left quadrant, False => right quadrant\n",
    "    centroid_v = centroid[1] < y_px/2  # True => top quadrant, False => bottom quadrant\n",
    "    \n",
    "    xmin, xmax = (np.min(x_coords), np.max(x_coords))\n",
    "    ymin, ymax = (np.min(y_coords), np.max(y_coords))\n",
    "\n",
    "    if preference == \"centroid\":\n",
    "        x, y = (centroid[0], centroid[1])\n",
    "        \n",
    "    elif preference == \"middle\":\n",
    "        if centroid[1] < 0.25 * y_px or centroid[1] > 0.75 * y_px:\n",
    "            x = centroid[0] - label_length_offset/2\n",
    "            y = ymin - label_height_offset if centroid[1]/y_px < 0.5 else ymax + label_height_offset\n",
    "        else:\n",
    "            y = centroid[1]\n",
    "            x = xmin - label_length_offset if centroid[0]/x_px < 0.5 else xmax + label_length_offset\n",
    "        \n",
    "    else:  # => preference = \"corner\"\n",
    "        if centroid_h:\n",
    "            x = xmin - label_length_offset\n",
    "        else:\n",
    "            x = xmax\n",
    "        \n",
    "        if centroid_v:\n",
    "            y = ymin - label_height_offset\n",
    "        else:\n",
    "            y = ymax + label_height_offset\n",
    "            \n",
    "    return clamp.clamp(x, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Running and training GNG"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we're running a single subject's image with 20 starting nodes, 500 maximum nodes and 100 iterations:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "environment.reproducible()\n",
    "\n",
    "gng = algorithms.GrowingNeuralGas(\n",
    "    n_inputs=2,\n",
    "    n_start_nodes=20,\n",
    "    shuffle_data=True,\n",
    "    verbose=False,\n",
    "    step=0.4,\n",
    "    neighbour_step=0.005,\n",
    "    max_edge_age=10,\n",
    "    max_nodes=500,\n",
    "    n_iter_before_neuron_added=3,\n",
    "    after_split_error_decay_rate=0.5,\n",
    "    error_decay_rate=0.995,\n",
    "    min_distance_for_update=0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ec6d9428a2ad4d088b5812d4c62277eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in tqdm(range(100)):\n",
    "    gng.train(data, summary = 'inline', epochs = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = plt.figure(figsize=(15, 11.52))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "for (n1, n2) in gng.graph.edges:\n",
    "    point_coordinates = np.concatenate(((n1.weight[0]*[1, -1]).round(), (n2.weight[0]*[1, -1]).round())).reshape(2, 2)\n",
    "    plt.plot(point_coordinates.T[0], point_coordinates.T[1], linewidth = 0.5, color = \"white\", alpha = 0.5)\n",
    "\n",
    "ax.imshow(hard_exudates)\n",
    "\n",
    "ax.annotate(f\"SUBJECT: {SUBJECT_ID:03}\", xy=(50, 50), color=\"white\")\n",
    "ax.annotate(f\"Lesion count: {len(extract_subgraphs(gng.graph))}\", xy=(50, 90), color=\"white\")\n",
    "ax.annotate(f\"Epochs: {gng.last_epoch}\", xy=(50, 110), color=\"white\")\n",
    "ax.annotate(f\"Final neuron count: {gng.graph.n_nodes}\", xy=(50, 130), color=\"white\")\n",
    "\n",
    "for idx, val in enumerate(extract_subgraphs(gng.graph)):\n",
    "    label_pos = calculate_ideal_label_position(val, preference=\"middle\")\n",
    "    ax.annotate(f\"Lesion {idx + 1}\", xy=label_pos, color=\"white\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculating lesion area"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The total lesion area allows a quantification of hard exudate severity. We can obtain this by calculating the convex hull area of the nodes in the pruned GNG graph:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def single_lesion_area(*nodes):\n",
    "    points = np.array([[node.weight[0][0], -node.weight[0][1]] for node in nodes])\n",
    "    return ConvexHull(points).volume"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def total_lesion_area(graph):\n",
    "    lesions = extract_subgraphs(graph)\n",
    "    \n",
    "    area = 0\n",
    "    \n",
    "    for lesion in lesions:\n",
    "        try:\n",
    "            area += single_lesion_area(*lesion)\n",
    "        except Exception:\n",
    "            pass\n",
    "        \n",
    "    return area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(f\"The total lesion percentage is {total_lesion_area(gng.graph)/(1152*1500)*100:.2f}%.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing all subjects' lesion areas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the following, we will be iterating over all 89 fundoscopy images, and calculate the relative area covered by hard exudates. The relative area is calculated with reference to the entire image frame of 1152x1500 = 1,728,000 pixels (this is computationally more efficient than taking it as a fraction of the circular image it encloses). In addition, we'll also plot a heatmap of "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_hard_exudate_image_and_mask(subject_id, **kwargs):\n",
    "    raw_image = img_as_float(io.imread(f\"assets/fundoscopies/image{subject_id:03}.png\"))\n",
    "    binary_mask = convert_to_binary_mask(img_as_float(io.imread(f\"assets/hard_exudate_masks/image{subject_id:03}.png\")))\n",
    "    masked_rgb = mask_over_image(binary_mask, raw_image)\n",
    "    \n",
    "    return filter_for_hard_exudates(masked_rgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_points(hard_exudates_bw, **kwargs):\n",
    "    return np.array(convert_bitmap_to_points(hard_exudates_bw, kwargs.get(\"included_point_fraction\", 0.5)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "hard_exudate_stats = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def plot_all_images(**kwargs):\n",
    "    global hard_exudate_stats\n",
    "    hard_exudate_stats = []\n",
    "    row_count = kwargs.get(\"rows\", 10)\n",
    "    col_count = kwargs.get(\"cols\", 9)\n",
    "    sub_count = kwargs.get(\"first_n\", 88)\n",
    "    iter_count = kwargs.get(\"train_epochs\", 50)\n",
    "    max_neuron_count = kwargs.get(\"max_nodes\", 500)\n",
    "\n",
    "    fig, axarr = plt.subplots(row_count, col_count)\n",
    "    fig.set_figheight(row_count * 2)\n",
    "    fig.set_figwidth(col_count * 2)\n",
    "    fig.subplots_adjust(hspace=0.5)\n",
    "    cmap = kwargs.get(\"cmap\", cm.plasma)\n",
    "    fig.suptitle(\n",
    "        f\"Hard exudate mapping of {sub_count + 1} subjects from DIABRETDB1\\n({iter_count} iterations, {max_neuron_count} neurons)\"\n",
    "    )\n",
    "\n",
    "    for _ in tqdm(range(row_count * col_count)):\n",
    "        row_id = math.floor(_ / col_count)\n",
    "        col_id = _ % (row_count - 1)\n",
    "        subject_id = _ + 1\n",
    "\n",
    "        if int(_) <= kwargs.get(\"first_n\", 88):\n",
    "            img, bwimg = get_hard_exudate_image_and_mask(subject_id=subject_id)\n",
    "            data = generate_points(bwimg, **kwargs)\n",
    "            if len(data) > 3:\n",
    "\n",
    "                environment.reproducible()\n",
    "\n",
    "                gng = algorithms.GrowingNeuralGas(\n",
    "                    n_inputs=2,\n",
    "                    n_start_nodes=kwargs.get(\"start_nodes\", 50),\n",
    "                    shuffle_data=True,\n",
    "                    verbose=False,\n",
    "                    step=0.4,\n",
    "                    neighbour_step=0.005,\n",
    "                    max_edge_age=kwargs.get(\"max_edge_age\", 4),\n",
    "                    max_nodes=kwargs.get(\"max_nodes\", 500),\n",
    "                    n_iter_before_neuron_added=kwargs.get(\n",
    "                        \"n_iter_before_neuron_added\", 3),\n",
    "                    after_split_error_decay_rate=0.5,\n",
    "                    error_decay_rate=0.995,\n",
    "                    min_distance_for_update=0.01)\n",
    "\n",
    "                gng.train(\n",
    "                    data,\n",
    "                    summary='inline',\n",
    "                    epochs=kwargs.get(\"train_epochs\", 50))\n",
    "                axarr[row_id, col_id].axis('off')\n",
    "                axarr[row_id, col_id].imshow(img)\n",
    "\n",
    "                lesion_count = len(extract_subgraphs(gng.graph))\n",
