{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "pretty-performer",
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors\n",
    "from PIL import Image, ImageDraw, ImageFont\n",
    "from library.analysis import testSets, generators"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "engaging-warehouse",
   "metadata": {},
   "source": [
    "# Constants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "crazy-taxation",
   "metadata": {},
   "outputs": [],
   "source": [
    "kScore = \"cohens kappa score\"\n",
    "f1Score = \"f1 score\"\n",
    "\n",
    "kScoreSd = kScore + \" - SD\"\n",
    "f1ScoreSd = f1Score + \" - SD\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "extensive-future",
   "metadata": {},
   "source": [
    "# Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "warming-department",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Repeater',\n",
       " 'ProWRAS',\n",
       " 'GAN',\n",
       " 'CTGAN',\n",
       " 'CTAB-GAN',\n",
       " 'ConvGeN-majority-5',\n",
       " 'ConvGeN-majority-full',\n",
       " 'ConvGeN-proximity-5',\n",
       " 'ConvGeN-proximity-full']"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ignoreSet = [\"ozone_level\", \"yeast_me2\"]\n",
    "\n",
    "gans = [g.replace(\"SimpleGAN\", \"GAN\") for g in generators.keys()]\n",
    "algs = [\"LR\", \"RF\", \"GB\", \"KNN\", \"DoC\"]\n",
    "\n",
    "gans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "edf5b592",
   "metadata": {},
   "outputs": [],
   "source": [
    "testSets = [t for t in testSets if t[0:7] == \"folding\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "needed-birmingham",
   "metadata": {},
   "outputs": [],
   "source": [
    "def cleanupName(name):\n",
    "    return name.replace(\"folding_\", \"\").replace(\"imblearn_\", \"\").replace(\"kaggle_\", \"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8106bddf",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = [cleanupName(d) for d in testSets]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "96a601db",
   "metadata": {},
   "outputs": [],
   "source": [
    "def ganName(name):\n",
    "    return name.replace(\n",
    "        \"ConvGeN-majority-5\", \"ConvGeN(5,maj)\").replace(\n",
    "        \"ConvGeN-majority-full\", \"ConvGeN(min,maj)\").replace(\n",
    "        \"ConvGeN-proximity-5\", \"ConvGeN(5,prox)\").replace(\n",
    "        \"ConvGeN-proximity-full\", \"ConvGeN(min,prox)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d35b280b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def filterConvGen(name):\n",
    "    return not name.startswith(\"ConvGeN\") or name == \"ConvGeN(min,maj)\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "selective-connecticut",
   "metadata": {},
   "source": [
    "# Load data from CSV files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "intended-watts",
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadDiagnoseData(ganType, datasetName):\n",
    "    fileName = f\"data_result/{ganType}/{datasetName}.csv\"\n",
    "    r = {}\n",
    "    try:\n",
    "        f1List = []\n",
    "        kList = []\n",
    "        indizes = [str(x) for x in range(100)]\n",
    "        with open(fileName) as f:\n",
    "            newBlock = True\n",
    "            n = \"\"\n",
    "            for line in f:\n",
    "                line = line.strip()\n",
    "                if newBlock:\n",
    "                    n = line\n",
    "                    if n == \"GAN\" or n == \"DoG\":\n",
    "                        n = \"DoC\"\n",
    "                    newBlock = False\n",
    "                elif line == \"---\":\n",
    "                    newBlock = True\n",
    "                    f1List = []\n",
    "                    kList = []\n",
    "                else:\n",
    "                    parts = line.split(\";\")\n",
    "                    if parts[0] == \"avg\":\n",
    "                        f1Avg = float(parts[5])\n",
    "                        f1Sd = math.sqrt(sum([(f1Avg - x)*(f1Avg - x) for x in f1List]) / len(f1List))\n",
    "                        kAvg = float(parts[6])\n",
    "                        kSd = math.sqrt(sum([(kAvg - x)*(kAvg - x) for x in kList]) / len(kList))\n",
    "                        r[n] = { f1Score: f1Avg, kScore: kAvg, f1ScoreSd: f1Sd, kScoreSd: kSd }\n",
    "                    elif parts[0] in indizes:\n",
    "                        f1List.append(float(parts[5]))\n",
    "                        kList.append(float(parts[6]))\n",
    "    except FileNotFoundError as e:\n",
    "        print(f\"Missing file: {fileName}\")\n",
    "    return r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "classical-rescue",
   "metadata": {},
   "outputs": [],
   "source": [
    "statistic = { }\n",
    "\n",
    "for gan in gans:\n",
    "    if ganName(gan) not in statistic:\n",
    "        statistic[ganName(gan)] = {}\n",
    "    \n",
    "    for ds in testSets:\n",
    "        if ds != \"Average\":\n",
    "            statistic[ganName(gan)][cleanupName(ds)] = loadDiagnoseData(gan, ds)\n",
    "            \n",
    "            d = cleanupName(ds)\n",
    "            if d not in dataset:\n",
    "                dataset.append(d)\n",
    "\n",
    "gans = list(statistic.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "unable-entrance",
   "metadata": {},
   "outputs": [],
   "source": [
    "for gan in gans:\n",
    "    f1 = { n: 0.0 for n in algs }\n",
    "    k =  { n: 0.0 for n in algs }\n",
    "    f1Sd = { n: 0.0 for n in algs }\n",
    "    kSd =  { n: 0.0 for n in algs }\n",
    "    c = 0\n",
    "\n",
    "    for ds in dataset:\n",
    "        if ds != \"Average\":\n",
    "            c += 1\n",
    "            for n in algs:\n",
    "                if n in statistic[gan][ds].keys():\n",
    "                    f1[n] += statistic[gan][ds][n][f1Score]\n",
    "                    k[n] += statistic[gan][ds][n][kScore]\n",
    "                    f1Sd[n] += statistic[gan][ds][n][f1ScoreSd]\n",
    "                    kSd[n] += statistic[gan][ds][n][kScoreSd]\n",
    "\n",
    "    avg = {}\n",
    "    for n in algs:\n",
    "        avg[n] = { f1Score: f1[n] / c, f1ScoreSd: f1Sd[n] / c, kScore: k[n] / c, kScoreSd: kSd[n] / c }\n",
    "    statistic[gan][\"Average\"] = avg"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "public-collins",
   "metadata": {},
   "source": [
    "# Show Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "extra-taiwan",
   "metadata": {},
   "outputs": [],
   "source": [
    "def drawTransparentRect(img, rect, color, opacity=1.0):\n",
    "    def mix(v, a, b):\n",
    "        return max(0, min(255, int((v * a) + ((1.0 - v) * b))))\n",
    "\n",
    "    def mixPixel(v, a, b):\n",
    "        return (mix(v, a[0], b[0]), mix(v, a[1], b[1]), mix(v, a[2], b[2]))\n",
    "\n",
    "\n",
    "    ((x0, y0), (x1, y1)) = rect\n",
    "    \n",
    "    for y in range(y0, y1):\n",
    "        for x in range(x0, x1):\n",
    "            p = (x, y)\n",
    "            c = mixPixel(opacity, color, img.getpixel(p))\n",
    "            img.putpixel(p, c)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5fec8e83",
   "metadata": {},
   "outputs": [],
   "source": [
    "def drawDiagram(size, rowNames, data, colNames=[], colors=None, border=20, barIndent=10, fontSize=20, markers=[0.25, 0.5, 0.75, 1.00], maxY=1.0):\n",
    "    silver = (204, 204, 204)\n",
    "    black = (0,0,0)\n",
    "    white = (255, 255, 255)\n",
    "    \n",
    "    defaultColors = [ (31,119,180)\n",
    "                    , (255,127,14)\n",
    "                    , (44,160,44)\n",
    "                    , (214,40,40)\n",
    "                    , (148,103,189)\n",
    "                    , (140,86,75)\n",
    "                    , (227,119,194)\n",
    "                    , (127,127,127)\n",
    "                    , (40,40,214)\n",
    "                    ]\n",
    "\n",
    "    defaultColors = [ (209,188,115) # Gold\n",
    "                    , (145,196,223) # bright blue\n",
    "                    , (83,119,202)  # blue\n",
    "                    , (224,138,86)  # orange\n",
    "                    , (131,202,112) # green\n",
    "                    , (199,102,99)  # red\n",
    "                    , (143,110,176) # violet\n",
