{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "16d92c9e",
   "metadata": {
    "id": "16d92c9e"
   },
   "source": "# KKB Quantum Readiness & Advantage — Proof of Concept\n### Hybrid quantum analytics and quantum-safe credit data exchange for a national credit bureau\n\n**Prepared by:** BVK Technology — Quantum Business Solutions · İTÜ ARI Teknokent, İstanbul<br>\n**Reference work:** İGA İstanbul Airport hybrid quantum gate-assignment simulator (v4, MARS model + Narrator Engine)<br>\n**Runtime:** Google Colab, CPU only (~5–8 minutes end-to-end) · Qiskit 2.x · FIPS 203/204 reference implementations\n\n---\n\nThis notebook demonstrates three tracks that the industry and literature review identified as relevant to Kredi Kayıt Bürosu (KKB):\n\n| Track | Question | Method | Maturity |\n|---|---|---|---|\n| **1. Quantum feature selection** | Can a quantum optimiser build a compact, non-redundant credit scorecard? | Information-value + redundancy QUBO, solved with QAOA, exact enumeration and classical annealing | PoC-ready |\n| **2. Quantum-kernel credit scoring** | Do quantum feature maps add predictive signal, especially when data is scarce (thin-file, SME, agricultural segments)? | Fidelity quantum kernel + SVM vs logistic scorecard, RBF-SVM and gradient boosting | Research |\n| **3. Quantum-safe credit report** | What does it take to protect a credit report against \"harvest now, decrypt later\"? | Hybrid X25519 + ML-KEM-768 (FIPS 203) encryption, ML-DSA-65 (FIPS 204) signature, Mosca risk model, crypto-inventory template | Ready now |\n\nA **Narrator Engine** (Stage 17) turns every number produced here into an executive Word report.\n\n> **Data notice.** No KKB data is used. Tracks 1–2 use the public UCI Statlog *German Credit* dataset (1,000 applicants, 20 attributes) as a stand-in for a credit-bureau scoring problem. In a real PoC the same code runs inside KKB's AI Safe Testing Environment on Kloudeks, keeping personal data in Türkiye (BDDK information-systems regulation).\n>\n> **Honesty notice.** Results are reported as measured. Where a classical method wins, the notebook says so — the literature shows QML on tabular credit data is at parity with, not ahead of, tuned classical models."
  },
  {
   "cell_type": "markdown",
   "id": "591d851f",
   "metadata": {
    "id": "591d851f"
   },
   "source": "## Stage 1 — Environment"
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0ce5ad42",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:36.299981Z",
     "iopub.status.busy": "2026-10-03T09:46:36.299578Z",
     "iopub.status.idle": "2026-10-03T09:46:41.222695Z",
     "shell.execute_reply": "2026-10-03T09:46:41.222124Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0ce5ad42",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791021950126,
     "user_tz": -180,
     "elapsed": 46424,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "97263bd8-32b3-457c-fc6c-6aed797943aa"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "\u001b[?25l     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/664.0 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K     \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m \u001b[32m655.4/664.0 kB\u001b[0m \u001b[31m28.1 MB/s\u001b[0m eta \u001b[36m0:00:01\u001b[0m\r\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m664.0/664.0 kB\u001b[0m \u001b[31m10.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25h  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
      "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
      "  Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9.8/9.8 MB\u001b[0m \u001b[31m56.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m327.8/327.8 kB\u001b[0m \u001b[31m15.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m237.1/237.1 kB\u001b[0m \u001b[31m9.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m253.0/253.0 kB\u001b[0m \u001b[31m15.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.4/2.4 MB\u001b[0m \u001b[31m43.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m54.9/54.9 kB\u001b[0m \u001b[31m2.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25h  Building wheel for docplex (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
      "✓ Qiskit 2.5.2 ready · seed 42\n"
     ]
    }
   ],
   "source": "!pip install -q \"qiskit>=2.0\" \"qiskit-algorithms>=0.4\" \"qiskit-optimization>=0.7\" kyber-py dilithium-py cryptography python-docx scikit-learn seaborn\n\nimport time, json, math, os, warnings, datetime, itertools\nimport numpy as np, pandas as pd\nimport matplotlib.pyplot as plt, seaborn as sns\nwarnings.filterwarnings('ignore')\n\nfrom sklearn.model_selection import train_test_split, StratifiedKFold, cross_val_score\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.svm import SVC\nfrom sklearn.ensemble import HistGradientBoostingClassifier\nfrom sklearn.preprocessing import MinMaxScaler, StandardScaler\nfrom sklearn.pipeline import make_pipeline\nfrom sklearn.metrics import roc_auc_score, roc_curve\n\nimport qiskit\nfrom qiskit.quantum_info import Statevector\nfrom qiskit.primitives import StatevectorSampler\nfrom qiskit.transpiler import generate_preset_pass_manager\nfrom qiskit_algorithms import QAOA\nfrom qiskit_algorithms.optimizers import COBYLA\nfrom qiskit_algorithms.utils import algorithm_globals\nfrom qiskit_optimization import QuadraticProgram\nfrom qiskit_optimization.algorithms import MinimumEigenOptimizer\ntry:\n    from qiskit.circuit.library import zz_feature_map          # Qiskit >= 2.1\n    def make_feature_map(n, reps=2): return zz_feature_map(n, reps=reps, entanglement='linear')\nexcept ImportError:\n    from qiskit.circuit.library import ZZFeatureMap\n    def make_feature_map(n, reps=2): return ZZFeatureMap(n, reps=reps, entanglement='linear')\n\nSEED = 42\nnp.random.seed(SEED); algorithm_globals.random_seed = SEED\n# Explicit pass manager: decomposes QAOA's Pauli-evolution gates (without it StatevectorSampler may stall in expm)\nPM = generate_preset_pass_manager(optimization_level=1, basis_gates=['rz', 'rx', 'h', 'cx', 'rzz', 'x', 'sx'])\n\nNAVY, BLUE, AMBER, TEAL, GRAY, LIGHT = '#182B5E', '#3D5FD0', '#D0820F', '#1F9A8A', '#7A879C', '#EEF4FB'\nplt.rcParams.update({'axes.spines.top': False, 'axes.spines.right': False, 'axes.titleweight': 'bold', 'figure.dpi': 110})\nFIG_DIR = 'kkb_figs'; os.makedirs(FIG_DIR, exist_ok=True)\ndef savefig(name):\n    path = os.path.join(FIG_DIR, name + '.png'); plt.savefig(path, dpi=170, bbox_inches='tight'); return path\n\ndef ks_stat(y, s):\n    fpr, tpr, _ = roc_curve(y, s); return float(np.max(tpr - fpr))\ndef score_metrics(y, s):\n    auc = roc_auc_score(y, s); return {'AUC': round(auc, 4), 'Gini': round(2 * auc - 1, 4), 'KS': round(ks_stat(y, s), 4)}\n\nprint(f'✓ Qiskit {qiskit.__version__} ready · seed {SEED}')"
  },
  {
   "cell_type": "markdown",
   "id": "f8120004",
   "metadata": {
    "id": "f8120004"
   },
   "source": "## Stage 2 — Credit data (public stand-in for bureau data)\nLoads UCI Statlog German Credit (via OpenML, then UCI mirror, then a synthetic fallback so the notebook always runs). Target: `default = 1` for a bad credit outcome (30% of applicants)."
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1a7de6a6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:41.226773Z",
     "iopub.status.busy": "2026-10-03T09:46:41.226355Z",
     "iopub.status.idle": "2026-10-03T09:46:41.325038Z",
     "shell.execute_reply": "2026-10-03T09:46:41.323611Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 312
    },
    "id": "1a7de6a6",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791021951225,
     "user_tz": -180,
     "elapsed": 645,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "c4e4af9b-d61e-4a06-d0e5-e8d38d5962e9"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Source: UCI Statlog German Credit (OpenML credit-g)\n",
      "Applicants: 1000 · attributes: 20 · default rate: 30.0%\n",
      "Train/test: 700/300 (stratified)\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "  checking_status  duration                  credit_history  \\\n",
       "0              <0         6  critical/other existing credit   \n",
       "1        0<=X<200        48                   existing paid   \n",
       "2     no checking        12  critical/other existing credit   \n",
       "3              <0        42                   existing paid   \n",
       "4              <0        24              delayed previously   \n",
       "\n",
       "               purpose  credit_amount    savings_status employment  \\\n",
       "0             radio/tv           1169  no known savings        >=7   \n",
       "1             radio/tv           5951              <100     1<=X<4   \n",
       "2            education           2096              <100     4<=X<7   \n",
       "3  furniture/equipment           7882              <100     4<=X<7   \n",
       "4              new car           4870              <100     1<=X<4   \n",
       "\n",
       "   installment_commitment     personal_status other_parties  residence_since  \\\n",
       "0                       4         male single          none                4   \n",
       "1                       2  female div/dep/mar          none                2   \n",
       "2                       2         male single          none                3   \n",
       "3                       2         male single     guarantor                4   \n",
       "4                       3         male single          none                4   \n",
       "\n",
       "  property_magnitude  age other_payment_plans   housing  existing_credits  \\\n",
       "0        real estate   67                none       own                 2   \n",
       "1        real estate   22                none       own                 1   \n",
       "2        real estate   49                none       own                 1   \n",
       "3     life insurance   45                none  for free                 1   \n",
       "4  no known property   53                none  for free                 2   \n",
       "\n",
       "                  job  num_dependents own_telephone foreign_worker  \n",
       "0             skilled               1           yes            yes  \n",
       "1             skilled               1          none            yes  \n",
       "2  unskilled resident               2          none            yes  \n",
       "3             skilled               2          none            yes  \n",
       "4             skilled               2          none            yes  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-3fc75a18-28c0-42a4-aa57-1e2d866def13\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>checking_status</th>\n",
       "      <th>duration</th>\n",
       "      <th>credit_history</th>\n",
       "      <th>purpose</th>\n",
       "      <th>credit_amount</th>\n",
       "      <th>savings_status</th>\n",
       "      <th>employment</th>\n",
       "      <th>installment_commitment</th>\n",
       "      <th>personal_status</th>\n",
       "      <th>other_parties</th>\n",
       "      <th>residence_since</th>\n",
       "      <th>property_magnitude</th>\n",
       "      <th>age</th>\n",
       "      <th>other_payment_plans</th>\n",
       "      <th>housing</th>\n",
       "      <th>existing_credits</th>\n",
       "      <th>job</th>\n",
       "      <th>num_dependents</th>\n",
       "      <th>own_telephone</th>\n",
       "      <th>foreign_worker</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>&lt;0</td>\n",
       "      <td>6</td>\n",
       "      <td>critical/other existing credit</td>\n",
       "      <td>radio/tv</td>\n",
       "      <td>1169</td>\n",
       "      <td>no known savings</td>\n",
       "      <td>&gt;=7</td>\n",
       "      <td>4</td>\n",
       "      <td>male single</td>\n",
       "      <td>none</td>\n",
       "      <td>4</td>\n",
       "      <td>real estate</td>\n",
       "      <td>67</td>\n",
       "      <td>none</td>\n",
       "      <td>own</td>\n",
       "      <td>2</td>\n",
       "      <td>skilled</td>\n",
       "      <td>1</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0&lt;=X&lt;200</td>\n",
       "      <td>48</td>\n",
       "      <td>existing paid</td>\n",
       "      <td>radio/tv</td>\n",
       "      <td>5951</td>\n",
       "      <td>&lt;100</td>\n",
       "      <td>1&lt;=X&lt;4</td>\n",
       "      <td>2</td>\n",
       "      <td>female div/dep/mar</td>\n",
       "      <td>none</td>\n",
       "      <td>2</td>\n",
       "      <td>real estate</td>\n",
       "      <td>22</td>\n",
       "      <td>none</td>\n",
       "      <td>own</td>\n",
       "      <td>1</td>\n",
       "      <td>skilled</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>no checking</td>\n",
       "      <td>12</td>\n",
       "      <td>critical/other existing credit</td>\n",
       "      <td>education</td>\n",
       "      <td>2096</td>\n",
       "      <td>&lt;100</td>\n",
       "      <td>4&lt;=X&lt;7</td>\n",
       "      <td>2</td>\n",
       "      <td>male single</td>\n",
       "      <td>none</td>\n",
       "      <td>3</td>\n",
       "      <td>real estate</td>\n",
       "      <td>49</td>\n",
       "      <td>none</td>\n",
       "      <td>own</td>\n",
       "      <td>1</td>\n",
       "      <td>unskilled resident</td>\n",
       "      <td>2</td>\n",
       "      <td>none</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>&lt;0</td>\n",
       "      <td>42</td>\n",
       "      <td>existing paid</td>\n",
       "      <td>furniture/equipment</td>\n",
       "      <td>7882</td>\n",
       "      <td>&lt;100</td>\n",
       "      <td>4&lt;=X&lt;7</td>\n",
       "      <td>2</td>\n",
       "      <td>male single</td>\n",
       "      <td>guarantor</td>\n",
       "      <td>4</td>\n",
       "      <td>life insurance</td>\n",
       "      <td>45</td>\n",
       "      <td>none</td>\n",
       "      <td>for free</td>\n",
       "      <td>1</td>\n",
       "      <td>skilled</td>\n",
       "      <td>2</td>\n",
       "      <td>none</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>&lt;0</td>\n",
       "      <td>24</td>\n",
       "      <td>delayed previously</td>\n",
       "      <td>new car</td>\n",
       "      <td>4870</td>\n",
       "      <td>&lt;100</td>\n",
       "      <td>1&lt;=X&lt;4</td>\n",
       "      <td>3</td>\n",
       "      <td>male single</td>\n",
       "      <td>none</td>\n",
       "      <td>4</td>\n",
       "      <td>no known property</td>\n",
       "      <td>53</td>\n",
       "      <td>none</td>\n",
       "      <td>for free</td>\n",
       "      <td>2</td>\n",
       "      <td>skilled</td>\n",
       "      <td>2</td>\n",
       "      <td>none</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3fc75a18-28c0-42a4-aa57-1e2d866def13')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-3fc75a18-28c0-42a4-aa57-1e2d866def13 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-3fc75a18-28c0-42a4-aa57-1e2d866def13');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "summary": "{\n  \"name\": \"display(raw_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"checking_status\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"<0\",\n          \"0<=X<200\",\n          \"no checking\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"duration\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 18,\n        \"min\": 6,\n        \"max\": 48,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          48,\n          24,\n          12\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"credit_history\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"critical/other existing credit\",\n          \"existing paid\",\n          \"delayed previously\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"purpose\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"education\",\n          \"new car\",\n          \"radio/tv\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"credit_amount\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2761,\n        \"min\": 1169,\n        \"max\": 7882,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          5951,\n          4870,\n          2096\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"savings_status\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"<100\",\n          \"no known savings\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"employment\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \">=7\",\n          \"1<=X<4\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"installment_commitment\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 2,\n        \"max\": 4,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          4,\n          2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"personal_status\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"female div/dep/mar\",\n          \"male single\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"other_parties\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"guarantor\",\n          \"none\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"residence_since\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 2,\n        \"max\": 4,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          4,\n          2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"property_magnitude\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"real estate\",\n          \"life insurance\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"age\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 16,\n        \"min\": 22,\n        \"max\": 67,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          22,\n          53\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"other_payment_plans\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"none\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"housing\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"for free\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"existing_credits\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 1,\n        \"max\": 2,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"job\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"unskilled resident\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"num_dependents\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 1,\n        \"max\": 2,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"own_telephone\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"none\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"foreign_worker\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"yes\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    }
   ],
   "source": "UCI_COLS = ['checking_status', 'duration', 'credit_history', 'purpose', 'credit_amount', 'savings_status', 'employment',\n            'installment_commitment', 'personal_status', 'other_parties', 'residence_since', 'property_magnitude', 'age',\n            'other_payment_plans', 'housing', 'existing_credits', 'job', 'num_dependents', 'own_telephone', 'foreign_worker']\n\ndef load_credit_data():\n    try:\n        from sklearn.datasets import fetch_openml\n        d = fetch_openml('credit-g', version=1, as_frame=True)\n        df = d.frame.copy(); y = (df.pop('class').astype(str) == 'bad').astype(int)\n        return df, y, 'UCI Statlog German Credit (OpenML credit-g)'\n    except Exception as e:\n        print('OpenML unavailable:', e)\n    try:\n        raw = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/statlog/german/german.data', sep=' ', header=None)\n        df = raw.iloc[:, :20]; df.columns = UCI_COLS; y = (raw.iloc[:, 20] == 2).astype(int)\n        return df, y, 'UCI Statlog German Credit (UCI mirror)'\n    except Exception as e:\n        print('UCI mirror unavailable:', e)\n    from sklearn.datasets import make_classification\n    X, y = make_classification(n_samples=1000, n_features=20, n_informative=8, n_redundant=6, weights=[0.7], random_state=SEED)\n    return pd.DataFrame(X, columns=UCI_COLS), pd.Series(y), 'Synthetic fallback (same shape as German Credit)'\n\nraw_df, y, DATA_SOURCE = load_credit_data()\nX_tr_raw, X_te_raw, y_tr, y_te = train_test_split(raw_df, y, test_size=0.30, stratify=y, random_state=SEED)\ny_tr, y_te = y_tr.reset_index(drop=True), y_te.reset_index(drop=True)\nX_tr_raw, X_te_raw = X_tr_raw.reset_index(drop=True), X_te_raw.reset_index(drop=True)\nprint(f'Source: {DATA_SOURCE}')\nprint(f'Applicants: {len(raw_df)} · attributes: {raw_df.shape[1]} · default rate: {y.mean():.1%}')\nprint(f'Train/test: {len(y_tr)}/{len(y_te)} (stratified)')\ndisplay(raw_df.head())"
  },
  {
   "cell_type": "markdown",
   "id": "c5cffa09",
   "metadata": {
    "id": "c5cffa09"
   },
   "source": "## Stage 3 — Credit-bureau feature engineering: Weight of Evidence (WOE) and Information Value (IV)\nScorecards at credit bureaus are typically built on WOE-encoded attributes. WOE and IV are fitted on the training set only."
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "76fd94bf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:41.327801Z",
     "iopub.status.busy": "2026-10-03T09:46:41.327471Z",
     "iopub.status.idle": "2026-10-03T09:46:42.378535Z",
     "shell.execute_reply": "2026-10-03T09:46:42.377104Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 817
    },
    "id": "76fd94bf",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791021952430,
     "user_tz": -180,
     "elapsed": 1055,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "d8795fb4-5155-4813-8ccd-f3c727d1b9eb"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 990x462 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "checking_status           0.630\n",
      "credit_history            0.252\n",
      "savings_status            0.221\n",
      "duration                  0.184\n",
      "property_magnitude        0.139\n",
      "employment                0.139\n",
      "purpose                   0.129\n",
      "other_payment_plans       0.118\n",
      "housing                   0.103\n",
      "foreign_worker            0.066\n",
      "credit_amount             0.064\n",
      "age                       0.047\n",
      "personal_status           0.039\n",
      "installment_commitment    0.033\n",
      "other_parties             0.026\n",
      "job                       0.024\n",
      "existing_credits          0.004\n",
      "own_telephone             0.003\n",
      "residence_since           0.002\n",
      "num_dependents            0.000\n"
     ]
    }
   ],
   "source": "def woe_fit(df, y, num_bins=5, eps=0.5):\n    maps = {}\n    for c in df.columns:\n        s = df[c]\n        if pd.api.types.is_numeric_dtype(s) and s.nunique() > num_bins:\n            edges = np.unique(np.quantile(s.astype(float), np.linspace(0, 1, num_bins + 1)))\n            edges[0], edges[-1] = -np.inf, np.inf\n            b = pd.cut(s.astype(float), edges, include_lowest=True).astype(str); kind = ('num', edges)\n        else:\n            b = s.astype(str); kind = ('cat', None)\n        tab = pd.crosstab(b, y)\n        good = tab.get(0, pd.Series(0, index=tab.index)) + eps; bad = tab.get(1, pd.Series(0, index=tab.index)) + eps\n        pg, pb = good / good.sum(), bad / bad.sum()\n        woe = np.log(pg / pb)\n        maps[c] = {'kind': kind, 'woe': woe.to_dict(), 'iv': float(((pg - pb) * woe).sum())}\n    return maps\n\ndef woe_transform(df, maps):\n    out = pd.DataFrame(index=df.index)\n    for c, m in maps.items():\n        kind, edges = m['kind']\n        b = pd.cut(df[c].astype(float), edges, include_lowest=True).astype(str) if kind == 'num' else df[c].astype(str)\n        out[c] = b.map(m['woe']).fillna(0.0).astype(float)\n    return out\n\nWOE = woe_fit(X_tr_raw, y_tr)\nX_tr, X_te = woe_transform(X_tr_raw, WOE), woe_transform(X_te_raw, WOE)\nIV = pd.Series({c: m['iv'] for c, m in WOE.items()}).sort_values(ascending=False)\n\nfig, ax = plt.subplots(figsize=(9, 4.2))\ncols = [NAVY if v >= 0.1 else (BLUE if v >= 0.02 else GRAY) for v in IV.values]\nax.barh(IV.index[::-1], IV.values[::-1], color=cols[::-1])\nax.axvline(0.1, color=AMBER, ls='--', lw=1); ax.text(0.102, 0.3, 'medium predictive power (IV ≥ 0.10)', color=AMBER, fontsize=8)\nax.set_xlabel('Information Value (train)'); ax.set_title('Attribute predictive power — Information Value')\nplt.tight_layout(); FIG_IV = savefig('01_information_value'); plt.show()\nprint(IV.round(3).to_string())"
  },
  {
   "cell_type": "markdown",
   "id": "8d3e0eb2",
   "metadata": {
    "id": "8d3e0eb2"
   },
   "source": "## Stage 4 — Classical baseline (the bar every quantum method must clear)\nLogistic-regression scorecard (industry standard), RBF-SVM and gradient boosting on all 20 WOE attributes. Metrics: AUC, Gini (= 2·AUC − 1) and KS, on the held-out test set plus 5-fold cross-validated AUC on the training set."
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0ecd42c6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:42.381030Z",
     "iopub.status.busy": "2026-10-03T09:46:42.380756Z",
     "iopub.status.idle": "2026-10-03T09:46:43.731468Z",
     "shell.execute_reply": "2026-10-03T09:46:43.729430Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 142
    },
    "id": "0ecd42c6",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791021954026,
     "user_tz": -180,
     "elapsed": 1399,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "3da26f7e-f352-4eb5-b2af-bff38f5359d9"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "         Feature set               Model  CV AUC    CV ±     AUC    Gini  \\\n",
       "0  All 20 attributes  Logistic scorecard  0.8016  0.0294  0.8013  0.6025   \n",
       "1  All 20 attributes             RBF-SVM  0.7892  0.0373  0.7733  0.5467   \n",
       "2  All 20 attributes   Gradient boosting  0.7624  0.0311  0.7851  0.5703   \n",
       "\n",
       "       KS  \n",
       "0  0.5222  \n",
       "1  0.4444  \n",
       "2  0.4476  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-9f3dc6a7-ed23-459b-a33c-5cd14d14585c\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Feature set</th>\n",
       "      <th>Model</th>\n",
       "      <th>CV AUC</th>\n",
       "      <th>CV ±</th>\n",
       "      <th>AUC</th>\n",
       "      <th>Gini</th>\n",
       "      <th>KS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>All 20 attributes</td>\n",
       "      <td>Logistic scorecard</td>\n",
       "      <td>0.8016</td>\n",
       "      <td>0.0294</td>\n",
       "      <td>0.8013</td>\n",
       "      <td>0.6025</td>\n",
       "      <td>0.5222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>All 20 attributes</td>\n",
       "      <td>RBF-SVM</td>\n",
       "      <td>0.7892</td>\n",
       "      <td>0.0373</td>\n",
       "      <td>0.7733</td>\n",
       "      <td>0.5467</td>\n",
       "      <td>0.4444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>All 20 attributes</td>\n",
       "      <td>Gradient boosting</td>\n",
       "      <td>0.7624</td>\n",
       "      <td>0.0311</td>\n",
       "      <td>0.7851</td>\n",
       "      <td>0.5703</td>\n",
       "      <td>0.4476</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9f3dc6a7-ed23-459b-a33c-5cd14d14585c')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-9f3dc6a7-ed23-459b-a33c-5cd14d14585c button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-9f3dc6a7-ed23-459b-a33c-5cd14d14585c');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_43af1ff5-f519-449a-a5bb-a04d112b9d0d\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('baseline')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_43af1ff5-f519-449a-a5bb-a04d112b9d0d button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('baseline');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "baseline",
       "summary": "{\n  \"name\": \"baseline\",\n  \"rows\": 3,\n  \"fields\": [\n    {\n      \"column\": \"Feature set\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"All 20 attributes\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Model\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Logistic scorecard\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CV AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.02003596765818912,\n        \"min\": 0.7624,\n        \"max\": 0.8016,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.8016\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CV \\u00b1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.004158124577258358,\n        \"min\": 0.0294,\n        \"max\": 0.0373,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.0294\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.01405750096330545,\n        \"min\": 0.7733,\n        \"max\": 0.8013,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.8013\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Gini\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.02801023622416161,\n        \"min\": 0.5467,\n        \"max\": 0.6025,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.6025\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"KS\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.044023175714616496,\n        \"min\": 0.4444,\n        \"max\": 0.5222,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.5222\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    }
   ],
   "source": "CV = StratifiedKFold(n_splits=5, shuffle=True, random_state=SEED)\ndef classical_models():\n    return {\n        'Logistic scorecard': LogisticRegression(C=1.0, max_iter=2000),\n        'RBF-SVM': make_pipeline(StandardScaler(), SVC(C=1.0, gamma='scale', class_weight='balanced')),\n        'Gradient boosting': HistGradientBoostingClassifier(max_depth=3, learning_rate=0.05, max_iter=200, random_state=SEED),\n    }\ndef scores_of(model, X):\n    return model.decision_function(X) if hasattr(model, 'decision_function') else model.predict_proba(X)[:, 1]\n\ndef evaluate(models, Xtr, Xte, label):\n    rows = []\n    for name, m in models.items():\n        cv = cross_val_score(m, Xtr, y_tr, cv=CV, scoring='roc_auc')\n        m.fit(Xtr, y_tr); met = score_metrics(y_te, scores_of(m, Xte))\n        rows.append({'Feature set': label, 'Model': name, 'CV AUC': round(cv.mean(), 4), 'CV ±': round(cv.std(), 4), **met})\n    return pd.DataFrame(rows)\n\nbaseline = evaluate(classical_models(), X_tr.values, X_te.values, 'All 20 attributes')\ndisplay(baseline)"
  },
  {
   "cell_type": "markdown",
   "id": "0289e260",
   "metadata": {
    "id": "0289e260"
   },
   "source": "---\n# Track 1 — Quantum feature selection for a compact scorecard\n\n## Stage 5 — QUBO formulation\nSelect exactly *k* attributes that maximise predictive power (IV) while penalising redundancy (pairwise |correlation|):\n\n$$E(x) = -\\sum_i r_i x_i \\;+\\; \\alpha \\sum_{i<j} |\\rho_{ij}|\\, x_i x_j \\;+\\; P\\Big(\\sum_i x_i - k\\Big)^2, \\qquad x_i \\in \\{0,1\\}$$\n\nA classical pre-screen keeps the 14 strongest attributes → **14 qubits**, small enough for exact verification and Colab simulation.\n\nTwo quantum formulations are compared: a **standard QAOA** (penalty term + X mixer) and a **constraint-preserving QAOA** (Hadfield et al., *Quantum Alternating Operator Ansatz*, 2019) whose XY mixer only moves between valid *k*-subsets, so no penalty is needed."