    "                lesion_area = total_lesion_area(\n",
    "                    gng.graph) / (1152 * 1500) * 100\n",
    "\n",
    "                hard_exudate_stats.append({\n",
    "                    \"subject\": _,\n",
    "                    \"lesion_count\": lesion_count,\n",
    "                    \"lesion_area\": lesion_area,\n",
    "                    \"neurons\": gng.graph.n_nodes\n",
    "                })\n",
    "\n",
    "                for (n1, n2) in gng.graph.edges:\n",
    "                    point_coordinates = np.concatenate(\n",
    "                        ((n1.weight[0] * [1, -1]).round(),\n",
    "                         (n2.weight[0] * [1, -1]).round())).reshape(2, 2)\n",
    "                    axarr[row_id, col_id].plot(\n",
    "                        point_coordinates.T[0],\n",
    "                        point_coordinates.T[1],\n",
    "                        linewidth=0.5,\n",
    "                        color=\"white\",\n",
    "                        alpha=0.3)\n",
    "\n",
    "                axarr[row_id, col_id].set_title(\n",
    "                    f\"Subject {subject_id}: \\n {lesion_count} @ {lesion_area:.3f}%\",\n",
    "                    fontsize=8,\n",
    "                    color='black')\n",
    "\n",
    "            else:\n",
    "                axarr[row_id, col_id].axis('off')\n",
    "                axarr[row_id, col_id].imshow(img)\n",
    "                axarr[row_id, col_id].set_title(\n",
    "                    f\"Subject {subject_id}: \\n no lesions\",\n",
    "                    fontsize=8,\n",
    "                    color='green')\n",
    "                hard_exudate_stats.append({\n",
    "                    \"subject\": _,\n",
    "                    \"lesion_count\": 0,\n",
    "                    \"lesion_area\": 0.0,\n",
    "                    \"neurons\": 0\n",
    "                })\n",
    "\n",
    "        else:\n",
    "            axarr[row_id, col_id].axis('off')\n",
    "\n",
    "        stat_table = pd.DataFrame.from_records(hard_exudate_stats)\n",
    "        axarr[row_count - 1, col_count - 1].set_title(\n",
    "            \"Lesion count vs area\", fontsize=8, color=\"black\")\n",
    "\n",
    "    y = stat_table[\"lesion_area\"]\n",
    "    x = stat_table[\"lesion_count\"]\n",
    "    nbins = kwargs.get(\"cva_bins\", 300)\n",
    "\n",
    "    k = kde.gaussian_kde([x, y])\n",
    "    xi, yi = np.mgrid[x.min():x.max(\n",
    "    ):nbins * 1j, y.min():y.max():nbins * 1j]\n",
    "    zi = k(np.vstack([xi.flatten(), yi.flatten()]))\n",
    "\n",
    "    axarr[row_count - 1, col_count - 1].pcolormesh(xi, yi,\n",
    "                                                   zi.reshape(xi.shape))\n",
    "\n",
    "    if \"json_output\" in kwargs:\n",
    "        with open(kwargs.get(\"json_output\"), 'w') as f:\n",
    "            json.dump(hard_exudate_stats, f)\n",
    "\n",
    "    if \"csv_output\" in kwargs:\n",
    "        with open(kwargs.get(\"csv_output\"), \"w\") as f:\n",
    "            stat_table.to_csv(f)\n",
    "\n",
    "    if \"output\" in kwargs:\n",
    "        plt.savefig(kwargs.get(\"output\"), dpi=kwargs.get(\"output_dpi\", 150))\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Warning - running this can take considerable time!\n",
    "plot_all_images(output=\"hard_exudates.png\", csv_output=\"hard_exudates.csv\", train_epochs=100, max_edge_age=8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](hard_exudates.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# References\n",
    "\n",
    "[<a id=\"cit-fritzke1995growing\" href=\"#call-fritzke1995growing\">1</a>] B. Fritzke, ``_A growing neural gas network learns topologies_'', Advances in neural information processing systems,  1995.\n",
    "\n",
    "[<a id=\"cit-bunce2006leading\" href=\"#call-bunce2006leading\">2</a>] Bunce Catey and Wormald Richard, ``_Leading causes of certification for blindness and partial sight in England \\& Wales_'', BMC public health, vol. 6, number 1, pp. 58,  2006.\n",
    "\n",
    "[<a id=\"cit-fong2004retinopathy\" href=\"#call-fong2004retinopathy\">3</a>] Fong Donald S, Aiello Lloyd, Gardner Thomas W <em>et al.</em>, ``_Retinopathy in diabetes_'', Diabetes care, vol. 27, number suppl 1, pp. s84--s87,  2004.\n",
    "\n",
    "[<a id=\"cit-kauppi2007diaretdb1\" href=\"#call-kauppi2007diaretdb1\">4</a>] Kauppi Tomi, Kalesnykiene Valentina, Kamarainen Joni-Kristian <em>et al.</em>, ``_DIARETDB1-standard diabetic retino-pathy database_'', IMAGERET---Optimal Detection and Decision-Support Diagnosis of Diabetic Retinopathy, vol. , number , pp. ,  2007.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Initialization Cell",
  "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": "3.6.7"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": false,
   "bibliofile": "biblio.bib",
   "cite_by": "number",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": false,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {},
   "toc_section_display": false,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "00518c0c147842f5a006c76e6dac565a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "00bb70333765441bbcd4ca067c744ac1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "015fb3bd6dfb4b89b36a5d8fa621d8e6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "019c92a2556c4ea4adc5b807f427f583": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "027adda42c304a79a0c1f61b7c32dda8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "02d7469796ed4348b5005ca26f3bf6e6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0303297ce1dd48c2bf42ce14d16149c7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "034b5b2f80a74be5ab25f5183eb677ff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_14d937ac8975463baf52f1860c2ef82f",
       "style": "IPY_MODEL_fe9209585e9b4a21bd87c71f9952be37",
       "value": "100% 90/90 [08:27&lt;00:00,  5.64s/it]"
      }
     },
     "035f4d0233a642f797ab26c2e24abba0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0433ac9d1a5e40b890f63485ac251d1c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_18d93412afc34729925c9496fa5bf56b",
       "style": "IPY_MODEL_9856a0f6e4f041a5a7b83e4c447670d9"
      }
     },
     "050535794586480bb38ed602422a576a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_ac875de57e5047649cbe2dc50473c9d5",
        "IPY_MODEL_4b9c8bfb425c4fd1a852da87eaada62e"
       ],
       "layout": "IPY_MODEL_3dd19c9edd674bf0a8c42b7ade38bdde"
      }
     },
     "0510cdcd983346e6b02a04d68a8a2088": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "058c3c1e035d4b52a3ee4175394ba337": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0730ec53f64f4be99d7759ab6f71d419": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0755420340cb4b588f03fc0f4a3fc65a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "08fab5952a154a40a0fb0abe83454f92": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_1431a9294b464b68a790214a3cc3e240",
       "max": 10,
       "style": "IPY_MODEL_c6754d09255743fbaaed124e7055b5ad"
      }
     },
     "0925e4c625154b2abe95f696a1fdbcba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "09a99109fd994737ba25b311aaaa3314": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_695c123a6c9846448a6a2329bfe8b819",
       "max": 10,
       "style": "IPY_MODEL_19a8cf0e3a93411089a23b35c77d457a",
       "value": 10
      }
     },
     "09eb006953344082b8a981531bf33789": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_09a99109fd994737ba25b311aaaa3314",
        "IPY_MODEL_2a8ca92234524e3b86309150905635d1"
       ],
       "layout": "IPY_MODEL_30d6864761cb429d95859b2f523d027f"
      }
     },
     "09f73f4a467e4839a42064c1e5b57466": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0aae941cebf645b295c001a23c3fc161": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0abc135fc1c3483281cacdd750204e64": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_027adda42c304a79a0c1f61b7c32dda8",
       "style": "IPY_MODEL_2e48129c00f0428fb4a6d5800244cc07"
      }
     },
     "0af6a8765044498092854f38ed428a5d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0b92273d260f4816b5b6c81791ed3878": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_0e676ef5da6f4af1926523806ae6dd2e",