    "                    , (134,99,66)   # brown\n",
    "                    , (207,131,189) # pink\n",
    "                    ]\n",
    "    \n",
    "    def y(v):\n",
    "        return v / maxY\n",
    "\n",
    "    def med(v):\n",
    "        if type(v) is tuple:\n",
    "            return v[0]\n",
    "        return v\n",
    "\n",
    "    \n",
    "    if colors is None:\n",
    "        colors = defaultColors\n",
    "\n",
    "    print((len(data[0]), len(colNames), len(data), len(rowNames)))\n",
    "\n",
    "    font = ImageFont.truetype(\"FreeSans\", fontSize)\n",
    "    \n",
    "    markerSize = 0\n",
    "    for m in markers:\n",
    "        markerSize = max(markerSize, font.getsize(f\"{m:0.2f}\")[0])\n",
    "\n",
    "    areaTop = 2 * border + markerSize\n",
    "\n",
    "    barStep = (size[0] - border - areaTop) // len(data)\n",
    "    barSize = max(border, barStep - border)\n",
    "    barIndent = min(barIndent, barSize / (1 + len(data[0])))\n",
    "    barIndent = barSize / (2 + len(data[0]))\n",
    "    \n",
    "    print((size[0], barSize, barSize * len(data)))\n",
    "    \n",
    "    # Create new Image\n",
    "    w = max(size[0], size[1])\n",
    "    img = Image.new(\"RGB\", (w,w))\n",
    "    d = ImageDraw.Draw(img)\n",
    "    \n",
    "    # Set background to white.\n",
    "    d.rectangle(((0,0), (w,w)), fill=white)\n",
    "    \n",
    "    # draw row names\n",
    "    height = size[1]\n",
    "    left = w - height\n",
    "    textSize = 0\n",
    "    for (n, name) in enumerate(rowNames):\n",
    "        s = font.getsize(name)\n",
    "        offset = int(border + barSize - s[1] + 1.5) // 2\n",
    "        textSize = max(textSize, s[0])\n",
    "        pos = (left + border, areaTop + offset + (barStep * n))\n",
    "        d.text(pos, name, fill=black, font=font)\n",
    "        \n",
    "        \n",
    "    # Calculate sizes for bar drawing.\n",
    "    barLength = height - (4 * border) - textSize\n",
    "    areaSize = (barLength, barSize)\n",
    "    areaLeft = left + (2 * border) + textSize\n",
    "    \n",
    "    # Draw Lines for bar height comparing.\n",
    "    markerPos = [areaLeft + int(y(v) * barLength) for v in markers]\n",
    "    for p in markerPos:\n",
    "        d.line(((p, border), (p, size[0] - border)), fill=silver)\n",
    "        \n",
    "    # Draw bars.\n",
    "    for (n, row) in enumerate(data):\n",
    "        area = ((areaLeft, areaTop + (n * barStep) + (border // 2)), areaSize)\n",
    "\n",
    "        indices = list(range(len(row)))\n",
    "        indices.sort(key= lambda i: 1.0 - med(row[i]))\n",
    "\n",
    "        for (n, i) in enumerate(indices):\n",
    "            v = y(med(row[i]))\n",
    "            c = colors[i]\n",
    "            offset = barIndent * n\n",
    "            tl = (area[0][0], area[0][1] + offset)\n",
    "            br = (tl[0] + int(v * area[1][0]), area[1][1] + tl[1] - offset)\n",
    "            rect = (tl, br)\n",
    "            d.rectangle(rect, fill=c, outline=black)\n",
    "\n",
    "    for (n, row) in enumerate(data):\n",
    "        area = ((areaLeft, areaTop + (n * barStep) + (border // 2)), areaSize)\n",
    "\n",
    "        indices = list(range(len(row)))\n",
    "        indices.sort(key= lambda i: 1.0 - med(row[i]))\n",
    "\n",
    "        for (n, i) in enumerate(indices):\n",
    "            v = y(med(row[i]))\n",
    "            c = colors[i]\n",
    "            offset = barIndent * n\n",
    "            tl = (area[0][0], area[0][1] + offset)\n",
    "            br = (tl[0] + int(v * area[1][0]), area[1][1] + tl[1] - offset)\n",
    "\n",
    "            if type(row[i]) is tuple and (row[i][0] != 0.0 or row[i][1] != row[i][0] or row[i][2] != row[i][0]):\n",
    "                vLower = y(row[i][1])\n",
    "                vUpper = y(row[i][2])\n",
    "                xCenter = (tl[1] + br[1]) // 2\n",
    "                yUpper = tl[0] + int(vUpper * area[1][0])\n",
    "                yLower = tl[0] + int(vLower * area[1][0])\n",
    "\n",
    "                #d.line(((br[0], tl[1]), (br[0], br[1])), fill=black, width=5)\n",
    "                d.line(((yLower, xCenter), (yUpper, xCenter)), fill=white, width=5)\n",
    "                d.line(((yLower, xCenter - border), (yLower, xCenter + border)), fill=white, width=5)\n",
    "                d.line(((yUpper, xCenter - border), (yUpper, xCenter + border)), fill=white, width=5)\n",
    "\n",
    "                #d.line(((br[0], tl[1]), (br[0], br[1])), fill=c, width=3)\n",
    "                d.line(((yLower, xCenter), (yUpper, xCenter)), fill=c, width=3)\n",
    "                d.line(((yLower, xCenter - border), (yLower, xCenter + border)), fill=c, width=3)\n",
    "                d.line(((yUpper, xCenter - border), (yUpper, xCenter + border)), fill=c, width=3)\n",
    "\n",
    "\n",
    "    # Draw axis.\n",
    "    d.line(((areaLeft, areaTop), (areaLeft, size[0] - border)), fill=black)\n",
    "    d.line(((areaLeft, areaTop), (w - border, areaTop)), fill=black)\n",
    "    \n",
    "    # Draw y-axis text.\n",
    "    img = img.rotate(90)\n",
    "    d = ImageDraw.Draw(img)\n",
    "\n",
    "    for (m, p) in zip(markers, markerPos):\n",
    "        d.text((border, size[1] - (p - left)), f\"{m:0.2f}\", fill=black, font=font)\n",
    "    \n",
    "    # Draw legend.\n",
    "    if len(colNames) > 0:\n",
    "        colNameWidth = 0\n",
    "        colNameHeight = fontSize * len(colNames)\n",
    "        for c in colNames:\n",
    "            colNameWidth = max(colNameWidth, font.getsize(c)[0])\n",
    "        \n",
    "        rWidth = int(fontSize * 0.75)\n",
    "        rHeight = fontSize // 2\n",
    "        rPadd = (fontSize - rHeight) // 2\n",
    "        \n",
    "        tl = (size[0] - (int(2.5 * border) + colNameWidth + rWidth), 0)\n",
    "        br = (size[0], int(1.2 * border) + colNameHeight)\n",
    "        drawTransparentRect(img, (tl, br), white, 0.75)\n",
    "\n",
    "        for (n, c) in enumerate(colNames):\n",
    "            t = border + (fontSize * n)\n",
    "            l = size[0] - border - colNameWidth\n",
    "            d.rectangle(((l - border - rWidth, t + rPadd - 1), (l - border, t + rPadd + rHeight)), fill=colors[n], outline=black)\n",
    "            d.text((l, t), c, fill=black, font=font)\n",
    "    \n",
    "    return img.crop((0, 0, size[0], size[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "038344f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def showDiagnose(algo, score):\n",
    "    def valueOf(d, g):\n",
    "        d = cleanupName(d)\n",
    "        if d not in statistic[g].keys():\n",
    "            print(f\"Missing '{d}' in '{g}'\")\n",
    "            return 0.0\n",
    "        \n",
    "        if algo in statistic[g][d].keys():\n",
    "            return statistic[g][d][algo][score]\n",
    "        else:\n",
    "            print(f\"Missing '{algo}' in ('{g}', '{d}')\")\n",
    "            return 0.0\n",
    "    \n",
    "    print(f\"{algo}: {score}\")\n",
    "    \n",
    "    data = [[valueOf(d, g) for g in gans] for d in testSets]        \n",
    "    img = drawDiagram((1024, 1024), [cleanupName(d) for d in testSets], data, colNames=gans)\n",
    "    img.save(f\"data_result/statistics/byAlgorithm/statistic-{algo}-{score}.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8904b4b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def showDiagnoseAverage(score, onlyOneBar=False):\n",
    "    def valueOf(g, algo):\n",
    "        if algo in statistic[g][\"Average\"].keys():\n",
    "            if True: #not g.startswith(\"ConvGeN\"):\n",
    "                return statistic[g][\"Average\"][algo][score]\n",
    "            else:\n",
    "                a = statistic[g][\"Average\"][algo][score]\n",
    "                sd = statistic[g][\"Average\"][algo][score + \" - SD\"]\n",
    "                return (a, a - sd, a + sd)\n",
    "        else:\n",
    "            return 0.0\n",
    "\n",
    "    print(f\"Average: {score}\")\n",
    "    \n",
    "    data = [[valueOf(g, algo) for g in gans] for algo in algs]\n",
    "    img = drawDiagram((1024, 1024), algs, data, colNames=gans, maxY=0.75)\n",