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "68c6281e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:43.734612Z",
     "iopub.status.busy": "2026-10-03T09:46:43.734339Z",
     "iopub.status.idle": "2026-10-03T09:46:43.750803Z",
     "shell.execute_reply": "2026-10-03T09:46:43.749388Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "68c6281e",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791021956870,
     "user_tz": -180,
     "elapsed": 2648,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "1596bc51-a6ff-4370-aaa7-4db6b9717f3c"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "QUBO: 14 binary variables (= qubits), 91 couplings, select k = 6, α = 0.6, P = 9.2\n",
      "Candidates: ['checking_status', 'credit_history', 'savings_status', 'duration', 'property_magnitude', 'employment', 'purpose', 'other_payment_plans', 'housing', 'foreign_worker', 'credit_amount', 'age', 'personal_status', 'installment_commitment']\n"
     ]
    }
   ],
   "source": "N_CAND, K_SELECT, ALPHA = 14, 6, 0.6\nCAND = list(IV.index[:N_CAND])\nr = (IV[CAND] / IV[CAND].max()).values                      # normalised relevance\nrho = np.abs(np.corrcoef(X_tr[CAND].values, rowvar=False))   # redundancy\nP_PEN = 2.0 * (1 + ALPHA * K_SELECT)                         # penalty > max objective swing\n\ndef qubo_matrix(r, rho, alpha, k, P):\n    n = len(r); Q = np.zeros((n, n))\n    for i in range(n):\n        Q[i, i] = -r[i] + P * (1 - 2 * k)\n        for j in range(i + 1, n):\n            Q[i, j] = alpha * rho[i, j] + 2 * P            # upper-triangular pair coefficients\n    return Q, P * k * k\n\nQ, Q_CONST = qubo_matrix(r, rho, ALPHA, K_SELECT, P_PEN)\ndef energy(x): x = np.asarray(x, float); return float(x @ np.triu(Q, 1) @ x + np.diag(Q) @ x + Q_CONST)\ndef objective(x):  # interpretable part only (no penalty)\n    x = np.asarray(x, float); return float(-r @ x + ALPHA * (x @ np.triu(rho, 1) @ x))\n\nqp = QuadraticProgram('KKB_feature_selection')\nfor i in range(N_CAND): qp.binary_var(f'x{i}')\nqp.minimize(constant=Q_CONST, linear={f'x{i}': float(Q[i, i]) for i in range(N_CAND)},\n            quadratic={(f'x{i}', f'x{j}'): float(Q[i, j]) for i in range(N_CAND) for j in range(i + 1, N_CAND)})\nprint(f'QUBO: {N_CAND} binary variables (= qubits), {N_CAND*(N_CAND-1)//2} couplings, select k = {K_SELECT}, α = {ALPHA}, P = {P_PEN}')\nprint('Candidates:', CAND)"
  },
  {
   "cell_type": "markdown",
   "id": "f9618cbd",
   "metadata": {
    "id": "f9618cbd"
   },
   "source": "## Stage 6 — Solve: exact enumeration, classical annealing, standard QAOA, constraint-preserving QAOA + classical local search"
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "07187d8b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:46:43.754194Z",
     "iopub.status.busy": "2026-10-03T09:46:43.753832Z",
     "iopub.status.idle": "2026-10-03T09:47:15.746363Z",
     "shell.execute_reply": "2026-10-03T09:47:15.745142Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 694
    },
    "id": "07187d8b",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022028794,
     "user_tz": -180,
     "elapsed": 71568,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "1a0d3782-9c4b-405a-f9fb-2469115ccb39"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                                            Solver Valid-sample prob.  \\\n",
       "0                Exact enumeration (16,384 states)                  —   \n",
       "1                    Classical simulated annealing                  —   \n",
       "2               Random valid selection (reference)    0.183 (uniform)   \n",
       "3           Standard QAOA (penalty + X mixer, p=1)              0.118   \n",
       "4       Constraint-preserving QAOA (XY mixer, p=3)              1.000   \n",
       "5  Hybrid: XY-QAOA top-32 + classical local search              1.000   \n",
       "\n",
       "  Expected objective          Best found Gap to exact % Time s  \n",
       "0                  —             -1.6834            0.0  0.013  \n",
       "1                  —             -1.6834            0.0  0.483  \n",
       "2             -0.765                   —          54.56      —  \n",
       "3            -0.7634  no valid in top-32              —   10.3  \n",
       "4            -1.4447             -1.5644           7.07   68.7  \n",
       "5                  —             -1.6834            0.0   68.7  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-906b0902-f509-45d8-a84a-30287479647d\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Solver</th>\n",
       "      <th>Valid-sample prob.</th>\n",
       "      <th>Expected objective</th>\n",
       "      <th>Best found</th>\n",
       "      <th>Gap to exact %</th>\n",
       "      <th>Time s</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Exact enumeration (16,384 states)</td>\n",
       "      <td>—</td>\n",
       "      <td>—</td>\n",
       "      <td>-1.6834</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Classical simulated annealing</td>\n",
       "      <td>—</td>\n",
       "      <td>—</td>\n",
       "      <td>-1.6834</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Random valid selection (reference)</td>\n",
       "      <td>0.183 (uniform)</td>\n",
       "      <td>-0.765</td>\n",
       "      <td>—</td>\n",
       "      <td>54.56</td>\n",
       "      <td>—</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Standard QAOA (penalty + X mixer, p=1)</td>\n",
       "      <td>0.118</td>\n",
       "      <td>-0.7634</td>\n",
       "      <td>no valid in top-32</td>\n",
       "      <td>—</td>\n",
       "      <td>10.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Constraint-preserving QAOA (XY mixer, p=3)</td>\n",
       "      <td>1.000</td>\n",
       "      <td>-1.4447</td>\n",
       "      <td>-1.5644</td>\n",
       "      <td>7.07</td>\n",
       "      <td>68.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Hybrid: XY-QAOA top-32 + classical local search</td>\n",
       "      <td>1.000</td>\n",
       "      <td>—</td>\n",
       "      <td>-1.6834</td>\n",
       "      <td>0.0</td>\n",
       "      <td>68.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-906b0902-f509-45d8-a84a-30287479647d')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-906b0902-f509-45d8-a84a-30287479647d button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-906b0902-f509-45d8-a84a-30287479647d');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_96dfbee4-3fce-41d6-8812-61144f475eb6\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('FS')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_96dfbee4-3fce-41d6-8812-61144f475eb6 button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('FS');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "FS",
       "summary": "{\n  \"name\": \"FS\",\n  \"rows\": 6,\n  \"fields\": [\n    {\n      \"column\": \"Solver\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"Exact enumeration (16,384 states)\",\n          \"Classical simulated annealing\",\n          \"Hybrid: XY-QAOA top-32 + classical local search\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Valid-sample prob.\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"0.183 (uniform)\",\n          \"1.000\",\n          \"\\u2014\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Expected objective\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          -0.765,\n          -1.4447,\n          \"\\u2014\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Best found\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"\\u2014\",\n          -1.5644,\n          -1.6834\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Gap to exact %\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          54.56,\n          7.07,\n          0.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Time s\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.483,\n          68.7,\n          \"\\u2014\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Constraint-preserving QAOA: 100% valid scorecards; expected quality 74% of the way from random to optimal; best of 32 most-likely outcomes within 7.07% of optimum; + local search → 0.0%.\n",
      "Standard penalty QAOA places 12% of its probability on valid selections (random sampling: 18%).\n",
      "Hybrid selection: ['checking_status', 'credit_history', 'savings_status', 'duration', 'purpose', 'housing']\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 1320x396 with 2 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    }
   ],
   "source": "from qiskit.quantum_info import SparsePauliOp\nfrom qiskit import QuantumCircuit\n\n# (a) Exact: enumerate all 2^14 = 16,384 selections (ground truth at this PoC size)\nt0 = time.time()\nBITS = ((np.arange(2 ** N_CAND)[:, None] >> np.arange(N_CAND)) & 1).astype(float)\nOBJ = -(BITS @ r) + ALPHA * ((BITS @ np.triu(rho, 1)) * BITS).sum(1)\nFEAS = BITS.sum(1) == K_SELECT\nI_OPT = int(np.where(FEAS)[0][np.argmin(OBJ[FEAS])]); x_exact = BITS[I_OPT]; t_exact = time.time() - t0\nE_UNIFORM = float(OBJ[FEAS].mean())                       # average quality of a random valid selection\nquality = lambda e: 100 * (E_UNIFORM - e) / (E_UNIFORM - OBJ[I_OPT])   # % of the way from random to optimal\ngap = lambda e: round(100 * (e - OBJ[I_OPT]) / abs(OBJ[I_OPT]), 2)\n\n# (b) Classical simulated annealing over valid k-subsets (swap moves)\ndef anneal(n_iter=4000, T0=0.5, seed=SEED):\n    rng = np.random.default_rng(seed); x = np.zeros(N_CAND); x[rng.choice(N_CAND, K_SELECT, replace=False)] = 1\n    best, best_e, e = x.copy(), objective(x), objective(x)\n    for t in range(n_iter):\n        T = T0 * (1 - t / n_iter) + 1e-4\n        i, j = rng.choice(np.where(x == 1)[0]), rng.choice(np.where(x == 0)[0])\n        x2 = x.copy(); x2[i], x2[j] = 0, 1; e2 = objective(x2)\n        if e2 < e or rng.random() < math.exp(-(e2 - e) / T):\n            x, e = x2, e2\n            if e < best_e: best, best_e = x.copy(), e\n    return best\nt0 = time.time(); x_sa = anneal(); t_sa = time.time() - t0\n\ndef output_distribution(res):\n    # exact output distribution of the optimised QAOA circuit (statevector, no shot noise)\n    circ = res.optimal_circuit.assign_parameters(res.optimal_parameters).remove_final_measurements(inplace=False)\n    return Statevector(circ).probabilities()\ndef best_of_top(pr, n=32):\n    top = [i for i in np.argsort(-pr)[:n] if FEAS[i]]\n    return min(top, key=lambda i: OBJ[i]) if top else None\n\n# (c) Standard QAOA: penalty QUBO + X mixer (p = 1); coefficients normalised for a well-conditioned landscape\nop_pen, _ = qp.to_ising(); op_pen = op_pen / np.abs(op_pen.coeffs).max()\nalgorithm_globals.random_seed = SEED; t0 = time.time()\nres_std = QAOA(sampler=StatevectorSampler(seed=SEED), optimizer=COBYLA(maxiter=200), reps=1, transpiler=PM).compute_minimum_eigenvalue(op_pen)\npr_std = output_distribution(res_std); t_std = time.time() - t0\n\n# (d) Constraint-preserving QAOA: objective only + XY ring mixer + neutral valid start (every other candidate)\nqp_obj = QuadraticProgram('objective_only')\nfor i in range(N_CAND): qp_obj.binary_var(f'x{i}')\nqp_obj.minimize(linear={f'x{i}': float(-r[i]) for i in range(N_CAND)},\n                quadratic={(f'x{i}', f'x{j}'): float(ALPHA * rho[i, j]) for i in range(N_CAND) for j in range(i + 1, N_CAND)})\nop_obj, _ = qp_obj.to_ising(); op_obj = op_obj / np.abs(op_obj.coeffs).max()\ndef xy_ring_mixer(n):\n    terms = []\n    for i in range(n):\n        j = (i + 1) % n\n        for P in ('XX', 'YY'):\n            lab = ['I'] * n; lab[i], lab[j] = P[0], P[1]; terms.append((''.join(reversed(lab)), 0.5))\n    return SparsePauliOp.from_list(terms)\nSTART = list(range(0, 2 * K_SELECT, 2))                   # neutral valid start, not informed by the IV ranking\ninit = QuantumCircuit(N_CAND)\nfor i in START: init.x(i)\nQAOA_REPS = 3; history = []\nalgorithm_globals.random_seed = SEED; t0 = time.time()\nres_xy = QAOA(sampler=StatevectorSampler(seed=SEED), optimizer=COBYLA(maxiter=300), reps=QAOA_REPS, transpiler=PM,\n              mixer=xy_ring_mixer(N_CAND), initial_state=init,\n              callback=lambda n, p, v, m: history.append(float(v))).compute_minimum_eigenvalue(op_obj)\npr_xy = output_distribution(res_xy); t_qaoa = time.time() - t0\n\n# (e) Hybrid: best of the 32 most likely QAOA outcomes + classical 1-swap local search\ndef local_search(x):\n    x = np.array(x, float); improved = True\n    while improved:\n        improved = False\n        for i in np.where(x == 1)[0]:\n            for j in np.where(x == 0)[0]:\n                x2 = x.copy(); x2[i], x2[j] = 0, 1\n                if objective(x2) < objective(x) - 1e-12: x, improved = x2, True; break\n            if improved: break\n    return x\ni_top_xy = best_of_top(pr_xy); i_top_std = best_of_top(pr_std)\nx_hybrid = local_search(BITS[i_top_xy])\n\nP_FEAS_STD, P_FEAS_XY = float(pr_std[FEAS].sum()), float(pr_xy[FEAS].sum())\nE_STD = float((pr_std[FEAS] * OBJ[FEAS]).sum() / max(P_FEAS_STD, 1e-12)); E_XY = float((pr_xy * OBJ).sum())\nsel = lambda x: [CAND[i] for i in range(N_CAND) if x[i] > 0.5]\nFS = pd.DataFrame([\n    {'Solver': 'Exact enumeration (16,384 states)', 'Valid-sample prob.': '—', 'Expected objective': '—', 'Best found': round(OBJ[I_OPT], 4), 'Gap to exact %': 0.0, 'Time s': round(t_exact, 3)},\n    {'Solver': 'Classical simulated annealing', 'Valid-sample prob.': '—', 'Expected objective': '—', 'Best found': round(objective(x_sa), 4), 'Gap to exact %': gap(objective(x_sa)), 'Time s': round(t_sa, 3)},\n    {'Solver': 'Random valid selection (reference)', 'Valid-sample prob.': f'{FEAS.mean():.3f} (uniform)', 'Expected objective': round(E_UNIFORM, 4), 'Best found': '—', 'Gap to exact %': gap(E_UNIFORM), 'Time s': '—'},\n    {'Solver': 'Standard QAOA (penalty + X mixer, p=1)', 'Valid-sample prob.': f'{P_FEAS_STD:.3f}', 'Expected objective': round(E_STD, 4), 'Best found': round(OBJ[i_top_std], 4) if i_top_std is not None else 'no valid in top-32', 'Gap to exact %': gap(OBJ[i_top_std]) if i_top_std is not None else '—', 'Time s': round(t_std, 1)},\n    {'Solver': f'Constraint-preserving QAOA (XY mixer, p={QAOA_REPS})', 'Valid-sample prob.': f'{P_FEAS_XY:.3f}', 'Expected objective': round(E_XY, 4), 'Best found': round(OBJ[i_top_xy], 4), 'Gap to exact %': gap(OBJ[i_top_xy]), 'Time s': round(t_qaoa, 1)},\n    {'Solver': 'Hybrid: XY-QAOA top-32 + classical local search', 'Valid-sample prob.': '1.000', 'Expected objective': '—', 'Best found': round(objective(x_hybrid), 4), 'Gap to exact %': gap(objective(x_hybrid)), 'Time s': round(t_qaoa, 1)},\n])\ndisplay(FS)\nQUALITY_XY = quality(E_XY)\nprint(f'Constraint-preserving QAOA: {P_FEAS_XY:.0%} valid scorecards; expected quality {QUALITY_XY:.0f}% of the way from random to optimal; '\n      f'best of 32 most-likely outcomes within {gap(OBJ[i_top_xy])}% of optimum; + local search → {gap(objective(x_hybrid))}%.')\nprint(f'Standard penalty QAOA places {P_FEAS_STD:.0%} of its probability on valid selections (random sampling: {FEAS.mean():.0%}).')\nprint('Hybrid selection:', sel(x_hybrid))\n\nfig, axes = plt.subplots(1, 2, figsize=(12, 3.6), gridspec_kw={'width_ratios': [1.4, 1]})\nbins = np.linspace(OBJ[FEAS].min(), OBJ[FEAS].max(), 30)\naxes[0].hist(OBJ[FEAS], bins=bins, weights=np.full(FEAS.sum(), 1 / FEAS.sum()), color=GRAY, alpha=0.7, label='Random valid selection')\naxes[0].hist(OBJ, bins=bins, weights=pr_xy, color=BLUE, alpha=0.75, label=f'Constraint-preserving QAOA (p={QAOA_REPS})')\naxes[0].axvline(OBJ[I_OPT], color=AMBER, ls='--', lw=2, label='Exact optimum')\naxes[0].set_xlabel('Scorecard objective (lower = better)'); axes[0].set_ylabel('Probability'); axes[0].legend(frameon=False, fontsize=8)\naxes[0].set_title('Quantum sampling concentrates on better scorecards')\nh = np.minimum.accumulate(history); axes[1].plot(range(1, len(h) + 1), h, color=BLUE, lw=2)\naxes[1].set_xlabel('COBYLA iteration'); axes[1].set_ylabel('Expected energy (best so far)'); axes[1].set_title(f'QAOA convergence ({N_CAND} qubits)')\naxes[1].grid(axis='y', color='#E3E8F0'); plt.tight_layout(); FIG_QAOA = savefig('02_qaoa_distribution'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "e580ec2d",
   "metadata": {
    "id": "e580ec2d"
   },
   "source": "## Stage 7 — Does the quantum-selected scorecard hold up?\nCompare the logistic scorecard on: all 20 attributes · the 6 QUBO-selected attributes · the naive top-6 by IV (redundancy-blind) · 6 random attributes (mean of 30 draws)."
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bec0da0c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:15.750889Z",
     "iopub.status.busy": "2026-10-03T09:47:15.750504Z",
     "iopub.status.idle": "2026-10-03T09:47:17.484181Z",
     "shell.execute_reply": "2026-10-03T09:47:17.480304Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 611
    },
    "id": "bec0da0c",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022034724,
     "user_tz": -180,
     "elapsed": 5923,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "69e20787-b945-4a16-f7d0-e9bf3537c335"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                          Scorecard  # attributes  CV AUC     AUC    Gini  \\\n",
       "0                 All 20 attributes            20  0.8016  0.8013  0.6025   \n",
       "1  QUBO-selected 6 (quantum hybrid)             6  0.7816  0.7940  0.5881   \n",
       "2               Top-6 by IV (naive)             6  0.7836  0.7768  0.5535   \n",
       "3             Random 6 (mean of 30)             6  0.6957  0.6822  0.3643   \n",
       "\n",
       "       KS  \n",
       "0  0.5222  \n",
       "1  0.4984  \n",
       "2  0.4635  \n",
       "3  0.3114  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-9015876c-7ecb-4696-9389-58b0568ae15b\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Scorecard</th>\n",
       "      <th># attributes</th>\n",
       "      <th>CV AUC</th>\n",
       "      <th>AUC</th>\n",
       "      <th>Gini</th>\n",
       "      <th>KS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>All 20 attributes</td>\n",
       "      <td>20</td>\n",
       "      <td>0.8016</td>\n",
       "      <td>0.8013</td>\n",
       "      <td>0.6025</td>\n",
       "      <td>0.5222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>QUBO-selected 6 (quantum hybrid)</td>\n",
       "      <td>6</td>\n",
       "      <td>0.7816</td>\n",
       "      <td>0.7940</td>\n",
       "      <td>0.5881</td>\n",
       "      <td>0.4984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Top-6 by IV (naive)</td>\n",
       "      <td>6</td>\n",
       "      <td>0.7836</td>\n",
       "      <td>0.7768</td>\n",
       "      <td>0.5535</td>\n",
       "      <td>0.4635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Random 6 (mean of 30)</td>\n",
       "      <td>6</td>\n",
       "      <td>0.6957</td>\n",
       "      <td>0.6822</td>\n",
       "      <td>0.3643</td>\n",
       "      <td>0.3114</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9015876c-7ecb-4696-9389-58b0568ae15b')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-9015876c-7ecb-4696-9389-58b0568ae15b button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-9015876c-7ecb-4696-9389-58b0568ae15b');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_fa110776-80d2-442e-856a-ff44d95fd8f0\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('SC')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_fa110776-80d2-442e-856a-ff44d95fd8f0 button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('SC');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "SC",
       "summary": "{\n  \"name\": \"SC\",\n  \"rows\": 4,\n  \"fields\": [\n    {\n      \"column\": \"Scorecard\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"QUBO-selected 6 (quantum hybrid)\",\n          \"Random 6 (mean of 30)\",\n          \"All 20 attributes\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"# attributes\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 7,\n        \"min\": 6,\n        \"max\": 20,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          6,\n          20\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CV AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.04747633621079031,\n        \"min\": 0.6957,\n        \"max\": 0.8016,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.7816,\n          0.6957\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.05521366829569166,\n        \"min\": 0.6822,\n        \"max\": 0.8013,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.794,\n          0.6822\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Gini\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.11046411181917862,\n        \"min\": 0.3643,\n        \"max\": 0.6025,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.5881,\n          0.3643\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"KS\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.09476734230032341,\n        \"min\": 0.3114,\n        \"max\": 0.5222,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.4984,\n          0.3114\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Mean pairwise |corr| — QUBO set: 0.081 · naive top-IV set: 0.094\n",
      "QUBO-selected: ['checking_status', 'credit_history', 'savings_status', 'duration', 'purpose', 'housing']\n",
      "Naive top-IV : ['checking_status', 'credit_history', 'savings_status', 'duration', 'property_magnitude', 'employment']\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 880x396 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    }
   ],
   "source": "SELECTED = sel(x_hybrid)\nTOP_IV = list(IV.index[:K_SELECT])\ndef lr_eval(features):\n    m = LogisticRegression(C=1.0, max_iter=2000)\n    cv = cross_val_score(m, X_tr[features].values, y_tr, cv=CV, scoring='roc_auc').mean()\n    m.fit(X_tr[features].values, y_tr); return cv, score_metrics(y_te, m.decision_function(X_te[features].values))\n\nrows = []\nfor label, f in [('All 20 attributes', list(X_tr.columns)), (f'QUBO-selected {K_SELECT} (quantum hybrid)', SELECTED), (f'Top-{K_SELECT} by IV (naive)', TOP_IV)]:\n    cv, met = lr_eval(f); rows.append({'Scorecard': label, '# attributes': len(f), 'CV AUC': round(cv, 4), **met})\nrng = np.random.default_rng(SEED); rnd = [lr_eval(list(rng.choice(X_tr.columns, K_SELECT, replace=False))) for _ in range(30)]\nrows.append({'Scorecard': f'Random {K_SELECT} (mean of 30)', '# attributes': K_SELECT, 'CV AUC': round(np.mean([c for c, _ in rnd]), 4),\n             'AUC': round(np.mean([m['AUC'] for _, m in rnd]), 4), 'Gini': round(np.mean([m['Gini'] for _, m in rnd]), 4), 'KS': round(np.mean([m['KS'] for _, m in rnd]), 4)})\nSC = pd.DataFrame(rows); display(SC)\nmean_red = lambda f: np.mean([abs(np.corrcoef(X_tr[a], X_tr[b])[0, 1]) for a, b in itertools.combinations(f, 2)])\nprint(f'Mean pairwise |corr| — QUBO set: {mean_red(SELECTED):.3f} · naive top-IV set: {mean_red(TOP_IV):.3f}')\nprint('QUBO-selected:', SELECTED); print('Naive top-IV :', TOP_IV)\n\nfig, ax = plt.subplots(figsize=(8, 3.6))\nbars = ax.bar(['All 20', f'QUBO {K_SELECT}\\n(quantum hybrid)', f'Top-{K_SELECT} IV\\n(naive)', f'Random {K_SELECT}'], SC['Gini'], color=[GRAY, BLUE, NAVY, '#C5CCD9'], width=0.55)\nfor b, v in zip(bars, SC['Gini']): ax.text(b.get_x() + b.get_width() / 2, v + 0.01, f'{v:.3f}', ha='center', fontweight='bold')\nax.set_ylabel('Test Gini'); ax.set_ylim(0, max(SC['Gini']) * 1.2); ax.set_title(f'Scorecard quality with {K_SELECT} of 20 attributes')\nax.grid(axis='y', color='#E3E8F0'); ax.set_axisbelow(True); plt.tight_layout(); FIG_SC = savefig('03_scorecard_gini'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "0abe773c",
   "metadata": {
    "id": "0abe773c"
   },
   "source": "---\n# Track 2 — Quantum-kernel credit scoring\n\n## Stage 8 — Quantum feature map and fidelity kernel\nEach applicant's 6 selected attributes are encoded into a 6-qubit state |ψ(x)⟩ with a ZZ feature map; the kernel is K(x, x′) = |⟨ψ(x)|ψ(x′)⟩|². A **bandwidth** parameter scales the inputs — the literature shows un-scaled quantum kernels concentrate exponentially and lose all signal."
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5fae6fda",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:17.488486Z",
     "iopub.status.busy": "2026-10-03T09:47:17.488095Z",
     "iopub.status.idle": "2026-10-03T09:47:50.328211Z",
     "shell.execute_reply": "2026-10-03T09:47:50.326363Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 745
    },
    "id": "5fae6fda",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022090049,
     "user_tz": -180,
     "elapsed": 55319,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "2064411a-bab7-462d-f39d-5a9aba1bcab1"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "6-qubit feature map, depth 25 · 9,000 statevectors in 52.7 s\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "   bandwidth  mean off-diag K  std off-diag K  CV AUC (C=1)\n",
       "0       0.02           0.8775          0.0779        0.7875\n",
       "1       0.03           0.7805          0.1222        0.7818\n",
       "2       0.05           0.5458          0.1897        0.7794\n",
       "3       0.10           0.1112          0.1786        0.7415\n",
       "4       0.20           0.0359          0.1093        0.7089\n",
       "5       0.35           0.0317          0.0948        0.7133\n",
       "6       0.50           0.0362          0.0873        0.7091\n",
       "7       0.75           0.0357          0.0879        0.7004\n",
       "8       1.00           0.0446          0.1218        0.6723"
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-f718da28-99c6-42ec-953c-9f3efa315922\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bandwidth</th>\n",
       "      <th>mean off-diag K</th>\n",
       "      <th>std off-diag K</th>\n",
       "      <th>CV AUC (C=1)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.8775</td>\n",
       "      <td>0.0779</td>\n",
       "      <td>0.7875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.03</td>\n",
       "      <td>0.7805</td>\n",
       "      <td>0.1222</td>\n",
       "      <td>0.7818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.05</td>\n",
       "      <td>0.5458</td>\n",
       "      <td>0.1897</td>\n",
       "      <td>0.7794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.10</td>\n",
       "      <td>0.1112</td>\n",
       "      <td>0.1786</td>\n",
       "      <td>0.7415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.20</td>\n",
       "      <td>0.0359</td>\n",
       "      <td>0.1093</td>\n",
       "      <td>0.7089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.35</td>\n",
       "      <td>0.0317</td>\n",
       "      <td>0.0948</td>\n",
       "      <td>0.7133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.50</td>\n",
       "      <td>0.0362</td>\n",
       "      <td>0.0873</td>\n",
       "      <td>0.7091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0357</td>\n",
       "      <td>0.0879</td>\n",
       "      <td>0.7004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.00</td>\n",
       "      <td>0.0446</td>\n",
       "      <td>0.1218</td>\n",
       "      <td>0.6723</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f718da28-99c6-42ec-953c-9f3efa315922')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-f718da28-99c6-42ec-953c-9f3efa315922 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-f718da28-99c6-42ec-953c-9f3efa315922');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "summary": "{\n  \"name\": \"plt\",\n  \"rows\": 9,\n  \"fields\": [\n    {\n      \"column\": \"bandwidth\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.3515679166249389,\n        \"min\": 0.02,\n        \"max\": 1.0,\n        \"num_unique_values\": 9,\n        \"samples\": [\n          0.75,\n          0.03,\n          0.35\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"mean off-diag K\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.3539726040874412,\n        \"min\": 0.0317,\n        \"max\": 0.8775,\n        \"num_unique_values\": 9,\n        \"samples\": [\n          0.0357,\n          0.7805,\n          0.0317\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"std off-diag K\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.04017076051060024,\n        \"min\": 0.0779,\n        \"max\": 0.1897,\n        \"num_unique_values\": 9,\n        \"samples\": [\n          0.0879,\n          0.1222,\n          0.0948\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CV AUC (C=1)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.041628399093781056,\n        \"min\": 0.6723,\n        \"max\": 0.7875,\n        \"num_unique_values\": 9,\n        \"samples\": [\n          0.7004,\n          0.7818,\n          0.7133\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Selected bandwidth = 0.02, C = 0.1\n",
      "Note: the best kernels sit in the low-bandwidth regime, where a quantum kernel behaves close to a classical one — consistent with Shaydulin & Wild (2022).\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 825x374 with 2 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    }
   ],
   "source": "N_QUBITS = len(SELECTED)\nFMAP = make_feature_map(N_QUBITS, reps=2)\nscaler01 = MinMaxScaler().fit(X_tr[SELECTED].values)\nZ_tr, Z_te = scaler01.transform(X_tr[SELECTED].values), np.clip(scaler01.transform(X_te[SELECTED].values), 0, 1)\n\ndef statevectors(Z, bw):\n    return np.array([Statevector(FMAP.assign_parameters(np.pi * bw * z)).data for z in Z])\ndef kernel(A, B): return np.abs(A.conj() @ B.T) ** 2\n\nBANDWIDTHS = [0.02, 0.03, 0.05, 0.1, 0.2, 0.35, 0.5, 0.75, 1.0]\nt0 = time.time(); SV = {bw: (statevectors(Z_tr, bw), statevectors(Z_te, bw)) for bw in BANDWIDTHS}; t_sv = time.time() - t0\nprint(f'{N_QUBITS}-qubit feature map, depth {FMAP.decompose().depth()} · {len(BANDWIDTHS)*(len(Z_tr)+len(Z_te)):,} statevectors in {t_sv:.1f} s')\n\ndef qsvm_cv(Ktr, C=1.0):\n    aucs = []\n    for a, b in CV.split(Ktr, y_tr):\n        m = SVC(kernel='precomputed', C=C, class_weight='balanced').fit(Ktr[np.ix_(a, a)], y_tr[a])\n        aucs.append(roc_auc_score(y_tr[b], m.decision_function(Ktr[np.ix_(b, a)])))\n    return float(np.mean(aucs))\n\nconc = []\nfor bw in BANDWIDTHS:\n    Ktr = kernel(SV[bw][0], SV[bw][0]); off = Ktr[~np.eye(len(Ktr), dtype=bool)]\n    conc.append({'bandwidth': bw, 'mean off-diag K': off.mean(), 'std off-diag K': off.std(), 'CV AUC (C=1)': qsvm_cv(Ktr)})\nCONC = pd.DataFrame(conc); display(CONC.round(4))\nBW = float(CONC.loc[CONC['CV AUC (C=1)'].idxmax(), 'bandwidth'])\nK_tr, K_te = kernel(SV[BW][0], SV[BW][0]), kernel(SV[BW][1], SV[BW][0])\nC_BEST = max([0.1, 0.3, 1.0, 3.0, 10.0], key=lambda c: qsvm_cv(K_tr, c))\nprint(f'Selected bandwidth = {BW}, C = {C_BEST}')\nprint('Note: the best kernels sit in the low-bandwidth regime, where a quantum kernel behaves close to a classical one — consistent with Shaydulin & Wild (2022).')\n\nfig, ax1 = plt.subplots(figsize=(7.5, 3.4))\nax1.plot(CONC['bandwidth'], CONC['CV AUC (C=1)'], 'o-', color=BLUE, lw=2, label='QSVM CV AUC'); ax1.set_xlabel('Feature-map bandwidth'); ax1.set_ylabel('CV AUC', color=BLUE)\nax2 = ax1.twinx(); ax2.plot(CONC['bandwidth'], CONC['std off-diag K'], 's--', color=AMBER, lw=1.5, label='kernel spread (std)'); ax2.set_ylabel('Std of kernel values', color=AMBER)\nax2.spines['right'].set_visible(True); ax1.set_title('Bandwidth tuning vs kernel concentration')\nplt.tight_layout(); FIG_BW = savefig('04_bandwidth_concentration'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "bc5796c8",
   "metadata": {
    "id": "bc5796c8"
   },
   "source": "## Stage 9 — Quantum kernel vs classical models on the same 6 attributes"
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c278202b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:50.331658Z",
     "iopub.status.busy": "2026-10-03T09:47:50.331369Z",
     "iopub.status.idle": "2026-10-03T09:47:51.758399Z",
     "shell.execute_reply": "2026-10-03T09:47:51.756921Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 743
    },
    "id": "c278202b",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022091624,
     "user_tz": -180,
     "elapsed": 1571,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "7453de90-be42-4205-e25f-104989eebc0b"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                                   Model  CV AUC     AUC    Gini      KS\n",
       "0          Quantum-kernel SVM (6 qubits)  0.7911  0.7833  0.5666  0.4619\n",
       "1                     Logistic scorecard  0.7816  0.7940  0.5881  0.4984\n",
       "2                                RBF-SVM  0.7574  0.7400  0.4801  0.3857\n",
       "3                      Gradient boosting  0.7554  0.7587  0.5175  0.4603\n",
       "4  Reference: logistic scorecard, all 20  0.8016  0.8013  0.6025  0.5222"
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-151e3b71-10e2-4f38-8108-3eed8d5d4c8a\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>CV AUC</th>\n",
       "      <th>AUC</th>\n",
       "      <th>Gini</th>\n",
       "      <th>KS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Quantum-kernel SVM (6 qubits)</td>\n",
       "      <td>0.7911</td>\n",
       "      <td>0.7833</td>\n",
       "      <td>0.5666</td>\n",
       "      <td>0.4619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logistic scorecard</td>\n",
       "      <td>0.7816</td>\n",
       "      <td>0.7940</td>\n",
       "      <td>0.5881</td>\n",
       "      <td>0.4984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>RBF-SVM</td>\n",
       "      <td>0.7574</td>\n",
       "      <td>0.7400</td>\n",
       "      <td>0.4801</td>\n",
       "      <td>0.3857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Gradient boosting</td>\n",
       "      <td>0.7554</td>\n",
       "      <td>0.7587</td>\n",
       "      <td>0.5175</td>\n",
       "      <td>0.4603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Reference: logistic scorecard, all 20</td>\n",
       "      <td>0.8016</td>\n",
       "      <td>0.8013</td>\n",
       "      <td>0.6025</td>\n",
       "      <td>0.5222</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-151e3b71-10e2-4f38-8108-3eed8d5d4c8a')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-151e3b71-10e2-4f38-8108-3eed8d5d4c8a button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-151e3b71-10e2-4f38-8108-3eed8d5d4c8a');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_96da2feb-7797-4684-a0b0-e56c048015d4\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('QK')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_96da2feb-7797-4684-a0b0-e56c048015d4 button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('QK');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "QK",
       "summary": "{\n  \"name\": \"QK\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"Model\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Logistic scorecard\",\n          \"Reference: logistic scorecard, all 20\",\n          \"RBF-SVM\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"CV AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.02046318645763657,\n        \"min\": 0.7554,\n        \"max\": 0.8016,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.7816,\n          0.8016,\n          0.7574\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"AUC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.025550009784733943,\n        \"min\": 0.74,\n        \"max\": 0.8013,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.794,\n          0.8013,\n          0.74\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Gini\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.051041825986145914,\n        \"min\": 0.4801,\n        \"max\": 0.6025,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.5881,\n          0.6025,\n          0.4801\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"KS\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.05173282323631682,\n        \"min\": 0.3857,\n        \"max\": 0.5222,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.4984,\n          0.5222,\n          0.3857\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 616x550 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    }
   ],
   "source": "qsvm = SVC(kernel='precomputed', C=C_BEST, class_weight='balanced').fit(K_tr, y_tr)\nq_scores = qsvm.decision_function(K_te)\nrows = [{'Model': f'Quantum-kernel SVM ({N_QUBITS} qubits)', 'CV AUC': round(qsvm_cv(K_tr, C_BEST), 4), **score_metrics(y_te, q_scores)}]\ncm = evaluate(classical_models(), X_tr[SELECTED].values, X_te[SELECTED].values, f'QUBO-selected {K_SELECT}')\nfor _, rw in cm.iterrows(): rows.append({'Model': rw['Model'], 'CV AUC': rw['CV AUC'], 'AUC': rw['AUC'], 'Gini': rw['Gini'], 'KS': rw['KS']})\nrows.append({'Model': 'Reference: logistic scorecard, all 20', 'CV AUC': baseline.iloc[0]['CV AUC'], 'AUC': baseline.iloc[0]['AUC'], 'Gini': baseline.iloc[0]['Gini'], 'KS': baseline.iloc[0]['KS']})\nQK = pd.DataFrame(rows); display(QK)\n\nfig, ax = plt.subplots(figsize=(5.6, 5))\nfor name, s, c in [(f'Quantum kernel ({N_QUBITS}q)', q_scores, BLUE)] + [(n, scores_of(m.fit(X_tr[SELECTED].values, y_tr), X_te[SELECTED].values), col) for (n, m), col in zip(classical_models().items(), [NAVY, TEAL, AMBER])]:\n    fpr, tpr, _ = roc_curve(y_te, s); ax.plot(fpr, tpr, lw=2, color=c, label=f'{name} (AUC {roc_auc_score(y_te, s):.3f})')\nax.plot([0, 1], [0, 1], ':', color=GRAY); ax.set_xlabel('False positive rate'); ax.set_ylabel('True positive rate')\nax.set_title('ROC — test set, 6 QUBO-selected attributes'); ax.legend(frameon=False, fontsize=8, loc='lower right')\nplt.tight_layout(); FIG_ROC = savefig('05_roc_comparison'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "e4013ea8",
   "metadata": {
    "id": "e4013ea8"
   },
   "source": "## Stage 10 — Small-data regime (thin-file, SME, agricultural segments)\nThe literature's most credible QML claims are in small-data settings. Train on 30 → 700 applicants (10 random stratified draws each) and evaluate on the fixed test set."