        "IPY_MODEL_b9e1a6247956495d97816f872caf479c"
       ],
       "layout": "IPY_MODEL_45fff1bb52db488a9e51e5b86868a6ae"
      }
     },
     "0bd4513d13694aedbec868b8e360625e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "0c9bf045ebcb46b59453d7b52b60dade": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_7de628796a044896af9e94eb6529e0eb",
        "IPY_MODEL_de8e6c2fa3c040fb99a54c63f406f549"
       ],
       "layout": "IPY_MODEL_2ae6faf9b449470187a78655c9251d2a"
      }
     },
     "0d4b082f3e75457cbe69542d73744af5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_ad753ca0d3ab4b2fb16d01b58816586f",
        "IPY_MODEL_98f7ba06cc9b48f28b2e455ae41c2636"
       ],
       "layout": "IPY_MODEL_fb2de3ebf4c144b08e2ab8945b1b8e5f"
      }
     },
     "0e676ef5da6f4af1926523806ae6dd2e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_56b685d612b54613bb915a720be48501",
       "max": 90,
       "style": "IPY_MODEL_803e0abfdcf44b4c82b6430db525d35f",
       "value": 90
      }
     },
     "0e9d7e781f0c4bc0a16e2d9a8286fb59": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_b71180b70dac49b1ac479f89f65e44a4",
        "IPY_MODEL_e0a26e285aa1437397fd0d4002b94cc5"
       ],
       "layout": "IPY_MODEL_a817519a23df423ba2f9d9f7549c4a18"
      }
     },
     "0f78f21c722144189749f994a8395491": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_ced51f4e76b24fec84bf66f3cb101463",
       "max": 88,
       "style": "IPY_MODEL_612348b659584a878c8f2fbe7b40d7b4",
       "value": 88
      }
     },
     "111bb75fc3534994b239c64d575f0d84": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_597401d973b34b9886e8ef9f6b189ca4",
       "max": 28,
       "style": "IPY_MODEL_17442701588a4b98b950b610d2de4e0c"
      }
     },
     "1431a9294b464b68a790214a3cc3e240": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "145e9df01826441a9da7dc4b18c0ff65": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "14d937ac8975463baf52f1860c2ef82f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "15d454b2144545f982fcdac5169ca8b7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_473d2836b04042b3b9d045d60ab5190f",
        "IPY_MODEL_5c0767c8f21e445a86ebb553a747eff3"
       ],
       "layout": "IPY_MODEL_a57579aaef93477e8676acade42f593f"
      }
     },
     "17112025f4fe4000bdeeed11c20e6d04": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_9fba272e5bf3437a9474f7fbb5f0591e",
       "max": 25,
       "style": "IPY_MODEL_9a13734ceaaa4b3b9a4668020dd4d3c7",
       "value": 25
      }
     },
     "17442701588a4b98b950b610d2de4e0c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1781494ed02740ca869bd1e064eb99bb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_4be104624cd44e0cbd2312834e2b652d",
       "style": "IPY_MODEL_8b406344a796486d9ff7e47831ee60b3",
       "value": "  0% 0/9 [00:00&lt;?, ?it/s]"
      }
     },
     "189d783df2944b29a352b86674403673": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "18aa8ef20501425499b18777348e6e3e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "18cad668f16c4d01a6527bbb37df2980": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9ace522cbb7d4303926bb85d64a3cca7",
       "style": "IPY_MODEL_71d7114256f54b1a89789066e32b5eaf",
       "value": "100% 10/10 [00:51&lt;00:00,  5.15s/it]"
      }
     },
     "18d93412afc34729925c9496fa5bf56b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "19a8cf0e3a93411089a23b35c77d457a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1a755127dae1468b91c16a0093d315cb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1ab0ac2d2b1b419ba4473189bd8bcb0f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1bc2c1a1aa454c4590b95b3059498f21": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1bda065d12db43c884f7938839e169fc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1c3c9b110e544016b44de79129dd6ffc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_8db922e7c07443bc8aba008ce0da946a",
       "max": 90,
       "style": "IPY_MODEL_fe99ea4107cf4a46914a447d8770b58c"
      }
     },
     "1c9491a21b5f4cd5a0800724f1a78825": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1ca0f884f0534d60b59ec7192a810cbe": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1d3a98fd72d74f8a9a579f8fc23ea6ae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1e009d8fb86445c5a21d62e5ae9df835": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "1ed2ed15dc754d21b7a49f0c81e76046": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1ca0f884f0534d60b59ec7192a810cbe",
       "style": "IPY_MODEL_189d783df2944b29a352b86674403673",
       "value": "100% 100/100 [14:12&lt;00:00,  8.56s/it]"
      }
     },
     "1f000f15cc6f41e082081f74dea1f555": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1f1ff649f3604cfeaecefb2cb0ed41c4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_90f483bac0204ef3bef71059f8431544",
        "IPY_MODEL_8efd3c760401458b94d42e5c7ddb276d"
       ],
       "layout": "IPY_MODEL_d086716c9a574e778ffea50e2535354c"
      }
     },
     "1f93bb88e9bc43a08316a7da921412a3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_d1bdf1fc287a45e8970002dd9acebdf8",
        "IPY_MODEL_647ee78851bb4422971b49ea656dd6f3"
       ],
       "layout": "IPY_MODEL_a93ecfb2df924818921dc833ad416535"
      }
     },
     "20952d855a014ab1a824ed09f0a67027": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "20bbe4e5dec849d58499f55f24582fc1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "222f429c25c0496bb009acb1d0b1cf00": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "22e2504ba9054f618f57cee5f2a09a0d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_af5025986ad84838bd55b263951fdf08",
        "IPY_MODEL_c4806627ddae446b972a697404a873b2"
       ],
       "layout": "IPY_MODEL_5cdcf66a594d45208df9448f4609d0e8"
      }
     },
     "22f93b3278374f5dacacf7dc14949088": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_fb01204cb4f24aec8a48c30c735b86fe",
       "max": 9,
       "style": "IPY_MODEL_145e9df01826441a9da7dc4b18c0ff65"
      }
     },
     "23f841d3843f479aba5ed9264c78c0f3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "24ed3d759d5d41f4a093343dc0303ca2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "25b4dd1981ed4667b672ad4164ec1e0a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "26fcd982a99343088ec4c77f28a95892": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2922b24b8e1a4e8ebafdd1320d1ea887": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "29756107fdd648adb95aa072ac8d1481": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_54bf007568bb4f4b96fc110f3e30e5e9",
        "IPY_MODEL_1ed2ed15dc754d21b7a49f0c81e76046"
       ],
       "layout": "IPY_MODEL_998ed64048194fde86d65f451e625807"
      }
     },
     "29cd4fe9727d482d91fbaa7ad7461caf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_943ba0e6aa8c4ae88de4cc494d463e2e",
       "max": 90,
       "style": "IPY_MODEL_f426a0d865e347759f68be85b90cfbb8",
       "value": 90
      }
     },
     "2a8ca92234524e3b86309150905635d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d83623f789a3408192e95f9b7a56fd83",
       "style": "IPY_MODEL_7e643c6844d64c9cb5964017b28ba56c",
       "value": "100% 10/10 [00:51&lt;00:00,  5.15s/it]"
      }
     },
     "2ac38c129f6d4c8380caaa15c539eb5d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2ae6faf9b449470187a78655c9251d2a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2ae8fc1c118341ba9461d0d9e59c3b5d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2b815be45b9c4228a115ac6f0e8e7ad0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_111bb75fc3534994b239c64d575f0d84",
        "IPY_MODEL_bc0bc4336e654c149ee1d3937b4fef6b"
       ],
       "layout": "IPY_MODEL_fb517b53d19a42f0991652b26df16e8e"
      }
     },
     "2c2b42f928064940a75ccb2d219f0b3b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2dcd6b64110b4831a65a25c47cdbb0b6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2e48129c00f0428fb4a6d5800244cc07": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "30458e818a364e41a4bdeae7082f09c9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_20bbe4e5dec849d58499f55f24582fc1",
       "max": 88,
       "style": "IPY_MODEL_540190125e87431687f1e418425e76f9",