    "    img.save(f\"data_result/statistics/average/statistic-Algo-Average-{score}.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d1f4961a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def showDiagnoseDataset(dataset):\n",
    "    print(f\"{dataset}\")\n",
    "    \n",
    "    def valueOf(algo, score, g):\n",
    "        if dataset in statistic[g]:\n",
    "            if algo in statistic[g][dataset]:\n",
    "                if score in statistic[g][dataset][algo]:\n",
    "                    return statistic[g][dataset][algo][score]\n",
    "                    \n",
    "        return 0.0\n",
    "    \n",
    "    scores = [f1Score, kScore]\n",
    "    \n",
    "    for score in scores:\n",
    "        data = [[valueOf(algo, score, g) for algo in algs] for g in gans]\n",
    "        img = drawDiagram((1024, 1024), gans, data, colNames=algs, maxY=0.75)\n",
    "        img.save(f\"data_result/statistics/byDataset/statistic-Classifier-{dataset}-{score}.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "magnetic-romance",
   "metadata": {},
   "outputs": [],
   "source": [
    "def showDiagnoseDatasetAverage():\n",
    "    print(\"Average\")\n",
    "    \n",
    "    dataset = \"Average\"\n",
    "    \n",
    "    def valueOf(algo, score, g):\n",
    "        if dataset in statistic[g]:\n",
    "            if algo in statistic[g][dataset]:\n",
    "                if score in statistic[g][dataset][algo]:\n",
    "                    if True: # algo != \"DoC\":\n",
    "                        return statistic[g][dataset][algo][score]\n",
    "                    else:\n",
    "                        a = statistic[g][dataset][algo][score]\n",
    "                        sd = statistic[g][\"Average\"][algo][score + \" - SD\"]\n",
    "                        return (a, a - sd, a + sd)\n",
    "                    \n",
    "        return 0.0\n",
    "    \n",
    "    scores = [f1Score, kScore]\n",
    "    \n",
    "    for score in scores:\n",
    "        data = [[valueOf(algo, score, g) for algo in algs] for g in gans]\n",
    "        img = drawDiagram((1024, 1024), gans, data, colNames=algs, maxY=0.75)\n",
    "        img.save(f\"data_result/statistics/byDataset/statistic-Classifier-{dataset}-{score}.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "57fe8925",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR: f1 score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "LR: cohens kappa score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "RF: f1 score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "RF: cohens kappa score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "GB: f1 score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "GB: cohens kappa score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "KNN: f1 score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "KNN: cohens kappa score\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "DoC: f1 score\n",
      "Missing 'DoC' in ('Repeater', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('ProWRAS', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('GAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('CTGAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('Repeater', 'abalone9-18')\n",
      "Missing 'DoC' in ('ProWRAS', 'abalone9-18')\n",
      "Missing 'DoC' in ('GAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('CTGAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('Repeater', 'car_good')\n",
      "Missing 'DoC' in ('ProWRAS', 'car_good')\n",
      "Missing 'DoC' in ('GAN', 'car_good')\n",
      "Missing 'DoC' in ('CTGAN', 'car_good')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'car_good')\n",
      "Missing 'DoC' in ('Repeater', 'car-vgood')\n",
      "Missing 'DoC' in ('ProWRAS', 'car-vgood')\n",
      "Missing 'DoC' in ('GAN', 'car-vgood')\n",
      "Missing 'DoC' in ('CTGAN', 'car-vgood')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'car-vgood')\n",
      "Missing 'DoC' in ('Repeater', 'flare-F')\n",
      "Missing 'DoC' in ('ProWRAS', 'flare-F')\n",
      "Missing 'DoC' in ('GAN', 'flare-F')\n",
      "Missing 'DoC' in ('CTGAN', 'flare-F')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'flare-F')\n",
      "Missing 'DoC' in ('Repeater', 'hypothyroid')\n",
      "Missing 'DoC' in ('ProWRAS', 'hypothyroid')\n",
      "Missing 'DoC' in ('GAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('CTGAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('Repeater', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('ProWRAS', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('GAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('CTGAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('Repeater', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('ProWRAS', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('GAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('CTGAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('Repeater', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('ProWRAS', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('GAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('CTGAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('Repeater', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('ProWRAS', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('GAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('CTGAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('Repeater', 'winequality-red-4')\n",
      "Missing 'DoC' in ('ProWRAS', 'winequality-red-4')\n",
      "Missing 'DoC' in ('GAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('CTGAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('Repeater', 'yeast4')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast4')\n",
      "Missing 'DoC' in ('GAN', 'yeast4')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast4')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast4')\n",
      "Missing 'DoC' in ('Repeater', 'yeast5')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast5')\n",
      "Missing 'DoC' in ('GAN', 'yeast5')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast5')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast5')\n",
      "Missing 'DoC' in ('Repeater', 'yeast6')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast6')\n",
      "Missing 'DoC' in ('GAN', 'yeast6')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast6')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast6')\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "DoC: cohens kappa score\n",
      "Missing 'DoC' in ('Repeater', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('ProWRAS', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('GAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('CTGAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'abalone_17_vs_7_8_9_10')\n",
      "Missing 'DoC' in ('Repeater', 'abalone9-18')\n",
      "Missing 'DoC' in ('ProWRAS', 'abalone9-18')\n",
      "Missing 'DoC' in ('GAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('CTGAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'abalone9-18')\n",
      "Missing 'DoC' in ('Repeater', 'car_good')\n",
      "Missing 'DoC' in ('ProWRAS', 'car_good')\n",
      "Missing 'DoC' in ('GAN', 'car_good')\n",
      "Missing 'DoC' in ('CTGAN', 'car_good')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'car_good')\n",
      "Missing 'DoC' in ('Repeater', 'car-vgood')\n",
      "Missing 'DoC' in ('ProWRAS', 'car-vgood')\n",
      "Missing 'DoC' in ('GAN', 'car-vgood')\n",
      "Missing 'DoC' in ('CTGAN', 'car-vgood')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'car-vgood')\n",
      "Missing 'DoC' in ('Repeater', 'flare-F')\n",
      "Missing 'DoC' in ('ProWRAS', 'flare-F')\n",
      "Missing 'DoC' in ('GAN', 'flare-F')\n",
      "Missing 'DoC' in ('CTGAN', 'flare-F')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'flare-F')\n",
      "Missing 'DoC' in ('Repeater', 'hypothyroid')\n",
      "Missing 'DoC' in ('ProWRAS', 'hypothyroid')\n",
      "Missing 'DoC' in ('GAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('CTGAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'hypothyroid')\n",