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0b410607",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:51.761317Z",
     "iopub.status.busy": "2026-10-03T09:47:51.761032Z",
     "iopub.status.idle": "2026-10-03T09:47:53.629474Z",
     "shell.execute_reply": "2026-10-03T09:47:53.628191Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 670
    },
    "id": "0b410607",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022093933,
     "user_tz": -180,
     "elapsed": 2304,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "3a3ae892-d911-4f55-e6b2-8f9e67b66e38"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "model  Logistic scorecard  Quantum kernel  RBF-SVM\n",
       "n                                                 \n",
       "30                 0.7364          0.7566   0.6965\n",
       "60                 0.7614          0.7697   0.7200\n",
       "120                0.7803          0.7743   0.7339\n",
       "240                0.7892          0.7756   0.7339\n",
       "480                0.7943          0.7808   0.7416\n",
       "700                0.7940          0.7833   0.7400"
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-857cfb9f-b735-4dde-917a-bec06120b53b\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>model</th>\n",
       "      <th>Logistic scorecard</th>\n",
       "      <th>Quantum kernel</th>\n",
       "      <th>RBF-SVM</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>0.7364</td>\n",
       "      <td>0.7566</td>\n",
       "      <td>0.6965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>0.7614</td>\n",
       "      <td>0.7697</td>\n",
       "      <td>0.7200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>0.7803</td>\n",
       "      <td>0.7743</td>\n",
       "      <td>0.7339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>0.7892</td>\n",
       "      <td>0.7756</td>\n",
       "      <td>0.7339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>480</th>\n",
       "      <td>0.7943</td>\n",
       "      <td>0.7808</td>\n",
       "      <td>0.7416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>700</th>\n",
       "      <td>0.7940</td>\n",
       "      <td>0.7833</td>\n",
       "      <td>0.7400</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-857cfb9f-b735-4dde-917a-bec06120b53b')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-857cfb9f-b735-4dde-917a-bec06120b53b button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-857cfb9f-b735-4dde-917a-bec06120b53b');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "summary": "{\n  \"name\": \"print('Quantum-kernel AUC minus best classical, by training size:', SMALL_GAP)\",\n  \"rows\": 6,\n  \"fields\": [\n    {\n      \"column\": \"n\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 266,\n        \"min\": 30,\n        \"max\": 700,\n        \"num_unique_values\": 6,\n        \"samples\": [\n          30,\n          60,\n          700\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Logistic scorecard\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.02294738910348335,\n        \"min\": 0.7364,\n        \"max\": 0.7943,\n        \"num_unique_values\": 6,\n        \"samples\": [\n          0.7364,\n          0.7614,\n          0.794\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Quantum kernel\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.009532558243549647,\n        \"min\": 0.7566,\n        \"max\": 0.7833,\n        \"num_unique_values\": 6,\n        \"samples\": [\n          0.7566,\n          0.7697,\n          0.7833\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"RBF-SVM\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.017053416080070296,\n        \"min\": 0.6965,\n        \"max\": 0.7416,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.72,\n          0.74,\n          0.7339\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 825x396 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Quantum-kernel AUC minus best classical, by training size: {30: 0.0202, 60: 0.0083, 120: -0.006, 240: -0.0137, 480: -0.0135, 700: -0.0107}\n"
     ]
    }
   ],
   "source": "SIZES, REPS = [30, 60, 120, 240, 480, 700], 10\nres = []\nXs_tr = StandardScaler().fit(X_tr[SELECTED].values)\nfor n in SIZES:\n    for rep in range(REPS):\n        idx = np.arange(len(y_tr)) if n >= len(y_tr) else train_test_split(np.arange(len(y_tr)), train_size=n, stratify=y_tr, random_state=rep)[0]\n        yy = y_tr[idx]\n        q = SVC(kernel='precomputed', C=C_BEST, class_weight='balanced').fit(K_tr[np.ix_(idx, idx)], yy)\n        res.append({'n': n, 'model': 'Quantum kernel', 'AUC': roc_auc_score(y_te, q.decision_function(K_te[:, idx]))})\n        rb = SVC(C=1.0, gamma='scale', class_weight='balanced').fit(Xs_tr.transform(X_tr[SELECTED].values[idx]), yy)\n        res.append({'n': n, 'model': 'RBF-SVM', 'AUC': roc_auc_score(y_te, rb.decision_function(Xs_tr.transform(X_te[SELECTED].values)))})\n        lr = LogisticRegression(max_iter=2000).fit(X_tr[SELECTED].values[idx], yy)\n        res.append({'n': n, 'model': 'Logistic scorecard', 'AUC': roc_auc_score(y_te, lr.decision_function(X_te[SELECTED].values))})\n        if n >= len(y_tr): break\nSMALL = pd.DataFrame(res).groupby(['n', 'model'])['AUC'].agg(['mean', 'std']).reset_index()\ndisplay(SMALL.pivot(index='n', columns='model', values='mean').round(4))\n\nfig, ax = plt.subplots(figsize=(7.5, 3.6))\nfor mdl, c in [('Quantum kernel', BLUE), ('RBF-SVM', TEAL), ('Logistic scorecard', NAVY)]:\n    d = SMALL[SMALL.model == mdl]; ax.plot(d['n'], d['mean'], 'o-', color=c, lw=2, label=mdl)\n    ax.fill_between(d['n'], d['mean'] - d['std'].fillna(0), d['mean'] + d['std'].fillna(0), color=c, alpha=0.12)\nax.set_xscale('log'); ax.set_xticks(SIZES); ax.set_xticklabels(SIZES); ax.set_xlabel('Training applicants (log scale)'); ax.set_ylabel('Test AUC')\nax.set_title('Learning curve — small-data regime'); ax.legend(frameon=False); ax.grid(axis='y', color='#E3E8F0')\nplt.tight_layout(); FIG_SMALL = savefig('06_small_data'); plt.show()\npiv = SMALL.pivot(index='n', columns='model', values='mean')\nSMALL_GAP = {int(n): round(float(piv.loc[n, 'Quantum kernel'] - max(piv.loc[n, 'RBF-SVM'], piv.loc[n, 'Logistic scorecard'])), 4) for n in SIZES}\nprint('Quantum-kernel AUC minus best classical, by training size:', SMALL_GAP)"
  },
  {
   "cell_type": "markdown",
   "id": "9abfc8c6",
   "metadata": {
    "id": "9abfc8c6"
   },
   "source": "---\n# Track 3 — Quantum-safe credit report\n\n## Stage 11 — Protect a credit report with hybrid post-quantum cryptography\nScenario: KKB sends a credit report to a member bank. Confidentiality uses a **hybrid key exchange** (classical X25519 + ML-KEM-768, FIPS 203) combined through HKDF into an AES-256-GCM key; authenticity uses **ML-DSA-65** (FIPS 204). Hybrid mode means an attacker must break *both* the classical and the post-quantum scheme."
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8e834313",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:53.632458Z",
     "iopub.status.busy": "2026-10-03T09:47:53.632118Z",
     "iopub.status.idle": "2026-10-03T09:47:53.790916Z",
     "shell.execute_reply": "2026-10-03T09:47:53.787395Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8e834313",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022094130,
     "user_tz": -180,
     "elapsed": 192,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "3bfa58a0-aa7e-4be5-f56a-4034840abdc0"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "✓ Report (457 B) sealed: envelope 1605 B + ML-DSA-65 signature 3309 B\n",
      "✓ Bank decrypted and verified the report · tamper test detected: True\n",
      "Per-message post-quantum overhead: 4,457 bytes\n"
     ]
    }
   ],
   "source": "from kyber_py.ml_kem import ML_KEM_768\nfrom dilithium_py.ml_dsa import ML_DSA_65\nfrom cryptography.hazmat.primitives.asymmetric import x25519, rsa, ec, padding\nfrom cryptography.hazmat.primitives import hashes, serialization\nfrom cryptography.hazmat.primitives.kdf.hkdf import HKDF\nfrom cryptography.hazmat.primitives.ciphers.aead import AESGCM\n\nreport = {  # synthetic, illustrative credit report — no real person\n    'report_id': 'KKB-DEMO-2026-000042', 'issued': datetime.datetime.now().isoformat(timespec='seconds'),\n    'subject': {'national_id_hash': 'sha256:9f2c…(pseudonymised)', 'birth_year': 1986},\n    'score': {'value': 1385, 'band': 'B', 'scale': '1–1900'},\n    'exposures': [{'type': 'consumer_loan', 'limit_try': 250000, 'balance_try': 118400, 'dpd_max_12m': 0},\n                  {'type': 'credit_card', 'limit_try': 60000, 'balance_try': 21450, 'dpd_max_12m': 15}],\n    'cheques': {'returned_24m': 0}, 'inquiries_3m': 2}\nPLAINTEXT = json.dumps(report, ensure_ascii=False).encode()\n\n# --- Member bank (recipient) long-term keys: classical + post-quantum\nbank_x_sk = x25519.X25519PrivateKey.generate(); bank_x_pk = bank_x_sk.public_key()\nbank_kem_ek, bank_kem_dk = ML_KEM_768.keygen()\n# --- KKB signing key (post-quantum)\nkkb_sig_pk, kkb_sig_sk = ML_DSA_65.keygen()\n\ndef kkb_seal(plaintext):\n    eph = x25519.X25519PrivateKey.generate()\n    ss_classical = eph.exchange(bank_x_pk)\n    ss_pq, kem_ct = ML_KEM_768.encaps(bank_kem_ek)\n    key = HKDF(algorithm=hashes.SHA256(), length=32, salt=None, info=b'KKB hybrid X25519+ML-KEM-768').derive(ss_classical + ss_pq)\n    nonce = os.urandom(12); ct = AESGCM(key).encrypt(nonce, plaintext, b'KKB-report-v1')\n    eph_pk = eph.public_key().public_bytes(serialization.Encoding.Raw, serialization.PublicFormat.Raw)\n    envelope = eph_pk + kem_ct + nonce + ct\n    return {'envelope': envelope, 'signature': ML_DSA_65.sign(kkb_sig_sk, envelope), 'sizes': (len(eph_pk), len(kem_ct), len(ct))}\n\ndef bank_open(msg):\n    env, sig = msg['envelope'], msg['signature']\n    if not ML_DSA_65.verify(kkb_sig_pk, env, sig): raise ValueError('signature invalid — report rejected')\n    eph_pk, kem_ct, nonce, ct = env[:32], env[32:32 + 1088], env[1120:1132], env[1132:]\n    ss_classical = bank_x_sk.exchange(x25519.X25519PublicKey.from_public_bytes(eph_pk))\n    ss_pq = ML_KEM_768.decaps(bank_kem_dk, kem_ct)\n    key = HKDF(algorithm=hashes.SHA256(), length=32, salt=None, info=b'KKB hybrid X25519+ML-KEM-768').derive(ss_classical + ss_pq)\n    return AESGCM(key).decrypt(nonce, ct, b'KKB-report-v1')\n\nmsg = kkb_seal(PLAINTEXT)\nassert bank_open(msg) == PLAINTEXT\ntampered = dict(msg, envelope=msg['envelope'][:-1] + bytes([msg['envelope'][-1] ^ 1]))\ntry: bank_open(tampered); TAMPER_DETECTED = False\nexcept ValueError: TAMPER_DETECTED = True\nPQ_OVERHEAD = len(msg['envelope']) + len(msg['signature']) - len(PLAINTEXT)\nprint(f'✓ Report ({len(PLAINTEXT)} B) sealed: envelope {len(msg[\"envelope\"])} B + ML-DSA-65 signature {len(msg[\"signature\"])} B')\nprint(f'✓ Bank decrypted and verified the report · tamper test detected: {TAMPER_DETECTED}')\nprint(f'Per-message post-quantum overhead: {PQ_OVERHEAD:,} bytes')"
  },
  {
   "cell_type": "markdown",
   "id": "d1302865",
   "metadata": {
    "id": "d1302865"
   },
   "source": "## Stage 12 — Cost of quantum safety: sizes and timings\nSizes are exact per the FIPS standards. ⚠️ Timings for ML-KEM/ML-DSA use **pure-Python reference code** and are far slower than production C/assembly implementations (liboqs, HSM firmware); they are shown for relative illustration only."
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "bdf87486",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:53.795733Z",
     "iopub.status.busy": "2026-10-03T09:47:53.795048Z",
     "iopub.status.idle": "2026-10-03T09:47:54.466353Z",
     "shell.execute_reply": "2026-10-03T09:47:54.464239Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 271
    },
    "id": "bdf87486",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022094858,
     "user_tz": -180,
     "elapsed": 695,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "2f97e677-5464-4a8a-ce63-a648e76556b0"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                     Scheme Quantum-safe  Public key B  \\\n",
       "0  RSA-2048 (key transport)           No           294   \n",
       "1             X25519 (ECDH)           No            32   \n",
       "2     ML-KEM-768 (FIPS 203)          Yes          1184   \n",
       "3  RSA-2048 PSS (signature)           No           294   \n",
       "4   ECDSA P-256 (signature)           No            91   \n",
       "5      ML-DSA-65 (FIPS 204)          Yes          1952   \n",
       "\n",
       "   Ciphertext / signature B  Op time ms  \n",
       "0                       256       0.049  \n",
       "1                        32       0.104  \n",
       "2                      1088       7.047  \n",
       "3                       256       0.792  \n",
       "4                        70       0.040  \n",
       "5                      3309      69.698  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-561eb43b-8307-4fe8-98fd-f0b1a643a3d5\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Scheme</th>\n",
       "      <th>Quantum-safe</th>\n",
       "      <th>Public key B</th>\n",
       "      <th>Ciphertext / signature B</th>\n",
       "      <th>Op time ms</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>RSA-2048 (key transport)</td>\n",
       "      <td>No</td>\n",
       "      <td>294</td>\n",
       "      <td>256</td>\n",
       "      <td>0.049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>X25519 (ECDH)</td>\n",
       "      <td>No</td>\n",
       "      <td>32</td>\n",
       "      <td>32</td>\n",
       "      <td>0.104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ML-KEM-768 (FIPS 203)</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1184</td>\n",
       "      <td>1088</td>\n",
       "      <td>7.047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>RSA-2048 PSS (signature)</td>\n",
       "      <td>No</td>\n",
       "      <td>294</td>\n",
       "      <td>256</td>\n",
       "      <td>0.792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ECDSA P-256 (signature)</td>\n",
       "      <td>No</td>\n",
       "      <td>91</td>\n",
       "      <td>70</td>\n",
       "      <td>0.040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>ML-DSA-65 (FIPS 204)</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1952</td>\n",
       "      <td>3309</td>\n",
       "      <td>69.698</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-561eb43b-8307-4fe8-98fd-f0b1a643a3d5')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-561eb43b-8307-4fe8-98fd-f0b1a643a3d5 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-561eb43b-8307-4fe8-98fd-f0b1a643a3d5');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_6d4e96fc-fe4e-4d50-b468-ce098491dd39\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('PQC')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_6d4e96fc-fe4e-4d50-b468-ce098491dd39 button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('PQC');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "PQC",
       "summary": "{\n  \"name\": \"PQC\",\n  \"rows\": 6,\n  \"fields\": [\n    {\n      \"column\": \"Scheme\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"RSA-2048 (key transport)\",\n          \"X25519 (ECDH)\",\n          \"ML-DSA-65 (FIPS 204)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Quantum-safe\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Yes\",\n          \"No\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Public key B\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 765,\n        \"min\": 32,\n        \"max\": 1952,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          32,\n          1952\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Ciphertext / signature B\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1271,\n        \"min\": 32,\n        \"max\": 3309,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          32,\n          3309\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Op time ms\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 27.932494854559625,\n        \"min\": 0.04,\n        \"max\": 69.698,\n        \"num_unique_values\": 6,\n        \"samples\": [\n          0.049,\n          0.104\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Extra bytes per query (KEM public key + ciphertext + larger signature): 5,511 B\n",
      "Upper-bound extra traffic at 20 M queries/day: 110 GB/day (≈ 0.01 Gbit/s average) — far less with session reuse\n"
     ]
    }
   ],
   "source": "def bench(fn, n=20):\n    t0 = time.perf_counter()\n    for _ in range(n): fn()\n    return (time.perf_counter() - t0) / n * 1000\n\nrsa_sk = rsa.generate_private_key(public_exponent=65537, key_size=2048); rsa_pk = rsa_sk.public_key()\nec_sk = ec.generate_private_key(ec.SECP256R1()); ec_pk = ec_sk.public_key()\nmsg32 = os.urandom(32)\nder = lambda pk: len(pk.public_bytes(serialization.Encoding.DER, serialization.PublicFormat.SubjectPublicKeyInfo))\noaep = padding.OAEP(mgf=padding.MGF1(hashes.SHA256()), algorithm=hashes.SHA256(), label=None)\npss = padding.PSS(mgf=padding.MGF1(hashes.SHA256()), salt_length=padding.PSS.MAX_LENGTH)\nrsa_sig = rsa_sk.sign(msg32, pss, hashes.SHA256()); ec_sig = ec_sk.sign(msg32, ec.ECDSA(hashes.SHA256()))\n_, kem_ct = ML_KEM_768.encaps(bank_kem_ek); dsa_sig = ML_DSA_65.sign(kkb_sig_sk, msg32)\n\nPQC = pd.DataFrame([\n    {'Scheme': 'RSA-2048 (key transport)', 'Quantum-safe': 'No', 'Public key B': der(rsa_pk), 'Ciphertext / signature B': 256, 'Op time ms': bench(lambda: rsa_pk.encrypt(msg32, oaep))},\n    {'Scheme': 'X25519 (ECDH)', 'Quantum-safe': 'No', 'Public key B': 32, 'Ciphertext / signature B': 32, 'Op time ms': bench(lambda: x25519.X25519PrivateKey.generate().exchange(bank_x_pk))},\n    {'Scheme': 'ML-KEM-768 (FIPS 203)', 'Quantum-safe': 'Yes', 'Public key B': len(bank_kem_ek), 'Ciphertext / signature B': len(kem_ct), 'Op time ms': bench(lambda: ML_KEM_768.encaps(bank_kem_ek))},\n    {'Scheme': 'RSA-2048 PSS (signature)', 'Quantum-safe': 'No', 'Public key B': der(rsa_pk), 'Ciphertext / signature B': len(rsa_sig), 'Op time ms': bench(lambda: rsa_sk.sign(msg32, pss, hashes.SHA256()))},\n    {'Scheme': 'ECDSA P-256 (signature)', 'Quantum-safe': 'No', 'Public key B': der(ec_pk), 'Ciphertext / signature B': len(ec_sig), 'Op time ms': bench(lambda: ec_sk.sign(msg32, ec.ECDSA(hashes.SHA256())))},\n    {'Scheme': 'ML-DSA-65 (FIPS 204)', 'Quantum-safe': 'Yes', 'Public key B': len(kkb_sig_pk), 'Ciphertext / signature B': len(dsa_sig), 'Op time ms': bench(lambda: ML_DSA_65.sign(kkb_sig_sk, msg32), n=5)},\n])\nPQC['Op time ms'] = PQC['Op time ms'].round(3); display(PQC)\n\n# Bandwidth impact at KKB scale (upper bound: a fresh hybrid handshake + signature on every query)\nDAILY_QUERIES = 20_000_000          # KKB public figure (~20 million queries/day)\nextra_per_query = (len(bank_kem_ek) + len(kem_ct)) + (len(dsa_sig) - len(ec_sig))\nGB_DAY = DAILY_QUERIES * extra_per_query / 1e9\nprint(f'Extra bytes per query (KEM public key + ciphertext + larger signature): {extra_per_query:,} B')\nprint(f'Upper-bound extra traffic at 20 M queries/day: {GB_DAY:,.0f} GB/day (≈ {GB_DAY*8/86400:.2f} Gbit/s average) — far less with session reuse')"
  },
  {
   "cell_type": "markdown",
   "id": "0c5e671b",
   "metadata": {
    "id": "0c5e671b"
   },
   "source": "## Stage 13 — \"Harvest now, decrypt later\": Mosca risk model\nMosca's inequality: data is at risk if **x + y > z**, where *x* = years the data must stay confidential, *y* = years to migrate to PQC, *z* = years until a cryptographically relevant quantum computer (CRQC). All inputs are adjustable assumptions."