       "value": 3
      }
     },
     "309369b6ad1645cebb279d52a0625b34": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "30d6864761cb429d95859b2f523d027f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "315243b4ba234789b24cabe78407a37f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_f9ee2e87a5004d5f80ac4c183fd3d0b3",
        "IPY_MODEL_034b5b2f80a74be5ab25f5183eb677ff"
       ],
       "layout": "IPY_MODEL_3fd143e7e4414516b4bf2d68c75ef906"
      }
     },
     "31b115c05027497d83188a0579e8c116": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_eeb510040e9c46fe95f34d82a0297333",
        "IPY_MODEL_d1d69545e01a407cb08df3bf0231eeef"
       ],
       "layout": "IPY_MODEL_25b4dd1981ed4667b672ad4164ec1e0a"
      }
     },
     "33141f46ec6546fe9685064cd20e2fc9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "366f659758b34c87afaf2147917ede65": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_5f0dea9b75884cc3bb5f11e025854688",
       "style": "IPY_MODEL_fcee3acd03ef4c57a2be82a8f4198059",
       "value": "100% 90/90 [08:33&lt;00:00,  5.71s/it]"
      }
     },
     "379d53ed1b394f22a7030319d48ad9e9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "37c9d3e60f4f422bb0a9c1408dd33aa1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "398c1836eac94eceb05b838000bbe607": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_90809530e68a483f8b8d61cfd17d6f32",
        "IPY_MODEL_6f6115de7dc5475db27295379d2dc510"
       ],
       "layout": "IPY_MODEL_c976210a1ce3463e9e8c9a4d1f1eb234"
      }
     },
     "3a943b7cacf24eafa28f2e9fa27ad13f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_6cb6e242728f4e70a027f6bc60f4430d",
        "IPY_MODEL_aa00cb24dc7d4313a047c7690925c919"
       ],
       "layout": "IPY_MODEL_9c0d418a97364580aeb54a86b91f054e"
      }
     },
     "3aa5427c81704e9ab37458c2adf9ae1a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3b2d377afc0f4617bfe8983590a518b5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_4da889a4aa1b476d9088614d0c0bbf11",
       "style": "IPY_MODEL_e36c4eb1e6e14d8bbc910b699cfeb015",
       "value": "100% 90/90 [05:09&lt;00:00,  3.44s/it]"
      }
     },
     "3dd19c9edd674bf0a8c42b7ade38bdde": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3dede50410a14b6f8b5f819277c343fa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "3e06aae64695403394fcd309df12ced8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_7da369be40d041cfa35b12648753f13e",
       "style": "IPY_MODEL_1c9491a21b5f4cd5a0800724f1a78825",
       "value": "  0% 0/25 [00:00&lt;?, ?it/s]"
      }
     },
     "3ea31605fc0041f6b7c48f4e56bebe2f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ed5a62d3d2d6413daef65ebed1609ca4",
       "style": "IPY_MODEL_b909f4b33ff74b28852566f818e51ad1",
       "value": "100% 88/88 [07:31&lt;00:00,  5.11s/it]"
      }
     },
     "3f392834ee304d1a8f3eea56009f65cd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "3fd143e7e4414516b4bf2d68c75ef906": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "409cc44278be4938adebef572a754646": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4121db4f1ea444d0948168829307fed8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "41bb17453e394ca7aa5e315bdab58bb2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "41bff7f42dc34229b50e32d96097bb7a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_17112025f4fe4000bdeeed11c20e6d04",
        "IPY_MODEL_aa196ac414294831b9c4f6be4411497d"
       ],
       "layout": "IPY_MODEL_97cf7328b6bd4109912a63705b168869"
      }
     },
     "41e1e763c865476080594157a084957f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "43b4f17e7b3d412e96ec8902d0f46b52": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_d8474593b70e4ce285f7aa1b29b53eae",
        "IPY_MODEL_dc66ee113db94d458385bce9951ebbe7"
       ],
       "layout": "IPY_MODEL_b555ed8e7544400288fb0dc7ddae6018"
      }
     },
     "45fff1bb52db488a9e51e5b86868a6ae": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "473d2836b04042b3b9d045d60ab5190f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_b924013709a44841a0761693753ce766",
       "max": 88,
       "style": "IPY_MODEL_7f09038fa39d4f46958017488887aea2",
       "value": 88
      }
     },
     "483de0f0f3c4423289ccee8612a78d27": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "493eed17a8a8424ba40aa7d924eaf6ff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_cc12d13e77814bafa7660677de3d2226",
        "IPY_MODEL_65d6f4c31e45473bba58eda6a3b49eba"
       ],
       "layout": "IPY_MODEL_20952d855a014ab1a824ed09f0a67027"
      }
     },
     "49d1db9d12ce47359beb75704a5baf16": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "4b456a75f3924eecbb6b764900caf834": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "4b9c8bfb425c4fd1a852da87eaada62e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b7d7f107533a4333a992849d74ec6b1a",
       "style": "IPY_MODEL_4bad4f0a087641c3a76cfbf8b3f29c88",
       "value": "100% 90/90 [08:30&lt;00:00,  5.67s/it]"
      }
     },
     "4bad4f0a087641c3a76cfbf8b3f29c88": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "4be104624cd44e0cbd2312834e2b652d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4c78a675f59c459a83606b69301a9949": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6c45294b310a46c69ce3015a5fc77fef",
       "style": "IPY_MODEL_d4196c89b27c4c26adda44727062c3a8",
       "value": "  0% 0/90 [00:00&lt;?, ?it/s]"
      }
     },
     "4da889a4aa1b476d9088614d0c0bbf11": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "4f44b167b17a449f974876a7278d81fd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_988ed663ca2d474ab8dbaa83ee7aeacb",
       "style": "IPY_MODEL_7dfc5809239847178a649016ca3cfe1b",
       "value": "100% 28/28 [00:00&lt;00:00, 1321.53it/s]"
      }
     },
     "506b01250ee949c69513899431d2193a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "50ddd8be2704467dbd1c5362fa27bc2e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "51657b0835704d6eb7660ff87f8f7d8f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_819f87f6c29b4a81bdbb7f0b89e1a0b2",
       "style": "IPY_MODEL_7b76779a459341f89c32bd3f38b9c4bc",
       "value": "100% 100/100 [14:30&lt;00:00,  8.71s/it]"
      }
     },
     "53817148b9b14f17b6c83f6f2a5204f1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_757f51f4d6264225ab39230260a0efad",
        "IPY_MODEL_b16a2b2f4d614ea2ad373a3cd71a6d44"
       ],
       "layout": "IPY_MODEL_09f73f4a467e4839a42064c1e5b57466"
      }
     },
     "53d612b7354c41dcabcc5cd9d2d752f4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "53d89bf0ac2f4707ab9386afd2d61235": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "540190125e87431687f1e418425e76f9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "547b5c9731f8425998590cc45e32cb17": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "54bf007568bb4f4b96fc110f3e30e5e9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_b765225ac3f243a08234498ce74a3d6c",
       "style": "IPY_MODEL_2922b24b8e1a4e8ebafdd1320d1ea887",
       "value": 100
      }
     },
     "55b2abb6090d4989a76ec32b2c77897b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_0433ac9d1a5e40b890f63485ac251d1c",
        "IPY_MODEL_8ab59ba210b148689f0782aaaf051010"
       ],
       "layout": "IPY_MODEL_c0c2a6850e0a4135a073c3ca3fb79398"
      }
     },
     "55c92282d2684c1cbbaec34d911fb8eb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "layout": "IPY_MODEL_5b10855ffbc34a1bb44c40895d44848e",
       "max": 90,
       "style": "IPY_MODEL_41bb17453e394ca7aa5e315bdab58bb2"
      }
     },
     "56b685d612b54613bb915a720be48501": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "56e9ede463924e29ab77a63e4b84be55": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_fa465225c61942198ecc234f7ef45a38",
       "style": "IPY_MODEL_2ac38c129f6d4c8380caaa15c539eb5d",
       "value": "  0% 0/28 [00:00&lt;?, ?it/s]"
      }
     },