      "Missing 'DoC' in ('Repeater', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('ProWRAS', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('GAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('CTGAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kddcup-guess_passwd_vs_satan')\n",
      "Missing 'DoC' in ('Repeater', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('ProWRAS', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('GAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('CTGAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kr-vs-k-three_vs_eleven')\n",
      "Missing 'DoC' in ('Repeater', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('ProWRAS', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('GAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('CTGAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'kr-vs-k-zero-one_vs_draw')\n",
      "Missing 'DoC' in ('Repeater', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('ProWRAS', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('GAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('CTGAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'shuttle-2_vs_5')\n",
      "Missing 'DoC' in ('Repeater', 'winequality-red-4')\n",
      "Missing 'DoC' in ('ProWRAS', 'winequality-red-4')\n",
      "Missing 'DoC' in ('GAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('CTGAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'winequality-red-4')\n",
      "Missing 'DoC' in ('Repeater', 'yeast4')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast4')\n",
      "Missing 'DoC' in ('GAN', 'yeast4')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast4')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast4')\n",
      "Missing 'DoC' in ('Repeater', 'yeast5')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast5')\n",
      "Missing 'DoC' in ('GAN', 'yeast5')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast5')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast5')\n",
      "Missing 'DoC' in ('Repeater', 'yeast6')\n",
      "Missing 'DoC' in ('ProWRAS', 'yeast6')\n",
      "Missing 'DoC' in ('GAN', 'yeast6')\n",
      "Missing 'DoC' in ('CTGAN', 'yeast6')\n",
      "Missing 'DoC' in ('CTAB-GAN', 'yeast6')\n",
      "(6, 6, 14, 14)\n",
      "(1024, 46, 644)\n",
      "Average: f1 score\n",
      "(6, 6, 5, 5)\n",
      "(1024, 165, 825)\n",
      "Average: cohens kappa score\n",
      "(6, 6, 5, 5)\n",
      "(1024, 165, 825)\n",
      "abalone_17_vs_7_8_9_10\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "abalone9-18\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "car_good\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "car-vgood\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "flare-F\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "hypothyroid\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "kddcup-guess_passwd_vs_satan\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "kr-vs-k-three_vs_eleven\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "kr-vs-k-zero-one_vs_draw\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "shuttle-2_vs_5\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "winequality-red-4\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "yeast4\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "yeast5\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "yeast6\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "Average\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n",
      "(5, 5, 6, 6)\n",
      "(1024, 134, 804)\n"
     ]
    }
   ],
   "source": [
    "gans = [g for g in statistic.keys() if filterConvGen(g)]\n",
    "\n",
    "for a in algs:\n",
    "    showDiagnose(a, f1Score)\n",
    "    showDiagnose(a, kScore)\n",
    "    \n",
    "showDiagnoseAverage(f1Score)\n",
    "showDiagnoseAverage(kScore)\n",
    "\n",
    "for t in testSets:\n",
    "    showDiagnoseDataset(cleanupName(t))\n",
    "\n",
    "showDiagnoseDatasetAverage()\n",
    "\n",
    "gans = list(statistic.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "63841c55",
   "metadata": {},
   "outputs": [],
   "source": [
    "def getValueOf(gan, dataset, algo, score):\n",
    "    if dataset not in statistic[gan].keys():\n",
    "        #print(f\"Missing '{dataset}' in '{gan}'\")\n",
    "        return None\n",
    "\n",
    "    if algo not in statistic[gan][dataset].keys():\n",
    "        #print(f\"Missing '{algo}' in ('{gan}', '{dataset}')\")\n",
    "        return None\n",
    "    \n",
    "    if score not in statistic[gan][dataset][algo].keys():\n",
    "        #print(f\"Missing '{score}' in ('{gan}', '{dataset}', '{algo}')\")\n",
    "        return None\n",
    "    \n",
    "    return statistic[gan][dataset][algo][score]\n",
    "    \n",
    "    \n",
    "    \n",
    "def calcTable(algo, score, ignore=[]):\n",
    "    table = []\n",
    "    \n",
    "    def calc(gc, g):\n",
    "        n = 0\n",
    "        for d in testSets:\n",
    "            d = cleanupName(d)\n",
    "            if d not in ignore:\n",
    "                vc = getValueOf(gc, d, algo, score)\n",
    "                v = getValueOf(g, d, algo, score)\n",
    "                if vc is not None and v is not None and vc >= v:\n",
    "                    n += 1\n",
    "        return n\n",
    "    \n",
    "    for gc in gans:\n",
    "        table.append([calc(gc, g) for g in gans])\n",
    "    return table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "177774b0",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[14, 3, 2, 10, 11, 2, 2, 2, 1],\n",
       " [11, 14, 9, 10, 11, 11, 6, 12, 3],\n",
       " [12, 5, 14, 11, 11, 10, 5, 10, 5],\n",
       " [5, 4, 3, 14, 6, 3, 1, 4, 2],\n",
       " [4, 3, 3, 9, 14, 3, 2, 3, 1],\n",
       " [12, 3, 4, 11, 11, 14, 0, 9, 1],\n",
       " [12, 11, 9, 13, 13, 14, 14, 14, 10],\n",
       " [12, 2, 5, 10, 11, 6, 0, 14, 0],\n",
       " [13, 11, 9, 12, 13, 13, 4, 14, 14]]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tables = {}\n",
    "ignore = [# \"webpage\"\n",
    "         #, \"mammography\"\n",
    "         #, \"protein_homo\"\n",
    "         #, \"ozone_level\"\n",
    "         #, \"creditcard\"\n",
    "         ]\n",
    "for a in algs:\n",
    "    tables[a + \" - \" + f1Score] = calcTable(a, f1Score, ignore)\n",
    "    tables[a + \" - \" + kScore] = calcTable(a, kScore, ignore)\n",
    "    \n",
    "tables[algs[0] + \" - \" + f1Score]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "453f491d",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR - f1 score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR - cohens kappa score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF - f1 score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF - cohens kappa score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GB - f1 score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GB - cohens kappa score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN - f1 score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN - cohens kappa score\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAccAAAGoCAYAAAAzXi49AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABEkElEQVR4nO3deZwdVbnu8d8TFCIzMomKRiDIUZAgAREQRQUBR9Qj4ogo6FEE9YjzveI9eNHDRQEnjKioh8kJRRQFlIAgUwIBwmRQUBFlVBFkDM/9o9aGYrM7Q6erqnf38/189qd3je/anU6/vd5atUq2iYiIiIdN6boBERER402SY0RERJ8kx4iIiD5JjhEREX2SHCMiIvokOUZERPRJcoyIiAlB0jck3Sxp/oBtH5RkSWstybmSHCMiYqI4Btilf6Wk9YGdgD8u6YmSHCMiYkKwfTZw+4BNnwc+BCzxrDdJjhERMWFJegXwZ9uXLs1xj2moPREREY8g7WK4dRnOMPcK4J7ailm2Z40cTysCHwd2XtpISY4REdGSW5HmjPpoW/fYnrkUh2wIPA24VBLAk4GLJW1t+6+LOjDJMSIiWjNlGS7mLVy4dPvbvhxYp7cs6Xpgpu3Fdl9zzTEiIlohVclxtK/Fn1/HA+cBT5d0g6S3j7at6TlGRERrqupmM2zvuZjt05b0XEmOERHRmmUpq7YpyTEiIlrTZM9xLCU5RkREK3rXHIdBkmNERLQmyTEiIqJPyqoRERE1KatGREQMkJ5jREREn/QcIyIialJWjYiIGCBl1YiIiD7pOUZERNRI6TlGREQ8SnqOERERfZIcoxXSWoZpncTecoW5ncR9yKZbdhZ6wYLOQgOw2mrdxV5ntY7/3e/pMPbUDmMDrNDNz/z111/PrbfeuswF0ZRVo0XTgDmdRJ6zfsc/5XO6+dwAu+3WWWgAdt21u9jv3bXjf/cu/zCZ3mFsgI26+ZmfOXPmmJ0rPceIiIg+6TlGRETUZBKAiIiIAZIcIyIi+qSsGhERUZOyakRExADpOUZERPRJzzEiIqImZdWIiIgBUlaNiIjoMyw9xyFp5tiStFDSPEnzJf1E0uotxV1d0rvbiBURMR715lcdzatNkzI5AnfbnmF7U+B24D0txV0dWKrkqMpk/XeKiAmkd81xtK825ZcunAc8CUDShpJ+LmmupF9L2qSsP0bSUWXdbyW9rKxfTtKhki6SdJmkd5b1K0v6paSLJV0u6ZUl1meADUuv9dCy74G14z9V1k2TdJWkLwMXA+u3+h2JiGjIsCTHSX3NUdJywIuAr5dVs4B32V4g6TnAl4EXlm3TgOcDGwJnStoIeAvwD9tbSVoBOFfSacCfgN1t3yFpLeB8SScDHwE2tT2jxN+Zap7/rQEBJ0vaAfgj8HTgbbYf1dOUtC+wb7X0lLH7hkRENCwDcsa3x0maR5Xw5gKnS1oZ2Bb4nh7+11uhdsx3bT8ILJD0e2ATYGfgWZJeW/ZZjSrZ3QD835LoHqTqma47oB07l9clZXnlcvwfgT/YPn9Q423PokrkSDO9VJ88IqIjuZVj/Lvb9gxJqwGnUF1zPAb4e69XN0B/EjJVb++9tn9R3yBpL2BtYEvb90u6nsGPSRVwiO2v9h0/DbhrKT5PRMRQGJae45Dk8GbY/gewP/BB4G7gOkn/Dg8NhNm8tvu/S5oiaUNgA+Aa4BfAf0h6bDlmY0krUfUgby6JcUfgqeUc/wRWqZ3zF8DepdeKpCdJWqepzxsR0bVccxwSti+RdCnweuCNwFckfQJ4LHACcGnZ9RrgLKry6Lts3yPpaKrS7MWqarG3AK8CjgV+ImkOMA+4usS6TdK5kuYDp9o+UNK/AeeVUu6dwJuAhY1/8IiIlqWsOs7ZXrlv+eW1xV1GOOxc2+/vO+5B4GPl1e+5I8R+Q9/yEcARA3bddIR2REQMrWEpq07K5BgREd1Iz3ECsb1X122IiBh2Xcx0M1pDksMjImIiaHJAjqRvSLq5jOvorTtU0tVlopWTlnS60CTHiIhoTcOjVY/h0eNGTqeafOVZwG+Bjy5RO5fiM0VERIzaskw6viTlWNtnU82XXV93mu0HyuL5wJOXpK255hgREa3peEDO3sCJS7JjkmNERLRmGQfkrFXuH++ZVabTXIK4+jjwANV96IuV5BgREa0Yg0kAbrU9c+nj6q3Ay4AX2V6i+aiTHCMiojVtl1Ul7QJ8GHi+7X8t6XFJjjFqurbbB4JsNL3T8J069dTuYu//qDn423XkkR3G3r+72ADTO/qZX7Bg7M7V5H2Oko4HXkBVfr0B+CTV6NQVqJ6+BHC+7Xct7lxJjhER0Yqm51a1veeA1V8fsG6xkhwjIqI1wzJDTpJjRES0JnOrRkRE1OSRVREREQOkrBoREdEnPceIiIiaYXpkVZJjRES0Jj3HiIiIPkmOERERNSmrRkREDJCeY0RERJ9h6TkOSQ4fW5IWSponab6k70lacSmOvUTSjPL+MZLukvSm2va5kp4taS9Jt5Q4V0t6f995tpBkSS/pW/9xSVdIuqwc+5xl/LgREeNCbxKA0b7aNCmTI3C37Rm2NwXuAx4xQ7uk5RZx7G+Abcv7zYFresuSVgI2AC4t20+0PQPYDvi4pPVr59kTOKd87cV9LtUzx55t+1nAi4E/jeYDRkSMR0mOw+PXwEaSXiDpTEnHAZdLmirpm5IuL73FHcv+5/JwctwWOAqYUZa3Bi62vbAewPZtwLXAegCqnpvyWmAvYGdJU8uu61E9zPPectyttm8c+48cEdGN3qCc0bzaNKmTo6THALsCl5dVWwMft/0M4D0Atjej6t19qySxes9xW+Bs4F5Jq5TlcwfEeQowFbisrNoOuM7274DZwG5l/WnA+pJ+K+nLkp4/hh83IqJTKauOf4+TNA+YA/yRh5/3daHt68r77YHvANi+GvgDsLHt64HlJT0B2ISqrHoR8Byq5PibWpw9JF0B/B44wvY9Zf2ewAnl/QllGdt3AlsC+wK3ACdK2qu/8ZL2lTRH0pxqt4iI4TAsPcfJOlr17nIt8CHlCdF31Vct4vjzqMqif7FtSedT9Qa3Bs6v7Xei7f3KtcSfSjqVKpu9BniFpI+XOGtKWsX2P0tJdjYwW9LlwFuBY+rBbc8CZlXtntntY9kjIpbCsNzKMSTN7MTZwBsBJG0MPIWqlwhV6fT9VEmS8vUtwF9t/73/RLbPo+qFHkA1yOZS2+vbnmb7qcAPgFdJerqk6bVDZ1D1WCMiht4wlVUna89xSXwZOKr03h4A9uoNlKFKjp+nJEfbfykjXH8z8EyVzwIXAxsDJ/Vt+wHwH8CVwBckrV5iXktVYo2ImBCG5T7HSZkcba88YN1sqnJmb/keqtGkg46/iL6yq+1pfcvHUCuHllGnTxjhfCcDJ5fFbQftExExEQxLWXVSJseIiOhGeo4RERE1vWuOwyDJMSIiWpPkGBER0Sdl1YiIiJqUVSMiIgZIzzEiIqJPeo4RERE1KatGREQMkLJqREREn/QcoxWbbgon9c/U2pKN9u/2T0Cd2t0DSY48srPQAGy0UXexd13kA2tacG13od97anexAejo527m3LE5TxePnhqtJMeIiGhNeo4RERF9khwjIiJqUlaNiIgYID3HiIiIPuk5RkRE1AzTJABD0syIiJgIpkwZ/WtxJH1D0s2S5tfWPV7S6ZIWlK9rLFE7R/8RIyIilk5vUM5oXkvgGGCXvnUfAX5pezrwy7K8WEmOERHRil5Ztameo+2zgdv7Vr8S+FZ5/y3gVUvS1lxzjIiI1izjNce1JM2pLc+yPWsxx6xr+y8Atv8iaZ0lCZTkGBERrVnG0aq32p45Rk1ZpJRVGyBpXUnHSfq9pLmSzpO0e237EZL+LGlKbd1ekh6U9KzauvmSprXc/IiIRjRdVh3BTZLWq+JrPeDmJTkoyXGMSRLwI+Bs2xvY3hJ4PfDksn0KsDvwJ2CHvsNvAD7eXmsjItrV8ICcQU4G3lrevxX48ZIclOQ49l4I3Gf7qN4K23+w/YWyuCMwH/gKsGffsacAz5T09FZaGhHRsoZv5TgeOA94uqQbJL0d+Aywk6QFwE5lebFyzXHsPRO4eBHb9wSOp/rr5f9Keqzt+8u2B4H/Bj7Gw3/pRERMCE1PAmC7v8PR86KlPVd6jg2T9CVJl0q6SNLywG7Aj2zfAVwA7Nx3yHHANpKetohz7itpjqQ5t99+S3ONj4gYYx2UVUclPcexdwXwmt6C7fdIWguYQ3Vz6mrA5dWlSVYE/gX8tLb/A5IOAz48UoAydHkWwGabzezuib8REUsp08dNXr8Cpkr6j9q6FcvXPYF32J5mexrwNGBnSSv2neMY4MXA2g23NSKiVcPSc0xyHGO2TTUDw/MlXSfpQqpZGT4JvIRH9hLvAs4BXt53jvuAI4Elulk1ImIYdHQrx6ikrNqAMhvD6wds+taAfV9dWzymtv5IqgQZETFhDEtZNckxIiJa0UV5dLSSHCMiojXpOUZERPRJzzEiIqKm6UkAxlKSY0REtCbJMSIiok/KqhERETUpq0ZERAyQnmNERESf9BwjIiJqUlaN1qzwj7lsdOqQ1CnGmH/W3efWbt0+DKXLz86p3YUGYNcOY3c9oeP0juKuMHanSlk1IiKiT3qOERERfdJzjIiIqMk1x4iIiAGSHCMiIvqkrBoREVGTsmpERMQA6TlGRET0Sc8xIiKiJmXViIiIAVJWjYiI6JOeY0RERI00PD3HIcnh3ZH0BEknSPqdpCslnSnpX5LmSbpd0nXl/Rll/+mSTin7zy3779B3zh9LOq9v3UHlvOvU1t3ZzqeMiGjHlCmjf7UpPcdFkCTgJOBbtl9f1s0AVrH9a0nHAKfY/n7ZNhX4KfBB2yeXdZsCM4Gzy/LqwLOBOyU9zfZ1tZC3Av8JfLj5TxcR0b5hKasOSTM7syNwv+2jeitsz7P96xH2fyNwXi8xlv3n2z6mts9rgJ8AJwCv7zv+G8Aekh4/Fo2PiBhPemXV0b7alOS4aJsCc5di/2cCFy9mnz2B48trz75td1IlyAMWdQJJ+0qaI2nOLSm8RsQQGZayapJjgySdJGm+pB+W5XWBjYBzbP8WeKCUXeuOBN4qadWRzmt7lu2ZtmeuvXJjzY+IGHPpOU4MVwBbLuX+z+4t2N4d2AvolUn3ANYArpN0PTCNvtKq7b8DxwHvHl2TIyLGp94kAOk5Dr9fAStI2qe3QtJWkp4/wv7HAdtJekVt3Yq193sCu9ieZnsaVeLtv+4I8DngnWTAVERMMEmOE4BtA7sDO5VbM64ADgJuHGH/u4GXAe+S9Ptyu8YngIMlTQOeApxf2/864A5Jz+k7z61Uo2RXGPMPFRHRoWEpq6Znshi2bwReN8K2vQasuxrYbYTTPWnA/r0y7AV96z8AfGBp2hoRMZ61MbeqpPcD7wAMXA68zfY9S3ue9BwjIqI1TfYcJT0J2B+YaXtTYDkGX7parPQcIyKiNS1cO3wM8DhJ91ON+Rh4GWxJThIREdG4psuqtv8s6f8BfwTuBk6zfdpozpWyakREtGYZy6pr9SZAKa99H3lurQG8Enga8ERgJUlvGk0703OMiIjWLGPP8VbbMxex/cXAdbZvASgTsGwL/M/SBkpyjIiI1jR8S8YfgW0krUhVVn0RMGc0J0pyjIiIVrRwzfECSd+nmuP6AeASYNZozpXkGBERrWl6tKrtTwKfXNbzJDlGRERr2p7pZrSSHCMiohVtzJAzVpIch90KVA/B6sKCjuL2TO8u9JFHdhcbQLu5s9g+suM//a/tMHaHP3MTRXqOERERfdJzjIiIqElZNSIiYoCUVSMiIvqk5xgREVHTxUOLRyvJMSIiWpOeY0RERE0G5ERERAyQsmpERESf9BwjIiL6pOcYERFRM0zXHIekmaMj6QmSTpD0O0lXSjpT0r8kzZN0u6Tryvszyv5bSLKkl/SdZ2HZ71JJF0vadhExp0s6pcScW2Lu0LfPjyWd17fuoNK2dWrr7hyb70RExPgwZcroX622s91w7ZEk4CRgtu0NbT8DeD/wEtszgJOBA23PsP3ictiewDnla93dZb/NgY8Ch4wQcyrwU2BWibkl8F5gg9o+qwPPBlaX9LS+U9wK/OdoP3NExHjXu9dxNK82TeSy6o7A/baP6q2wPW+knUsyfS2wE/BrSVNt3zNg11WBv41wmjcC59k+uRZzPjC/ts9rgJ8ANwGv55GJ9hvAXpI+a/v2RXy2iIihk7Lq+LApMHcp9t8OuM7274DZwG61bY8rZdWrgaOB/xrhHM8ELl5MnD2B48urv4d6J1WCPGAp2h0RMTSGpec4kZPj0toTOKG8P4FHJq5eWXUTYBfg26WnuUiSTpI0X9IPy/K6VE9fPMf2b4EHJG3ad9iRwFslrbqI8+4raY6kObfcscSfLyKiY3ORNOpXmyZycrwC2HJJdpS0HFW5839Luh74ArCrpFX697V9HrAWsLakT5ce5bxazGfX9t0d2At4fFm1B7AGcF2JM42qtFo//9+B44B3j9Re27Nsz7Q9c+0RU2hExDj04DK8WjSRk+OvgBUk7dNbIWkrSc8fsO+LgUttr297mu2nAj8AXtW/o6RNgOWA22x/vPQoZ5TNxwHbSXpF7ZAVa+/3BHYpMaZRJe9HJMfic8A7mdjXhCNiMvIyvFo0YZOjbQO7AzuV2yquAA4Cbhyw+55UI1vrfgC8obx/XK2HeCLwVtsLB8S8G3gZ8C5Jvy+3a3wCOFjSNOApwPm1/a8D7pD0nL7z3Fras8JSfeiIiPHMDE3PcUL3TGzfCLxuhG17DXpfW3cy1e0e2F5uKWJezSMH89Q9acD+vTLsBX3rPwB8YEnjRkQMhZZ7gKM1oZNjRESMMy33AEcryTEiItrRK6sOgSTHiIhoT8qqERERfdJzjIiIqOnglozRSnKMiIj2pOcYERFRkwE5ERERA6SsGhER0Sc9x4iIiD7pOUYb/sGWnMqcTmLvem3LD1jrc+qC7v6XLVjQWWgAjjyyu9jav9vfbv5Zhz93u3b8m/0L3f6fW2a55hgRETFAkmNERESflFUjIiJqhqisOmGf5xgREeNQww87lrS6pO9LulrSVZKeO5pmpucYERHtab7neATwc9uvlbQ8sOJoTpLkGBER7Wi4rCppVWAHYC8A2/cB943mXCmrRkREe5otq24A3AJ8U9Ilko6WtNJompnkGBER7XlwGV6wlqQ5tde+fWd/DPBs4Cu2twDuAj4ymmamrBoREe1Y9kdW3Wp75iK23wDcYPuCsvx9Rpkc03OMiIj2LFvPcZFs/xX4k6Snl1UvAq4cTTPTc4yIiPY0P1r1vcCxZaTq74G3jeYkjfccJT1B0gmSfifpSkk/k7TxGJ5/ZUlfKee/RNJcSfsswXGWdFht+YOSDqotv0/SW8r7gyT9WdK88tptrNpfi/cKSR8p7/eTNKp/0IiIcWtZBuMsYTnW9jzbM20/y/arbP9tNE1tNDlKEnASMNv2hrafAXwMWHcMwxwN/A2YXi7A7gI8fgmOuxd4taS1+jdIegywN3BcbfXnbc8or5+NQbsfwfbJtj9TFr8B7D/WMSIiOtdgWXUsNd1z3BG43/ZRvRW25wHnSDpU0nxJl0vaA0DSCyTNrs1ucKwqu0r6bu8cZb+fSNoQ2Br4hO0Hy/lvsf3Z2r4HSrpI0mWSPlVr2wPALOD9A9r9QuBi2w8s6QctbTpL0ncl/VbSZyS9UdKF5TNuWPZ7uaQLSi/3DEnrlvV7Sfpi+Qz/Aq6XtPWSxo+IGAoN9xzHStPJcVNg7oD1rwZmAJsDLwYOlbRe2bYF8D7gGVT3rGwHnA5sU7tfZQ/gROCZwKW9xNhP0s7AdKoEOgPYUtIOtV2+BLxR0mp9h243oN37lQT7DUlrjPB5NwcOADYD3gxsbHtrqt7te8s+5wDblF7uCcCHRjjXHOB5I2yLiBg+vUkA0nMc0fbA8bYX2r4JOAvYqmy70PYNJeHNA6aVHtzPgZeXkudLgR/3n1TSx8s1wRvLqp3L6xLgYmATqmQJgO07gG/z6BLmelQ3kvZ8BdiQKsH+BTiMwS6y/Rfb9wK/A04r6y8HppX3TwZ+Iely4ECqBD/IzcATB22QtG/vPp877rhl0C4REeNTkiMAVwBbDli/qCd23lt7v5CHR9SeCLyOquR5ke1/Ug3R3VzSFADbn7Y9A1i1FueQ2rXCjWx/vS/e4cDbgfosCncDU3sLtm8qifxB4GtUPdHFtf3B2vKDtc/xBeCLtjcD3lmP02dqacej2J5VLjjPXHXVtUc4PCJiHEpZFYBfASvUR49K2opqAM0ekpaTtDbVXHgXLuZcs6lmPtiHKlFi+1qq8uPBkpYr55/Kw8n3F8DeklYu254kaZ36SW3fDnyXKkH2XAVsVGvzerVtuwPza+f75WLa3W814M/l/VsXsd/GvTgRERNCyqoV26ZKJjuVWy2uAA6iGgV6GXApVQL9ULl5c1HnWgicAuxavva8A1gTuFbSXOAM4MPlmNNKrPNKGfP7wCoDTn8YUB+1eipVwu757zKo5jKqQUa9QTzrUQ3sWRoHAd+T9Gvg1v6PWXu/XfksERETx5D0HBufBMD2jVTl0H4Hlld939lUPcTe8n592/cD+tfdQVWeHCn+EVSPMOlfv3Lt/U3UHmti+w+SbpM03fYC228e4fTbUA3qGdT2Fwz6XLZ/zIDrpVQJ/nYASVsAV9juT54REcNtSB52nBlyRvYRqp7hgpF2sP3FsQgk6V1Uj1h5dVm1FvC/xuLcERHjRsOPrBpLSY4jsH0NcE1LsY4C6veCnt5G3IiI1rVcHh2tJMeIiGhPeo4RERF90nOMiIioyTXHiIiIAZIcIyIi+qSsGhERUZOyakRExADpOUZERPRJzzEiIqImZdVoy733wrXXdhT8Zx3FLU49tbvY06cvfp8mLRhxUsPm/azjf3ft1l1d7sgjOwsNwK67dvPZ7501c+xOlrJqREREn/QcIyIiajp49NRoJTlGRER70nOMiIjok+QYERFRk7JqRETEAOk5RkRE9ElyjIiIqElZNSIiYoD0HCMiIvqk5xgREVEzRHOrTlnaAyQ9QdIJkn4n6UpJP5O08Vg1SNLKkr5Szn+JpLmS9lmC4yzpsNryByUdVFt+n6S3LGVbfrNUjV9G5Xu5uqTlJZ0tKX+8RMTE8uAyvFq0VMlRkoCTgNm2N7T9DOBjwLpj2Kajgb8B021vAewCPH4JjrsXeLWktfo3lCSzN3Dc0jTE9rZLs/+ysr2b7b/bvg/4JbBHm/EjIhrnZXi1aGl7jjsC99s+qrfC9jzgHEmHSpov6XJJewBIeoGk2ZK+L+lqSceqsquk7/bOUfb7iaQNga2BT9h+sJz/Ftufre17oKSLJF0m6VO1tj0AzALeP6DdLwQutv1AOcdsSZ8vvbOrJG0l6YeSFkg6uBbrzkV9jv4gS3HeH5Ue8RWS9q2tv76W3H8EvHEx/x4REcOjV1Ydgp7j0pbtNgXmDlj/amAGsDmwFnCRpLPLti2AZwI3AucC2wGnA1+VtJLtu6h6SCeW/S7tJcZ+knYGplMlUAEnS9rBdi/Wl4DLJP1336HbDWj3fbZ3kHQA8GNgS+B24HeSPm/7tr79B32OcwY0c0nOu7ft2yU9rnyvfjAg3nxgqxG+D/sC+wKsscZTBu0SETE+DcmAnKW+5jiC7YHjbS+0fRNwFg//Yr/Q9g0l4c0DppUe3M+Bl5eS50upEskjSPq4pHmSbiyrdi6vS4CLgU2okiUAtu8Avg3s33eq9YBb+tadXL5eDlxh+y+27wV+D6w/4DM+6nOM8L1YkvPuL+lS4Pyy7lFPB7S9ELhP0ioDts2yPdP2zJVXXnuEZkREjEMTtOd4BfDaAesfVWKsubf2fmEt5onAe6h6VRfZ/qekK4HNJU2x/aDtTwOf7pU3S5xDbH91EfEOp0qc36ytuxuYOkK7Huxr44MM/r6M9DlG2m/geSW9AHgx8Fzb/5I0e0DbelYA7hlhW0TEcJnAo1V/BaxQHz0qaSuqATR7SFpO0trADsCFiznXbODZwD5UiRLb1wJzgIMlLVfOP5WHk+8vgL0lrVy2PUnSOvWT2r4d+C7w9trqq4CNlvKzLhFJh0jafSkOWQ34W0mMmwDbjHDeNYFbbN8/Fu2MiBgXWhiQU3LRJZJOGW0zlyo52jawO7BTudXiCuAgqlGglwGXUiXQD9n+62LOtRA4Bdi1fO15B7AmcK2kucAZwIfLMaeVWOdJuhz4PvCosiNwGNW1z55TqRJ2EzYDFvlZ+/ycqgd5GfBfVKXVut6PwI7Az5a9eRER40g7ZdUDqDpFo7bU99HZvhF43YBNB5ZXfd/ZVD3E3vJ+fdv3A/rX3QG8cxHxjwCOGLB+5dr7m4AVa8t/kHSbpOm2F9h+wSLaWN+28hJ8jsfaPm/AsSOel+oPgkcoPeVVgDvKqjcAH+3fLyJiaLVwS4akJ1ONY/k08IHRnmesBuQMg49QDcwZU7ZfMkanugI42vb9kpYHfmT7mjE6d0TE+LBsPce1JM2pvfZ91PmrcScfYhmvbk6aGVhKohm3ycb2JrX391GNuo2ImFiWbUDOrbZnjrRR0suAm23PLYMfR23SJMeIiOhY82XV7YBXSNqN6i6AVSX9j+03Le2JJlNZNSIiutbggBzbH7X9ZNvTgNcDvxpNYoT0HCMiok1DMkNOkmNERLSjxUkA+u8YWFpJjhER0Z4hmSEnyTEiItqTsmpERETNEM2tmuQYERHtSc8x2rDCCrBRI1OqL4H+B4O1bHqH8Rcs6C42wPRHPeRscsQG8JGLeghQs7R/t7/Zu/rsK/xjDE+WnmNERERNyqoREREDpKwaERHRJz3HiIiIPuk5RkRE1OSaY0RExABJjhEREX1SVo2IiKhJWTUiImKA9BwjIiL6pOcYERFRM0Rl1SlNnFTSEySdIOl3kq6U9DNJG4/h+VeW9JVy/kskzZW0zxIcZ0mH1ZY/KOmg2vL7JL2lvD9I0p8lzSuv3caq/UvQzpdJ+lRb8SIiWuNleLVozJOjJAEnAbNtb2j7GcDHgHXHMMzRwN+A6ba3AHYBHr8Ex90LvFrSWv0bJD0G2Bs4rrb687ZnlNfPRttYScst5SE/BV4hacXRxoyIGJceXIZXi5roOe4I3G/7qN4K2/OAcyQdKmm+pMsl7QEg6QWSZkv6vqSrJR2ryq6Svts7R9nvJ5I2BLYGPmH7wXL+W2x/trbvgZIuknRZXw/sAWAW8P4B7X4hcLHtB5b0g5Y2nS3ppNJDPkrSlLLtTkn/R9IFwHMlfaB89vmS3lf2+YCkb5T3m5VtK9o2MBt42ZK2JSJi3FuWXuOw9xyBTYG5A9a/GpgBbA68GDhU0npl2xbA+4BnABsA2wGnA9tIWqnsswdwIvBM4NJeYuwnaWdgOlUCnQFsKWmH2i5fAt4oabW+Q7cb0O79SoL9hqQ1Rvi8WwP/CWwGbFg+J8BKwHzbzwHuBt4GPAfYBthH0hbA4cBGknYHvgm80/a/yvFzgOeNEDMiYjhN4p7jSLYHjre90PZNwFnAVmXbhbZvKAlvHjCt9OB+Dry8lDxfCvy4/6SSPl6uCd5YVu1cXpcAFwObUCVLAGzfAXybRz+NcD3gltryV6iS3QzgL8BhDHah7d/bXggcXz4nwELgB7XPfpLtu2zfCfwQeF75vHsB3wHOsn1u7bw3A08cFFDSvpLmSJpzxx23DNolImJ8msTJ8QpgywHrF/WUzntr7xfy8CjaE4HXUZU8L7L9T+BKYPNe+dL2p23PAFatxTmkdq1wI9tf74t3OPB2qt5dz93A1N6C7ZtKIn8Q+BpVD3GQ/s5+b/mekjB7bRrJdOBOHp0Ip5Y2PTqgPcv2TNszV1117UWcOiJiHJnkZdVfASvUR49K2opqAM0ekpaTtDawA3DhYs41G3g2sA9VosT2tVQlx4N7A10kTeXhBPQLYG9JK5dtT5K0Tv2ktm8HvkuVIHuuAjaqtXm92rbdgfm18/2ytm1rSU8ryXoP4JwBn+Ns4FWSVixl4t2BX5fS7hHle7GmpNfWjtm4FzMiYsIYkp7jmN/naNvlGtrhkj4C3ANcT3VNcWXgUqq/AT5k+6+SNlnEuRZKOoWq9PjW2qZ3AIcC10q6naqH9eFyzGmS/g04rxo4y53Am6jKlHWHAfvVlk+lKm/2/LekGaWt1wPvLOvXoxrY03Me8Bmqa45nU43U7f8cF0s6hof/GDja9iVlMM6Xbf9W0tuBMyWdbftmqoFNHx3pexMRMZQm8ww5tm+kKof2O7C86vvOpuoh9pb369u+H49MYr3rhu9kBLaPoOqR9a9fufb+JmDF2vIfJN0mabrtBbbfPMLpt6Ea1NPzL9t7LCpWWf4c8Lm+dXvX3v+J0nOVtC7wONuXj/QZIyKGzhBNApAZch7pI1Q9wwUj7WD7iy204ylUI2AjIiaWJMfhY/sa4Jql2H82tV7vGLbjorE+Z0TEuDCZy6oRERGPkrJqRETEAOk5RkRE9EnPMSIioiZl1YiIiAFSVo2IiOiTnmNERESf9BwjIiJqcs0xJoVduw1/6qndxd5oo8Xv06QFI87hNPG9973ddT2O7CxyRft39dlnjt2phiQ5tvk8x4iImMwafmSVpPUlnSnpKklXSDpgtE1NzzEiItrTbM/xAeA/y5OQVgHmSjrd9pVLe6L0HCMioj0N9hxt/8X2xeX9P6me0/uk0TQzPceIiGhHiwNyJE0DtgAuGM3xSY4REdGeZUuOa0maU1ueZXtW/06SVgZ+ALyvPP93qSU5RkREe5ZtwO2tthc5dFbSY6kS47G2fzjaQEmOERHRjobLqpIEfB24yvbnluVcGZATERHtaXBADrAd8GbghZLmldduo2lmeo4REdGeBnuOts8BNBbnSnKMiIh2ZPq4iIiIAYZk4vEluuYo6QmSTpD0O0lXSvqZpI3HqhGSVpb0lXL+SyTNlbTPEhxnSYfVlj8o6aDa8vskvWUp2/KbpWp8QySdIWmNrtsRETGmHlyGV4sWmxzL6J+TgNm2N7T9DOBjwLpj2I6jgb8B021vAewCPH4JjrsXeLWktfo3SHoMsDdw3NI0xPa2S7P/kihtWVrfAd491m2JiOhUswNyxsyS9Bx3BO63fVRvhe15wDmSDpU0X9LlkvYAkPQCSbMlfV/S1ZKOVWVXSd/tnaPs9xNJGwJbA5+w/WA5/y22P1vb90BJF0m6TNKnam17AJgFvH9Au18IXGz7gXKO2ZI+L+nsMintVpJ+KGmBpINrse5c1OfoD1L2OVzSb8r3Yuuy/iBJsySdBnxb0lMl/bJ8hl9Keoqk1SRdI+np5Zjjaz3mk4E9l+DfJyJiOPSuOU6EniOwKTB3wPpXAzOAzYEXA4dKWq9s2wJ4H/AMYAOq4bWnA9tIWqnsswdwIvBM4NJeYuwnaWdgOlUCnQFsKWmH2i5fAt4oabW+Q7cb0O77bO8AHAX8GHhP+Xx7SVpzQPhBn2OQlUqP893AN2rrtwReafsNwBeBb9t+FnAscKTtfwD7AcdIej2whu2vAdj+G7DCCO2KiBhOEyg5jmR74HjbC23fBJwFbFW2XWj7hpLw5gHTSg/u58DLS5nxpVQJ6hEkfbzcm3JjWbVzeV0CXAxsQpUsAShTA30b2L/vVOsBt/StO7l8vRy4okxSey/we2D9AZ/xUZ9jhO/F8aUtZwOrSlq9F8/23eX9c3m4xPsdqu8ftk8v7fkS8I6+894MPLE/mKR9Jc2RNOeOO/o/YkTEODaByqpXUPWA+i3qXpJ7a+8X8vCo2BOB11GVPC8qs6ZfCWwuaQqA7U/bngGsWotziO0Z5bWR7a/3xTsceDuwUm3d3cDUEdr1YF8bH2TwyN2RPke//n+23vJdI+z/0D7lc/9baW//ddapZf0jD7Rn2Z5pe+aqq669iBAREePIBCur/oqqvPfQ6FFJW1ENoNlD0nKS1gZ2AC5czLlmA88G9qFKlNi+FpgDHCxpuXL+qTycfH8B7F0mkkXSkyStUz+p7duB71IlyJ6rgEae1y7pEEm711b1rrduD/yjlEv7/QZ4fXn/RuCc8v79pa17At8o8wL2BkI9Abh+zD9ARERXJkrP0baB3YGdyq0WVwAHUZUILwMupUqgH7L918WcayFwCrBr+drzDmBN4FpJc4EzgA+XY04rsc6TdDnwfWCVAac/DKiPWj2VKmE3YTOg/ln/Vm4BOYpHJui6/YG3SbqManqjA8rtMO+gejjnr4GzgU+U/bcEzu8NKIqImBCGpOe4RLcY2L6Rqhza78Dyqu87m6qH2Fver2/7flSDUOrr7gDeuYj4RwBHDFi/cu39TcCKteU/SLpN0nTbC2y/YBFtrG9beQk+x2Ntn1db/oHtj/a17aC+5eupysn9/q22zwdq698MfHnA/hERw2mIZsiZ6BOPf4RqYM6Ysv2SsT7nAPNt/7KFOBER7RmSsuqEnj7O9jXANQ3HeEFD5/1aE+eNiOjUkPQcJ3RyjIiIcaSDHuBoJTlGRER70nOMiIjok+QYERFRk7JqRETEAOk5RkRE9EnPMSIiomaIJgFIcoyIiPYkOUZERPQZkrKqqnnFY1hJugX4wzKcYi3g1jFqzjDF7jr+ZI3ddfzJGntZ4z/V9jI/H2/m8vKcJ4z+eP2JubZnLms7lkR6jkNuWX9gJc1p64dtPMXuOv5kjd11/MkaezzEf8iQ9MeSHCMioj255hgREVGT0aoxRGZN0thdx5+ssbuOP1ljj4f4lSEpq2ZATkREtGLmY+U5a4z+eN2SATkRETERDUl/LMkxIiLaMUTXHKd03YBolyrrd92OiJikHlyGV4vSc5xkbFvSj4At24wraSvgT7b/WpbfAryGagKDg2zf3mDspyxqu+0/NhU7xg9JKwH32F7YdVvaImkN4InA3cD1trvvt6WsGuPY+ZK2sn1RizG/CrwYQNIOwGeA9wIzqEbRvbbB2D+l+i+p2joDawPrAMs1FVjSP3n410Evvqn+7y1vu9H/g13HL22YCTyPh39JzwfOaPIPohJ3CvB64I3AVsC9wAplVqmfAbNsL2gw/lTgZTz6s//U9hUNxl0NeA+wJ7A8cAswFVhX0vnAl22f2VT8RRqismqS4+S0I/AuSdcDd1H90rTtZzUYc7naL8M9qH4x/QD4gaR5DcbF9mb1ZUnTgA9TJev/23DsVfpirwK8G3gncFKTsbuOL2kvYH/gOmAucA3VL+ntgQ9Lmg/8rwZ77mcCZwAfBeb3ek2SHk/1f+Azkk6y/T9jHVjSQcDLgdnABcDNVJ994xJ3KvCfti8b69jA94FvA8+z/fe+dm0JvFnSBra/3kDsxUvPMcaxXTuIuZykx9h+AHgRsG9tWys/h5KmAx8HngMcBuxv+/6WYq8OvA94C3AcsJXt29qI3WH8lYDtbN89QptmANOBppLjiwf9+5Y/0np/mD22odgX2T5ohG2fk7QOsMhy/2jZ3mkR2+ZS/aHSnSHpOWZAziRk+w/A+sALy/t/0fzPwvHAWZJ+TFVe+jWApI2AfzQZWNKmko6n+oV4BrCp7aPbSIyS1pJ0CHAx8ACwhe1PtJUYu4xv+0sjJcayfZ7tXzbYhMdB1VMc8FpD0nJN/QzY/mmJPbV/m6S1bN9se04TsWtx3t63vJykTzYZc7F6ZdUhGJCTSQAmofIfZCbwdNsbS3oi8D3b2zUcdxtgPeA023eVdRsDK9m+pMG4C4E/UV17fNRgDNv7Nxj7LqprPt8E/jkg9ueait11fEkfsv3fkr7Ao4tpBm4H/sf27xqKf4rtl0m6jkdfcwZYGfia7Y81Eb+04XJgH9vnl+XXAIfY3ripmLXYxwGrA28H1qT6GTjL9gebjj2SmVPkOcuP/njdm0kAolm7A1tQ9SawfWO5FtWo3i+InjJ6cGuqgQMvbTD03g2ee3EO5eHE0P89buMv0y7jX1W+jtRDWhP4IbB5E8Ftv6x8fdqg7ZKWoxog01hyBN4AfEPSbKpBOWsCL2ww3kNsv0HSHsDlVNWhPW2f20bsRWq4ByhpF+AIqoF2R9v+zGjOk+Q4Od1XbukwPJSkWiFpeWA3ql8au1CVOo9qMqbtb43QlqlUgyaajH3QSNvK7S2N6jK+7Z+UrwO//6UNdzXZhlqcNaiubz5U5rR9NvBvTca1fbmkTwPfoeq572D7hiZj9pRr7AdQ/R/7N6qBOJfY/lcb8Qcyjf5JVv7g+RKwE3ADcJGkk21fubTnSnKcnL4r6avA6pL2oepZHd1kQEk7UfUQX0I1ivA7wNa239Zk3AHtWA7YudaWXwPfazH+M6huL9iT6lprq8/X6yK+pLWpRgc/g0cmpxfa/moL8d9BlSSeDMwDtgHOo4UenKSvAxsCz6IaqfoTSV+0/aWmYwM/AfazfYYkAR8ALgKe2ULskTXbc9wauNb27wEknQC8EkhyjMWz/f9KsroDeDrwv22f3nDYX1Alou1tXwcg6YiGYz6k3Fv5Bqry7YXAdsDT2vgrWtJTqZLRnlSDYp4KzLR9fdOxx0N84FjgRKrv/buAt1JdB23LAVT3OZ5ve0dJmwCfain2fOAdrgZ3XFeuuzd6nblma9t3QHWfFnCYpJNbij2yZpPjk6jGF/TcQDU6faklOU5Ckj5r+8PA6QPWNWVLqh7LGZJ+D5xAgzff10m6gep2ga8AB9r+p6TrWkqMvwFWo/q8r7W9oMS+vunY4yF+sabtr0s6wPZZVKOWz2ox/j2275GEpBVsXy3p6W0Etv15SZtLel5Z9Wvbb1/kQWNnVUnforqv9EHgHKo/FDozF34hWGsZTjFVUv0a9izb9Udx9Q+6glEWcpMcJ6edqMpcdbsOWDdmymjUS6hu/t6OMnuHpFOBk/p+wMfaD4BXUU0+sLDcTtLWMO1bqMp561LNyLOgxdjjIT5A73aJv0h6KXBjaVNbbij3ef4IOF3S30obGidpf6p7en9YVv2PpFm2v9BC+G9S3dP672X5TWXdiPdBNs32Lg2HuIHqNrWeJzPKf+vcyjGJSPoPqtlRNgDqw+dXAc61/aaW2zOFapaafWz/++L2X8ZYopoVZU+qAUGrUQ1x/6ntOxuOvRrVPLJ7AhtRDa9/ie0Lm4w7juK/jKqkvj7wBWBV4FO2Wy/xSXo+1b/9z23f10K8y4Dn1m5dWgk4r+HZqHqx59me0bfuUtuNjA4eDyQ9Bvgt1UQjf6a6xvoGj2K6viTHSaT8klwDOAT4SG3TP93wPJcl/pOo7nO8zPZ9ZZaQ9wF72X5i0/Fr7Xgs1UjZPYGdbS9LmWdpY69L1YN9PbC+7VafkNJ1/K6U0arrU6uW2b64hbiXU81GdE9Znko1e85miz5yTGKfARxDNQEHVD/vb7P9oqZjd0nSbsDhVJdtvmH706M6T5Lj5FWSU330YGNPp5D0Pqqp264FVqC6D+lzVHNA/rftvzQY+5XAk3sjBCVdQDXhOFSDkb7TVOzFtOupZYaiTrQVX9LTqCaZn8Yjk9Mrmo5d4v8XsBfwex4eDmLbbYxWfX+J3ZvH9lXAMbYPbyH2U4AvAs+lKqX/Bjigy5+5YZLkOAlJejlVYnoi1YTITwWust3YEG9JV1KNVL29/Ke9luqer/MXc+hYxD4XeL3tP5XleVRll5WAbzb5l7Sk7YENbH+7LH8feHzZfLDtXzUVezzELzEvBb5OdTP6Q2MVy+Ccxkm6BtisjTJqX9wpVLeN3EM1KEbA2U3OBlWLvRzwrbYvlUwkGZAzOR1M9Z/2DNtbSOpdi2vSPb3Sre0/SvptG4mxWL6XGItzXM0telsLEyB8iqrX1PN0qp7ESlQzszSdnLqOD9W//ZEtxBnJfKrrrDe3GdT2g5IOs/1cymxULcZeKGltScu3/UfBRJHkODndb/s2SVMkTbF9pqTPNhzzyZLqvyDXqS+7wflNqa6zPsT2frXFtRuMC7Bq3+wcC1w9GQFVE4I3rev4AEeoms/3NKpnKgLtXPMrDgEuUfWIrHr8Nsq6p6maT/WHbr9Mdz1wbrm38aGZiNzwfL4TRZLj5PR3SStTjSA8VtLNVDeHN+nAvuU2H5tzgaR9bH+tvlLSO