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0ce88ee2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:54.469987Z",
     "iopub.status.busy": "2026-10-03T09:47:54.469679Z",
     "iopub.status.idle": "2026-10-03T09:47:54.908528Z",
     "shell.execute_reply": "2026-10-03T09:47:54.906045Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 585
    },
    "id": "0ce88ee2",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022095349,
     "user_tz": -180,
     "elapsed": 487,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "ec1a6a90-ebfc-45c7-8495-cf2f84dcae9b"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                             Data class  x (yrs)  Protected until  \\\n",
       "0               Real-time query traffic        2             2033   \n",
       "1               Credit reports & scores       10             2041   \n",
       "2  Credit histories (KRS / Risk Center)       15             2046   \n",
       "3           Identity & demographic data       30             2061   \n",
       "4                 Archives & DC backups       20             2051   \n",
       "\n",
       "  Aggressive (2030) Regulatory planning (2035) Conservative (2040)  \n",
       "0           AT RISK                         ok                  ok  \n",
       "1           AT RISK                    AT RISK             AT RISK  \n",
       "2           AT RISK                    AT RISK             AT RISK  \n",
       "3           AT RISK                    AT RISK             AT RISK  \n",
       "4           AT RISK                    AT RISK             AT RISK  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-948c2be0-f0a6-47fe-8710-0a38d2a32002\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Data class</th>\n",
       "      <th>x (yrs)</th>\n",
       "      <th>Protected until</th>\n",
       "      <th>Aggressive (2030)</th>\n",
       "      <th>Regulatory planning (2035)</th>\n",
       "      <th>Conservative (2040)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Real-time query traffic</td>\n",
       "      <td>2</td>\n",
       "      <td>2033</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>ok</td>\n",
       "      <td>ok</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Credit reports &amp; scores</td>\n",
       "      <td>10</td>\n",
       "      <td>2041</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Credit histories (KRS / Risk Center)</td>\n",
       "      <td>15</td>\n",
       "      <td>2046</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Identity &amp; demographic data</td>\n",
       "      <td>30</td>\n",
       "      <td>2061</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Archives &amp; DC backups</td>\n",
       "      <td>20</td>\n",
       "      <td>2051</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "      <td>AT RISK</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-948c2be0-f0a6-47fe-8710-0a38d2a32002')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-948c2be0-f0a6-47fe-8710-0a38d2a32002 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-948c2be0-f0a6-47fe-8710-0a38d2a32002');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_6057824d-8bf3-4a52-a6ed-2cdbb1718269\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('MOSCA')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_6057824d-8bf3-4a52-a6ed-2cdbb1718269 button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('MOSCA');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "MOSCA",
       "summary": "{\n  \"name\": \"MOSCA\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"Data class\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Credit reports & scores\",\n          \"Archives & DC backups\",\n          \"Credit histories (KRS / Risk Center)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x (yrs)\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10,\n        \"min\": 2,\n        \"max\": 30,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          10,\n          20,\n          15\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Protected until\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10,\n        \"min\": 2033,\n        \"max\": 2061,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          2041,\n          2051,\n          2046\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Aggressive (2030)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"AT RISK\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Regulatory planning (2035)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"AT RISK\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Conservative (2040)\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"AT RISK\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 935x374 with 1 Axes>"
      ],
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA60AAAFrCAYAAAA6i29fAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAAq9xJREFUeJzs3XdYFNf7NvB76WVZli69iL13Y4sl9oYlil2MGjUmlsQv1ohRE2OsURNLjL0ktlixxRpj1Bh71wiCoNJhlbqc9w9e5rcLu4gFl8D9ua69rtmdM2eemZ2d3WfPmTMyIYQAERERERERUTFkZOgAiIiIiIiIiPRh0kpERERERETFFpNWIiIiIiIiKraYtBIREREREVGxxaSViIiIiIiIii0mrURERERERFRsMWklIiIiIiKiYotJKxERERERERVbTFqJiIiIiIio2GLSSkRERERERMUWk1YiogKEhIRAJpNBJpOhefPmb1RX8+bNpbpCQkLeSnxvy+DBg6XYBg8ebOhwXktYWJi0DTKZDGFhYYYOKZ+pU6dCJpPByMgIV69eNXQ49Bbs3btXOuZWr15t6HCKrbS0NEyfPh0VKlSApaWltM8uX778WufZtWvXSsv4+PgUaeyvy8fHR4px7dq1r13PiRMntM5tmjRfP3HixJsF/JqKy/fHf+E74FU0bdoUMpkM7u7uSE9Pl15/nePK0L8/6tSpA5lMBn9/f2RkZLxWHUxaid4RzS/Y3Mf+/fvzlYuOjoaZmZlWueKW4BAZwokTJxASEoKQkJA3+gFoKJGRkViwYAEAoEuXLqhevbo0L++PUl2Pn376yVChUwE6d+6MmjVrAgCmTZuG58+fGzagYmrKlCn46quvcPfuXaSlpRk6nDcWFhYmnY+K43d07p8BISEhWLRokaHDoVe0Y8cO/PHHHwCACRMmwNzc3MARvZlp06YBAB48eIBly5a9Vh0mbzMgIno1ixYtQseOHbVeW7ZsGTIzMw0UEZVWU6ZMwdChQwEALi4uBo5GtxMnTmDGjBkAgPfffz/fP/qurq44ffq01vPiZO7cuUhNTQUAfPrppwaOht6mTz/9FB999BGio6OxatUqjB071tAhFTvbt2+Xpj/66CMMHDgQRkZGKFeuHIYMGYIPPvgAAGBra2uoEF9JWFiYdD4CoDNx3b59u5Sgly9fvkji0DznVatWTZq+fPmyFJ+3t3epOCaL+3fAq8h97ywtLfHRRx8ZOJo317VrV3h7eyM8PBxz5szBJ598AjMzs1eqg0krkQEdPXoUN27cQJUqVQDkdJ9asWKFgaOi0qhcuXIoV65cka8nMzMTQohX/rIqDHNzczRp0uSt1/s2PH/+HOvWrQMAuLu7o0WLFnrLBgUFYciQIflefxfvT14qlQpyufydr/dteP78Oaytrd/Junr06IFRo0YhPT0dK1aseKMEISQkBDNmzMCgQYP+kz0K9ImIiJCm+/Xrh2bNmknPra2t4eXlZYiwilTdunWLfB3F9Zz3LmVnZyM9PR2WlpYlYn/88ccfuHbtGoCcZM/GxsbAEb05mUyGPn36YM6cOXj27Bl27tyJwMDAV6qD3YOJDMDY2BhWVlYAgMWLF0uvb9q0CbGxsQAAhUJRYB1btmxBmzZt4OTkBFNTUzg6OuKDDz7A5s2b85V98uQJRo8ejfLly8PS0hLm5uZwc3ND48aNMWbMGDx58kSrfGxsLL788kvUqVMHtra2MDc3h6enJ7p164bLly9L5VasWIFOnTqhbNmyUCqVMDU1hb29PRo3boylS5ciKyurUPsj73Uo165dw6effgoXFxfI5XJ88MEH0gn8l19+Qc2aNWFhYQFPT09Mnz4darU6X503b97E0KFD4efnBwsLC8jlclSvXh2TJ09GXFxcvvIJCQkYOXIkXFxcYGlpiYYNGyI0NPSlsf/222/o2LEjXFxcpPehU6dO+P333wu17ble57okzeta1qxZg0WLFqFcuXKwsLBAjRo1sHPnTgDAP//8gzZt2kAul8Pe3h79+/fPtw8KuiZJ1745cOCA3uvQ8nZ1jY6OxuDBg+Hs7Axzc3PcvHkTSUlJGDVqFBo1agR3d3dYWVlJx9mHH36IP//8U6ov9/jQbNU4efJkvmu8XnY906scE3mvl4uIiMCgQYPg5OQECwsL1KtX75Xe49DQUCQnJwMA2rdvDyMj/V+/Xl5eaNKkSb5HYVrAp0yZIsXdq1evfPM7d+4szf/qq68A5L9u+8KFC2jdujUUCgU8PT0BABkZGfjmm29Qp04dKBQKmJqawsnJCTVr1sSQIUPw119/FXpf6JOdnY3vvvsO/v7+sLCwQPny5bFw4UIcO3asUNfzHT58GNOnT4efnx9MTEywatUqqdylS5cwaNAg+Pj4wMLCAgqFAo0aNcLKlSuRnZ39RvsPyGkdbNq0KQDg9u3bWudJQwkPD8e4ceNQtWpVyOVyWFpaws/PD3379kV4eLhW2b/++gt9+vSBp6cnzMzMoFAoUK9ePcyZMydfd+e8x8vNmzfRrVs32NnZwdLSEi1atMClS5ek8rnnFiGE9FrLli21rkUt6JrWtLQ0TJo0CZ6entK5bf369S/d/hMnTqBnz57w8PCAmZkZ7Ozs0LJlS2zbti1f2bznvz///BOtW7eGXC6HjY0NOnfujIcPH0rlfXx88v3xpOscru/aw2PHjiEwMBBVqlSRvsNzz0fBwcGIj49/6fYVtF6ZTIagoCCpTHh4uFa5y5cvQy6XS89v3rypVefFixeleQqF4pW7vP/9999o164dbGxsdO6/XHFxcZg6dSpq1KgBuVwOCwsL+Pn5YciQIfliynt+v3r1KsaMGQN3d3eYmpri0KFDer8DCnP5Rd6W8oMHD6Jr165wdXWVjp8mTZrgxx9/zPfb5lWPn5fZunWrNN2pU6eXlt+0aRNq1aoFCwsLuLq6YsyYMUhJSSnUugr67i/oOtj09HQsXrwYjRo1glKphJmZGby8vDBkyBDcu3dP57o0t0VzGwtNENE7sWbNGgFAABDm5uZixIgRAoCwtLQUsbGxQgghqlWrJgCI2rVri/fff18qP336dKketVotevfuLc3T9ejdu7dQq9VCCCEyMzNF+fLlCyx/9uxZqf5Lly4JZ2dnvWXXrFkjldWMUdejZ8+ehdo3Dx8+1FquQoUK+eqyt7cXU6ZM0bmeb7/9Vqu+HTt2CHNzc71xubm5ibt370rlX7x4IWrUqJGvnEwmEzVr1pSev//++9Iy2dnZIigoqMDtnzNnjlZc+t5TIYTWcsePHy/UfvP29i5wnwEQs2bNEhYWFvleb9++vVZdgwYNkuYNGjSoUPumVq1aOvfN8ePHtcqWK1dO6/mlS5fyved5H0ZGRmLPnj06jw9dD13lHj58+NrHhObnVaFQCBcXl3zLWFhYiPDw8EK9V6NHj5aWW716db75mvvMyclJ2NjYCDMzM+Ht7S0GDRokrl27Vqj1RERECBMTEwFAmJmZiWfPnknzEhMThZmZmbR/c2OfPn26tG53d3dhaWkpPbe1tRVCiJce6998802h4ivI8OHDddZdp06dfO91roKOs4ULFwohhFi1apW0T3Q9OnfuLDIzM197/+WaOnWqVOeCBQteez/kvh+an8NXdejQIWFjY6N3mzXPMd9//70wMjLSW7Zy5cpa+0HzeHF1dRVyuTzfMs7OziIlJUUIoX1uyfvw9vbOV2fe82y7du10Llu7du189eT68ssvCzxeR44cqVVeM0ZfX1+dx0vVqlWl71XNc29B+1eznOZ3p+b26nr4+/uLxMREqXzec6omXestqG4g59z48ccfS8/Hjh2rVef//vc/ad7QoUMLPNby7r/y5ctLnxN9+08IIe7evSvc3Nz0xmhubi527Nghlc97fs/7ed+1a5fe74C8+0/XQ/M7+YsvviiwbLNmzcSLFy9e+/h5mapVq0rLPXjwIN98zeNK83Og+WjYsKHIyMiQltH3+0Pfd39By8THx2t9/+d9WFtbi99//z1f3C9evBCmpqYCgLCzsyv0/sjFpJXoHcmbtN66dUvIZDIBQHz99dfi999/l+avW7dO78li2bJl0utGRkZi4sSJIjQ0VEycOFHrh8eyZcuEEEJcuHBBes3T01Ns3bpV/P7772Lz5s1i2rRpom7duuLcuXNCCCHS09OFn5+fVF6pVIqvv/5aHDx4UGzcuFH069dPbNy4UYplxYoVYtWqVWLPnj3i+PHj4vfffxerV68Wjo6OUh1///33S/dN3i8aOzs7sXr1arFp0yZhbW2tNa93795i//79omvXrtJrrq6uUl1Pnz7V+hHVrFkz8dtvv4l169YJDw8P6fX69etLy8yaNUt63djYWHz55ZfiwIED+X5Ea/6YWrFihfS6XC4X8+bNE0eOHBELFy6UfvTLZDKtPwSKMmk1MjIS06dPF/v37xeVKlXSqq9mzZrit99+0/pRDUDcuXNHqkvfF5fmvjEyMhJTpkwRBw4cEEOHDtW7b/L+QDA3NxdfffWVOHTokFi5cqWIiIgQcXFxYsaMGWLLli0iNDRUnDhxQhw4cECMGzdO60teCCHS0tLE6dOntRKnmjVritOnT0sPXcdR7g+W1zkmND+vAISfn5/YsmWLWLNmjbC1tZVenzRpUqHeK833/o8//sg3/2U/qszMzMTevXsLta6ePXtKy3333XfS62vXrpVeb9u2rfR63h/Q7u7uYtWqVeLw4cNi8eLFQgghfQ5NTEzEkiVLxLFjx8SOHTvEvHnzRJs2baQE8XX98ccfWjH07NlT7Nu3T3zzzTf5fgBryrufgoKCxL59+8Svv/4qTp06JW7evCmMjY2l+Z9++ql0PtM812n+wfSq+y/Xhg0bpPmDBw9+7X3xpklrbGysUCqVUixubm5i8eLF4tChQ2Lt2rWiU6dO4uTJk0IIIa5evar1vdG9e3exb98+8cMPP2jV8eGHH+aLL/dRp04dsXPnTrFo0SLpBykAsWLFCiFETnJy+vRprWW+//57cfr0aXHhwoV8dWqeSzZu3Ki13CeffCIOHDggJk2apBW3ZtJ66NAh6XUTExPx5ZdfisOHD4uVK1cKJycnad7WrVulZfIm1q1btxZ79uzJt62HDh0SQuR8r37//fda8zTPR7kJp76k9eDBg2LRokVi165d4ujRo+L48eNi27Ztol69elL5efPmSeVfNWk9ffq0mDx5svR6mTJltOJLS0sT165dk+Y7ODiItLQ0qU5fX19pnuZ3mD6vuv+EEKJ+/frS656enmLdunXit99+E02bNpVel8vl4unTp0KI/Od3IyMj8fnnn4vQ0FCxYcMGcfXqVb3fAYmJiVrbf/r0aa0//y0tLcWZM2eEEELs379fq47hw4eLAwcOiDlz5mj98TlhwoQ32n591Gq1dGybmJjoLKN5XMlkMvHFF1+I0NBQERwcLP2uBKB1Xn6bSWufPn2k18uVKyfWrVsnDh8+rPXnrJOTk/THlaayZctKZcLCwl66PzQxaSV6R/ImrUII6R9kd3d30b59ewFAuLi4iPT0dL0nC81WhwEDBmito2/fvtK8unXrCiGEuHfvnvRatWrVxMWLF7X+IdSU92R97NixArcpIiJCfPbZZ6JSpUrCysoq3w9IAGLJkiUv3Td5v2h++OEHaV6HDh20fnzltohoJuMARHJyshBCiCVLlmh94Wn+W3348GGtZa5fvy6EEKJ69ep6T9qaX6yaP6bq1q0rvf7xxx9rfRkGBARI84KCgqRlCkpaX4fmF1evXr2k1+fOnau1nTdv3hRC5HwZara+5LZkCqH/iyu39R+A6N+/v9b6NfdBQUnr999/rzP+0NBQ0aVLF+Hm5qb1Y1fX+yqE/h+2ufT9YHmdYyJv0nr+/HlpmdxeEkDOj/zCqFy5srTMrVu38s0/efKkaNSokfj222/Frl27RGhoqJgyZYpWq6ezs7Pez27eunKXqVChgvS65mfpl19+kV7X3K8ymUxcuXIlX53u7u7Sj7uDBw+K+Pj4Qm13YX366adSDD4+PiIrK0ual7fVQ5Pm67reC81l69evr/U5nTFjhjTP19dXWuZV91+uAwcOSPM7dOhQqO0uTAuQ5qMwf2hp/rFpZmYmbt++rbfs559/rne/r1y5UppnbGwsfW40jxdTU1MRGRkpLaPZKjp+/HitdRW0Hfo+2126dNH7me/Vq5c0TzNp1fzToUuXLlrvueZnt0WLFtIymuc/R0dHrc9ZxYoVdZ7LCkokc+lLWl+8eCHmzJkj6tevL5RKpc6W7h49ehRqXfr2q+Y5LG9LdK4WLVpIZTZv3iyEEOL8+fPSa5UrV9a5XF6vuv80E2YAWq1y8fHxWn8yLl26VAiR//ye9/jSVUazt40mzWPbzMxMhIaGSvN69OghzWvatKnWcpp/BDg5OYns7OzX2v6CPHv2TCrv4uKis4zmcaX5h5IQ2sd/gwYNpNffVtKamJio1ZK8fPlyrc+YZmzr1q3LF7vmbyrN79XC4DWtRAY0ZswYAMDjx4+l6ydHjhxZ4CA1mtd55B1wQHNgi9xyZcuWRfv27QEA165dQ506dWBtbQ1vb2/06NFDuu4RAG7cuCFNOzg4FDhYzNOnT1G3bl18//33uHXrFl68eKGz3Ktcm5OrUaNGWnHkqlevHkxMcsaPc3R01Lkezf1TrVo1rZEoc+95liu37N27d3WuGwAaN26sM0bNfbVixQo0bdpUevz222/SvNxrcYuavn1ma2uLSpUqAQCMjIxgb28vzSvMe6N5bUruNXu5CjvgRY8ePfK9tnbtWrRv3x579uxBVFSU3hGzExISCrWOgrzOMaHJxsYG9erVk55r7t/XOb6FxrV9uZo1a4YzZ87gf//7HwICAtCuXTvMmjUL8+fPl8o8e/ZM61pffZo1a4YaNWoAAO7cuYNTp04hMTERR44ckeIPCAjQuay/v7/WrXhyjR49GgCQmpqKdu3awd7eHk5OTmjVqhUWL178xrcw0TzO3nvvPRgbG0vP3+Q40/ycnj9/XutzOn36dGnew4cPpev2Xnf/6XpfDUFzm6tVq4YKFSroLat5vOfd75rfJ2q1Wus8matixYpwd3eXnr/pZyOvNz0379mzR+s9X758uTRP37n5vffeg6WlpfT8bW+TEAIdO3bExIkTcf78eSQmJmpdV/021/Uyn332mTSdew34r7/+Kr2ma0C4lynM/tM87mQymdZ7aWdnh6pVq0rPdZ2TAd2f98LYtGkTRowYASBnfJEtW7agXbt2OtdX0G+smJgYaQwSTW/z+CnMOSVvjJrPdX1m39SdO3e0rukdMWKE1mdM83p5XZ+xNzlPMmklMqC2bdtKCQWQM/rpyJEj3+o6ZDIZfvvtN/z888/o3bs3qlWrBktLSzx69Ag7d+5Ejx49XuueWatXr8bTp08B5Iz8+P333+P48eM4ffq01rD7ur6MX0YzqdAcsEapVOpd5l3/YMw7IIw+KpWqiCPJUZz3ma7bDnz99dfSdP369bF9+3acPn0aW7Zs0Sr3OsfP26aZ6AOQ/jgBCr8PnZycpOlX+eGS94+C3M/cy2jeUmfVqlXYtWuX9MdA//799f4xpu8WERMnTsTBgwcxbNgw1K9fH/b29oiNjcWxY8cwduxYDBgwoFBxFSVdsRf2cwpof1ZfZ/9pDubl7OxcqHXWqlULp0+f1nrkDqDTvn37fPNq1apV6O15F97GZ+Nte9Nzc1Fv09mzZ3H8+HHp+dixY3H48GGcPn0aAwcOlF5/F+e+zp07w9vbG0DOYEX37t2TBqoyNTV9rc/1uzomXud2Nr/99hsGDx6M7OxsyGQy/Pzzz+jevftbiwl48+13cHCQvsPfxp+2L6P5eck7wFRMTEyB5V9G12fsdc6TuZi0EhmQTCbT+qezT58+L/0Qaya5Z86c0ZqneX+y3HLi/99eJCgoCFu3bsXVq1ehUqm0koONGzcCACpXriy9FhcXh5MnT+Zbf+5J99GjR9Jr7dq1w6efformzZujevXqiIyMLHAbipLm/rl+/TqSkpKk53/88YfWl0ZuWc1biZw9e1arPn0tW5rr+eabbyByLrfQeqjVaq1RNP+LNO8tmHff5N74/GV0fclpHj/Tpk1Djx490KRJkwJHnNZMxl/lB93rHBNvm+YfOXfu3Mk3/9y5czq3Ke8+dnNzK9T6+vbtK/142r59O3766SdpXkGtJ/p+kAgh0LZtW6xcuRLnzp1DXFwc7ty5I90OZ+fOndI9aF+H5nF2/vx5rX2heV4riK7YNd/Ptm3b6vycCiGgUqm0Rmd+nf13+/ZtaVrz/S6Ira1tvlGic2/94uzsnG9eYe5hqnkev3btms7WltxjXnP//PXXX1ojsWvud2Nj4yK7z2hB3vTc/PHHH+t9z3W1kr2KvCOAF/acpHnuc3BwwMKFC9G6dWs0adIEjx8/fqOY9MWnLzZjY2OMGjUKQM4xMWLECKmlrFOnTq+cVBSW5nskhNB6LxMTE3H9+nWdZTW9SvIEAIcPH0bv3r2l75ilS5dq/Umga30F/cZycnLK1+PrbTAyMpI+w5mZmfj3338LLJ83Rs3nhblNmp2dnTSt+dvt3r17Or+rypUrp9Uj4+zZs3rPqT/++KPWsqmpqdLxb2trK41OX1i8TyuRgQ0cOBBPnz6FEAL9+vV7afkhQ4bgn3/+AZDTzcXDwwPNmjXDqVOntBLR3B9WT58+xXvvvYeAgADUqFEDrq6ukMlkOHTokFQ298dm69at4evrKw3N3q1bNwQHB6NWrVqIi4vDoUOH8MEHH6B///7w8/OTlv/999+xYcMG2NraYt68ee/k30F9evfujUmTJuH58+dISUlBQEAAxo0bh+TkZEyePFkqV79+fakLUq9evaRuLBs2bIC3tzcaNmyI3bt349y5czrXM3ToUFy8eBEAMH36dDx//lzqvhYREYErV65gz5492LBhQ75bOOii+QV8/PjxQi3zLvTu3RtXr14F8H/7pkGDBti5cyf+/vvv167Xz88Pt27dAgAsXLgQpqamePDgAaZOnap3Gc1uVlevXsXOnTvh7OwMpVKp1Z1M1za86jHxtjVv3hxLly4FkJOU5U18goODERkZib59+6J+/fowNTXFmTNntLoHe3p66u0SmZelpSWGDh2KuXPnIi0tTfpRWLduXZ3df1+mcePGKFu2LJo0aQI3NzdYW1vj4sWL0mUB2dnZSEtLg6WlJcLCwuDr6yst+/DhQ+nWJvr07t0b33//PQDgwYMH6N+/PwYMGICrV69iyZIlrxxvrqCgICxcuBDZ2dk4dOgQ+vXrhw8//BByuRyPHz/GnTt3EBoaioCAAK3uwq+z/zTPFe+///5rx/ymevfujSlTpiApKQkZGRlo1aoVgoODUaFCBTx58gQ7duzAuHHj8P7772Pw4MFYtGgRsrOz8fDhQwQGBmLw4MGIiIjQ+mx07969UAnz29arVy/s3bsXQE5L4JgxY9C+fXv88ccfOm9dA+Scm3fs2AEAWLlyJczMzNCmTRuYmZkhIiICN2/exN69ezF58uR8t/h4FZrnIyDnPFa/fn0YGRkV+DnV/O6Mi4vD7NmzUbduXWzfvv2Vb5VW2PiioqKwfv16+Pn5wdLSEnXq1JHmDR06FCEhIUhNTcWxY8ek1z/66KO3Fkte1apVQ7169XDhwgUAOZ/TmTNnwtbWFgsWLJBa6ORyuc5bT72qs2fPolu3bsjIyACQczxXr15d609BLy8v6ZYtuZdNnTp1CiNHjkTXrl1x7do1rfNx7q1iikLz5s2lxP38+fNax0xe27dvx8SJE9G8eXOcOnVKOvaBnHPBy2hePnDq1CmMHz8eXl5e+P7773XeTlCpVKJnz5745ZdfAAA9e/ZEcHAwKlWqBJVKhUePHuHs2bPYv3+/dJu3XBcvXpT+NGjatGmBt37T6ZWugCWi16ZrIKaCFHTLG80BKHQ9NG95Ex0dXWBZANIIoUIIcfHiRa3Rf/M+cgeTePLkibC3t88339XVVWvggcIMOFTQ4An6BgkoaJnt27e/0u1Nnj9/rjXgkOZD81YymgOBqNVq0b9//5fuW82BMYpy9GDNQT4KGoBD3zKvc8sbzdcLGohJF82BMDQfLVu21Pu+3r59W+eAJa1atRJCvN1joqB9+LIBoXR5/vy5NOqwp6enNIBHrpfdPsrW1lbnqMMFCQ8P1xo5F4D48ccf85UrzPbou6VS7qNbt25S2cIOhpKXvlve5L21gqbCfG5+/PHHfPsh70PXeaqw+08IIRISEqTjq3z58oXaXn3exi1vDhw4kG/kdX37avHixa99y5u8x0tBg7oU9F7pq1OtVovWrVvrjEvzmMz7GdUcMEffozDnPyH0n7ezsrKEl5dXvnqNjY2lMrrOt2q1WjRp0iTfciYmJqJZs2avfE7Vt18TEhJ0HgNly5YVeeUdDd7NzU1rUK6XeZ39d+fOnTe65Y2u84q+Mi+7xVDe2DQHKNP1KOiWN4Xd/oJojqbet2/ffPM1jyvN31uaj/r164v09PSXxpGcnKw1qnbuQ6lUCk9PT53LxMbGag1gqe+Rl+atlDZt2lSofaGJ3YOJ/mOMjIzwyy+/YNOmTfjggw9gb28PY2Nj2Nvbo1WrVti4cSO2bt0q/YOlVCoxa9YsdOjQAT4+PpDL5TA2NoaTkxM++OADbN26VauLcu3atXHjxg1MnjwZNWvWhFwuh5mZGTw8PNC1a1fUrFkTAODi4oJTp06hbdu2UCqVUCqV6N69O86cOaPVzc4QevTogYsXL2LIkCHw8fGBmZkZLC0tUbVqVUycOBFXrlzR6jZjZWWF48ePY/jw4XBycoK5uTnq1KmDbdu2ITAwUOc6jIyMsGHDBuzYsQMdO3aEi4sLTExMYG9vjypVqmDQoEHYvn07GjZs+K42u0hYWlri+PHjGDFiBJycnGBhYYG6deti9+7dWq3B1tbWr1TvsGHDsGrVKlSpUgUWFhbw8fHBV199hRUrVuhdpkKFCti0aROqV69e4GBlurzqMfG2WVlZSV3RIiIicOLECa358+fPx5QpU/Dee+/Bw8MDZmZmsLa2RtWqVfH555/jxo0bhW5lzeXl5YUuXbpIzy0tLdGnT5/Xin/SpEkIDAxEhQoVoFQqYWRkJA1Q9c0332j18tD8d14mk8Hc3LxQ6/jxxx/x7bffws/PD2ZmZihbtizmzp2LKVOmSGWsrKxeOfYRI0bg3LlzGDhwIHx9fWFubg65XI5y5cqhc+fOWLFiBT755JN8y73K/tuxYwfS09Ol9Rla+/btce3aNXz22WeoVKkSrKyspM9ZYGCgVkv4Z599hj/++AO9e/eGu7s7TE1NIZfLUbt2bXz99dc4f/681jXZ75KRkRF2796NCRMmwM3NDWZmZqhcuTJ++OEHTJw4Ue9ys2fPxrFjx9CzZ09pm2xtbVGxYkX06tULGzZseONrGY2NjfHbb7+hRYsWUjf5V9mmoUOHwtXVFVZWVmjUqBEOHTpU4OCHr0qpVOK3335DgwYNYGFhUWBZzWu4AWDQoEFaXUCLQvny5XH16lVMnjwZ1apVg5WVFczMzODt7Y2goCBcvHjxrV9vWljz5s3DgQMH0LlzZzg7O8PExAS2trZo1KgRli1bht9//11rsKW3rXHjxtIlBrt37y5wbIzg4GCsWLECVatWhZmZGVxcXDB69GgcOXKkUN+TNjY2CA0Nxfvvvw8LCwvY2tqiR48eBbbwOjg44Ny5c1i4cCEaN24MpVIJU1NTlClTBnXq1MH48ePzfccJIbB161YAOV2re/bsWci9oV0JERGRXnlbBXNf07ypua7bD5C2iIgI6RY2AQEB72Sd69evl96jvLcsKipbt26V1jlq1KhCL6frOBNCiLFjx0r11a5d+22FWSiF2X/Z2dlSrwNXV1ehUqneaYxEb4PmfYvv3btn6HBKvR07dkjvx6JFiwwdzhvbuXOntD0LFix4rTrY0kpERAUKDAzE3Llz8ddffyEyMhLnz5/HgAEDpGurjYyMdA5oQdo8PDzw+eefA8j59zz3WuG3Tfz/QTAePnyo1XL9tkcm1yf31jC5LaWF9c0332Ds2LE4duwYHj16hOvXr+Prr7/WGt08d2TdovSq+2/v3r24cuUKAGDmzJmv3OuAyFDS09ORmJiIjRs3SgP+tG7dGv7+/gaOjLp37y7dvmbu3LlST47/qlmzZgHI+V7Q1bOlMGRCFJMbixERUbHUpEmTfCMU5jIyMsLChQu1upiTYeUdCAkAAgICsGvXrneyfm9vb0RGRuLkyZOFvscqAEydOhWzZ8/WOz8wMBAbN24s8m6Lht5/RO/K4MGDsW7dOum5iYkJzp49i7p16xowKiLdOHowEREVaMCAATA3N8etW7cQFxcHY2NjeHh4oEmTJvjkk0+0RqKk4kMmk6FMmTLo3r075syZ887Wq3lz+VfRsmVL3Lx5E5cuXcKzZ8+QlZUFR0dH1K1bF4MHD0a3bt3ecqQFM9T+I3rXLC0tUa1aNcycOZMJKxVbbGklIiIiIiKiYovXtBIREREREVGxxaSViIiIiIiIii0mrURERERERFRsMWkl+g/IyspCZGQksrKyDB0KEREREdE7xaSV6D/gyZMn8PT0xJMnTwwdSrGVHvcAV2YqcWWmEulxDwwdTon3b/gTuFYdBNeqg/BvOI/LovYwKQE+q76Dz6rv8DApwdDhEBERvVNMWomIiIiIiKjYYtJKRERERERExZaJoQMgInobZCZmsHStJU1T0TIzM0WNKr7SNBUtM2NjVHcsI00TERGVJjIhhDB0EERUsMjISHh6eiIiIgIeHh6GDoeIiIiI6J1h92AiIiIiIiIqtpi0EhERERERUbHFa1qJqETIeh6LZ2cXAwCc3xsDE2tHA0dUssXGJ+OHnw8AAEYN6QBHe4WBIyrZ4lJfYMXV8wCAj6vXh4OllYEjIiIieneYtBJRiaBOS0LM2SUAAIdag5m0FrHklBf4cW0oAKD/h82ZtBax5Ix0rLx6AQDQp2INJq1ERFSqsHswERERERERFVtMWomIiIiIiKjYYtJKRERERERExRavaSX6D/l22RYobO0NHUaxJBcx6PD/pxes3AaVzOmN6jt9u+6bB/WWnVpbw9AhEBEREb1zbGklIiIiIiKiYotJKxERERERERVb7B5MRCWCgAwZsJCmqWgZyWRQ2FhJ01S0jGSAjZm5NE1ERFSaMGklohLhucwRv5nONnQYpYaPlwvunP3R0GGUGt4KO1wb9JmhwyAiIjIIdg8mIiIiIiKiYotJKxERERERERVb7B5MRCWCqUiFp7gCAIiQ1UCmzNLAEZVsySkvsOfQeQBAl7b1petbqWgkZ6Rj/7+3AQAd/SpC8f+vbyUiIioNmLQSUYlgDhXqqrcBAJ6ZlEUmmLQWpdj4ZEwIWQMAaFSvIpPWIhaX+gKTTh8GADR09WLSSkREpQq7BxMREREREVGxxaT1P8rHxwcHDx4sdPlHjx5BLpcjPT29CKMqXYYNGwZ7e3tUqFABALBnzx54e3tDLpfjxIkTqFKlCo4ePWrgKImIiIiI/tuYtBah5s2bw8LCAnK5HA4ODmjXrh3u3r37TtYtk8lw+/Zt6bmXlxdUKhXMzUt3l7KwsDDIZDKkpaW9UT1nzpzB3r17ER4ejjt37gAAPv/8c8ybNw8qlQrNmzfHjRs38MEHH7yNsImIiIiISi0mrUVs0aJFUKlUiIiIgJOTE4YMGWLokEqUrKwsg9T58OFD+Pj4wMbGRuu1atWqvfV4iIiIiIhKMyat74iVlRUCAwNx6dIl6bUnT56gd+/ecHFxgaenJ0JCQpCdnQ0gJwFq1aoVHBwc4OjoiD59+iAhIaFQ62rUqBEAoE6dOpDL5Vi5cmW+FsbmzZtjypQpeP/992FtbY3mzZvj2bNn+Pzzz2Fvbw8/Pz+cOHFCqjM5ORkjRoyAh4cHypQpg9GjR+ttrczOzsakSZPg5OQELy8vrF69GjKZDGFhYdK6ly9fLpU/ePAgfHx8CrVf1q5di4YNGyI4OBjOzs745JNPUK1aNfz6669aMVSuXBnbt2/Xu28cHR0hl8tx8OBBnDhxAmXKlMGiRYvg7u6OTp06AQD69OkDV1dX2NraomnTprh27RoAYOXKlRg6dCguXLgAuVyOYcOGQS6XQ61Wo06dOnB0dASg3YU7Ozsb8+fPR/ny5WFjY4MqVargn3/+efmbSURERERUyjFpfUdSUlKwadMm+Pv7A8hJYrp06QJ/f3+Eh4fj3Llz2L17N1avXg0AEEIgODgYUVFRuH37NqKjozFt2rRCrevPP/8EAFy8eBEqlQrDhw/XWW7Tpk1YsWIFYmJioFar0bBhQ1SpUgUxMTEYNWqU1nJBQUFIS0vDzZs3cfv2bdy7dw8zZ87UWe/q1auxbds2nDt3Djdu3MDevXsLvZ9etl8A4O+//4ajoyMeP36MRYsWISgoCOvXr5fmX7hwAU+fPkWXLl307pvY2FioVCq0a9dOev7w4UM8ePAAu3btAgC0adMGd+7cwbNnz1C/fn306dMHADB8+HAsX74c9erVg0qlwqpVq6BSqaR9Hhsbm2+933//PVatWoWdO3ciOTkZv/32GxwcHPTuh+TkZERGRkqP6OjoQu9DIiIiIqKShElrERs/fjxsbW2hUChw/vx5bNy4EUBO4hUREYFZs2bBwsICbm5uGD9+PLZs2QIA8PPzQ5s2bWBubg5HR0eMGzcOJ0+efKuxDR48GBUrVoSVlRW6desGMzMzDBkyBMbGxujbty/u3bsHlUqFZ8+eYc+ePViyZAkUCgWUSiWmTp0qxZrXli1bMGbMGPj5+cHGxgYzZswodEwv2y8A4OzsjC+++AKmpqawtLRE//79cezYMTx79gwAsH79egQGBsLMzKzQ6xVC4JtvvoGFhQUsLXNulRIUFASFQgFzc3N8+eWXuHHjBuLi4gpdp6bly5dj1qxZqFq1KmQyGcqVKwdvb2+95RcsWABPT0/pUb9+/ddaLxERERHRfx3v01rEFixYgBEjRuD+/fvo0KED7t27h2rVqiEsLAwxMTGws7OTymZnZ8PT0xMA8PTpU4wZMwanT59GSkoKsrOzta6f1CSXy6Xp0NBQNG3atFCxlSlTRpq2srLK9xwAVCoVHj16BLVaLcUG5CR5arVaZ71RUVHw8vKSnheUnOX1sv0CAJ6enpDJZNJzZ2dntGnTBps3b8Ynn3yCrVu3Yv/+/YVeJ5DTXTh3mwFArVZjypQp2LZtG2JiYmBklPP/TmxsbIEtpPo8evRIamUvjPHjx2Po0KHS8+joaCauL/Ec9thtkvMHSQbv0VrkvNydcO3UEgCAUmFt4GhKPk8bW1zs/wkAwNbcwsDREBERvVtMWt8Rf39/LF68GB999BHat28PLy8veHh4SNd55jV58mRkZ2fj6tWrcHBwwO7du/Hxxx/rLJvbNbWoeHl5wcTEBM+ePStU66WbmxsePXokPdecBnKS7BcvXkjPnzx5orWugvYLAK2ENVdQUBBmzJgBX19fODk56U3wdC2r6/XNmzdj586dOHLkCHx9fZGcnAylUgkhhN64CuLl5YX79++jZs2ahSqvUCigUChea12llZAZIx3ylxekt8LExBiO9jxG3xUTIyM4WFq9vCAREVEJxO7B71D79u3h4uKCFStWoF69enBycsLMmTPx/PlzZGdn4969e1IX4JSUFFhbW8PW1hZRUVGYN2/eK63LxcUFDx48eCtxlylTBh07dsSYMWOQkJAAIQQiIiL03ie2d+/e+P777/Hw4UOkpKQgJCREa36tWrWwfft2aVTlJUuWSPNetl/06dixIx4/fowvv/wSgwYN0lvOyckJRkZGL903KSkpMDc3h4ODA1JTUzF16tQCy7/M8OHD8eWXX+LmzZsQQuDevXsIDw9/ozqJiIiIiEoDJq3vWHBwML799ltkZGRg7969uHfvHsqVKwc7Ozv06tVLGnBn+vTpuHr1KpRKJTp06ICAgIBXWs+MGTMwdOhQKJVKrFq16o3jXrduHUxNTVGzZk3Y2tqibdu2eu85O3ToUHTr1g3169dH5cqV0bFjR63548aNg62tLVxdXdGtWzf069dPmmdsbFzgftHHxMQE/fv3x/Xr19G/f3+95aysrDB16lS0aNECSqUShw4d0llu4MCB8PPzg7u7OypVqoR69eoVuP6XGTNmDAYOHIjOnTvDxsYG3bt3R3x8/BvVSdqMRQacsu/DKfs+jEWGocMp8V6kpuPP87fw5/lbeJGabuhwSrzUrEycjXqEs1GPkJqVaehwiIiI3imZeN3+jkSFlJaWBktLS+nepkXl+++/x/79+/Umov9lkZGR8PT0xOiJc6GwtTd0OMWSXMSgQ9YcAMABk4lQyZzeqL7Tt+u+jbDeqlNraxg6BMm/4U/QuGMwAODM/m/h513mJUvQm3iYlIAWv/4EADjeayh8be1esgQREVHJwZZWKhGSk5OxYsUKjBgxwtChEBERERHRW8Sklf7z1qxZA1dXV9SsWfOVu1ETEREREVHxxtGDqchZWFi89qi7hREUFISgoKAiq5+IiIiIiAyHLa1ERERERERUbDFpJSIiIiIiomKLSSsREREREREVW0xaiYiIiIiIqNjiQExEVCKkQonfjUdL01S0XF3ssXv9FGmaipartRzbO/eRpomIiEoTJq1EVCKoZaaIk/kaOoxSw9LCDPVrlzd0GKWGhYkp6pbxMHQYREREBiETRXkvEiJ6KyIjI+Hp6YmIiAh4ePCHKxERERGVHmxpJaISQWRnIUv1DABgIneGzIint6KUlaVGTFwSAMDJwRYmJsYGjqhky8rORmzqcwCAo6U1TIw4JAUREZUe/NYjohIhIyEcNxdXxs3FlZGREG7ocEq8R49jULvVONRuNQ6PHscYOpwSLyIlCQ03L0fDzcsRkZJk6HCIiIjeKSatREREREREVGwxaSUiIiIiIqJii0krERERERERFVtMWomIiIiIiKjYYtJKRERERERExRaTViIiIiIiIiq2mLQSERERERFRsWVi6ACIiN4GE7kTPLssk6apaDk52GLRrKHSNBUtR0srfPd+e2maiIioNJEJIYShgyCigkVGRsLT0xMRERHw8PAwdDhERERERO8MuwcTERERERFRscWklYiIiIiIiIotJq1EVCKkx/+La3PccG2OG9Lj/zV0OCXew0dP4VdvGPzqDcPDR08NHU6JF5aUgEprFqHSmkUIS0owdDhERETvFAdiIqKSQQhkZ76QpqloCSGQmpohTVPREgBSszKlaSIiotKELa1ERERERERUbDFpJSIiIiIiomKLSSsREREREREVW0xaiYiIiIiIqNhi0kpERERERETFFpNWIiIiIiIiKrZ4yxsiKhGMLZVwbvy5NE1Fy1Zhjc+GdZamqWjZmpvjk5oNpWkiIqLSRCZ4gz2iYi8yMhKenp6IiIiAh4eHocMhIiIiInpn2D2YiIiIiIiIii12Dyb6D/l22RYobO0NHUapcPp2XUOHQFSkTq2tYegQiIiICoVJKxGVCFYiHo2z1gAAzpgE4YWMyX1RykxPRtT93QAAN/+uMDVXGDiiki3LPBUxFS8DAJxu14RJuqVhAyIiInqHmLQSUYlgBDXsECVNU9ESQo2M1BhpmoqWkGUjU54iTRMREZUmvKaViIiIiIiIii0mrURERERERFRsMWklIiIiIiKiYotJKxERERERERVbTFqJiIiIiIio2GLSSkRERERERMUWb3lDRCVCFswRIasuTVPRMjIyhdyunDRNRctIbQLLWBdpmoiIqDThN99/TGBgICpWrIiQkBCcPn0agwcPxoMHDwwdFpHBpckUOGsyyNBhlBomZnK4lu1s6DBKDeNMczjdqWHoMIiIiAyC3YPfsm3btqFhw4aQy+VwdnZG8+bNsXfv3iJZV9OmTbUSVh8fHxw8eLBI1vW2rF27Fg0bNnzt5TMzMzF8+HDY29vDwcEB3bt3f4vRERERERFRccOk9S1avHgxRo4cifHjxyM6OhrR0dGYNm0afvvtN53ls7Ky3m2ABl7/21jfhg0bcPz4cdy5cwePHz/Gxx9//BYiezOGfh+JiIiIiEoyJq1vSXJyMqZMmYJly5ahV69esLGxgbGxMVq1aoXVq1cD+L9WxuDgYDg7O+OTTz6BEAILFixA+fLlYW9vjw4dOiAyMlKq9/jx46hSpQpsbGwwcOBAZGRkSPNOnDiBMmXKAAD69OmDR48eoVu3bpDL5ZgyZUq+GMPCwiCTybB27Vr4+vqievWc6/8OHTqEunXrQqlUonbt2jh9+rS0TPPmzTFp0iQ0atQINjY2aNWqFSIiIqT558+fx3vvvQdbW1tUq1YN+/btk+aFhISge/fuGDJkCJRKJb755huMGDECFy5cgFwuh1wux/Pnz/H333+jQYMGUCgUcHJyQr9+/fTuZxMTE9ja2sLR0REWFhZo27btS9+bDRs2wN/fHzY2NvD09MTChQuleaGhoahbty5sbW3h6emJtWvXAgAyMjIwYcIEeHh4wMXFBYMHD0ZSUpK0nEwmww8//ICKFStCqVQCAC5cuIBmzZrBzs4OlSpVws6dO6XyBw8eRLVq1WBjY4MyZcpgwoQJL42bXo2FSMZ7WevwXtY6WIhkQ4dT4mVlqBD9YC+iH+xFVobK0OGUeGrTdMRUuIKYClegNk03dDhERETvFJPWt+TPP/9EWlraS7ur/v3333B0dMTjx4+xaNEiLF26FJs2bcLhw4fx9OlT1K5dG4GBgQCA+Ph4dO3aFRMnTkRCQgJatWqFPXv26Kx3y5Yt8PLywq5du6BSqTB79my9MRw8eBBXrlzBxYsXceXKFfTt2xeLFi1CfHw8vvrqKwQEBCA2NlYq//PPP2Pp0qWIiYmBv78/+vfvDwBISEhAu3btMGTIEMTFxWH+/Pno3bs3bt26JS27d+9etGnTBvHx8fjiiy+wfPly1KtXDyqVCiqVCtbW1vj000/RpUsXJCYmIiIiAqNGjdIbe506dXD58mWMHTu2wP2c6/nz5xgyZAjWrFmDlJQUXLlyBc2bNwcAXLx4Eb169cKMGTMQHx+PS5cuoUaNnGvGvv76a/z+++84f/487t69i/j4+Hxx/frrrzh16hRiYmIQHR2Ndu3aYfz48YiNjcXatWsxdOhQaV8EBQXhf//7H1JSUnD//n307NmzwLiTk5MRGRkpPaKjowu1vaWZCdLhKa7CU1yFCfijvqhlZ2dClXAPqoR7yM7ONHQ4JV62cRZSHZ8i1fEpso3Zu4OIiEoXJq1vSVxcHBwdHWFqWvAoms7Ozvjiiy9gamoKS0tL/Pjjj5g1axZ8fHxgamqKkJAQXLhwAY8ePcK+fftQrlw5DBgwACYmJhg0aBCqVav2xrGGhIRAoVDA0tISK1aswNChQ9GkSRMYGRmhU6dOqFmzJg4cOCCV79+/P2rXrg0LCwt8++23OH36NCIjI7F//354e3tj2LBhMDExQZs2bdC5c2ds3rxZWrZOnToIDAyEkZERLC0tdcZjZmaG8PBwREVFwcLCAo0bN9ZZLj4+Hu3bt8fu3btx/fp1jBo1CkIIAECnTp3www8/6FzO1NQUN2/eRHJyMuzt7VGrVi0AwKpVqzBo0CB07NgRxsbGcHR0lOZt3LgR06dPh5ubG2xtbTF37lz88ssvWi3dEydOhLOzMywtLbFhwwZ88MEHCAgIgLGxMRo0aIBu3bph27Zt0jbev38fsbGxkMvlaNCgQYHv0YIFC+Dp6Sk96tevX2B5IiIiIqKSiknrW+Lg4IDY2FhkZhbc4uDp6QmZTCY9DwsLQ+/evaFUKqFUKuHo6AgjIyNERkYiKioKXl5eWst7e3u/cayadYSFhWHJkiXS+pVKJf766y9ERUVJZTRjUCqVUCgUePz4MR4/fgwfHx+tun18fPD48eNXivfnn3/GixcvULt2bVSpUgU///yzznLbtm1DuXLl0LFjR+zZswdXrlzB0KFD8eLFC1y8eBFt2rTJt4y1tTX27NmDXbt2wdPTE++//z7Onj0LAHj06BH8/f11rivvtvn4+ECtVuPJkyc6ty0sLAy7d+/W2o+//PKL1EK6a9cuXLt2DeXKlUO9evW0ulHrMn78eEREREiP8+fPF1ieiIiIiKik4i1v3pJGjRrBwsICu3btQq9evfSW00xYgZyEcPny5VKXVU3379/Ho0ePtF579OgRatasWai6CxODl5cX/ve//yEkJERvec0YEhMTkZycDHd3d7i7uyMsLEyrbFhYGMqVK6c3Jl0xli1bFhs3boQQAidPnkSbNm3QrFmzfAllVlYW0tNzun1aW1vjwIEDaNWqFWrXro1WrVrpTUA/+OADfPDBB8jIyMCiRYvQq1cvREREwMvLC/fv39e5TO625XYXDgsLg5GRkXQNcd5t8fLyQmBgoHRNbF61a9fGzp07oVar8csvv6Bnz56Ii4uDtbW1zvIKhQIKhULnPCIiIiKi0oQtrW+JQqHA7NmzMXr0aGzfvh0qlQpqtRonT57EsGHD9C43cuRITJkyRbp1TUJCAn799VcAQMeOHXH37l1s3rwZWVlZ2LBhA65du6a3LhcXl1e+Z+vw4cOxcuVK/PHHH8jOzkZqaiqOHz+uNRjUpk2bcPnyZaSlpWHSpElo3LgxPDw80KFDB4SFheHnn39GVlYWjh49ir1796Jv374Fxvj48WMp+QSA9evX49mzZ5DJZFAqlZDJZDA2Ns63bMeOHXH16lUsXLgQ6enpEEKgZcuWuHPnDmxsbHSu7+nTp/jtt9+gUqlgamoqDZAFAEOHDsX69esRGhoKtVqN2NhYXL58GQDQr18/zJw5E9HR0UhOTsbEiRPRu3dvmJmZ6VxP//79ERoair179yIrKwsZGRk4d+4cbt26hYyMDGzYsAEJCQkwNjYucBuJiIiIiEgbk9a3aMyYMVi6dCnmzZsHFxcXuLq6IiQkBAEBAXqX+fTTTxEYGIhOnTpBoVCgRo0aOHToEICcLse7du3CrFmzYGdnhyNHjqBz585665o0aRLmzp0LpVKJadOmFSrm2rVrY926dZgwYQIcHBzg7e2N+fPnIzs7WyozePBgjBo1Ck5OTrhz5w42bdoEALC3t8eBAwewYsUKODg4YOzYsdi8eTMqVaqkd30tW7ZEzZo14erqCqVSiefPn+PIkSOoVq0a5HI5PvzwQ/zwww/w9fXNt6yPjw+OHDmC3bt3w8XFBeXKlUNsbCwuXryIvXv34ptvvsm3THZ2NhYtWgQPDw8olUqsXbtWuua2bt262LRpEyZPniyNnHzlyhUAwOTJk/H++++jbt268Pf3h0Kh0HvNLAB4eHjgwIEDWLRoEVxcXODm5oZJkyZJyfnmzZtRtmxZ2NjYIDg4GL/++issLCwK8Q4REREREZVuMpE7kg2RDs2bN0dgYCBGjBhh6FBKtcjISHh6emL0xLlQ2NobOpxiSS5i0CFrDgDggMlEqGROb1Tf6dt130ZYJVZGWgLCr68BAHhXDYKZhZ2BIyrZMi2eI7rOGQCA68XGME3TfWnBqzi1tsYb10FERPQu8JpWIioRsmGMBLhJ01S0ZDJjmFk6SdNUtGTCCKYqG2maiIioNGHSSkQlwguZPY6Yfm7oMEoNU3MFvKsMMHQYpYZJuiVcr7xn6DCIiIgMgkkrFejEiROGDoGIiIiIiEox9jEiIiIiIiKiYostrURUIpgJFcpnnwIA3DVqhgyZ3MARlWzqzFQkPL0IALBzqQNjU0sDR1SyqU0ykOIWDgCwifKGcZbu228RERGVRExaiahEMEMqKmf/DgAIM6qHDDBpLUpqdRoSnpwHACgcqzBpLWLZJplI9nwIALB+5saklYiIShV2DyYiIiIiIqJii0krERERERERFVsyIYQwdBBEVLDIyEh4enoiIiICHh4ehg6nWEqPe4DbP9QBAFQcdRHmDmUNHFHJ9m/4EzTuGAwAOLP/W/h5lzFwRCXbw6QEtPj1JwDA8V5D4WtrZ+CIiIiI3h22tBIREREREVGxxaSViIiIiIiIii0mrURERERERFRs8ZY3RFQyyGQwMrWSpqloyWQyWFqaSdNUtGQALE1MpWkiIqLShAMxEf0HcCAmIiIiIiqt2D2YiIiIiIiIii0mrURERERERFRs8ZpWIioR1OnJSLq9FwBgW7EzjM0VBo6oZEtRpeLA0b8BAB0+qAsbuaWBIyrZUjLScTDsHgCgnU852JiZGzgiIiKid4dJKxGVCFmqGETs+QQAYO3RkElrEYuJS8LYqT8BAOrVKsektYjFpr7AhJOhAIC6Lu5MWomIqFRh92AiIiIiIiIqtpi0EhERERERUbHFpJWIiIiIiIiKLSatREREREREVGwxaSUiIiIiIqJii0krERERERERFVtMWomIiIiIiKjYkgkhhKGDIKKCRUZGwtPTExEREfDw8DB0OMWSyM5CluoZAMBE7gyZEW9DXZSystSIiUsCADg52MLExNjAEZVsWdnZiE19DgBwtLSGiRH/cyYiotKDv+qIqESQGZnAVOFm6DBKDRMTY7i62Bs6jFLDxMgIZaxtDB0GERGRQfCvWiIiIiIiIiq22NJKRCVCdmYqUp9cAQBYlqkBI1NLA0dUsqWmZeDazTAAQLXKPrC0MDNsQCVcWlYmrsc+BQBUdXSBhYmpgSMiIiJ6d9jSSkQlQmZyFO6vbYf7a9shMznK0OGUeNFP49F14Gx0HTgb0U/jDR1OiRf9XIWee7eg594tiH6uMnQ4RERE7xSTViIiIiIiIiq2mLQSERERERFRscWklYiIiIiIiIotDsRE9B/y7bItUNjyNiO6yEUMOvz/6QUrt0Elc3qj+k7frvvmQZVgGWkJ0nTf4Nsws3hqwGhKvkyL50CdnOm+wbdgmmZt2IDolZ1aW8PQIRAR/WexpZWIiIiIiIiKLSatREREREREVGwxaSUiIiIiIqJii9e0ElGJkApbHDceKU1T0TIxlcO9/IfSNBUt4wwLOF+rK00TERGVJkxaiahEUMvMECPzN3QYpYaRsSmsFJ6GDqPUMMo2hkUyB2EjIqLSid2DiYiIiIiIqNhiSysRlQgyoYYZUgEAGbCEkBkbOKKSTYhsqLPSAADGJhaQyfgfaFESyEa2SRYAwCjLBDL+50xERKUIk1YiKhGsEY8OWXMAAAdMJkKFN7tPKxUsMz0J4dfXAAC8qwbBzMLOwBGVbFkWqYiucwYA4HqxMe/TSkREpQr/qiUiIiIiIqJiq1gmrYGBgQgJCQEAnD59GmXLli30smFhYZDJZEhLS9M5f9OmTWjZsuXbCLNQ5HI57t69+9brXbduHYYNG/bW631Xvv76awwePPil5QYPHoyJEycWfUDvyPbt29G3b19Dh0FERERE9J/xWknrtm3b0LBhQ8jlcjg7O6N58+bYu3fv244NANC0aVM8ePBAeu7j44ODBw++dn39+vXDsWPHXlouJCQEgYGBr72eXCqVCuXLl3/jejRlZWVh+vTpWslc3v1y+PBh2NnZ4ddff5XmW1paSu/Zhx9+iCdPnkjlk5KSMGzYMLi5ucHGxga+vr4YN25cgXHcvHlT77a9bH2TJ0/G2rVrX2fzC/TPP/+gS5cusLe3h1KpRM2aNbFw4UKo1eo3rvtNjz0A6NGjBy5duoRr1669cTxERERERKXBKyetixcvxsiRIzF+/HhER0cjOjoa06ZNw2+//aazfFZW1pvG+J9UlNu9b98+eHh46G2B3rVrF3r16oUNGzagV69eWq+rVCrcvn0bCQkJ+OKLL6R548aNQ1xcHK5evYrk5GQcO3YMtWrVemkcnTp10ju/oPUVhXPnzqFZs2aoX78+bt26hcTERGzevBnnzp1DSkpKka67MLKysiCTydCvXz+sWLHC0OEQEREREf0nvFLSmpycjClTpmDZsmXo1asXbGxsYGxsjFatWmH16tUAgLVr16Jhw4YIDg6Gs7MzPvnkEwghsGDBApQvXx729vbo0KEDIiMjpXqPHz+OKlWqwMbGBgMHDkRGRoY078SJEyhTpgwAoE+fPnj06BG6desGuVyOKVOm6I1127Zt8PX1hZ2dnVaLYW58ACCEwIQJE+Di4gKFQoGKFSvixIkT2LdvH77++mvs2LEDcrkcPj4+0vZ/9NFHKFOmDDw8PDB+/Hikp6cD+L9uyWvXroWvry+qV68OAJDJZLh9+zYAICMjA5MnT4avry8cHR3Rt29fJCQkAADS0tIwePBgODg4wNbWFjVq1MDNmzd1btu+ffv0dnHesGEDgoKCsHPnTr0Jpb29Pbp164ZLly5Jr50/fx59+vSBo6MjZDIZfH19MXDgQL37NzeOgpLWgtan2ZJd2G1PS0tDt27d0KdPH2RmZuabP2HCBAwYMABTp06Fi4sLAKBy5crYunUrlEolAODChQto1qwZ7OzsUKlSJezcuVNafvDgwRg1ahS6d+8OGxsbVK9eHZcvXwag/9i7d+8e2rdvD0dHR5QtWxY//PCD1jZ2794dQ4YMgVKpxHfffQcAaNGiBfbt2/fS/UZERERERK+YtP75559IS0tD9+7dCyz3999/w9HREY8fP8aiRYuwdOlSbNq0CYcPH8bTp09Ru3ZtKWGJj49H165dMXHiRCQkJKBVq1bYs2ePznq3bNkCLy8vqQVv9uzZemM4evQorl+/josXL2LNmjU6uwQfPnwYW7duxeXLl5GcnIzQ0FB4eXmhU6dOmDx5Mnr06AGVSoWwsDAAwGeffYbHjx/j9u3b+Pvvv3HmzBl89dVXWnUePHgQV65cwcWLF/Otb9KkSfjnn39w7tw5REREwMzMDKNHjwaQc43qtWvX8ODBAyQmJmLr1q2wt9d9I/mrV6+iYsWK+V5fvXo1xo4di9DQ0AKv242JicGOHTvg7+8vvdaoUSNMmzYNP/30k95kWVNCQgJu3LiBpk2bvrSsrvVpKsy2JyUloV27dnB1dcWmTZtgamqqNf/Fixc4c+YMPvzwQ71xREdHo127dhg/fjxiY2Oxdu1aDB06FLdu3ZLKbN68GePHj0diYiJatmyJzz77DIDuY+/Fixdo1aoVunTpgujoaBw4cABz5szBkSNHpPr27t2LNm3aID4+HmPHjgUAVKpUCeHh4dIfFrokJycjMjJSekRHR+stS0RERERUkr1S0hoXFwdHR8d8CUNezs7O+OKLL2BqagpLS0v8+OOPmDVrFnx8fGBqaoqQkBBcuHABjx49wr59+1CuXDkMGDAAJiYmGDRoEKpVq/ZGGwUAM2bMgLW1Nfz8/NCsWTP8888/+cqYmZkhLS0NN27cQGZmJnx9feHn56ezPrVajS1btuDbb7+FUqlEmTJlMGPGDKxfv16rXEhICBQKBSwtLbVeF0Jg+fLlWLhwIZydnWFpaYmZM2di27ZtUKvVMDMzQ0pKCm7fvg0hBCpVqiS1MOeVkJAAhUKR7/VDhw6hVq1aqFu3rs7levbsCYVCAWdnZyQnJ2Pp0qXSvO+//x5BQUFYuXIlatasCQ8Pj3zbpik0NBStWrUq8FgoaH2aXrbtT548QfPmzdG0aVP88MMPMDLKf9gmJCQgOzsbbm5ueuPZsGEDPvjgAwQEBMDY2BgNGjRAt27dsG3bNqlMQEAAmjRpAmNjYwwcOFDncZNr3759cHV1xciRI2FqaooKFSpg2LBh2LJli1SmTp06CAwMhJGRkXRM2NjYSDHrs2DBAnh6ekqP+vXr6y1LRERERFSSvVLS6uDggNjYWJ1dMzV5enpCJpNJz8PCwtC7d28olUoolUo4OjrCyMgIkZGRiIqKgpeXl9by3t7erxKWTppJj5WVFVQqVb4yLVq0wIwZMzB58mQ4OTkhMDAQUVFROuuLjY1FRkaG1FUYyBmYJzo6GkKIl8YeExODFy9e4L333pP2Q7Vq1WBkZIQnT55gwIABGDRoEIYPHw5nZ2cMHz4cycnJOuuys7PTOe+HH35ATEwMevXqpfM92r59O5KTk3HhwgU8efIEjx8/luZZWFggODgY58+fR2JiIiZMmICgoCCtVkhN+/fvf2nX4ILWp+ll2x4aGopnz55JLZW62NnZwcjISO/7B+Qch7t375b2v1KpxC+//KLVipn3uHn+/HmB9V26dEmrvrlz52oNOKXreMi9vtbOTv99LcePH4+IiAjpcf78eb1lKUc65Pjb+EP8bfwh0iE3dDglnrGJFZy9P4Cz9wcwNrEydDglnnGmGezvV4b9/cowzjQzdDhERETv1CslrY0aNYKFhQV27dpVYDnNhBUAvLy8sGfPHiQmJkqP1NRUNGrUCG5ubnj06JFW+bzPC6r7TY0aNQrnz5/Hw4cPkZWVheDgYJ3rcXR0hJmZmdRVGMhJWlxdXbXK6ovP0dERlpaWuHz5stZ+SEtLg7u7O0xMTDBlyhRcvXoV165dw82bN6VrIPOqXr26dJ1s3nX8/vvvuH//PgIDA/UOBlW3bl1MmTIFo0aN0kq4c1lZWWHMmDGws7PT2VVYrVbj8OHDaN++vc76X3V9L9v2oKAgBAYG4oMPPkB8fLzOdVhZWaFx48bYvn273ji8vLwQGBiotf9VKhV+/PHHQm2HruO6UaNGWvWlpKTgwIEDepcBgFu3bsHb27vApFWhUMDDw0N6uLq6FirG0ixTZol/jRriX6OGyJRZvnwBeiPGJuawdaoOW6fqMDYxN3Q4JZ6R2hTypx6QP/WAkbrg3k5EREQlzSslrQqFArNnz8bo0aOxfft2qFQqqNVqnDx5ssB7ho4cORJTpkyRbl2TkJAg3YqlY8eOuHv3LjZv3oysrCxs2LChwNuBuLi4aN0C501cuHABZ8+eRUZGBqysrGBlZQVjY2NpPWFhYcjOzgYAGBsbIzAwEJMmTUJiYiKePn2KGTNmYMCAAYVal5GRET7++GOMGzdOatl79uwZdu/eDSBnMKorV65ArVZDLpfD3NxciiWvjh074sSJEzrnOTo64tixY7hz5w769OmjN3EdMmQIoqOjpfV/9dVXOHfuHNLT05GZmYmff/4