     "577acaa6b1274256915c4479152c1284": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "57ff1eda83e848bb9c2fc8959ba97daa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "584704dd472e44f5b51c450f0a7fda25": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "58baeff11a034988aecdd80d5cac4a03": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "597401d973b34b9886e8ef9f6b189ca4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5b10855ffbc34a1bb44c40895d44848e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5b14010168f84b358a04d5e873fd0f32": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_0f78f21c722144189749f994a8395491",
        "IPY_MODEL_3ea31605fc0041f6b7c48f4e56bebe2f"
       ],
       "layout": "IPY_MODEL_0aae941cebf645b295c001a23c3fc161"
      }
     },
     "5b87853341e743b69e09a26e35ae37ea": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "5c0767c8f21e445a86ebb553a747eff3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_c70636eb78c1478a9f5db863eccd930b",
       "style": "IPY_MODEL_76f5b57cc7d742abbf2e698fc905e0ed",
       "value": "100% 88/88 [07:30&lt;00:00,  5.12s/it]"
      }
     },
     "5c6cb1dca2cb406485b70228083024cf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_0abc135fc1c3483281cacdd750204e64",
        "IPY_MODEL_9ade876972b643dcb829782ff633d978"
       ],
       "layout": "IPY_MODEL_9a4e74b25d474f2eac80510e9e6b46e1"
      }
     },
     "5ccbfc6d40a843bea5749e4de71c64b3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_55c92282d2684c1cbbaec34d911fb8eb",
        "IPY_MODEL_8a941b9c242948fca37ffbc4fa44f036"
       ],
       "layout": "IPY_MODEL_759ff96b27354a5c92789805964ed1b7"
      }
     },
     "5ccccd970fc44595ac93c33d3cc310fd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "5cda27c7def047d591f0dee574a6d7bb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5cdcf66a594d45208df9448f4609d0e8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5f0dea9b75884cc3bb5f11e025854688": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "60a2001ef02d477d90e959583af749ff": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "612348b659584a878c8f2fbe7b40d7b4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6142c5cee965411c86da96afbd620933": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6160eb9f9de74da0bf94bc7b1b8e49d0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "646435b6956b4f119ccd43be89a6d4d7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b900b5d56f854e47bffd581f7c12b13b",
       "style": "IPY_MODEL_bb55af2679c04ca195cee18aa74c5a07",
       "value": "  0% 0/10 [00:00&lt;?, ?it/s]"
      }
     },
     "647ee78851bb4422971b49ea656dd6f3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d799e98fa6204f1c83f8f21d27254eae",
       "style": "IPY_MODEL_5b87853341e743b69e09a26e35ae37ea",
       "value": "  6% 6/100 [16:25&lt;4:20:34, 166.32s/it]"
      }
     },
     "64fa9b445c4b437a8a5b03373e61cfbe": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_cc7955fa6d8c4e2688c84cb23f1210a2",
       "max": 90,
       "style": "IPY_MODEL_8bbef66d358f4fe39ca312fe94bcb019",
       "value": 90
      }
     },
     "65d6f4c31e45473bba58eda6a3b49eba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_53d89bf0ac2f4707ab9386afd2d61235",
       "style": "IPY_MODEL_3dede50410a14b6f8b5f819277c343fa",
       "value": "100% 25/25 [02:08&lt;00:00,  5.15s/it]"
      }
     },
     "662203e107a34036a68fc427d5706bf9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6650faf722e54650886cf9a889976613": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "665801d996dd433e9fd6cf35178bfa7a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "67890f5565f84132b5afb3508fcb1f50": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "688b10a2664a420cad021abd21216866": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_1bda065d12db43c884f7938839e169fc",
       "max": 10,
       "style": "IPY_MODEL_49d1db9d12ce47359beb75704a5baf16"
      }
     },
     "68c8aa674d9b40ffb6cabf5daee6f2fb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "695c123a6c9846448a6a2329bfe8b819": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6bf1ea47bdbb4c978d7d1ad9954b8706": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_cb9e90fe5368457e8948f5dcb1d0a9bd",
        "IPY_MODEL_18cad668f16c4d01a6527bbb37df2980"
       ],
       "layout": "IPY_MODEL_74781be418cb4a32b07a982932546865"
      }
     },
     "6c45294b310a46c69ce3015a5fc77fef": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6cb6e242728f4e70a027f6bc60f4430d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_2c2b42f928064940a75ccb2d219f0b3b",
       "max": 25,
       "style": "IPY_MODEL_6142c5cee965411c86da96afbd620933"
      }
     },
     "6f6115de7dc5475db27295379d2dc510": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_eedb1f3627ac4ed5946a02f825f80d5d",
       "style": "IPY_MODEL_665801d996dd433e9fd6cf35178bfa7a",
       "value": "100% 10/10 [00:51&lt;00:00,  5.14s/it]"
      }
     },
     "6fbce630aaa94a1a94f4ea834d7a4dab": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "71d7114256f54b1a89789066e32b5eaf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "74781be418cb4a32b07a982932546865": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "757f51f4d6264225ab39230260a0efad": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_858b3232322a4ca5a3545a0511b2daf6",
       "style": "IPY_MODEL_e40c930925b44edd9f5d855d0a51ad3d",
       "value": 100
      }
     },
     "759ff96b27354a5c92789805964ed1b7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "75f703b6e13344c48c6842c31c1881a7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_ca17c544a2ba4e3e8fc97845ce158389",
       "max": 90,
       "style": "IPY_MODEL_e02d675509084f77828138f14141f831",
       "value": 90
      }
     },
     "76d618d738da49ccadbe9535e8cbefc1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_95b1f69f054f472296a2315e1fbaa43b",
       "max": 25,
       "style": "IPY_MODEL_b53f4441deda48b594c6fa747ec56e76",
       "value": 25
      }
     },
     "76f5b57cc7d742abbf2e698fc905e0ed": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7a7ffae8791f4d618153661b5997cdbd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f5281ad0c9b848dbafa428b8ceffeb68",
       "style": "IPY_MODEL_6160eb9f9de74da0bf94bc7b1b8e49d0",
       "value": "  3% 3/88 [00:15&lt;07:19,  5.17s/it]"
      }
     },
     "7b76779a459341f89c32bd3f38b9c4bc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7b8cbc5bf55a48efb5d23c3699993d4b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7da369be40d041cfa35b12648753f13e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7de628796a044896af9e94eb6529e0eb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_d6403c7209a04062b78ae000955a302c",
       "max": 10,
       "style": "IPY_MODEL_a086e9a66d0e4c88ac11d2e23e60590c"
      }
     },
     "7dfc5809239847178a649016ca3cfe1b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7e13c14b58384b9883dca857ec979778": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7e643c6844d64c9cb5964017b28ba56c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "7ef8ccd630374243abb0a27bdce1d00e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7f09038fa39d4f46958017488887aea2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "803e0abfdcf44b4c82b6430db525d35f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "809d2374470541f1b9189c487405ad1f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "80bb2579f1674b46b3da3c57b2a7f099": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_015fb3bd6dfb4b89b36a5d8fa621d8e6",
       "style": "IPY_MODEL_67890f5565f84132b5afb3508fcb1f50",
       "value": 100
      }
     },
     "819f87f6c29b4a81bdbb7f0b89e1a0b2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "82655ff32e8341fb8e6c0b95bbb329a5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "82b43b4fc8c1403397c8a1f817bb8472": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8493dd76d772428dbae6d61acd69e237": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "858b3232322a4ca5a3545a0511b2daf6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8688d276953a4209ac8d04d6004ef39b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_a49a19fd5b834097a45f680c2348e68e",
        "IPY_MODEL_51657b0835704d6eb7660ff87f8f7d8f"
       ],
       "layout": "IPY_MODEL_50ddd8be2704467dbd1c5362fa27bc2e"
      }
     },
     "887bfad434104221afa6a93d1f455ef3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "88f156f74c144e32b17542595b838e9c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "89372cf9870242d0afd62f481fb3e7b8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8a941b9c242948fca37ffbc4fa44f036": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f1eda9bb799543208e2c7e4168c35516",