6kmBGjS6vUF26+uLa7bcOzxEB9gM+DNVDPSPHTNj5bmGAW+BXyWvrJuSz5A1UtfKOmess62V20h9o3lNYVHz6sbi5FrjpNQKSXeTfWf5o1UQ9uPdYOPMVL17L5LO/jruTfw6EdUvYZeb2VLqoFBr7J9U4OxfwIc5fIIo9r6lwH/YbvJCdc7j19iXQ08q6vynqSzbD+/i9jjgaRVqRLyo57KEiNLcpykypRi013Nu7gisFyT/3nKrBZPo0pO51KNnDu/N71VGyS9kIfnlbyipcEo04FTqD5vPTFvC7zM9m8bjr8R1aO6Oolf2nAi8F7brV7zq8X/HNUfRifTQVlX0qupBuSYaoacH7UUdybVTf+9XuM/gL17ZfVYtCTHSUjVZOP7Ao+3vWH5BX5U0/c/lSS8NdUv5m2p5rv8K9UEBO9uMnaXJK1A1UN/KDEDx/XufZsE8WdTTbx9Ee1f80PSmQNWt3Urx5epJl7o3Wu4B/A72+9pIfZlwHts9x4svj3w5TYmIJgIkhwnoXIrw9bABba3KOsub+PG5BJrJarRstsBbwGm2N6gjdhtk3Sa7Z27bkeXyqw0j9LWrRxdknQFsGnvckK5vePyJm+bqsU+130PMB+0LgbLgJzJ6d4yQw3w0JRLjf6VJOkNVL3FGVS9h4uAC6juffxrk7E71vRo2KUm6WLbz24r3nhLgmVSiL/avqCFcNcATwF6N96vD1zWQlyAC1U9mu54qv/fewCzJT0bWh0tPJSSHCensyR9DHicqkdXvZvq2W9NmgVcTfVg47PbuNY1TqxWrjkNZPuHI21r0KAnF7QXvJrW7H7gS7ZP6aAJzwE2k/QY27s2HGtN4CpJvVHRWwHnldsrmi4tzyhfP9m3flvaHS08lFJWnYRKaeftVA/9FdWzFo9uciRpmbFjcx6+3vh04C9UD509r43BMV2QdBvwY0Z4lI7tvVtuEpIOtv2JtuPW4j+Rao7dbdzOQ387M1JJuWe89arjYUmOk5Sk5YFNqP6CvKaDqbXWBV4LvJ/qocOtPNuxbW2XMJeEpLWA27q4raYrkrbl0XO7fruzBnWg5XLy0JvSdQOifaqeqfc74EiqiYmvldRoeUnSsyS9S9K3JV1Ldc1xB6pHGI3qSd1D4umqnl/5CJKeJ2nDpoNL2kbSbEk/lLRFmSVmPnCTpEafrVc/v6TVJH1d0mWSjit/HLVC0neA/0d1O8VW5TWzrfh9bTlD0qnlPtO2PQf4hKpnqMZipOc4CZWbsl9m+9qyvCHVcw03aTBm/f7G30yWJwNI+hXwPtuX9a2fCXzS9ssbjj+Hag7V1aiu++5q+3xJmwDH90YrNxT7oV6zpKOpbtv5GvBq4Pm2X9VU7L52XAU8Yzz0lCdTSXnYZUDO5HRzLzEWv6fhSZnrpUVJy0vatCxeY/v+EQ6bCNbpT4wAtudImtZC/MfYPg1A0v/pTfZu++reaOWWzPTDD979vKS3thh7PvAEqmvcnbLdm9KtlRvxU04evSTHyekKST8Dvkt1zfHfgYt6oyqbHEEp6QVUc11eTzVIZX1Jb7V9dlMxOzZ1Edse10L8+lyid/dta7ontY6kD1D9O68qSbXeW5uXdNYCriwjRluZhEDSLrZ/Xt6vRvWIuK2oEvX7m5yysNaG7wAbAvOAhWW1qZ6hGouRsuokJOmbi9jc6AhKSXOBN9i+pixvTFXe27KpmF2SdDzwqwGTnr8d2Nn2Hg3HX0j1RAZRJeN/9TYBU20/tsHY/bcQfNn2LZKeQPWA67c0FbuvHa1PQjAeSsrjqZw8jJIco1WSLuufvmrQuomiDDw5CbiPh0tpM4Hlgd0n+AQIk1ZfcpxXKyk/arnBNnwP2N925+XkYZSy6iRUemtfAda1vamkZwGvsH1wC+HnSvo68J2y/EbafXxVq0r5bFtVD5TuXWf9aZf3dZbp+15F1YNv7Kkckt5ENYfrwMdElYFg69k+p6H459jeXtI/eWQJWTT/2KjxUFJuvZw8kaTnOAlJOovq+Ypfrc2tOt/2pos+ckxirwC8h2pYvYCzqcpt9y7ywFgm5b7W3YA3ALsAP6B6AG9jMyNJOgDYm+qPn7nALVTXYDcCng/cCnzE9oKm2tCV8VBSnsxz2o6FJMdJSNJFtreSdEktOTZe6ikz81zWRhKOSpkecE/gJcCZwInAF2xPayn+clTTlG1HdQvD3cBVwKm2/9hGG0o71qCa17Q+ajNzi8aIUladnG4tJa3ekwJeSwvD3G0/KOlSSU9p8xfjJPcL4NdUE7xfByDpiLaC214InF5enZD0X8BeVLcs9Uq8jc4t2mVJueNy8oSR5Dg5vYfqhvBNJP0ZuI7q2l8b1qO6leRCqlGUQK6DNGhL4PXAGZJ+D5wAtDJVn6T/vYjNtv1fbbQDeB2wYctTJK4JXFJGZ49YUm4isO3ty9dVFrdvjCxl1UmsDMyYQlXq2sP2sS3EzHWQjpRp7PYEXkN179tJtmc1GO8/B6xeiWrS+zVtr9xU7L52/AD4D9uNTnQxIG7nJeWUk0cvyXESkbQqVa/xSVRPijijLH8QuNT2KxuMPRV4F9VfzpcDX7f9QFPxoqIBD1su1353Al5v+20ttWMV4ACqxPhd4LC2klWZqu/HVDfgT5pRmyOVk23nUVVLIMlxEpH0Y+BvVI+JehGwBtX9dgfYntdw7BOpnuH3a2BX4A+2D2gyZkB90FVH8R8PfICqbP8t4Ajbf2u5DVcAX6X6o+yha4ANTwLQeUlZ0jXAZi2XkyeMXHOcXDawvRk8NGvHrcBTbP+zhdjPqMX+OnDhYvaPsdHZw5YlHUo1I8wsql/SdzYVazFutX1kyzHvGrDuoZIy0Mb11vnA6jQ8b/JElZ7jJKK+Zwv2L0/U2JOZOnzYsqQHqcqYD9DhqElJnyvtOJlHllVbufbWVUl5spaTx0qS4yRSm2cTHjnXZuO/rLqMPZnljxCQdOaA1Y1fe+u6pNxFOXkiSXKMmMAk3UU1wfm5feufB9xo+3cttWN7YLrtb0paC1ild9/lRNRXUv5SFyVlSWfZHjg6PBYvyTFiAlPHD1susT5JNdn6021vrOqBv9+zvV3Dcbu8Eb/zknLX5eRhlwE5ERNb1w9bBtgd2AK4uMS+sVyHa1qXN+K3+bzKkfRGKW9TW9fozEATSZJjxMTW9cOWAe6zbUm96QpXaiOo7SMkfZGHb8R/Fg/fiP/mFm/E76SkbHvHpmNMZEmOERPbRZL28eCHLbf1qLDvSvoqsLqkfaie1PG1xRwzJrqe27VeUga+SXVf8f9QJeumYnb6qLCJItccIyaw8fKw5fJ0kJ2prrn9wnbjyWqc3Ig/j1JSrj0Bp9GHe0/mR4WNpSTHiEmg72HLV7jDhy23ZTzM7SrpQttb926pKSXl85pMjiVu5/O6Drskx4hoRCndPt72oWX5BmBVqt7jh2x/pcW2dHUj/geB6VRz2R5C1aM7zvYXmo4dyybJMSIaIekiYBfbt5XlS2xvUSahP832Di20YTzM7dpqSXk8lJMnggzIiYimTOklxuJ7ALbvkdT4SNnxMrdrSYZtDggaD/O6Dr30HCOiEZKutb3RgPVTgGttb9Bw/M5uxB8vJeUuHxU27MbDjaoRMTGdJungAev/D3Ba08FtT7H9ONur2F619lqlhRlq3gV8o7Z8S4m5NtUDpxsl6fHle38ZVYXw2bY/nMS45FJWjYimHAgcLela4NKybnNgDvCONhvSwY34nZWUx0s5edilrBoRjZK0AfDMsnhlW5Od1+K3PrdrlyXl8TCv60SQsmpENO3zwCrAr9pOjMXuwCsoA1Vs31ja06TOSsodl5MnjCTHiGja54DtgSslfU/Sa8vtHG25z1WJrM25XQ8ENpR0raQflNe1VLPUfLCF+EBVTpb0tvJ+LUlPayv2sEtZNSJaUZu1ZR+q+x9b6cV0eSN+lyXlrh4VNlEkOUZE48oglJcDewDPBk6x/d4W47c+t2uJ+2PgRODHtgfdf9hk7Hm0PK/rRJLRqhHRKEknAs8Bfg58CZg90hMjmtLBjfg9n6P6g+AQSRdSJcpTbN/TQuxOHhU2UaTnGBGNkrQLcHp5fFSbccfFjfgldusl5czrumySHCOicZK2BaZRq1bZ/nbDMTuf27XE7ayk3FU5eSJIWTUiGiXpO8CGwDyg13s00GhypOO5XaH7knKH5eShl55jRDRK0lXAM9zyL5uu53YtsVovKY+ncvIwS88xIpo2H3gC8JeW454m6WDbn+hb38rcrgC2fy5pW0nTaK+k/C5gl9ryLbaf3CsnA0mOSyDJMSKathbVBAAXUk1rBoDtVzQct/O5XTsqKXdeTp4IUlaNiEZJev6g9bbPail+lzfit15SHg/l5Ikg08dFRKNKEryaaj7TVYCr2kqMRZdzu/ZKym3q9FFhE0V6jhHRKEmvAw4FZlMNCnkecKDt77cU//lUt1G8FGj1RnxJZwIzStxWSsrlZv+jga0YUE7OI6yWTJJjRDRK0qXATr0H7UpaGzjD9uYtt6OLG/E7Kyl3/aiwYZeyakQ0bUrfE+hvo+XfPWUgymuoRnJuBXyrjbgdl5S7flTYUEtyjIim/VzSLyTtJWkv4KfAz9oKXm7Ev4qq1/glYMMWZ6h5HVVJ9d+B1wEXSHptG7Hp/lFhQy1l1YhohKSNgHVtnyvp1VS/qAX8DTi2rd5MV3O7ltidl5S7elTYsMt9jhHRlMOBjwHY/iHwQwBJM8u2l7fRiI5uxO/ptKQ8YF7XVsrJE0GSY0Q0ZZrty/pX2p5TElUrOpzbFUpJGTi+LO9BSyXlrud1HXZJjhHRlEVd32pzppaZtH8jfq+kfGBfSfk84NiWmvFN4A1dlJMngiTHiGjKRZL2sf21+soyMfbcFtvRxdyuh9NxSbnjcvLQy4CciGiEpHWBk4D7eDgZzgSWB3a3/deW2tHFjfjzbW86wrbLbW/WVOxanIHlZNv7Nx17IkjPMSIaYfsmYFtJOwK9RPFT279quSkHtRwPxkdJufVy8kSS5BgRjbJ9JnBmh/HPKr3YrcqqC/tGkDZhPJSUu3pU2ISQsmpETGhdzO06HkrKXZSTJ5Ikx4iY0Lq8Eb+vpHxFmyXlrh8VNuySHCNiQusfAFOea3hpG4NiutZBOXnCyNyqETHRdTq3a1c6ntd16KXnGBET0niZ27Ur42Fe12GWnmNETFSHA/+E6kZ82x+w/X6qXuPhHbarLZ0/KmyY5VaOiJioxsXcrh3qbF7XiSDJMSImqvFwI37rxsm8rkMvXeyImKgukrRP/8oO5nZt2+FM7nLymMiAnIiYkMbDjfhdGA/zuk4EKatGxIQ0juZ2bdukLCePtfQcIyImEEnHA78aYV7XnW3v0U3LhkuSY0TEBDJZy8ljLckxImIC6nJe14kgyTEiIqJPbuWIiIjok+QYERHRJ8kxIiKiT5JjREREnyTHiIiIPv8f5CN4+9UGDqsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 504x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DoC - f1 score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DoC - cohens kappa score\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cmap = matplotlib.colors.ListedColormap([\n",
    "    (1.0, x / 255.0, 0.0)\n",
    "    for x in range(256)\n",
    "    ] + [\n",
    "    ((255 - x) / 255.0, (255 - x) / 255.0, 1.0) # x / 255.0)\n",
    "    for x in range(256)\n",
    "    ])\n",
    "\n",
    "#cmap.set_extremes(bad=cmap(0.0), under=cmap(0.0), over=cmap(1.0))\n",
    "\n",
    "for k in tables.keys():\n",
    "    print(k)\n",
    "    labels = list(gans)\n",
    "    t = tables[k]\n",
    "    if k[0:3] == \"DoC\":\n",
    "        #continue\n",
    "        labels = labels[-4:]\n",
    "        t = [r[-4:] for r in t[-4:]]\n",
    "        f = plt.figure(figsize=(5, 4))\n",
    "        f.add_axes([0.4, 0.45, 0.6, 0.5])\n",
    "    else:\n",
    "        f = plt.figure(figsize=(7, 6))\n",
    "        f.add_axes([0.27, 0.25, 0.7, 0.74])\n",
    "    p = plt.imshow(t, cmap=cmap)\n",
    "    plt.colorbar(p)\n",
    "    plt.xticks(range(len(labels)), labels, rotation=\"vertical\")\n",
    "    plt.yticks(range(len(labels)), labels)\n",
    "    plt.savefig(f\"data_result/statistics/successCount/statistic-{k}.pdf\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e4397945",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Table:\n",
    "    def __init__(self, heading):\n",
    "        self.heading = [str(h) for h in heading]\n",
    "        self.sizes = [len(h) for h in self.heading]\n",
    "        self.rows = []\n",
    "        \n",
    "    def add(self, row):\n",
    "        row = [str(r) for r in row]\n",
    "        self.rows.append(row)\n",
    "        self.sizes = [max(a,len(b)) for (a, b) in zip(self.sizes, row)]\n",
    "        \n",
    "    def separator(self):\n",
    "        return \"|\".join([\"-\" * n for n in self.sizes])\n",
    "    \n",
    "    def showRow(self, row):\n",
    "        def pad(n, t):\n",
    "            while len(t) < n:\n",
    "                t += \" \"\n",
    "            return t\n",
    "        \n",
    "        return \"|\".join([pad(n, t) for (n,t) in zip(self.sizes, row)])\n",
    "    \n",
    "    def show(self):\n",
    "        print(self.showRow(self.heading))\n",
    "        print(self.separator())\n",
    "        for row in self.rows:\n",
    "            print(self.showRow(row))\n",
    "            \n",
    "    def showLatex(self, caption, key):\n",
    "        \n",
    "        columnConfig = \"|\".join([\"l\"] + [\"@{\\\\hskip3pt}c@{\\\\hskip3pt}\" for h in self.heading[1:]])\n",
    "\n",
    "        text = \"\\\\begin{table*}[ht]\\\\scriptsize\"\n",
    "        text += \"\\\\caption{\" + caption + \"}\\\\label{\" + key + \"}\"\n",