ZKpUKtWvXzrfsn3/+CX9/fzg5ORVq23WtT1Nhtn3+/Plo0aJFgYnr3LlzsWHDBnzzzTeIiYkBANy5cwd9+/ZFYmIi+vfvj9DQUOzduxdZWVnIyMjAuXPn9LYm55X32OvUqRPCwsKwevVqpKenIysrC9euXcOFCxcKrOfEiRPo0KFDodZJRERERFTavfItb8aMGYOlS5di3rx5cHFxgaurK0JCQhAQEKB3mU8//RSBgYHo1KkTFAoFatSogUOHDgHI6XK8a9cuzJo1C3Z2djhy5Ag6d+6st65JkyZh7ty5UCqVmDZt2quGryU5ORkjRoyAg4MD3NzckJSUhDlz5gAAPvzwQ5iamsLBwUG6tcz3338PFxcXVKhQAbVq1UL9+vXx5ZdfFnp9c+bMQY0aNdC0aVPY2NigUaNGUrfPJ0+eoHfv3rC1tUXZsmXh7u6OCRMm6Kync+fOCA8Px8OHD3XOd3JywrFjx3Dz5k3069dP5z1Kzc3NMWbMGMyYMQNCCBgZGWHo0KFwcnKCs7MzVq1ahV27dsHX1zffsoUdNbig9Wkq7LYvWLAA77//vt7EtWHDhjhx4gT+/PNPlCtXDkqlEr1790aDBg1gY2MDDw8PHDhwAIsWLYKLiwvc3NwwadIkaQTol8l77Mnlchw5cgR79uyBp6cnnJycCuzWDeRc27xp0yZ8/PHHhVonEREREVFpJxO6+mtSsbd27VqcOXMGq1ateufrrlKlCjZv3owaNWq883X/1+3YsQPbt2/XGqypMCIjI+Hp6YnRE+dCYat7VOnSzlrEonXWQgDAEZNxeC5zfKP6Tt/WPaAZ5chIS0TErU0AAM9K/WBmoTRsQCVcpsULPKnxFwCgzJWGME3jdcT/NafW8juTiOh1mRg6AHo9gwcPxuDBg9/5ejMyMtCvXz8mrK+pR48e6NGjh6HDKJFkEDBDmjRNRU0gW50uTVNRExAmWdI0ERFRafLK3YOpdDMzM8PkyZMNHQYREREREZUSTFqJiIiIiIio2GLSSkRERERERMUWk1YiIiIiIiIqtpi0EhERERERUbHFpJWIiIiIiIiKLd7yhohKhExY4rZRc2maipaRsTmULnWkaSpaRlmmsHnsLU0TERGVJkxaiahESJfJcdW4s6HDKDVMTK3g5Pm+ocMoNYyzzGAXVsHQYRARERmETAjBu5QTFXORkZHw9PREREQEPDw8DB0OEREREdE7w2taiYiIiIiIqNhi0kpEJUJGUgTu/tQCd39qgYykCEOHU+JFRsehXe8QtOsdgsjoOEOHU+I9ViWjy64N6LJrAx6rkg0dDhER0TvFa1qJqEQQWRlIjb4kTVPRysjIxJUbD6VpKloZajWuxj6RpomIiEoTtrQSERERERFRscWklYiIiIiIiIotJq1ERERERERUbDFpJSIiIiIiomKLSSsREREREREVW0xaiYiIiIiIqNjiLW+IqEQwMrOConwHaZqKlpWlOdq2rC1NU9GyMjFFa29/aZqIiKg0kQkhhKGDIKKCRUZGwtPTExEREfDw8DB0OERERERE7wy7BxMREREREVGxxaSViIiIiIiIii0mrURUImSqniJ81zCE7xqGTNVTQ4dT4j2LTcSo4OUYFbwcz2ITDR1OiffshQpjju3DmGP78OyFytDhEBERvVNMWomoRMhOVyHx+jYkXt+G7HT+qC9qqudp2LX/LHbtPwvV8zRDh1PiPc/MxO4Ht7D7wS08z8w0dDhERETvFJNWIiIiIiIiKraYtBIREREREVGxxaSViIiIiIiIii0mrURERERERFRsMWklIiIiIiKiYotJKxERERERERVbJoYOgIjobZAZm8LcsaI0TUXL1NQE5cu6SdNUtEyNjFBO6SBNExERlSYyIYQwdBBEVLDIyEh4enoiIiICHh4ehg6HiIiIiOid4d/jRP8h3y7bAoWtvaHDKBVO365r6BCIiIiIdDq1toahQ3in2MeIiIiIiIiIii22tBJRiWAmnqNc9h8AgHtGTZAhszZwRCWbOisViU8vAQCULrVgbGJp4IhKNrVJBlJcHwEAbKK9YJxlZuCIiIiI3h0mrURUIpjhBapkHwYAhBvVRgaYtBYldVYa4qP/AgDYOFRi0lrEsk0ykez1LwDAOsaVSSsREZUq7B5MRERERERExRaTViIiIiIiIiq2mLQSERERERFRscWklYiIiIiIiIotJq1ERERERERUbDFpJSIiIiIiomKrVCStZcqUwYkTJ4qk7k2bNqFly5ZFUvfrOnjwIHx8fAwdxjvRvHlzLF++XO98uVyOu3fvvvF6AgMDERIS8sb1UNHKhhGyS8dprXiQGeU86N3IluU8iIiISpkS8WvjZYnL23LixAmUKVNG67V+/frh2LFj0nOZTIbbt2+/0XqmTp0KZ2dnKJVKtGjRAs+fP3+j+kozlUqF8uXLv9N1vqvjkbSpZE7Ybvodtpt+B5XMydDhlHhmFnYoV2csytUZCzMLO0OHU+KZplnD62xreJ1tDdM03oOYiIhKFxNDB0Dajh07hh9++AF///03PD09cebMGRgZlYj/FvLJysqCiQkPQSIiIiIi0q9EZkMLFy6Eu7s7nJ2dMX/+fK15QggsWLAA5cuXh729PTp06IDIyEhpvkwmw8qVK1GxYkXY2tqif//+yMjIQFJSEtq3b49nz55BLpdDLpfj1q1bWLt2LRo2bAgAaNSoEQCgTp06kMvlWLlyJTp37oy5c+dqxdC+fXvMmzdPZ+wmJiawsLCAu7s7TE1N0bx5c1haWha4vQkJCejevTtsbW1RvXp1XL58WWv+kydP0Lt3b7i4uMDT0xMhISHIzs4GACn+iRMnwt7eHh4eHti3bx8OHTqEChUqQKlU4osvvtDaf3PnzoWfnx8cHBwQEBCAqKgoaf7Vq1dRr1492NjYoFOnThgxYgQGDx4MAAgLC4NMJsPatWvh6+uL6tWrAwDGjx8PLy8v2NjYoHbt2jh58qRUX0hICLp3747+/fvDxsYGVapU0ZoPAI8fP0aLFi1gY2ODxo0bIzw8XOv9zG35Tk9Px6RJk+Dr6wsbGxvUrVsXEREROvfp8ePHUaVKFdjY2GDgwIHIyMiQ5iUnJ6Nz585wdnaGnZ0dOnTogEePHgEAgoODcfr0aYwdOxZyuRwDBgwAAHz33Xfw9/eHjY0NKlWqhB07dhT4nhIRERERUY4Sl7QeOXIEs2bNwt69e/Ho0SPcu3cPsbGx0vylS5di06ZNOHz4MJ4+fYratWsjMDBQq47t27fjzJkzuHfvHs6ePYuNGzfC1tYWoaGhcHZ2hkqlgkqlQqVKlbSW+/PPPwEAFy9ehEqlwvDhwxEUFIQNGzZIZZ48eYLjx4+jf//+OuMvX748UlNT0a9fP2RmZhZqm0ePHg21Wo3IyEjs2rULP//8szQvOzsbXbp0gb+/P8LDw3Hu3Dns3r0bq1evlspcvHgR3t7eePbsGYKDgxEUFIQ1a9bg/Pnz+Oeff7By5Upp29atW4cffvgBBw4cQGRkJFxdXdGrVy8AQGZmJrp27YqePXsiPj4e48ePx8aNG/PFe/DgQVy5cgUXL14EkJPk//PPP0hISMDAgQPx4Ycf4sWLF1L5PXv2oG3btkhISMD//vc/dO3aFQkJCdL89evXY8mSJYiLi4OrqyumTZumcz8FBwfj1KlTOHHiBJKSkrB69WpYWVnlKxcfH4+uXbti4sSJSEhIQKtWrbBnzx6tfTpo0CCEhYUhIiICCoUCo0aNAgB8++23aNq0KRYtWgSVSiW9976+vjh58iSSkpIwc+ZMDBgwQOvPkrySk5MRGRkpPaKjo/WWpRwmIg1e2RfhlX0RJiLN0OGUeGp1OpLjbiE57hbU6nRDh1PiZRtn4blTNJ47RSPbOMvQ4RAREb1TJS5p3bJlCwYPHozatWvDwsICc+bMkVoVAeDHH3/ErFmz4OPjA1NTU4SEhODChQtSSxkATJ48GQ4ODnB2dkbHjh3xzz//vHY8nTt3xpMnT6Q6Nm3ahFatWuW7NhbISfratGmD+fPnw8LCAj179pRa+EaPHo3//e9/+ZZRq9XYtm0bZs2aBRsbG5QtWxaffPKJNP/vv/9GREQEZs2aBQsLC7i5uWH8+PHYsmWLVMbd3R0jR46EiYkJ+vXrh9jYWIwdOxa2trbw8/NDw4YNpfg3btyIcePGoWLFirC0tMR3332Hc+fO4cGDBzh79ixevHiBCRMmwNTUFC1btkS7du3yxRwSEgKFQiG1IPfr1w+Ojo4wMTHB2LFjkZmZiVu3bknla9SogQEDBsDExASDBg2Cr68v9u/fL80PCgpC1apVYWZmhr59++p8v7Kzs7Fy5UosXrwY3t7eMDIyQo0aNeDg4JCv7L59+1CuXDmtdVarVk2ar1Qq0bNnT1hZWUEul2PSpEn5Wn/z6tmzJ9zd3WFkZISePXuiQoUKOHfunN7yCxYsgKenp/SoX79+gfUTYIEUNFRvRkP1ZlggxdDhlHjqzBd4+jAUTx+GQp354uUL0BtRm6Yjrvw1xJW/BrUp/yQgIqLSpcQlrVFRUfDy8pKeK5VKKBQK6XlYWBh69+4NpVIJpVIJR0dHGBkZabV6aSaUVlZWUKlUrx2Pqakp+vXrh/Xr1wPIaRUcNGiQzrLHjh3D8+fPMWTIEKxbtw4mJiYICAhAamoq/vjjD50JYExMDDIzM7W22dvbW2t7Y2JiYGdnJ23zJ598gqdPn+rdXl2v5e6Dx48fa41MLJfL4eDggMePHyMqKgpubm5a1+B6enrmi1kzPgCYN28eKlWqBFtbWyiVSiQlJWm1jmtuW+7yjx8/1hu/rvcrNjYWqamp8Pf3zzcvr7zHUN6YX7x4gY8//hje3t5QKBRo2rQpVCoV0tP1/5Bcv349atasKb0H165d09rGvMaPH4+IiAjpcf78+ZfGTURERERUEpW4pNXNzU2r1TQxMRHJycnScy8vL+zZsweJiYnSIzU1VboetSAy2evdaiAoKAhbtmzBxYsX8ejRI3Tt2lVnuaysLGRmZkIIAWNjY6k1tFatWrCxsdF5ax0nJyeYmppqbbPmtJeXFzw8PLS2Nzk5GTdu3HitbXF3d0dYWJj0XKVSIS4uDu7u7nBzc0NUVJRWy7aua0Y19+Pp06fxzTff4JdffkFCQgISExNha2sLIYTO7cl97u7u/kpxOzo6wtLSEvfv339p2bzHUN4Y5s+fj5s3b+Kvv/5CcnIyTp8+DQBSzHmPk/DwcAwdOhTff/894uLikJiYiGrVqmltY14KhQIeHh7Sw9XVtdDbSkRERERUkpS4pLV3795Yt24dLl++jLS0NEyePFmr5W/kyJGYMmUKHjx4ACBnEKNff/21UHW7uLggISFB63pKXWVy685Vo0YNuLu7Y8iQIQgMDIS5ubnOZZs2bQohBCZMmACVSoXMzEy0bdsWd+7cga2tLdRqdb5ljI2N0aNHD3z55ZdISUnBv//+i2XLlknz69WrBycnJ8ycORPPnz9HdnY27t2799LurPr069cPixYtwt27d5GWlobg4GDUr18fZcuWxXvvvQcLCwvMnz8fWVlZOHHiBA4ePFhgfSkpKTAxMYGjoyOysrIwe/ZsrT8ZAODKlSvYvHkzsrKysGHDBjx48AAdOnR4pbiNjIwwdOhQjBs3Do8ePYIQAleuXEFcXFy+sh07dsTdu3e11nnt2jWtmC0tLaFUKpGQkICZM2dqLZ/3GMi9ZZGTU85tWNavX4/r16+/UvxERERERKVViUta27Zti0mTJqFjx47w8vJC2bJl4ejoKM3/9NNPERgYiE6dOkGhUKBGjRo4dOhQoequWLEi+vfvD39/fyiVSq3rLnPNmDEDQ4cOhVKpxKpVq6TXg4KCcPXqVb1dg4Gc1rUjR47g9u3bUgvp6dOnpRba3MF+8lq6dCmEEPDw8EBAQACCgoKkecbGxti7dy/u3buHcuXKwc7ODr169XrtgX0GDRqE4cOHo23btnBzc0NERAR++eUXADldoXfv3o1ffvkFSqUS3333HXr37q03SQdy3q+OHTuiYsWK8Pb2hqmpab4uxV26dMGBAwdgZ2eHb775Brt27YK9vf0rxz537lw0aNAATZo0ga2tLYYOHYrU1NR85RwcHLBr1y7MmjULdnZ2OHLkCDp37izNHzt2LDIyMuDk5IQGDRqgdevWWsuPGTMGe/bsgZ2dHQYNGoTKlStjwoQJaNy4MVxcXHD58uVCtewTEREREREgEwX1UaS3Zs+ePZgwYQLu3Llj6FDeqa5du6JevXqYOnXqay0fEhKC27dvY+vWrW85sv+WyMhIeHp6YvTEuVDYvnrCXhrIRQw6ZM0BABwwmQiVzOmN6jt9u+7bCKvEykhLQPj1NQAA76pBMLOwM3BEJVumxXNE1zkDAHC92BimadYGjoiIiAzp1Noahg7hnSpxLa3FUVpaGpYsWYIRI0YYOpQid+rUKTx+/BhqtRp79uzBwYMH0a1bN0OHRURERERE/1FMWovY77//DgcHBwghMHLkSEOHU+Tu37+PunXrQqFQIDg4GD///DOqVKli6LCIiIiIiOg/it2Dif4D2D345WRCDUskAQBSYQshM36j+tg9uGAiW42szJzbS5mYyiEzerP9TQUTsmyozXJuq2WcYQ6Z4H/ORESlWWnrHmxi6ACIiN4GITPGCzChf1dkRsYwNbc1dBilhkwYwSTd0tBhEBERGQT/qiUiIiIiIqJiiy2tRFQiGIlM2IqcWzklyVyRLTM1cEQlW3Z2FjJSYwEAZpaOMDLi10lREjI1MqxzumObPZdDJtgdm4iISg+2tBJRiWCFRLRWL0Zr9WJYIdHQ4ZR4WRkpiLi1GRG3NiMrI8XQ4ZR4WeZpeFrjHJ7WOIcs8zRDh0NERPROMWklIiIiIiKiYotJKxERERERERVbvOUN0X9A7i1vIiIi4OHhYehwiqX0uAe4/UMdAEDFURdh7lDWwBGVbP+GP0HjjsEAgDP7v4WfdxkDR1SyPUxKQItffwIAHO81FL62dgaOiIiI6N1hSysREREREREVW0xaiYiIiIiIqNhi0kpERERERETFFpNWIiIiIiIqtK+//hqDBw8ukrrv37+P6tWrQ61WF0n9r6JTp044fPiwocMgMGklohLCVOEKv3674NdvF0wVroYOp8Qr42yHrSsnYOvKCSjjzEGBiloZazk2tP8QG9p/iDLWckOHQ/RObNu2DQ0bNoRcLoezszOaN2+OvXv3AgBOnDgBIyMjyOVy2NjYwM/PD/PmzctXx8aNG1GzZk1YWVnByckJgwcPxpMnT7TKqFQqTJgwAb6+vrC2toaPjw8++ugjhIWFFen2jR07FjKZDJcvXy7S9RSFyZMnY+3atUVS95dffolx48bB2NgYADBhwgSUL18eNjY28Pf3x8KFC7XKR0REoE2bNrC2toavry+2bt0qzfv333/RoEED2NvbQ6lUolGjRvjjjz+0ll+6dCnc3d0hl8vRo0cPJCQkaG3n5MmTi2Q76RUJIir2IiIiBAARERFh6FCIiIiK3KJFi4SDg4P45ZdfRHJyssjKyhJHjx4VQ4YMEUIIcfz4ceHi4iKV/+uvv4SVlZU4cuSIVh2Ojo5iz549Ij09XTx+/Fj0799flC1bViQmJgohhEhPTxcNGjQQLVq0EFevXhVZWVkiPj5eLF68WKxatarIti8jI0M4OjoKe3t7MWbMmCJbj1qtFtnZ2UVW/9v29OlToVAoRHJysvTa1KlTxfXr14VarRbXr18X7u7uYuvWrdL8xo0bi5EjR4oXL16I48ePC7lcLq5duyaEECI5OVncu3dP2g87duwQSqVSpKenCyGEOHz4sLC3txcXL14UycnJ4sMPPxS9evXSiqls2bLiwoUL72DrqSBMWon+A5i0EhFRaZGUlCSsra21EpO88iatQghRt25dMXfuXKkOuVwutmzZolVGrVaLSpUqiRkzZgghhFi9erVwcnLSSpLehdzkac2aNcLR0VFkZGRoxRgcHCwcHR2Fp6enWL16tQAgHj58KIQQIi4uTgQEBAiFQiGqVasm5syZI7y9vaXlvb29xbfffitq164tzM3NRVRUlLh7965o166dcHBwEH5+fmLZsmVS+QsXLoj69esLGxsb4ejoKPr27SuEECI7O1t88cUXwtnZWdjY2IgKFSqI48ePCyGEmD59uujdu7cQQoj27dtL+z1X+/btxbfffiuEECI6Olr06tVLODs7Cw8PDzF9+nShVqt17pcNGzaIpk2bFrjvRo4cKUaPHi2EEOLu3bvCxMRExMfHS/P79u0rvvjii3zLqdVq8dtvvwkA4vHjx1LZzz//XCqTW1/unxpCCDF06FDx5ZdfFhgTFT12DyaiEkFkq6FOS4Q6LREi2/DXwZR0anU2kpKfIyn5OdTqbEOHU+Kps7ORlJ6GpPQ0qLO5v6lopMc9KPCRlZqgVT4jMbzA8pmqp68Vx59//om0tDR07969UOWFEDhz5gxu3LgBf39/AMDZs2eRmpqKHj16aJU1MjJCz549cejQIQDA4cOH0a5dO9jY2LxWrK9rzZo16NmzJ3r37o309HSp2zMArF69Gjt27MCFCxdw8+ZNhIaGai376aefAgAeP36M3bt3Y926dfnqX79+PbZt24aUlBRYW1ujVatW6NKlC6Kjo3HgwAHMmTMHR44ckerr0qULEhMTERERgVGjRgHI2Tdbt27F5cuXkZycjNDQUHh5eeVbV//+/bFp0ybpeUxMDH7//Xf07dsX2dnZ6NKlC/z9/REeHo5z585h9+7dWL16tc79cvXqVVSsWFHvfsvOzsbp06dRtWpVAMD169fh7e0NO7v/u0ylZs2auH79utZy3t7eMDc3R0BAAIKCguDm5iYtX7NmTalcuXLlYGZmhtu3b0uvVapU6T/ZhbukMTF0AEREb0NGQhhu/1AHAFBx1EWYO5Q1cEQlW3jkMzTuGAwAOLP/W/h5lzFwRCXbo5QktPj1JwDA8V5D4WvL64jp7cs9h+pTpvlUuDT9Qnr+YFM3ZMT/q7e8XfU+8Or64yvHERcXB0dHR5iamhZY7tmzZ1AqlUhLS0N6ejqCg4MREBAAAIiNjdVbh6urK2JiYqR11alT8Ha/bU+ePMHBgwdx9OhRWFpaonv37lizZo2UpG/ZsgVjxoyBj48PAGD69OnYvn07AECtVmPbtm34559/IJfLIZfLMWrUqHzX844ePRp+fn4AgF27dsHV1RUjR44EAFSoUAHDhg3Dli1b0Lp1a5iZmSE8PBxRUVHw8PBA48aNAQBmZmZIS0vDjRs34OjoCF9fX53bExAQgI8//hg3btxAlSpV8Msvv6Bx48bw8PDA+fPnERERgVmzZkEmk8HNzQ3jx4/HmjVrMGzYsHx1JSQkwNbWVu++mzBhAkxNTaVBoFQqFZRKpVYZpVKJlJQUrdfCw8ORlpaGrVu3QiaTSa8XZnkbGxut61zJMNjSSkRERETFhoODA2JjY5GZmVlgOWdnZyQmJiIlJQUzZ87E8ePHpWUcHR311hEdHQ0nJydpXVFRUYWObfXq1VKy+LKHvtF1N2zYAHd3dzRr1gwAMGDAABw8eFAaICoqKgqenp5Sec3pmJgYZGZm6p2fy9vbW5oOCwvDpUuXoFQqpcfcuXOl9f3888948eIFateujSpVquDnn38GALRo0QIzZszA5MmT4eTkhMDAQJ37ysrKCt26dZNaWzdt2oT+/ftL646JiYGdnZ207k8++QRPn+puhbezs0NycrLOebNmzcK+fftw4MABmJubAwDkcjmSkpK0yiUlJelsObewsMDgwYMxa9YsXLlypdDLp6SkaLXkkmGwpZWIiIiIUHHUxQLnG1vZaz0v228XhDpLb3kj89cb6bpRo0awsLDArl270KtXr5eWNzU1xdSpU7Fv3z78+OOPGDNmDN577z1YWFhgx44dCAwMlMpmZ2dj+/btUqtmmzZtMGnSJKhUKsjlL4/3o48+wkcfffRa25VrzZo1ePLkCVxdc0a6F0IgKysLGzZswIQJE+Dm5oaIiAipvOa0k5MTTE1NERERIbVIas7Ppdma6OXlhUaNGuHEiRM64ylbtiw2btwIIQROnjyJNm3aoFmzZvD398eoUaMwatQoJCQkYNiwYQgODsaGDRvy1dG/f38MHz4cQ4YMwZUrV9CzZ09p3R4eHoUeibl69epYuXJlvtfnzJmDtWvX4uTJkyhT5v969lStWhXh4eFITEyUWkwvX74sdR/WJSMjA//++y9q1KiBqlWr4vLly+jXrx+AnNvtpKena3VRvnXrllYXYjIMtrQSEREREcwdyhb4MLHUbm0yU3oXWN5U7vJacSgUCsyePRujR4/G9u3boVKpoFarcfLkSZ1dSnNNmjQJc+bMQWpqKhQKBb766it8+umn2LdvHzIyMhAVFYVBgwYhMTFRui60f//+8PHxQUBAAG7cuAG1Wo2kpCQsW7ZM73WXb+LcuXO4ffs2Tp8+jcuXL+Py5cu4cuUKJk6ciDVr1gAAevfujSVLliA8PBwqlQozZ86Uljc2NkaPHj0wffp0qFQqhIeH48cfC+6C3alTJ4SFhWH16tVIT09HVlYWrl27hgsXLgDIuf712bNnkMlkUCqVkMlkMDY2xoULF3D27FlkZGTAysoKVlZW0m1o8mrVqhUyMjIwatQodOrUCQqFAgBQr149ODk5YebMmXj+/Dmys7Nx7949nDx5Umc9rVu3xpUrV6BSqaTX5s6di+XLl+P333+Hu7u7Vvly5cqhXr16mDp1KlJTU3Hq1Cns2bMHgwYNAgAcO3YMFy5cQFZWFl68eIGvvvoK8fHxaNCgAQBg8ODBWLNmDS5dugSVSoWpU6eie/fuWl2UT548iY4dOxa4j6noMWklIiIiomJlzJgxWLp0KebNmwcXFxe4uroiJCREumZVly5dusDOzg4rVqwAAIwfPx7z58/HlClTYGtrC3d3d9y4cQMnT56Ei0tOQm1mZoajR4+iZs2a6NixIxQKBapXr47Lly+jdevWb3271qxZg3bt2qFevXooU6aM9Pjss89w//59nDt3DkOHDkWXLl1Qt25dVKpUCS1btgQAqUvs0qVLkZWVBXd3d3Tu3Bl9+vSR5ukil8tx5MgR7NmzB56ennBycsLw4cOlbrhHjhxBtWrVIJfL8eGHH+KHH36Ar68vkpOTMWLECDg4OMDNzQ1JSUmYM2eOznUYGxsjMDAQR44ckboG576+d+9e3Lt3D+XKlYOdnR169eqF6OhonfW4uLigQ4cO0jW8ABAcHIzo6GhUqVJF6no9YsQIaf7WrVtx584dODg4YODAgVixYoXU0pqSkoLBgwdDqVTC09MTJ06cQGhoqDQQU+vWrRESEoKOHTuiTJkyyMjIwPLly6W6//rrL8jlctSvX1//m0rvhEwIIQwdBBEVLDIyEp6enoiIiICHh4ehwymW0uMecCCmd+jf8CcciOkdepiUwIGYiN7Qnj17MGTIEBw7dgzVq1c3dDiFdunSJTRo0ABpaWkwMsrf3rRw4ULs378fR48eNUB0b9+9e/fQo0cPXLp0SW/L7rvSuXNnjB49Gm3btjVoHMSWViIiIiIqBbp06YKffvoJZ8+eNXQoBcrIyMDevXuRlZWFmJgYTJo0CQEBAVLCeufOHVy6dAlCCFy7dg2LFy8u9O2B/gvKlSuHq1evGjxhBYC9e/cyYS0mOBATEREREZUKBXUvLi6EEJg1axb69u0LMzMztGzZEkuXLpXmP3/+HIGBgYiMjISDg4M0CBJRScaklYhKBBNrB7i3+06apqLlYGeDr6cMkKapaNlbWOKrRq2kaSIquczNzXHu3Dm982vXro27d+++w4iIDI/XtBL9B/CaViIiIiIqrXhNKxERERERERVbTFqJiIiIiIio2GLSSkQlQnpCGK7PL4vr88siPSHM0OGUeOERz1Cl6WhUaToa4RHPDB1OifcoORG1NyxF7Q1L8Sg50dDhEBERvVMciImISoZsNdQv4qRpKlrq7GzEJ6RI01S01EIgPi1VmiYiIipN2NJKRERERFQKNG/eHMuXLzd0GK8lJCQEgYGBRb6eR48eQS6XIz09vUjqT0hIQLly5ZCcnFwk9b+KefPmYfLkyYYOo1DY0kr0H/Ltsi1Q2NobOoxiSS5i0OH/Ty9YuQ0qmdMb1Xf6dt03D6oEy0hLkKb7Bt+GmcVTA0ZT8mVaPAfq5Ez3Db4F0zRrwwZE9A7EhB3G4xsb8DzxPoxNLGBl6wf3KgPh4Pk+Ep9cwLVDw2FkYgEZZDC1sINrhV7wqDoIAHBqbQ0AwMaNGzFv3jzcvXsX1tbW6NixI+bMmYMyZcpI61GpVJgxYwa2b9+OZ8+ewcnJCa1atcK0adPg4+Pz1rdr8ODB2Lx5M8zMzGBiYoKqVati/vz5aNCgwVtf15vw8fHB8uXL0a5dO0OH8k55eXlBpVIVWf3fffcd+vbtC4VCIT1fv349wsLCYGdnhwEDBuCrr76CsbExACAxMRHDhw9HaGgobGxs8L///Q9jx44FkHPsdujQATdv3kRGRgb8/PwwY8YMdO3aFQAQFhYGX19fWFv/33dG//79pT8uRo0aBX9/f4wbNw5OTm/2u6mosaWViIiIiIqVxzc34f5fX8O9Sn806HUUDT48Cs8awxH36LhUxtTCHo37nUWjfn+iQrM5CL/yIxKi/pLmL168GOPGjcPMmTORmJiIK1euQK1Wo0mTJkhKSgIAZGRk4IMPPsDFixexZ88eJCcn49KlS6hRowaOHj1aZNs3fvx4qFQqxMTEoEmTJujZs2eRrctQsrKyDB1CsZOZmYmffvoJAwYMkF7Lzs7GmjVrEB8fj9OnT2Pfvn2YN2+eNH/06NFIT0/H48ePcejQIXz99dcIDQ0FkHNP3+XLl+Pp06dITk7Gjz/+iP79++Px48da642NjYVKpYJKpdJqabeyskKHDh2wbt26It7yN8eklYiIiIiKjawMFcIuLYV/g0lw8mkLE1NryIyMYefaAOUbh+hcRuFUDVa2fngefxsAkJycjKlTp2LJkiXo3LkzzMzM4ObmhnXr1sHMzAyLFy8GkNMS+++//2L37t2oVq0ajI2NYWdnh88++wxDhw4t8m01NTXFgAEDEBkZiZiYGCn2ESNGwMPDA2XKlMHo0aORlpYmLbNw4UK4u7vDxcUFCxYsQJkyZXDixAkAOa24EydOlMrevn0bMplM57ofPnyIVq1awcHBAY6OjujTpw8SEnJ60fTp0wePHj1Ct27dIJfLMWXKFADA+fPn8d5778HW1hbVqlXDvn37pPpCQkLQvXt3DBkyBEqlErNnz4ajoyPOnz8vlUlPT4ednR3+/vvvfPGcOHECZcqUwXfffQdnZ2d4eHhI75Muffr0gaurK2xtbdG0aVNcu3ZNmjd48GCMGjUK3bt3h42NDapXr47Lly9L8318fDB//nzUqVMHCoUCHTp0kLY9LCwMMplM2ufNmzfHtGnT0KJFC9jY2KBx48YIDw+X6jp27BgqV64MGxsbDBw4EN27d0dISIjOmM+dOwcLCwv4+/tLrwUHB6Nu3bowNTWFt7c3+vXrhz/++AMA8Pz5c2zbtg2zZ8+GQqFAtWrVMGzYMPz8888Aco6fypUrw9jYGEIIyGQyZGZmasX3Mi1atMDevXsLXd5QmLQSERERETLSEgp8qLNStcpnpicVWD4r8/lrxZEccwXZ6gw4eLcsVHkhBJKeXcKLxH9hYeMFADh79ixSU1PRo0cPrbJGRkbo2bMnDh06BAA4fPgw2rVrBxsbm9eK9U2lpaVh7dq1cHBwgJ2dHQAgKCgIaWlpuHnzJm7fvo179+5h5syZAIAjR47g66+/xv79+xEeHo6HDx8iNjb2tdYthEBwcDCioqJw+/ZtREdHY9q0aQCALVu2wMvLC7t27YJKpcLs2bORkJCAdu3aYciQIYiLi8P8+fPRu3dv3Lp1S6pz7969aNOmDeLj4/G///0Pffv2xfr166X5e/bsgbu7O+rW1X0JTmxsLB4+fIhHjx5h9+7dmDFjBo4dO6azbJs2bXDnzh08e/YM9evXR58+fbTmb968GePHj0diYiJatmyJzz77TGv+xo0bsWvXLkRFRSExMRELFy7Uu6/Wr1+PJUuWIC4uDq6urtJ+io+PR0BAAKZOnSrtn4ISwKtXr6JixYp65wPAyZMnUbVqVQDA3bt3kZ2dLT0HgJo1a+L69etayzRt2hQWFhZ477330KxZs3xdzf39/eHm5obAwEBERERozatUqZJWQl9c8ZpWIiIiIkL49TUFzndwawx7t//7Mfz47g5kpifqLW/jUBllfF/9esis9ESYmithZGRaYLnMtHj8ubkJstUZENkZ8KgaBAevFgBykh9HR0eYmuavw9XVVWrVjIuLQ506dV45xje1aNEiLF++HMnJyXB0dMSOHTtgYmKCZ8+eYc+ePYiPj5cS6alTp2LQoEGYPXs2tmzZgkGDBqFmzZoAgNmzZ2PZsmWvFYOfnx/8/PwA5HQzHTduHKZOnaq3/P79++Ht7Y1hw4YByEkaO3fujM2bN0tJdZ06daTBkiwtLREUFITWrVtj4cKFMDU1xfr16zFo0CC968jOzsY333wDCwsL1KlTB4MGDcLmzZvRsmX+PzCCgoKk6S+//BILFixAXFwcHBwcAAABAQFo0qQJAGDgwIH46aeftJYfM2YMvLxy/uTo2bOn3uQ4d125iWPfvn3x5ZdfAgD27duH8uXLo2/fvtK8BQsW6K0nISFBupZVlyVLluDatWtSoq9SqWBra6tVRqlUIiUlReu106dPIyMjA6Ghobh//750PayjoyMuXLiAmjVrIjExEcHBwejcuTMuXrwolbGxsUFycjKys7NhZFR82zOLb2RERK8gE5a4a9QUd42aIhOWhg6nxDMyNofSuRaUzrVgZGxu6HBKPKMsU9hEecEmygtGWQX/kCf6rzMxVyIzPRHZ2ZkFljO1sEejvn+gUb8z8K75CZKeXIDIzrmO0tHREbGxscjMzF9HdHS0NOiMg4MDoqKiCh3b6tWrIZfLC/UYPHiw3nrGjh2LxMREREVFwc/PT2rpCgsLg1qthqenJ5RKJZRKJTp16oRnz3Luhx0VFQVPT0+pHoVCkS+pKaynT58iMDAQ7u7uUCgU6NevX4Gtto8fP843MJWPj4/W9ZPe3t5a82vVqgUPDw/s378fMTExOHr0KPr37693Hba2tlrb4+3tne/6TABQq9WYOHEiypYtC4VCIa1XM37NwbasrKzw/Ll2y3/e+QUNvqSvbN73A0C+55rs7Oz0jhq8YcMGfP311zh8+LCUeMvl8nzlk5KSdPYMMDMzQ9euXXHgwAHs2bNHWr5u3bowMTGBo6MjfvjhB9y8eRP37t2TlktJSYFCoSjWCSvAllYiKiHSZXJcNg4wdBilhompFZz+f4sGFT3jLDPYPSy4SxnRm/KuGlTgfGMTC63n7uV7QAj992k2MjZ7rTgUTjVgZGyGuEfH4OTT9qXljYxM4VVjGOIjTyH6zq9wr9wP7733HiwsLLBjxw6t26RkZ2dj+/bt6N69O4Cc1sJJkyZBpVJBLpe/dF0fffQRPvroo9faLl3KlCmD1atXo0GDBujVqxe8vLykFlczs/z7z83NTat7Z3JysjSoFJCTpLx48UJ6/uTJE73rnjx5MrKzs3H16lU4ODhg9+7d+Pjjj6X5ea+FdXd3R1hYmNZrYWFhKFeunN5lgJxWyvXr1yM8PBzNmzeHq6ur3piSkpKQnJwstUY+evQI7u7u+cpt3rwZO3fuxJEjR+Dr64vk5GQolUqId3wf67zvBwBERESgRo0aOstXr14dc+bMyff6pk2bMGHCBBw9elSr+3D58uUhk8lw48YNVKlSBQBw+fJlre7CeWVlZeHBgwc658lkMshkMq39dOvWLanlvjgr3ik1EREREb0TZhZ2BT6MTbR7sZia2xZY3sT09W7NZGImh0+t0Xhwbg5iwo5AnfkCIluNxCd/496fM/Qu51ltCCKur4E6Kw0KhQJfffUVPv30U+zbtw8ZGRmIiorCoEGDkJiYiE8//RRAzu0/fHx8EBAQgBs3bkCtViMpKQnLli3D6tWrXyv+V1WlShW0bt0a3377LcqUKYOOHTtizJgxSEhIgBACEREROHjwIACgd+/eWL9+Pa5evYr09HRMmzZNq4WsVq1aOHDgAGJiYhAfH49vv/1W73pTUlJgbW0NW1tbREVFaY1YCwAuLi5ayU+HDh0QFhaGn3/+GVlZWTh69Cj27t0rdY3Vp1+/fjhy5AiWL19eYOszkHPN8eTJk5Geno5Lly5h3bp1+a5VzY3d3NwcDg4OSE1NLbBbc1Hq2LEj7t69i19++QVZWVnYunUrrly5ord8/fr1kZqain///Vd6bcuWLRg7dixCQ0PzJaPW1tbo2bMnpkyZgpSUFFy/fh0//fQThgwZAgD4+++/cfz4caSnpyMjIwOrV6/G2bNn0bx5cwA5Az/dunUL2dnZSExMxOjRo+Hv74/y5ctL6zhx4gQ6duz4FvdK0WDSSkRERETFinvlfijbYCIe31iPv35tiXPbPsCjK8ula1Z1sfdsDhMzBZ7c3Q4g57Yy8+fPx5QpU2Brawt3d3fcuHEDJ0+ehIuLC4CcLpVHjx5FzZo10bFjRygUCmmk2datW7+TbQVyRpBduXIloqOjsW7dOpiamqJmzZqwtbVF27ZtcffuXQBA27Zt8b///Q/t2rWDp6cnvLy8YGtrC3PznMs0+vfvjwYNGsDf3x+NGzeWWpR1mT59Oq5evQqlUokOHTogICBAa/6kSZMwd+5cKJVKTJs2Dfb29jhw4ABWrFgBBwcHjB07Fps3b0alSpUK3DZHR0e0adMG0dHR0v1DCyrr7e0NT09PdOrUCVOnTkWrVq3ylRs4cCD8/Pzg7u6OSpUqoV69egXWW1QcHBywc+dOhISEwM7ODgcOHEC7du2k9yMvMzMzDB06VGtwqsmTJyMxMRFNmzaVupa3b99emr9s2TKYmprC1dUVrVu3xsSJE6X5mZmZ+Pzzz+Ho6AgXFxesWrUK27dvR61atQAA//77Lzp27AgbGxtUrFgR8fHx2Ldvn3Q9a2pqKvbv3//SPxOKA5l41+3oRPTKIiMj4enpidET50Jha2/ocIolS5GA99QbAQBnjfsjVWb3RvWdvq17ZEPKkZmRgicP9gMAypTtCFMzw4y8WVpkmaUhtsJVAIDjneowybB4yRJEpduptfm7Z+7ZswdDhgzBsWPHUL16dQNE9fYlJCTA3t4eERER8PDwMHQ4eo0fPx7Pnz/HihUr9JY5ceIEAgMDC+zS/F9Qo0YNTJgwQe+1u/Hx8WjQoAH++ecfg41anWv+/PmIjY3FN998Y9A4CoMtrcWYj4+P1B0kr0ePHkEulyM9Pf0dR1U8rF27Fg0bNvzP1EtFzxhZcBRhcBRhMAZvaF7URHYW0p5HIe15lDTwCRUdYaRGhiIRGYpECCO1ocMh+k/q0qULfvrpJ5w9e9bQobyRnTt3Ij09HSkpKRg3bhzq169frBPWJ0+eYOPGjRgxYoShQykShw4dQnx8PDIzM7F8+XI8ePAAbdvqvxbb3t4e9+7dM3jCCgCff/75fyJhBZi0FomAgACYmZlJw6kXBS8vL6hUKr3dD4rKyZMnUb16ddjY2KBChQoIDQ0tsHzuDZrlcjkUCgWcnZ3Rpk0b7NixI1/ZY8eOoWXLllAoFHBwcECDBg2wdu3aItoSIiIiKm0CAgK0Bhv6L/r555/h7OwMT09PREdHY/PmzYYOSa8ZM2bA398fQUFBUpfVkubChQsoX7487OzssGLFCuzatUsanZreHiatb9mzZ8+wf/9+2NjYYNOmTQWWzcr677VODBgwAEOGDEFycjJCQ0MLHNZbU2xsLJKTk/Hw4UMMHToUI0aMwKxZs6T5u3btQteuXdGrVy+Eh4cjNjYWS5Yswb59+4pqU4iIiIj+c/bt24ekpCQkJibi0KFDKFu2rKFD0mv69OlQqVQFDgiVq3nz5v/JrsFTp05FbGwsVCoVLl269E6vhS5NmLS+ZRs3boS/vz8mTJiANWu0b9IdEhKC7t27Y8iQIVAqlfjuu++Qnp6OSZMmwdfXFzY2Nqhbt67W0Nk3btxAnTp1oFAo0KFDByQkJAD4vxbMtLQ0/Prrr9Iw2Ll+/fVXVK5cGQAghMCCBQtQvnx52Nvbo0OHDoiMjJTmTZgwAS4uLlAoFKhYsSJOnDihd/tMTEzg5+cHmUwGPz+/Aofc1sXa2hq9evXC0qVLMWvWLMTHx0MIgbFjx2Lq1KkYMWIE7OzsIJPJUL9+fWzfvr3A+saPHw+lUomyZcvi119/lV4PDQ1F7dq1oVAo4OnpiWnT/l979x0WxbX/D/y9Cyx96UgHkaDGgprEkhgV7OUbezdqjL3jNRGjBlsUsV9NJE3RK4oXbxSjV0ViicaCaLAjNkKRKIpLkbILO78/+DE3K2Asy7LC+/U8+8SdOXvmnE/mWeazc+acBRqfi4uLQ7t27WBjYwMnJ6dKh0aEhISgadOmuH//Pjp06ICwsDBx36FDhzTWK/Py8sLy5cvRpEkTWFlZoX///lAoFACAwsJCjB49GnZ2drCysoKfnx+uX7/+UrEjIiIiIqqNmLRq2ZYtWzBixAgMHz4cV65cwcWLFzX2//zzz+jSpQuysrIwc+ZMzJkzB7/++iuOHz+O7Oxs/PjjjzAzMxPLb9++HXv27MH9+/ehUCiwdu3acsf86KOPkJaWJi5MDZSu91T2APjGjRsRERGBmJgYPHjwAC1atBDXLIuJiUFkZCQSEhLEu6ceHh4V9k0QBLRu3Rrjx4/HjRs3XitOffv2hUqlwrlz53Dz5k2kpKRg4MCBL1VHfHw87O3tkZmZiU2bNmHUqFHiYsnm5uYIDw+HQqHAf//7X3E2NaB0UqPOnTtjzJgxePDgAZKSkhAQEFCur7NmzcK+fftw4sQJuLi4vFCbwsPDER0djbS0NBQVFWH69OkAgK1bt+LKlSu4c+cOFAoFIiMjYWtb+YRKOTk5SEtLE18ZGRkvFRsiIiIiopqCSasWxcfH49q1axg+fDjc3d3Rrl27cndb33nnHQwZMgRSqRTGxsb47rvvsH79enh6ekIqlcLPzw92dnZi+RkzZsDDwwMWFhYYMGBAuSQYAExMTNC/f39xOHJWVhYOHz6M4cOHAwA2bdqEpUuXwsvLC0ZGRli4cCHOnz+PlJQUyGQyFBYW4tq1a1CpVKhbty68vb0r7F9oaCgyMzMRFhaGzp074/Lly2K/HR0dXypWMpkM9vb2yMrKwuPHjwHghRPDMra2tggKCoKRkRG6dOmCbt26YdeuXQCAdu3aoWnTppBKpWjSpAmGDh2KEydOACj9IeDDDz/E6NGjIZPJIJfL0apVK7He4uJijBo1CtevX8eRI0dgY/Pis9BOnToV3t7esLS0xFdffYVdu3ZBrVZDJpMhNzcXiYmJEAQBDRs2hJOTU6X1rFmzBu7u7uKrZcuWLxUbIiIiIqKagkmrFm3ZsgVt27YVh4x+/PHH2LFjh8YMv56enuK/Hz16hIKCAvj4+FRa518TGzMzM+Tl5VVYbsSIEdi5cyfUajWioqLQsmVL8VjJyckYPHgwrK2tYW1tDXt7e0ilUqSlpcHf3x+LFi3CF198AQcHBwwZMgT379+v8Bjr1q3Dl19+iT59+mDt2rXo3LkzLl68iJMnT6Jbt24vHCcAUCqVePToEWxtbcUkvbLjVsbV1VVjQW1PT0+kp6cDKF1M2d/fHw4ODrCyssKmTZvw6NEjAKUzLz8v5nfv3sWuXbsQHBwMc/OXWxj9r3epPT09oVQqkZmZiY8//hijRo3C+PHj4ejoiPHjxyMnJ6fSembNmoXU1FTxFRcX91LtICIiIiKqKZi0aklRURF27tyJCxcuwMnJCU5OTpgzZw6ysrKwb98+sZxEIhH/bW9vD1NTU9y+ffu1j9+hQwdIJBIcP34c27dv11gbysPDA/v27YNCoRBfBQUFeP/99wEAkydPRlxcHO7du4fi4mLMmTOnwmMUFxeLCfjAgQOxcuVKdOrUCWvXrkVQUNBLtXfv3r0wMjJCq1atUL9+fXh4ePzt86vPSk9Ph1qtFt+npKTA1dUVADBs2DD07NkTKSkpyM7OxqRJk1C2JLGHh8dzY+7r64udO3eid+/euHDhgrjdwsIC+fn54vuKJgtISUnR+LeRkREcHBxgaGiIefPm4fLly7hy5QquX7+OlStXVtoGuVwONzc38eXs7PwCEandSiDDfUlD3Jc0RAlk1d2cGk8qNYSZVV2YWdWFVGpY3c2p8SRqA5hk2cMkyx4StUF1N4eIiEinmLRqyd69e6FUKnHlyhUkJCQgISEBV69exfDhw8sNES4jlUoxduxYBAYGIiUlBYIg4NKlS+Jw2ZchlUoxdOhQLF++HPHx8RrPh06aNAnz5s3DnTt3AJQuRF02adH58+dx5swZKJVKmJmZwczMDAYGFV8QDR06FJ999hkSExOhVqvRoEEDWFpaIjc3t9LPPCs/Px+7d+/GtGnTMHfuXNja2kIikWDt2rVYunQpvvvuOygUCgiCgIsXL2Lw4MGV1pWVlYXQ0FCoVCrExsbi4MGDYr9zc3NhY2MDU1NTxMfHa0wHP3z4cPz666/417/+BZVKhZycHJw7d06j7n79+mHTpk3o0aOHOCS7efPm2L17N/Ly8pCamooNGzaUa9M333yDe/fuITc3F/Pnz8fgwYMhlUpx7NgxXLp0CSUlJbCwsICxsfELx4xeTIHECqcMx+KU4VgUSKyquzk1nqHMEq5v9YXrW31hKKv+teZqOkOlCRxvtIDjjRYwVJpUd3OIiIh0ikmrlmzZsgUjR46Et7e3eKfVyckJgYGBiImJqXToa2hoKFq1aoW2bdvCysoKY8eORUFBwSu1YcSIEYiNjUX37t01nsOcNm0ahgwZgl69ekEul8PPzw+HDx8GUDrhz8SJE2FnZwcXFxdkZ2cjJCSkwvpXrlyJHj16oGvXrrC0tMSECROwYsUKzJw5E926dXvuZEH29vaQy+Xw9PREWFgYNmzYgODgYHF/v379sGfPHkRGRsLd3R329vaYPHkyevXqVWmd7777Lh4+fAgHBweMHz8emzdvRv369QGUJo+LFy+GpaUlFi5cqJHEu7u749ChQwgLC4ODgwPq169f4YzJ/fv3x9dff43u3bvj4sWLCAwMhJWVFZydndG3b1/xmeG/GjlyJD766CO4ubnBwMAA69evB1B6V3bw4MGwsrJCvXr14Orqis8++6zSvhERERHVJCdPnqyy5XlUKhUaN26M5OTkKqn/ZaxatQpffPFFdTejxpEIZWMmiei1eHl5ISws7KWf730RaWlpcHd3x9SgUMitKp91mLTnZOK71d0EIqJaLTM5BunX/oWnitswMDSBmZU3XBuNhJ17eyj+PI8rh8dDamgCCSQwMrGBc/1BcGs8CgDwa7gfgNLJF1etWoWkpCSYm5ujZ8+eCAkJ0ZgzJC8vD4sWLcLu3bvFH8M7duyIBQsWaCxtp01Hjx7F0qVLER8fDyMjI/j4+GDSpEkYPXp0lRxP15KTk1G3bl0UFBTAxKTqR4d89913+O2337B161YApas2bNy4Ufz/3rt3b6xcuRIWFhYASudWmT59OiIjI2FoaIhx48Zh2bJlGo/xAcDNmzfh5+eHPn36IDIyUtx+4sQJTJkyBXfv3kWjRo3www8/wM+v9JzLz8+Hj48PLl26BAcHhyrve23BO61EVCOYCDl4r3gn3iveCROh8kmuSDuKVU/x571D+PPeIRSrnlZ3c2q8EqMiPPa5isc+V1FiVPT3HyB6w6Vfj8Dts8vg2mgEWg2KRauBsXD3G4/HKcfEMkYmtvhg+Bm8P/w06rcLwR+XNuHJ/bPi/vXr1yMwMBBLliyBQqEQH9Np27YtsrOzAZQmL506dcKFCxewb98+5OTk4Pfff4efnx9iY2OrpG979uxB7969MWjQIPzxxx949OgRNmzYgP3791fJ8V5FcXFxdTfhpXzzzTf4+OOPxff5+flYtWoVHj58iMuXL+PWrVsaI9wWL16MixcvIikpCRcvXsRPP/2EsLCwcvVOmjSp3AoOjx8/Ru/evfH555/jyZMnGDp0KD766CNx3hczMzP06NFDTKBJO5i0ElGNYIgi1BXiUVeIhyF4UV/V1CVK5D6+jtzH16EuUVZ3c2o8tUExnta5j6d17kNt8GZdTNKbQ2Xy9LmvEkOVRvli44Lnl3/FH1iKlXlI/n0jfFrNhYNXVxgamUMiNYCNcyv4frCwws/IHZrAzMobT7MSAZQ+/jR//nxs2LAB//d//weZTAYXFxds3boVMplMfHxn+/btuHv3LqKjo9GkSRMYGBjAxsYG06dPx9ixY1+p/c8jCAJmzpyJ+fPnY+LEibCxsYFEIkHLli01JqTcunUrGjRoAGtrawQEBCAxMVHc16FDByxYsAD+/v6wtLTEBx98gD/++EOs/7PPPkOdOnUgl8vRoEED8REoQRCwZs0a+Pr6wtbWFj169EBaWppYr0QiwTfffCMeNzQ0FD179tRof2hoKHr06AEAOHjwIFq0aAG5XA53d3csWLBALFc22ae9vT0sLCxw6NAhHD9+XLzD/Xd1K5VKfPHFF6hbty7s7e0xbNgwPHnypMKYpqWl4fr162jbtq24bdKkSWjfvj2MjY1hb2+P8ePH49SpU+L+LVu24Msvv4SjoyM8PDwwe/ZsbN68WaPerVu3ok6dOggICNDY/tNPP8HHxwcjR46EsbExAgMDoVarNX7k8Pf3x88//1xhe+nVMGkl0pLk5OQqGRpMRESkCxnv/PbcV55Tqkb5h40uPLe8wvPWK7UjJ/MS1CVK2HkG/H1hlCZj2Q9/R77iLkwsS5eeO3PmDAoKCtC/f3+NslKpFAMGDBDn9oiJiUG3bt1gaambCeVu3ryJlJQUjbk2nnX8+HHMnDkT4eHhePjwITp27IiePXtqLKG4bds2bNiwAY8fP4azs7OYMMbExCAyMhIJCQnIycnBwYMHxeX4Nm7ciIiICMTExODBgwdo0aIFhgwZonHsf//73/j111+RmZmJYcOGITY2FpmZmeL+iIgIcYUKc3NzhIeHQ6FQ4L///S++//57MfE+ffo0gNLlHfPy8spdH/1d3XPnzsXFixdx7tw5pKamQiaTYerUqRXG6/Lly/Dy8nruMOQTJ06gcePGAEonJL1//z6aNWsm7m/WrBmuXr0qvn/8+DEWL16MNWvWlKvr6tWrGp+VSCRo2rSpxucbNmyIhISESttDL49JKxERERHpjeIiBYyMrSGVGj23nKowC6d3tMVv21vh8sFP4NJwKOw8/AGUJkv29vYwMipfh7Ozs5gsPX78GC4uLtrvRCXKVoh43jG3b9+O0aNHo3Xr1pDJZJg7dy4KCgo07hR+8sknaNy4MWQyGYYNGyaudCCTyVBYWIhr165BpVKhbt268Pb2BgBs2rQJS5cuhZeXF4yMjLBw4UKcP39eY7m+oKAgODo6wtTUFG5ubvjggw/EFSeuXr2Ku3fvok+fPgCAdu3aoWnTppBKpWjSpAmGDh2KEydOvFAcnle3IAgICwvD2rVrxbYsWbIEUVFRKCkpKVfXkydPIJfLKz1WdHQ0du7ciSVLlgAofYYZAKytrcUy1tbWKCwsFIdFf/bZZ5g+fXqFSw7m5eVpfLbs87m5ueJ7S0tL5OTkaCzNSK+Hi+sREREREZwvfPDc/dJizTWwHa+9A0FS+UW5tOTVLjMNja2hKlJArVY9N3E1MrFF68G/QK1WIe1KOLLSTkBQF0NiYAR7e3s8evQIKpWqXOKakZEhTpBjZ2dX6QoPFfnxxx8xY8aMFyo7YMAAhIeHa2yzs7MDANy/f19MJp+Vnp6ucWdSKpXCw8MD6enp4ra/TiRlZmYmJmL+/v5YtGgRvvjiCyQlJaFbt25Ys2YNXFxckJycLC7F99e609LSxLuxnp6eGm0ZMWIEfvjhB0yZMgURERHo27cvzMzMAADnzp1DUFAQrl69CqVSiaKiIvTt2/eFYvO8uh8+fIj8/Hy0adNGo7xUKsWff/4JV1dXje02NjbIyal4LovY2Fh8+umniI6Oho+PDwCIkzFlZ2dr/NvExASGhoY4efIkfv/9d3z//fcV1mlhYSE+E10mOztb4259bm4u5HK5Rqzp9TCSRERERASjQvPnvgyKNZM/wyLT55dXGb9SO+QOfpAayPA45egLlZdKjeDhNw6ABBk3S+/ctWnTBiYmJvjPf/6jUVatVmP37t3w9y+9I9ulSxccPnxYTPr+zqeffoq8vLwXej2bsAJA/fr14eHhofH86rNcXV01lm5Rq9VITU0tl6xVZvLkyYiLi8O9e/dQXFyMOXPmAAA8PDywb98+KBQK8VVQUCA+fwqg3Oy5AwYMQEJCAm7fvo0dO3aIw3eB0iG+PXv2REpKCrKzszFp0iSULUrybD0Vqaxue3t7mJqaIiEhQaOthYWFFcagadOmSE5O1hg+DZTO0DxkyBBERUXhww8/FLfb2NjAxcVFY/huQkKCOHz4l19+we3bt+Hq6gonJyesWrUK0dHR4rKKjRs31visIAi4fPmy+HkAuHHjhsYQYnp9TFqJiIiISG8Yyizg1Xwq7pwLQWbyEZSo8iGoS6D4Mx63Ti+q9HPuTcYg9eoWlBQXQi6XY/HixZg2bRr2798PpVKJ+/fvY9SoUVAoFJg2bRqA0rt9Xl5e6NOnD65du4aSkhJkZ2fj66+/xo8//qj1vkkkEqxduxZLly7Fd999B4VCAUEQcPHiRQwePBgAMHz4cISHhyMuLg4qlQorVqyATCbTmGioMufPn8eZM2egVCphZmYGMzMzGBgYACidnGjevHm4c+cOgNJhtWXDcysjl8vRq1cvTJ48GSqVCh07dhT35ebmwsbGBqampoiPj8eOHTvEfQ4ODpBKpeKxXqZuqVSKCRMmIDAwEBkZGQCAhw8fIjo6usJ63Nzc0LBhQ/z222/ituPHj2PAgAGIiIgQf6D4q9GjR2PJkiXIzMxEamoqVq9ejTFjxgAAZs+ejVu3biEhIQEJCQmYOHEiunbtipMnTwIA+vXrh1u3bmH79u1QKpXipF6dOnXSOP6zE03R62HSSkRERER6xfXt4ajXKgjp17bh7L8DcC6qE1IuhYnPrFbE1r0DDGVy/JlUehdz1qxZWL16NebNmwcrKyu4urri2rVrOHHiBOrUqQOg9BnQ2NhYNGvWDD179oRcLkfTpk2RkJCAzp07V0nf+vXrhz179iAyMhLu7u6wt7fH5MmT0atXLwClQ3xXr16Njz/+GA4ODoiJicGBAwdgbPz3d65zcnIwceJE2NnZwcXFBdnZ2QgJCQEATJs2DUOGDEGvXr0gl8vh5+cnTkj1PCNGjMCRI0cwZMgQMQEGSpeZWbx4MSwtLbFw4UKNyaXMzMwwf/58+Pv7w9rautLjVFZ3SEgI/Pz88OGHH8LS0hLvv/8+4uLiKm3j5MmTsW3bNvH9okWLkJOTg/79+8PCwgIWFhZo1KiRuD84OBh+fn5466230KxZM/Tu3RsTJ04EUDr818nJSXxZWFjAxMQEjo6OAEqHeO/duxchISGwsrJCREQE9u3bJ/7/KSgowIEDB2rMmrv6QiKU3ccnIr2VlpYGd3d3TA0KhdzKtrqbo5fMhCy0L/4WAHDCcALyJa8Xp5OJ72qjWTWWqigb6Umlw+5cffvDyNiqmltUsxUbF+BhowsASp8jNCwyreYWEem3X8P9ym3bt28fxowZg6NHj6Jp06bV0CqqKiqVCs2aNcOBAwfg5eVVrW1ZvXo1Hj16hOXLl1drO2oaJq1Eb4CypDU1NRVubm7V3RwiIqI30t69e/HgwQNMmDChuptCRC+BswcTERERUa1QtlwLEb1Z+EwrERERERER6S3eaSWiGqG44Akex5fO9Gj37qcwNLWp5hbVbE+y87A1snQ5ilFDAmBjZVHNLarZFIUF2H4jAQAwomEzWJvwmVYiIqo9mLQSUY1Qkp+FP48vBQBYv92XSWsVe6LIw4oNpRMxfdStJZPWKvakqBCr4k8BAHp6N2DSSkREtQqHBxMREREREZHeYtJKREREREREeotJKxEREREREektJq1ERERERESkt5i0EhERERERkd5i0kpERERERER6i0veEL0BiouLAQAZGRnV3BL9VfQkAw+y1QAAi/sZMC4wruYW1WwZGQ9Ronr6//99HzKD4mpuUc2WkaNA8RNF6b/T02GU+7R6G0RERLWek5MTDA11k05KBEEQdHIkInpl58+fR8uWLau7GUREREREAIDU1FS4ubnp5FhMWoneAIWFhbhy5QocHBx09ovWmygjIwMtW7ZEXFwcnJ2dq7s5NR7jrVuMt+4x5rrFeOsW461bNTHeurzTyqtfojeAiYkJ3nvvvepuxhvD2dlZZ7/8EeOta4y37jHmusV46xbjrVuM96vhRExERERERESkt5i0EhERERERkd5i0kpENYZcLkdwcDDkcnl1N6VWYLx1i/HWPcZctxhv3WK8dYvxfj2ciImIiIiIiIj0Fu+0EhERERERkd5i0kpERERERER6i0krERERERER6S0mrURERERERKS3mLQSkV6IiopC79694ebmBnNzczRr1gybN2/Gs3PF/fjjj/D19YWJiQn8/Pywf/9+jf13795Fr1694ObmBhMTE7i4uGDgwIFISkoqd8zExER07twZ5ubmcHJywueffw6lUlml/dQXuo53eHg4JBJJuVdQUFCV91UfaCvez+rTpw8kEglWrVpVbt/p06fRpk0bmJqawtPTEytWrCh3vJpK1/FeuHBhhed3WFiY1vumr7QV8+Tk5Apj2bp163LH5Dmuu3jX9nNc298pZ8+eRadOnWBpaQm5XI7WrVsjISFBo0xtvkapiGF1N4CICADWrFkDLy8vrF69Gg4ODjhy5AjGjRuH1NRUBAcHAwAiIyMxbtw4zJs3DwEBAdi1axf69u2LkydPin9g8/Ly4OTkhOXLl8Pd3R0ZGRlYvnw5/P39cenSJdjb2wMAnjx5goCAALz11lv46aefkJ6ejlmzZiE/Px8bN26stjjoiq7jXebQoUOwsrIS37u6uuqu09VIW/H+q4MHD+Ls2bMVHu/27dvo2rUrOnfujKVLl+Ly5csICgqCgYEBZs+eXaV91Qe6jjcAmJqa4ujRoxrbvL29tdsxPabtmC9btgz+/v7ie0tLS439PMd1G2+gdp/j2oz30aNH0aNHD4wZMwZz5syBSqVCXFwc8vPzxTK1/RqlQgIRkR7IzMwst23cuHGCXC4XSkpKBEEQBF9fX2Ho0KEaZdq0aSN07979uXUnJSUJAISIiAhx27JlywRzc3Ph8ePH4rZvv/1WMDAwENLT01+nK28EXcd7y5YtAoAKj1sbaDvehYWFgo+Pj7B582YBgLBy5UqN/ePHjxc8PT2FoqIicdvcuXMFa2trobCwUBtd0mu6jndwcLBgbm6uxR68ebQV83v37gkAhKioqOcej+e4buNd289xbcVbpVIJXl5ewueff/7c49X2a5SKcHgwEemFZ+/IAUDz5s2Rk5ODp0+f4u7du0hKSsKgQYM0ygwZMgS//PILioqKKq3bzs4OADSG1Rw8eBCdOnWCra2tuG3QoEFQq9WIiYl53e7oPV3Hu7bTdrxXrVoFGxsbjB49usLjHTx4EH369IFMJtOoS6FQ4MyZM6/fIT2n63hT1X6nVITnuG7jXdtpK96xsbFITk7G9OnTn3u82n6NUhEmrUSkt06dOgVXV1dYWloiMTERANCgQQONMg0bNoRSqcS9e/c0tqvVaqhUKiQnJ2Pq1Klwd3dH3759xf2JiYnl6rK2toazs7N4rNqmKuNdplGjRjAwMIC3tzeWL1+OkpKSquuQnnvVeKekpGD58uX45z//CYlEUq7ep0+fIjU1tVxdDRo0gEQi4fmt5XiXKSgogIODAwwNDfH222/j+++/r5qOvEFe5ztl0qRJMDAwgKOjI8aNG4esrCxxH8/xilVVvMvwHNf0KvE+e/Ys7OzsEB8fj/r168PQ0BC+vr7Ytm2bxud4jVIen2klIr106tQpREZGYvXq1QBKn+8ASr+0/8rGxgYAyv2BHTlyJCIiIgAA9erVQ2xsrMazlE+ePClXV1l9Ff2xrumqOt7Ozs5YtGgRWrVqBYlEgn379mH+/PlIT0+vlc/nvE68AwMD0a9fvwqfuwQAhUJRYV0ymQxmZmY8v6HdeAOAj48PVqxYgebNm6OwsBA7duzA+PHjkZ2dXSuer6zIq8bc2NgYkyZNQteuXWFtbY1z587hq6++Qnx8POLi4mBkZMRzvAJVGW+A5/izXjXef/75J54+fYpPPvkEixcvxttvv40dO3Zg1KhRqFOnDrp27SrWx2sUTUxaiUjvpKWlYfDgwfD39//bITSVWbJkCWbMmIGUlBSsW7cOnTp1wqlTp+Dh4aHl1r75dBHvrl27in+MAaBLly4wNTXF2rVrMW/ePDg7O2ulL2+C14l3TEwMYmJicPPmzSpqXc2ji3iPGDFC433Pnj2hVCqxdOlSzJgxQ7zwry1eJ+bOzs745ptvxPft27dHo0aN0KtXL+zZs6fc8EvSTbx5jv/P68RbrVajsLAQK1aswNSpUwEAAQEBSExMxFdffaXxd5I0cXgwEekVhUKB7t27w87ODv/5z38glZZ+TZX9Wpmdna1RvuzXzb8+9wEAdevWxXvvvYf+/fvj8OHDKCkpQWhoqLjfxsamXF1l9T1bV02mq3hXZNCgQSgpKSk3zX9N9rrxnj59OqZPnw4zMzMoFArxjlNhYWG5u0/P1qVUKpGfn8/zG9qNd2UGDRqE7Oxs3L59W4s90n/a+k75qx49esDc3BwXLlwAwHP8r3QR78rUxnP8deNdVi4gIECjXMeOHXHt2jXxPa9RymPSSkR6o6CgAL169UJ2djYOHjyoMby07NmOZ5/lSExMhEwme+60+2ZmZmjYsKHGH9YGDRqUqys7OxsZGRnlniOpqXQZb9JOvG/evIlly5bBxsZGfAHAggULYGNjg8LCQpibm8Pd3b1cXTdv3oQgCDy/od140/9U1XfKs3iOl9JVvKmUNuLdqFGjSuv/6/cJr1HKY9JKRHqhuLgYgwYNwo0bN3Do0KFy63d6e3vD19cXUVFRGtt37dqFjh07aswg+aycnBxcvnxZ44909+7dERsbq3G3JCoqClKpFF26dNFOp/SYruNdkcjISBgYGKB58+av3pE3hLbifezYsXIvAJg4cSKOHTsmluvevTuio6OhUqk06rK2tsb7779flV3VC7qOd0UiIyNhbW0NHx8fLfdOP1Xld8r+/fvx9OlTvPfee+I2nuO6jXdFatM5rq14d+3aFUZGRoiNjdUod+TIEbzzzjvi+9p+jVKh6l5zh4hIEErXOwMgrF69Wjhz5ozGq2zNvR07dggSiUT48ssvhWPHjgkTJ04UDA0NhdOnT4v1BAcHCzNnzhSioqKE48ePC+Hh4UKLFi0EuVwu3Lx5UyyXlZUlODs7C+3btxcOHz4sbN68WbC2thamTJmi875XB13Hu0uXLkJISIhw4MAB4cCBA8KECRMEiUQizJw5U+d9rw7aindFUMG6obdu3RLMzc2F/v37C7/88ouwbt06QSaTlStXU+k63i1atBDWr18vHD58WNi7d68wcOBAAYCwbt26KuujvtFWzGfNmiXMnj1b2L17txAbGyssW7ZMsLS0FN59911BpVKJ5XiO6zbetf0c1+Z3yj/+8Q/B3NxcWLdunXD48GFh9OjRgkQiEQ4fPiyWqe3XKBVh0kpEesHT01MAUOHr3r17YrkffvhB8PHxEWQymdCkSRPh559/1qgnOjpaaN++vWBnZycYGxsL3t7ewujRo4U7d+6UO+b169eFjh07CqampoKjo6Mwe/ZsjYXqazJdx3v69OnCW2+9JZiamgrGxsZCkyZNhPXr1wtqtVoX3a122op3RSpKogRBEH777TehVatWgrGxseDm5iYsX76c8a6ieA8aNEjw8vISTExMBFNTU6Fly5bC9u3btd0tvaatmP/www/iD1+GhoaCp6enMHPmTCE7O7vcMXmO6y7etf0c1+Z3ikqlEoKDgwVXV1dBJpMJfn5+QnR0dLlytfkapSISQRAELd64JSIiIiIiItIaPtNKREREREREeotJKxEREREREektJq1ERERERESkt5i0EhERERERkd5i0kpERERERER6i0krERERERER6S0mrURERERERKS3mLQSERERERGR3mLSSkRERG8Uf39/NGzYEEqlsty+gQMHwt3dHXl5edXQMiIiqgpMWomIiOiNEhYWhnv37iE0NFRj+6FDh7B7925s2LABFhYW1dQ6IiLSNokgCEJ1N4KIiIjoZSxatAghISG4du0avL29UVhYiMaNG6Nx48bYu3dvlR23oKAApqamVVY/ERGVxzutRERE9MaZO3cuPD09MWXKFADAsmXL8ODBA2zcuBFpaWkYMWIE7O3tYWpqinbt2uHChQsan9+2bRvatm0LW1tb2NjYoEOHDoiLi9Mos3DhQlhYWCAuLg5t2rSBiYkJvv76a531kYiIShlWdwOIiIiIXpZMJsO3336LDh06YOnSpQgNDUVISAjMzc3RvHlzWFhYYMOGDbCyssKGDRsQEBCAW7duwdHREQCQnJyMkSNHol69elAqldi5cyfatWuHy5cvw9fXVzyOUqnEsGHDEBgYiGXLlsHOzq66ukxEVGtxeDARERG9scaMGYMtW7agRYsWiIuLw+LFi7F+/XokJSWJCWpRURF8fX0xePDgcs/BAoBarYZarUbjxo3Rr18/LFu2DEDpndZFixYhMjISgwcP1mm/iIjofzg8mIiIiN5YQUFBAIB//OMfMDAwQExMDPz9/WFra4vi4mIUFxfDwMAA7du3x/nz58XP3bhxA3379kWdOnVgYGAAIyMj3Lx5E0lJSeWO0bNnT531h4iIyuPwYCIiInpjyWQyjf8+evQIZ8+ehZGRUbmy9erVAwDk5uaiS5cucHBwwJo1a+Dp6QkTExOMHTsWhYWFGp8xMzPjTMRERNWMSSsRERHVGLa2tujWrRuWLFlSbp+xsTEA4MyZM0hLS8P+/fvh5+cn7s/Ozoabm5vGZyQSSdU2mIiI/haTViIiIqoxOnXqhO3bt6Nhw4YwNzevsExBQQGA/92dBYDTp08jOTkZjRo10kk7iYjoxTFpJSIiohpj1qxZiIiIQPv27TFjxgx4eHggMzMT586dg4uLCwIDA9G6dWtYWFhgypQpCAoKQnp6OoKDg+Hq6lrdzSciogpwIiYiIiKqMezs7HD27Fk0a9YMc+bMQZcuXRAYGIjk5GS0atUKAFCnTh1ERUXh4cOH6N27N9atW4dvv/0WPj4+1dx6IiKqCJe8ISIiIiIiIr3FO61ERERERESkt5i0EhERERERkd5i0kpERERERER6i0krERERERER6S0mrURERERERKS3mLQSERERERGR3mLSSkRERERERHqLSSsRERERERHpLSatREREREREpLeYtBIREREREZHeYtJKREREREREeotJKxEREREREektJq1ERERERESkt/4fuGTaEgqqwpgAAAAASUVORK5CYII=\n"
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "4 of 5 data classes are exposed under the 2035 planning scenario if migration starts in 2026.\n"