       "style": "IPY_MODEL_24ed3d759d5d41f4a093343dc0303ca2",
       "value": "  0% 0/90 [00:00&lt;?, ?it/s]"
      }
     },
     "8a970e858a17453ba8cfddc0ed76c991": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_2ae8fc1c118341ba9461d0d9e59c3b5d",
       "max": 90,
       "style": "IPY_MODEL_0bd4513d13694aedbec868b8e360625e",
       "value": 90
      }
     },
     "8ab59ba210b148689f0782aaaf051010": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_7b8cbc5bf55a48efb5d23c3699993d4b",
       "style": "IPY_MODEL_a1ff6d8e230544d89b0b24dbe78e35fd",
       "value": "  0% 0/100 [00:00&lt;?, ?it/s]"
      }
     },
     "8ab741ff74ad4d0bbb83667676ceb105": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8b406344a796486d9ff7e47831ee60b3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8bbef66d358f4fe39ca312fe94bcb019": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "8db922e7c07443bc8aba008ce0da946a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "8df2c7f932bf4e2fb314fbe2dab70825": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9a063cbddab5438caadd6252c2fc70da",
       "style": "IPY_MODEL_2dcd6b64110b4831a65a25c47cdbb0b6",
       "value": "100% 90/90 [5:48:29&lt;00:00, 232.33s/it]"
      }
     },
     "8efd3c760401458b94d42e5c7ddb276d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_9bc966462eb546299239bce7be1b59a1",
       "style": "IPY_MODEL_e0d636b29b3a4641b125db3d27fe696c",
       "value": "100% 25/25 [02:08&lt;00:00,  5.17s/it]"
      }
     },
     "8f1582049ca64ee9b0b15138baa35a4d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_00518c0c147842f5a006c76e6dac565a",
       "style": "IPY_MODEL_1e009d8fb86445c5a21d62e5ae9df835",
       "value": "  0% 0/9 [00:00&lt;?, ?it/s]"
      }
     },
     "8f6fa134fe4c4d61a3f1dfaf7e14915e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "90809530e68a483f8b8d61cfd17d6f32": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_058c3c1e035d4b52a3ee4175394ba337",
       "max": 10,
       "style": "IPY_MODEL_82b43b4fc8c1403397c8a1f817bb8472",
       "value": 10
      }
     },
     "909bda28c1d4492e9d3470eacd430d98": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_eb8efefa18e44c7ca789776b18658160",
        "IPY_MODEL_1781494ed02740ca869bd1e064eb99bb"
       ],
       "layout": "IPY_MODEL_1bc2c1a1aa454c4590b95b3059498f21"
      }
     },
     "90e05c2cbc3046c899c17d613aeccfcc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_29cd4fe9727d482d91fbaa7ad7461caf",
        "IPY_MODEL_8df2c7f932bf4e2fb314fbe2dab70825"
       ],
       "layout": "IPY_MODEL_4121db4f1ea444d0948168829307fed8"
      }
     },
     "90f483bac0204ef3bef71059f8431544": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_0510cdcd983346e6b02a04d68a8a2088",
       "max": 25,
       "style": "IPY_MODEL_887bfad434104221afa6a93d1f455ef3",
       "value": 25
      }
     },
     "911531db34964ba495a851298ce02a74": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9346cb026615479e8b1776f3e71a887d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "943ba0e6aa8c4ae88de4cc494d463e2e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "95b1f69f054f472296a2315e1fbaa43b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "97cf7328b6bd4109912a63705b168869": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9856a0f6e4f041a5a7b83e4c447670d9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "988ed663ca2d474ab8dbaa83ee7aeacb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "98f7ba06cc9b48f28b2e455ae41c2636": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b15faa81899c44d5a946a56d38e66ff8",
       "style": "IPY_MODEL_d4efc94cd337496ab9cf5a84254afda0",
       "value": " 84% 21/25 [01:48&lt;00:20,  5.15s/it]"
      }
     },
     "998ed64048194fde86d65f451e625807": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9a063cbddab5438caadd6252c2fc70da": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9a13734ceaaa4b3b9a4668020dd4d3c7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "9a4e74b25d474f2eac80510e9e6b46e1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9ace522cbb7d4303926bb85d64a3cca7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9ad7b812cb9b47d887b2c6657bf1cb0e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9ade876972b643dcb829782ff633d978": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1ab0ac2d2b1b419ba4473189bd8bcb0f",
       "style": "IPY_MODEL_7e13c14b58384b9883dca857ec979778",
       "value": "  0% 0/100 [00:00&lt;?, ?it/s]"
      }
     },
     "9b44b7140fc84627b1b124db44ff85b3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9bc966462eb546299239bce7be1b59a1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9c0d418a97364580aeb54a86b91f054e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9d7fb34e5e4c41b8961d0b7a6e72fdba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "9fba272e5bf3437a9474f7fbb5f0591e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a086e9a66d0e4c88ac11d2e23e60590c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a1b4612bb49741c3bdbd3d1704a17104": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_1c3c9b110e544016b44de79129dd6ffc",
        "IPY_MODEL_4c78a675f59c459a83606b69301a9949"
       ],
       "layout": "IPY_MODEL_6650faf722e54650886cf9a889976613"
      }
     },
     "a1df2afe6ea24698a55d13d59bd51ee7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a1ff6d8e230544d89b0b24dbe78e35fd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a49a19fd5b834097a45f680c2348e68e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_68c8aa674d9b40ffb6cabf5daee6f2fb",
       "style": "IPY_MODEL_33141f46ec6546fe9685064cd20e2fc9",
       "value": 100
      }
     },
     "a57579aaef93477e8676acade42f593f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a5fc2c1c1ab74173af6705948e16e6b2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_f577ceaefdc64d87b930c1f97324bd9f",
        "IPY_MODEL_4f44b167b17a449f974876a7278d81fd"
       ],
       "layout": "IPY_MODEL_cea73cba106144c8beae5fbcbc586fb5"
      }
     },
     "a61e64444efa4aaf8550cfed4dfe491a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a817519a23df423ba2f9d9f7549c4a18": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a81c0e5b00be4ee494a48af7c3e825ea": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a8f51f671143498a89418c569a2da675": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_e0ccad31d26b4b96a4e767f6b05f712b",
        "IPY_MODEL_8f1582049ca64ee9b0b15138baa35a4d"
       ],
       "layout": "IPY_MODEL_409cc44278be4938adebef572a754646"
      }
     },
     "a93ecfb2df924818921dc833ad416535": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "aa00cb24dc7d4313a047c7690925c919": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_809d2374470541f1b9189c487405ad1f",
       "style": "IPY_MODEL_547b5c9731f8425998590cc45e32cb17",
       "value": "  0% 0/25 [00:00&lt;?, ?it/s]"
      }
     },
     "aa196ac414294831b9c4f6be4411497d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d44ce48c2dc546f68bc93d5877b5d36e",
       "style": "IPY_MODEL_483de0f0f3c4423289ccee8612a78d27",
       "value": "100% 25/25 [02:08&lt;00:00,  5.16s/it]"
      }
     },
     "ab1d6ba6e0d743f69ea3d04cc866a56d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ac875de57e5047649cbe2dc50473c9d5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_222f429c25c0496bb009acb1d0b1cf00",
       "max": 90,
       "style": "IPY_MODEL_57ff1eda83e848bb9c2fc8959ba97daa",
       "value": 90
      }
     },
     "aca7bcca156b445a8cad20927a353865": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_b51a890a55cc4858aedb43b0a5ce3b09",
        "IPY_MODEL_56e9ede463924e29ab77a63e4b84be55"
       ],
       "layout": "IPY_MODEL_b3eb977be2794300b654eafae5f6eecf"
      }
     },
     "ad753ca0d3ab4b2fb16d01b58816586f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_a61e64444efa4aaf8550cfed4dfe491a",
       "max": 25,
       "style": "IPY_MODEL_89372cf9870242d0afd62f481fb3e7b8",
       "value": 21
      }
     },
     "af5025986ad84838bd55b263951fdf08": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_9b44b7140fc84627b1b124db44ff85b3",