    "        text += \"\\\\centering\\\\tabularnewline\\n\"\n",
    "\n",
    "        text += \"\\\\begin{tabular}{\" + columnConfig + \"}\\\\hline\\n\"\n",
    "        text += \" & \".join([\"\\\\textbf{\" + h + \"}\" for h in self.heading])\n",
    "        text += \"\\n\\\\tabularnewline\\n\\\\hline\\n\"\n",
    "        \n",
    "        for row in self.rows:\n",
    "            text += \" & \".join(row)\n",
    "            text += \"\\n\\\\tabularnewline\\n\"\n",
    "            \n",
    "        text += \"\\hline\\end{tabular}\\end{table*}\\n\"\n",
    "        \n",
    "        return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d0e2faa0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def tableRow(algo, dataset, myGans):\n",
    "    row = []\n",
    "    for gan in myGans:\n",
    "        v = getValueOf(gan, dataset, algo, f1Score)\n",
    "        w = getValueOf(gan, dataset, algo, kScore)\n",
    "        row.append((v, w))\n",
    "    return row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "generic-juice",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "def p(f, bold=False):\n",
    "    if f is None:\n",
    "        text = \"?\"\n",
    "    else:\n",
    "        text = f\"{f:0.3f}\"\n",
    "        \n",
    "    if bold:\n",
    "        return \" \\\\textbf{\" + text + \"} \"\n",
    "    else:\n",
    "        return \" \" + text + \" \"\n",
    "\n",
    "def latex(text):\n",
    "    r = \"\"\n",
    "    for x in text:\n",
    "        if x == \"_\" or x == \"-\":\n",
    "            r += \" \"\n",
    "        else:\n",
    "            r += x\n",
    "    return r\n",
    "\n",
    "def pairMax(row):\n",
    "    a = 0.0\n",
    "    b = 0.0\n",
    "    \n",
    "    for (x,y) in row:\n",
    "        a = max(a, x)\n",
    "        b = max(b, y)\n",
    "    \n",
    "    return (a, b)\n",
    "    \n",
    "with open(\"data_result/statistics/Tables.tex\", \"w\") as latexFile:\n",
    "    for algo in algs:\n",
    "        latexFile.write(\"\\n\")\n",
    "        latexFile.write(\"% ### \" + algo + \"\\n\")\n",
    "        latexFile.write(\"\\n\")\n",
    "        f = \"\"\n",
    "        heading = [\"dataset ($f_1~$score$~/~\\\\kappa~$score)\"]\n",
    "        \n",
    "        myGans = [x for x in gans if not x.startswith(\"ConvGeN\") or x == 'ConvGeN(min,maj)']\n",
    "        if algo[0:3] == \"DoC\":\n",
    "            continue\n",
    "            myGans = list(gans)[-4:]\n",
    "\n",
    "        for g in myGans:\n",
    "            heading.append(latex(g))\n",
    "        table = Table(heading)\n",
    "        \n",
    "        avg = [[0.0, 0.0] for h in heading[1:]]\n",
    "        mx  = [[0.0, 0.0] for h in heading[1:]]\n",
    "        cnt = 0\n",
    "\n",
    "        for d in testSets:\n",
    "            d = cleanupName(d)\n",
    "            if d not in ignore:\n",
    "                cnt += 1\n",
    "                row = tableRow(algo, d, myGans)\n",
    "                line = [latex(d)]\n",
    "\n",
    "                #print((algo, d, myGans))\n",
    "                #print(row)\n",
    "                \n",
    "                m = pairMax(row)\n",
    "\n",
    "                for (n, r) in enumerate(row):\n",
    "                    line.append(f\"{p(r[0], r[0] == m[0])} / {p(r[1], r[1] == m[1])}\")\n",
    "                    avg[n][0] += r[0] or 0.0\n",
    "                    avg[n][1] += r[1] or 0.0\n",
    "                    mx[n][0] = max(mx[n][0], r[0] or 0.0)\n",
    "                    mx[n][1] = max(mx[n][1], r[1] or 0.0)\n",
    "                table.add(line)\n",
    "\n",
    "        m = pairMax(avg)\n",
    "        table.add([\"\\\\hline Average\"] + [f\"{p(a / cnt, a == m[0])} / {p(b / cnt, b == m[1])}\" for (a,b) in avg])\n",
    "        #table.add([\"maximum\"] + [f\"{p(a)} / {p(b)}\" for (a,b) in mx])\n",
    "\n",
    "        latexFile.write(table.showLatex(algo, \"tab:results:\" + algo + \":A\") + \"\\n\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "bronze-pearl",
   "metadata": {},
   "outputs": [],
   "source": [
    "    \n",
    "with open(\"data_result/statistics/TablesConvGeN.tex\", \"w\") as latexFile:\n",
    "    for algo in algs:\n",
    "        latexFile.write(\"\\n\")\n",
    "        latexFile.write(\"% ### \" + algo + \"\\n\")\n",
    "        latexFile.write(\"\\n\")\n",
    "        f = \"\"\n",
    "        heading = [\"dataset ($f_1~$score$~/~\\\\kappa~$score)\"]\n",
    "        \n",
    "        myGans = [x for x in gans if x.startswith(\"ConvGeN\")]\n",
    "\n",
    "        for g in myGans:\n",
    "            heading.append(latex(g))\n",
    "        table = Table(heading)\n",
    "        \n",
    "        avg = [[0.0, 0.0] for h in heading[1:]]\n",
    "        mx  = [[0.0, 0.0] for h in heading[1:]]\n",
    "        cnt = 0\n",
    "\n",
    "        for d in testSets:\n",
    "            d = cleanupName(d)\n",
    "            if d not in ignore:\n",
    "                cnt += 1\n",
    "                row = tableRow(algo, d, myGans)\n",
    "                line = [latex(d)]\n",
    "\n",
    "                #print((algo, d, myGans))\n",
    "                #print(row)\n",
    "                \n",
    "                m = pairMax(row)\n",
    "\n",
    "                for (n, r) in enumerate(row):\n",
    "                    line.append(f\"{p(r[0], r[0] == m[0])} / {p(r[1], r[1] == m[1])}\")\n",
    "                    avg[n][0] += r[0] or 0.0\n",
    "                    avg[n][1] += r[1] or 0.0\n",
    "                    mx[n][0] = max(mx[n][0], r[0] or 0.0)\n",
    "                    mx[n][1] = max(mx[n][1], r[1] or 0.0)\n",
    "                table.add(line)\n",
    "\n",
    "        m = pairMax(avg)\n",
    "        table.add([\"\\\\hline Average\"] + [f\"{p(a / cnt, a == m[0])} / {p(b / cnt, b == m[1])}\" for (a,b) in avg])\n",
    "\n",
    "        latexFile.write(table.showLatex(algo, \"tab:results:\" + algo + \":A\") + \"\\n\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bearing-strengthening",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "% ### LR\n",
      "% ### RF\n",
      "% ### GB\n",
      "% ### KNN\n",
      "% ### DoC\n"
     ]
    }
   ],
   "source": [
    "for algo in algs:\n",
    "    print(\"% ### \" + algo)\n",
    "    heading = [\"dataset\"]\n",
    "    for g in gans:\n",
    "        if filterConvGen(g):\n",
    "            heading.append(g)\n",
    "    table = []\n",
    "    \n",
    "    avg = [[0.0, 0.0] for h in heading[1:]]\n",
    "    cnt = 0\n",
    "    \n",
    "    for d in testSets:\n",
    "        d = cleanupName(d)\n",
    "        if d not in ignore:\n",
    "            cnt += 1\n",
    "            row = tableRow(algo, d, heading[1:])\n",
    "            table.append([(d,d)] + row)\n",
    "\n",
    "            for (n, r) in enumerate(row):\n",
    "                avg[n][0] += r[0] or 0.0\n",
    "                avg[n][1] += r[1] or 0.0\n",
    "    \n",
    "    table.append([(\"Average\", \"Average\")] + [(a / cnt, b / cnt) for (a,b) in avg])\n",
    "    \n",
    "    with open(f\"data_result/statistics/{algo}-f1.csv\", \"w\") as f:\n",
    "        f.write((\";\".join(heading)) + \"\\n\")\n",
    "        for row in table:\n",
    "            f.write((\";\".join([str(x[0]) for x in row])) + \"\\n\")\n",
    "\n",
    "    with open(f\"data_result/statistics/{algo}-kappa.csv\", \"w\") as f:\n",
    "        f.write((\";\".join(heading)) + \"\\n\")\n",
    "        for row in table:\n",
    "            f.write((\";\".join([str(x[1]) for x in row])) + \"\\n\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "581f1134",
   "metadata": {},
   "source": [
    "def dsWeight(d, score):\n",
    "    w = 0.0\n",
    "    for g in gans:\n",
    "        for a in algs:\n",
    "            x = getValueOf(g, d, a, score)\n",
    "            if x is not None:\n",
    "                w += x\n",
    "    return w\n",
    "\n",
    "\n",
    "\n",
    "dataNames = [cleanupName(d) for d in testSets]\n",
    "dataNames.sort(key=lambda d: dsWeight(d, f1Score) + dsWeight(d, kScore))\n",
    "for d in dataNames:\n",
    "    w = dsWeight(d, f1Score) + dsWeight(d, kScore)\n",
    "    print(f\"{w:0.3f} - {d}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}