       "max": 25,
       "style": "IPY_MODEL_c65f44d80f304c728e2f538e4b897b3a",
       "value": 25
      }
     },
     "afa5e83a39a244ecabe2482b0ca593e3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_577acaa6b1274256915c4479152c1284",
       "max": 90,
       "style": "IPY_MODEL_0925e4c625154b2abe95f696a1fdbcba",
       "value": 17
      }
     },
     "b0f9bc537c1b4954bddc4a11d83713f5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_08fab5952a154a40a0fb0abe83454f92",
        "IPY_MODEL_d03589774c4d4f44af6481b1ed6f343b"
       ],
       "layout": "IPY_MODEL_b9f21dcfb64b4222b4d154a49782d495"
      }
     },
     "b15faa81899c44d5a946a56d38e66ff8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b16a2b2f4d614ea2ad373a3cd71a6d44": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_309369b6ad1645cebb279d52a0625b34",
       "style": "IPY_MODEL_662203e107a34036a68fc427d5706bf9",
       "value": "100% 100/100 [14:14&lt;00:00,  8.43s/it]"
      }
     },
     "b1909ee6073e443780891d989501f92f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b1b29246f2a74c30b18df1afd48f1edb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b25634b44d8945a98ecf0c00288638fe": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_e4c64948dff0460b89f3d8175f7ff1a2",
       "style": "IPY_MODEL_9d7fb34e5e4c41b8961d0b7a6e72fdba",
       "value": "100% 90/90 [05:10&lt;00:00,  3.45s/it]"
      }
     },
     "b3b71b8f0b9146d8bed3ace0cd0ac985": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b3d6e5c9eb6a44929739e5a3d10f498d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b3eb977be2794300b654eafae5f6eecf": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b3fc19c236504832a686abbb0f1523bd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b51a890a55cc4858aedb43b0a5ce3b09": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_b3d6e5c9eb6a44929739e5a3d10f498d",
       "max": 28,
       "style": "IPY_MODEL_a1df2afe6ea24698a55d13d59bd51ee7"
      }
     },
     "b53f4441deda48b594c6fa747ec56e76": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b555ed8e7544400288fb0dc7ddae6018": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b71180b70dac49b1ac479f89f65e44a4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_8ab741ff74ad4d0bbb83667676ceb105",
       "style": "IPY_MODEL_506b01250ee949c69513899431d2193a",
       "value": 34
      }
     },
     "b765225ac3f243a08234498ce74a3d6c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b769d77d8fd744198bfb61f9a9c5ba32": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_80bb2579f1674b46b3da3c57b2a7f099",
        "IPY_MODEL_d2e1632478b54dfc9ad014edfe7740a8"
       ],
       "layout": "IPY_MODEL_fcdff191fd304320be1ab613d399185a"
      }
     },
     "b7d7f107533a4333a992849d74ec6b1a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b808b47be13348298c57426f84169311": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b900b5d56f854e47bffd581f7c12b13b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b909f4b33ff74b28852566f818e51ad1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b924013709a44841a0761693753ce766": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b983a237a8804b5d9940981bc986e6cf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_30458e818a364e41a4bdeae7082f09c9",
        "IPY_MODEL_7a7ffae8791f4d618153661b5997cdbd"
       ],
       "layout": "IPY_MODEL_60a2001ef02d477d90e959583af749ff"
      }
     },
     "b9e1a6247956495d97816f872caf479c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_3aa5427c81704e9ab37458c2adf9ae1a",
       "style": "IPY_MODEL_de2f18616d584ca4a7d4a7a168046c66",
       "value": "100% 90/90 [3:32:10&lt;00:00, 141.45s/it]"
      }
     },
     "b9f21dcfb64b4222b4d154a49782d495": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bb55af2679c04ca195cee18aa74c5a07": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "bc0bc4336e654c149ee1d3937b4fef6b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_88f156f74c144e32b17542595b838e9c",
       "style": "IPY_MODEL_18aa8ef20501425499b18777348e6e3e",
       "value": "  0% 0/28 [00:00&lt;?, ?it/s]"
      }
     },
     "be40d655b9164d23b20d5f1a0075ea2c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_0755420340cb4b588f03fc0f4a3fc65a",
       "max": 25,
       "style": "IPY_MODEL_8493dd76d772428dbae6d61acd69e237"
      }
     },
     "bfcae9cbfc904e54bb1431768d81b2fb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_0af6a8765044498092854f38ed428a5d",
       "max": 10,
       "style": "IPY_MODEL_eeaa46ab062a4e5c9e0bd0d5b5ad993e",
       "value": 10
      }
     },
     "c0c2a6850e0a4135a073c3ca3fb79398": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c11a4231c93b4ea49d9cd79a2736ca81": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c3fbdaeed3d54b72a289282aae31fc50": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c40816fda2f94b6caa178b74094061c5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c4806627ddae446b972a697404a873b2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_b3fc19c236504832a686abbb0f1523bd",
       "style": "IPY_MODEL_0730ec53f64f4be99d7759ab6f71d419",
       "value": "100% 25/25 [02:08&lt;00:00,  5.16s/it]"
      }
     },
     "c5d615275cf54f5993587bc6fcb2d23b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_bfcae9cbfc904e54bb1431768d81b2fb",
        "IPY_MODEL_d224da74a8814b97aee4b0ade8a9322a"
       ],
       "layout": "IPY_MODEL_de0aeadf9a7643fb8ccb311974de1e1a"
      }
     },
     "c65f44d80f304c728e2f538e4b897b3a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c6754d09255743fbaaed124e7055b5ad": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "c70636eb78c1478a9f5db863eccd930b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c8b996fd40ac47faab23924c8acc0c44": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_afa5e83a39a244ecabe2482b0ca593e3",
        "IPY_MODEL_ddf23fa78d244fbfa61efb3fa5f0c1cf"
       ],
       "layout": "IPY_MODEL_9ad7b812cb9b47d887b2c6657bf1cb0e"
      }
     },
     "c976210a1ce3463e9e8c9a4d1f1eb234": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ca17c544a2ba4e3e8fc97845ce158389": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cab298f49df742518365c70c6b1db565": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_688b10a2664a420cad021abd21216866",
        "IPY_MODEL_646435b6956b4f119ccd43be89a6d4d7"
       ],
       "layout": "IPY_MODEL_1a755127dae1468b91c16a0093d315cb"
      }
     },
     "cb9e90fe5368457e8948f5dcb1d0a9bd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_c3fbdaeed3d54b72a289282aae31fc50",
       "max": 10,
       "style": "IPY_MODEL_4b456a75f3924eecbb6b764900caf834",
       "value": 10
      }
     },
     "cc12d13e77814bafa7660677de3d2226": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_f67e0ffb7d8e46a2a195f4b0df342cc7",
       "max": 25,
       "style": "IPY_MODEL_c40816fda2f94b6caa178b74094061c5",
       "value": 25
      }
     },
     "cc7955fa6d8c4e2688c84cb23f1210a2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "cea73cba106144c8beae5fbcbc586fb5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ced51f4e76b24fec84bf66f3cb101463": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d03589774c4d4f44af6481b1ed6f343b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_584704dd472e44f5b51c450f0a7fda25",
       "style": "IPY_MODEL_e5c6cf08fe8345c29edfc90d58f19faa",
       "value": "  0% 0/10 [00:00&lt;?, ?it/s]"
      }
     },
     "d086716c9a574e778ffea50e2535354c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d1bdf1fc287a45e8970002dd9acebdf8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_b1909ee6073e443780891d989501f92f",
       "style": "IPY_MODEL_53d612b7354c41dcabcc5cd9d2d752f4",
       "value": 6
      }
     },
     "d1d69545e01a407cb08df3bf0231eeef": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ab1d6ba6e0d743f69ea3d04cc866a56d",
       "style": "IPY_MODEL_c11a4231c93b4ea49d9cd79a2736ca81",
       "value": "100% 15/15 [01:17&lt;00:00,  5.16s/it]"
      }
     },
     "d224da74a8814b97aee4b0ade8a9322a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_26fcd982a99343088ec4c77f28a95892",
       "style": "IPY_MODEL_00bb70333765441bbcd4ca067c744ac1",
       "value": "100% 10/10 [00:51&lt;00:00,  5.15s/it]"
      }
     },
     "d2e1632478b54dfc9ad014edfe7740a8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d667d53eef84400c81a3872e14bb9880",
       "style": "IPY_MODEL_1d3a98fd72d74f8a9a579f8fc23ea6ae",
       "value": "100% 100/100 [14:01&lt;00:00,  8.48s/it]"
      }
     },
     "d4196c89b27c4c26adda44727062c3a8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d44ce48c2dc546f68bc93d5877b5d36e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d4efc94cd337496ab9cf5a84254afda0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d6403c7209a04062b78ae000955a302c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d667d53eef84400c81a3872e14bb9880": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d748428f1cb84da3a92e6da5132ad895": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d799e98fa6204f1c83f8f21d27254eae": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d83623f789a3408192e95f9b7a56fd83": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d8432f0d2bc5414895ffb5a22765b452": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d748428f1cb84da3a92e6da5132ad895",
       "style": "IPY_MODEL_fb66944f02624c3ba4a49e7c20945ce0",
       "value": "100% 25/25 [02:08&lt;00:00,  5.15s/it]"
      }
     },
     "d8474593b70e4ce285f7aa1b29b53eae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_3f392834ee304d1a8f3eea56009f65cd",
       "style": "IPY_MODEL_58baeff11a034988aecdd80d5cac4a03"
      }
     },
     "dc66ee113db94d458385bce9951ebbe7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_f578b3110f3a4cf6840b2a00f129faa2",
       "style": "IPY_MODEL_0303297ce1dd48c2bf42ce14d16149c7",
       "value": "  0% 0/100 [00:00&lt;?, ?it/s]"
      }
     },
     "ddb660f2302a4549a8ebff52148d7fb6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ddf23fa78d244fbfa61efb3fa5f0c1cf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_1f000f15cc6f41e082081f74dea1f555",
       "style": "IPY_MODEL_f9469c6b3a8646129b9b0da6f10f7810",
       "value": " 19% 17/90 [3:35:09&lt;20:44:49, 1023.15s/it]"
      }
     },
     "de0aeadf9a7643fb8ccb311974de1e1a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "de2f18616d584ca4a7d4a7a168046c66": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "de8e6c2fa3c040fb99a54c63f406f549": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_ddb660f2302a4549a8ebff52148d7fb6",
       "style": "IPY_MODEL_82655ff32e8341fb8e6c0b95bbb329a5",
       "value": "  0% 0/10 [00:00&lt;?, ?it/s]"
      }
     },
     "e02d675509084f77828138f14141f831": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e0a26e285aa1437397fd0d4002b94cc5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_02d7469796ed4348b5005ca26f3bf6e6",
       "style": "IPY_MODEL_5ccccd970fc44595ac93c33d3cc310fd",
       "value": " 34% 34/100 [19:49&lt;38:43, 35.20s/it]"
      }
     },
     "e0ccad31d26b4b96a4e767f6b05f712b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_8f6fa134fe4c4d61a3f1dfaf7e14915e",
       "max": 9,
       "style": "IPY_MODEL_37c9d3e60f4f422bb0a9c1408dd33aa1"
      }
     },
     "e0d636b29b3a4641b125db3d27fe696c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e0f96d7ee0e34747a8f38f937dba0ce4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_22f93b3278374f5dacacf7dc14949088",
        "IPY_MODEL_f307048d8eb249bf8e31fa1067d5feae"
       ],
       "layout": "IPY_MODEL_a81c0e5b00be4ee494a48af7c3e825ea"
      }
     },
     "e36c4eb1e6e14d8bbc910b699cfeb015": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e40c930925b44edd9f5d855d0a51ad3d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e4c64948dff0460b89f3d8175f7ff1a2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e5c6cf08fe8345c29edfc90d58f19faa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e7183cedfc854be2b7378bc062c92e70": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_75f703b6e13344c48c6842c31c1881a7",
        "IPY_MODEL_3b2d377afc0f4617bfe8983590a518b5"
       ],
       "layout": "IPY_MODEL_9346cb026615479e8b1776f3e71a887d"
      }
     },
     "e7819519ba394835b38b047416ac628a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_be40d655b9164d23b20d5f1a0075ea2c",
        "IPY_MODEL_3e06aae64695403394fcd309df12ced8"
       ],
       "layout": "IPY_MODEL_7ef8ccd630374243abb0a27bdce1d00e"
      }
     },
     "e799e3d88dc14b5d8edd4708c833513a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_76d618d738da49ccadbe9535e8cbefc1",
        "IPY_MODEL_d8432f0d2bc5414895ffb5a22765b452"
       ],
       "layout": "IPY_MODEL_23f841d3843f479aba5ed9264c78c0f3"
      }
     },
     "eb8efefa18e44c7ca789776b18658160": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "danger",
       "layout": "IPY_MODEL_5cda27c7def047d591f0dee574a6d7bb",
       "max": 9,
       "style": "IPY_MODEL_f7a52564e7ba4f5bb151899ecbb10ede"
      }
     },
     "ed5a62d3d2d6413daef65ebed1609ca4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ed9a8e625c8a42dc93db0c144297448f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "eeaa46ab062a4e5c9e0bd0d5b5ad993e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "eeb510040e9c46fe95f34d82a0297333": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_379d53ed1b394f22a7030319d48ad9e9",
       "max": 15,
       "style": "IPY_MODEL_035f4d0233a642f797ab26c2e24abba0",
       "value": 15
      }
     },
     "eedb1f3627ac4ed5946a02f825f80d5d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f1eda9bb799543208e2c7e4168c35516": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f307048d8eb249bf8e31fa1067d5feae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_911531db34964ba495a851298ce02a74",
       "style": "IPY_MODEL_b1b29246f2a74c30b18df1afd48f1edb",
       "value": "  0% 0/9 [00:00&lt;?, ?it/s]"
      }
     },
     "f3b0f8dce0ca4aae9ce37be9c9df73b8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_8a970e858a17453ba8cfddc0ed76c991",
        "IPY_MODEL_366f659758b34c87afaf2147917ede65"
       ],
       "layout": "IPY_MODEL_b808b47be13348298c57426f84169311"
      }
     },
     "f426a0d865e347759f68be85b90cfbb8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f5281ad0c9b848dbafa428b8ceffeb68": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f577ceaefdc64d87b930c1f97324bd9f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_41e1e763c865476080594157a084957f",
       "max": 28,
       "style": "IPY_MODEL_b3b71b8f0b9146d8bed3ace0cd0ac985",
       "value": 28
      }
     },
     "f578b3110f3a4cf6840b2a00f129faa2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f67e0ffb7d8e46a2a195f4b0df342cc7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "f7a52564e7ba4f5bb151899ecbb10ede": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f8a07f8007d246c19a5b02ad930d2d32": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_64fa9b445c4b437a8a5b03373e61cfbe",
        "IPY_MODEL_b25634b44d8945a98ecf0c00288638fe"
       ],
       "layout": "IPY_MODEL_019c92a2556c4ea4adc5b807f427f583"
      }
     },
     "f9469c6b3a8646129b9b0da6f10f7810": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f9ee2e87a5004d5f80ac4c183fd3d0b3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "layout": "IPY_MODEL_6fbce630aaa94a1a94f4ea834d7a4dab",
       "max": 90,
       "style": "IPY_MODEL_ed9a8e625c8a42dc93db0c144297448f",
       "value": 90
      }
     },
     "fa465225c61942198ecc234f7ef45a38": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fb01204cb4f24aec8a48c30c735b86fe": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fb2de3ebf4c144b08e2ab8945b1b8e5f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fb517b53d19a42f0991652b26df16e8e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fb66944f02624c3ba4a49e7c20945ce0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "fcdff191fd304320be1ab613d399185a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.1.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "fcee3acd03ef4c57a2be82a8f4198059": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "fe9209585e9b4a21bd87c71f9952be37": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "fe99ea4107cf4a46914a447d8770b58c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.4.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": ""
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}