     ]
    }
   ],
   "source": "THIS_YEAR = 2026\nDATA_CLASSES = {  # x = confidentiality horizon (years) — assumptions for discussion\n    'Real-time query traffic': 2, 'Credit reports & scores': 10, 'Credit histories (KRS / Risk Center)': 15,\n    'Identity & demographic data': 30, 'Archives & DC backups': 20}\nMIGRATION_YEARS = 5                  # y — typical enterprise PQC migration estimate\nCRQC_SCENARIOS = {'Aggressive (2030)': 2030, 'Regulatory planning (2035)': 2035, 'Conservative (2040)': 2040}\n\nrows = []\nfor d, x in DATA_CLASSES.items():\n    exposed_until = THIS_YEAR + MIGRATION_YEARS + x\n    rows.append({'Data class': d, 'x (yrs)': x, 'Protected until': exposed_until,\n                 **{k: ('AT RISK' if exposed_until > z else 'ok') for k, z in CRQC_SCENARIOS.items()}})\nMOSCA = pd.DataFrame(rows); display(MOSCA)\n\nfig, ax = plt.subplots(figsize=(8.5, 3.4))\nfor i, (d, x) in enumerate(DATA_CLASSES.items()):\n    ax.barh(i, MIGRATION_YEARS, left=THIS_YEAR, color=GRAY, height=0.55)\n    ax.barh(i, x, left=THIS_YEAR + MIGRATION_YEARS, color=BLUE, height=0.55)\nfor (k, z), c in zip(CRQC_SCENARIOS.items(), [AMBER, NAVY, TEAL]):\n    ax.axvline(z, color=c, ls='--', lw=1.5, label=f'CRQC — {k}')\nax.legend(frameon=False, fontsize=8, loc='lower right')\nax.set_yticks(range(len(DATA_CLASSES))); ax.set_yticklabels(DATA_CLASSES.keys(), fontsize=8); ax.invert_yaxis()\nax.set_xlabel('Year'); ax.set_title(f'Mosca model: migration ({MIGRATION_YEARS} yrs, grey) + confidentiality horizon (blue)')\nplt.tight_layout(); FIG_MOSCA = savefig('07_mosca'); plt.show()\nAT_RISK_2035 = int((MOSCA['Regulatory planning (2035)'] == 'AT RISK').sum())\nprint(f'{AT_RISK_2035} of {len(DATA_CLASSES)} data classes are exposed under the 2035 planning scenario if migration starts in {THIS_YEAR}.')"
  },
  {
   "cell_type": "markdown",
   "id": "c18bcc87",
   "metadata": {
    "id": "c18bcc87"
   },
   "source": "## Stage 14 — Cryptographic inventory (CBOM) prioritisation template\nEvery regulatory roadmap (NIST, EU, NCSC, G7, MAS) starts with a cryptographic inventory. The systems below are a **generic, illustrative template** for a credit bureau — not a description of KKB's systems — to be populated during a discovery phase."
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "000f4f19",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:54.914069Z",
     "iopub.status.busy": "2026-10-03T09:47:54.913571Z",
     "iopub.status.idle": "2026-10-03T09:47:55.416629Z",
     "shell.execute_reply": "2026-10-03T09:47:55.414064Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 747
    },
    "id": "000f4f19",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022095812,
     "user_tz": -180,
     "elapsed": 459,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "d55f5aee-c682-4331-af01-c35399c4690f"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "                        System (illustrative)          Typical current crypto  \\\n",
       "0       Member-bank query API (bureau ↔ bank)  TLS 1.2/1.3, ECDHE + RSA certs   \n",
       "1                PKI root & issuing CAs, HSMs          RSA-4096 / ECDSA roots   \n",
       "2              Archives & data-centre backups          AES-256 + RSA key wrap   \n",
       "3  E-government / public-registry integration            TLS + RSA signatures   \n",
       "4          Primary ↔ secondary DC replication              IPsec IKEv2 (ECDH)   \n",
       "5                Consumer portal & mobile app          TLS 1.3, ECDHE + ECDSA   \n",
       "6                 Fintech / partner SaaS APIs        TLS + OAuth2 JWT (RS256)   \n",
       "7              Document & report e-signatures         RSA / ECDSA e-signature   \n",
       "8             Code signing & software updates             RSA-3072 signatures   \n",
       "9                       Internal service mesh                    mTLS (ECDSA)   \n",
       "\n",
       "   Data horizon yrs  Exposure  Criticality  HNDL risk  Priority score  \\\n",
       "0                15         5            5        3.8            4.52   \n",
       "1                20         2            5        5.0            4.10   \n",
       "2                20         2            5        5.0            4.10   \n",
       "3                15         4            4        3.8            3.92   \n",
       "4                15         3            5        3.8            3.92   \n",
       "5                10         5            4        2.5            3.70   \n",
       "6                10         5            3        2.5            3.40   \n",
       "7                10         4            4        2.5            3.40   \n",
       "8                10         3            4        2.5            3.10   \n",
       "9                 5         1            3        1.2            1.68   \n",
       "\n",
       "               Wave  \n",
       "0  Wave 1 (2026–28)  \n",
       "1  Wave 1 (2026–28)  \n",
       "2  Wave 1 (2026–28)  \n",
       "3  Wave 2 (2028–30)  \n",
       "4  Wave 2 (2028–30)  \n",
       "5  Wave 2 (2028–30)  \n",
       "6  Wave 2 (2028–30)  \n",
       "7  Wave 2 (2028–30)  \n",
       "8  Wave 3 (2030–35)  \n",
       "9  Wave 3 (2030–35)  "
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-636aa0cb-22ac-41d6-b42c-954dff638728\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>System (illustrative)</th>\n",
       "      <th>Typical current crypto</th>\n",
       "      <th>Data horizon yrs</th>\n",
       "      <th>Exposure</th>\n",
       "      <th>Criticality</th>\n",
       "      <th>HNDL risk</th>\n",
       "      <th>Priority score</th>\n",
       "      <th>Wave</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Member-bank query API (bureau ↔ bank)</td>\n",
       "      <td>TLS 1.2/1.3, ECDHE + RSA certs</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>3.8</td>\n",
       "      <td>4.52</td>\n",
       "      <td>Wave 1 (2026–28)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PKI root &amp; issuing CAs, HSMs</td>\n",
       "      <td>RSA-4096 / ECDSA roots</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.10</td>\n",
       "      <td>Wave 1 (2026–28)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Archives &amp; data-centre backups</td>\n",
       "      <td>AES-256 + RSA key wrap</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.10</td>\n",
       "      <td>Wave 1 (2026–28)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>E-government / public-registry integration</td>\n",
       "      <td>TLS + RSA signatures</td>\n",
       "      <td>15</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>3.8</td>\n",
       "      <td>3.92</td>\n",
       "      <td>Wave 2 (2028–30)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Primary ↔ secondary DC replication</td>\n",
       "      <td>IPsec IKEv2 (ECDH)</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>3.8</td>\n",
       "      <td>3.92</td>\n",
       "      <td>Wave 2 (2028–30)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Consumer portal &amp; mobile app</td>\n",
       "      <td>TLS 1.3, ECDHE + ECDSA</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.70</td>\n",
       "      <td>Wave 2 (2028–30)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Fintech / partner SaaS APIs</td>\n",
       "      <td>TLS + OAuth2 JWT (RS256)</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.40</td>\n",
       "      <td>Wave 2 (2028–30)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Document &amp; report e-signatures</td>\n",
       "      <td>RSA / ECDSA e-signature</td>\n",
       "      <td>10</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.40</td>\n",
       "      <td>Wave 2 (2028–30)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Code signing &amp; software updates</td>\n",
       "      <td>RSA-3072 signatures</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.10</td>\n",
       "      <td>Wave 3 (2030–35)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Internal service mesh</td>\n",
       "      <td>mTLS (ECDSA)</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1.2</td>\n",
       "      <td>1.68</td>\n",
       "      <td>Wave 3 (2030–35)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-636aa0cb-22ac-41d6-b42c-954dff638728')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-636aa0cb-22ac-41d6-b42c-954dff638728 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-636aa0cb-22ac-41d6-b42c-954dff638728');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "  <div id=\"id_528ff14f-d75a-4e87-982c-524de70b8a8c\">\n",
       "    <style>\n",
       "      .colab-df-generate {\n",
       "        background-color: #E8F0FE;\n",
       "        border: none;\n",
       "        border-radius: 50%;\n",
       "        cursor: pointer;\n",
       "        display: none;\n",
       "        fill: #1967D2;\n",
       "        height: 32px;\n",
       "        padding: 0 0 0 0;\n",
       "        width: 32px;\n",
       "      }\n",
       "\n",
       "      .colab-df-generate:hover {\n",
       "        background-color: #E2EBFA;\n",
       "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "        fill: #174EA6;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate {\n",
       "        background-color: #3B4455;\n",
       "        fill: #D2E3FC;\n",
       "      }\n",
       "\n",
       "      [theme=dark] .colab-df-generate:hover {\n",
       "        background-color: #434B5C;\n",
       "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "        fill: #FFFFFF;\n",
       "      }\n",
       "    </style>\n",
       "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('CBOM')\"\n",
       "            title=\"Generate code using this dataframe.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "    <script>\n",
       "      (() => {\n",
       "      const buttonEl =\n",
       "        document.querySelector('#id_528ff14f-d75a-4e87-982c-524de70b8a8c button.colab-df-generate');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      buttonEl.onclick = () => {\n",
       "        google.colab.notebook.generateWithVariable('CBOM');\n",
       "      }\n",
       "      })();\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "CBOM",
       "summary": "{\n  \"name\": \"CBOM\",\n  \"rows\": 10,\n  \"fields\": [\n    {\n      \"column\": \"System (illustrative)\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 10,\n        \"samples\": [\n          \"Code signing & software updates\",\n          \"PKI root & issuing CAs, HSMs\",\n          \"Consumer portal & mobile app\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Typical current crypto\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 10,\n        \"samples\": [\n          \"RSA-3072 signatures\",\n          \"RSA-4096 / ECDSA roots\",\n          \"TLS 1.3, ECDHE + ECDSA\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Data horizon yrs\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4,\n        \"min\": 5,\n        \"max\": 20,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          20,\n          5,\n          15\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Exposure\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1,\n        \"min\": 1,\n        \"max\": 5,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          2,\n          1,\n          4\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Criticality\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 3,\n        \"max\": 5,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          5,\n          4,\n          3\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"HNDL risk\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.2240188814810915,\n        \"min\": 1.2,\n        \"max\": 5.0,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          5.0,\n          1.2,\n          3.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority score\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.7870507536931075,\n        \"min\": 1.68,\n        \"max\": 4.52,\n        \"num_unique_values\": 7,\n        \"samples\": [\n          4.52,\n          4.1,\n          3.1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Wave\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Wave 1 (2026\\u201328)\",\n          \"Wave 2 (2028\\u201330)\",\n          \"Wave 3 (2030\\u201335)\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 935x396 with 1 Axes>"
      ],
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5wAAAGBCAYAAAAKZb69AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAAzYpJREFUeJzs3Xd4FFX78PHvpgd2k03vBQhNpEPoGJAaivSW0Ds2yCMdMQgKIk3hAR4VCb0K0kNRKYqAVCnSlJBGSUhCEkrqvH/wZn5ZUlGWKN6f65rr2p1z5pwzZyabvfecmdEoiqIghBBCCCGEEEI8ZyYl3QAhhBBCCCGEEC8nCTiFEEIIIYQQQhiFBJxCCCGEEEIIIYxCAk4hhBBCCCGEEEYhAacQQgghhBBCCKOQgFMIIYQQQgghhFFIwCmEEEIIIYQQwigk4BRCCCGEEEIIYRQScAohhBBCCCGEMAoJOIUQQgghhBBCGIUEnEIIIYo0YMAANBoNGo2GAQMGlHRz/pSIiAh1HzQaDRERESXdJPE3kfu8OHjwoLo+ICBAXR8aGqquDw0NVdcHBAS88Pb+HYWFhal94uvrW9LNEcX0Mny2i78/s5JugBBC/B2FhYUxcODAPOtNTU2xt7enevXqDBgwgD59+qDRaPLkO378OF988QU//vgjMTExZGZm4ujoSJ06dejVqxfdu3fH1NQ037ozMzPZsmULGzZs4OTJk9y9exdzc3Pc3d2pWbMmPXr0oEOHDpiZyUd4bgcPHlSDBV9fX/nyJF5KCxYsICkpCYBOnTpRo0aNF1JvWFiY+iNNQECABNpGVFLH+GUl/xtKnnxbEUKIZ5CVlUVcXBwHDhzgwIED7Ny5k3Xr1hmkv/POOyxevDjPtjExMcTExLBt2zY+++wztmzZgpubm0GeP/74gx49enDq1CmD9Y8fP+bKlStcuXKF9evXc+PGjRc6ijB58mSGDBkCgIuLywur91kcPHiQadOmAfDaa6/l+VLh5ubGkSNHDN4LARicF1WrVi3BlhRtwYIF3Lx5E3jy5flFBpyHDh1S3z8dcAYGBqr9aGVl9ULa9LJ6kcf4n/DZ/lcV9b9BGJ8EnEIIUQybNm3C1dWVxMREFi5cyP79+wFYv349ffr0oUOHDgCMGzfOINh84403GDBgAKVKlSI8PJzPP/+crKwsjh07Rvv27Tl27Bjm5uYAxMXF8frrr6ujCCYmJvTr148OHTpgZ2dHTEwMe/fuZdOmTS9254Hy5ctTvnx5o9eTkZGBoihYWFg897ItLS1p3Ljxcy9X/PP9W86L1NRUtFqtUcp2dnbG2dnZKGWXtEePHmFpaYmJyct3JdqL+mz/NzHm39k/liKEECKP5cuXK4C63LhxQ0178OCBYmlpqaa9++67iqIoypUrVxRTU1N1fefOnfOUO3fuXINy//e//6lpI0aMMEhbtWpVvm27ceOGkpycXOQ++Pj4qGV9/fXXyvz58xU/Pz/F0tJSqVatmvLNN98oiqIop06dUlq2bKmULl1asbOzU4KCgpT4+HiDsvr376+W1b9/f4O0hIQEZcSIEYqzs7NiZWWl1KtXT9m1a5fywQcfqNu89tprav4ffvjBYD9jY2OV/v37K05OTopGo1HOnDmjJCUlKSNHjlQaNGiguLu7K9bW1oqFhYXi6empdOvWTfnpp58M+iN3efkt+eXLfUwVRVEuXryoDB48WClTpoxiaWmplC5dWqlataoyceLEPP2R+/zw8fFRIiMjlX79+imOjo6KpaWlUqdOHeXAgQNFHiNFUZSWLVuqZX311Vfq+pkzZ6rr58yZo65fsmSJur5t27bq+hkzZigtW7ZUfHx8FJ1Op5iZmSmOjo5K8+bN85xLffr0UcsYO3ZsnjZVrVrV4NzJkZSUpHz44YdKrVq1FJ1Op1hYWCjlypVT3n77bSU2NjZPOfv371cCAwMVFxcXxczMTNHpdEqZMmWUDh06KJ9//nmx+udFyH1e/PDDD+r61157TV3/wQcfqOsLOrefPi9yK2gbRVGUDRs2KM2aNVMcHBwUMzMzxdbWVilfvrzSrVs3Zc2aNXm2z2/J+bt8up4TJ04oLVq0UHQ6naLX6xVFUZQzZ84offv2VapVq6Y4Ozsr5ubmSqlSpZSKFSsqI0eOVG7evJnvPuW35OxLfvt+5coVdZ2lpaWSkJBgsN/ffPONmu7r66tkZ2eraT/88IPStWtXxcPDQzE3N1f0er3SrFkzZePGjcU8qv9nz549SufOnRUPDw/FwsJC0ev1Ss2aNZXQ0NACj92VK1eUzp07K3q9XgGU2bNnq+l169bNU8dbb72lpg8cODDfMiMjI5WgoCDFwcFBsbKyUurXr6/s27dPLaO4x1hRnvwfmjVrllK3bl1Fp9Mp5ubmiqenp9KjRw/l6NGjxe6bgj7bn/Uz7osvvii0f95++201vV+/fur6rKwsZfny5Urz5s0Ve3t7xdzcXHF1dVW6d++unDx5Mk85uf+vLV++XFm+fLlSo0YNxdLSUnFyclJGjRqlPHz4UFGU4v9vyLFnzx6lY8eOiqurq3rONWrUSFm8eLGSkZFRaL+Fh4crDRo0UEqVKqVUr169yM/YV199VU1fsWJF8Q7WP5gEnEIIkY/CAk5FURRbW1s1beTIkYqiKMr06dMNtjl9+nSech89eqQ4OjqqeZo3b64oiqKkp6crNjY26vpmzZr95X3I/Y+5YsWK+f6znTFjhmJlZZVnfe5ARlEK/lLy8OFDpXr16nm212g0Ss2aNfP9gv10wFm+fHmD92fOnCnyi4KJiYmyfft2RVGeT8D5zTffGPyI8PTi7u6uXL16Vc2f+/ywsbFRXFxc8mxjZWVl8MW9IJ988km+X8TatGmjru/QoYO6vmfPnur6uXPn5nu881vee++9fI+Bh4eHkpWVpaadPXvWYN8ePHig9p+vr2+B5Ts7Oyvnz583qEOj0RSYv2LFikX2zYuSu10vOuBcsWJFocetdevWebbPb8kv4PTw8FCsra3V97a2tnnamd/i6OioREREFCtvYQHn0324dOlSgz7p1KmTwWdRjqlTpxZaZ85nblGys7OV4cOHF1hO7nbmbr+tra3i5ORkkDcmJkaxs7NT3589e1bdNisrS3Fzc1PTfvzxxzxl6vV6xd3dPU8bTE1Nla1btz7TMb57967yyiuvFJjPxMSk2D/oFCfgLM5nXEpKiqLVatW0y5cvq2VlZGQY9GdO/zx+/Fhp3bp1gfthZmamrF692qC9uT/nnv7f8fT58SwB53vvvVdovqZNm6qB7NP9VrZsWcXExER9X716deXnn39W3zs7Oyvp6enqtpcuXTLo25zP2JfZyzc3QAghjCgpKYkpU6Zw//59dV3NmjUBOHfunLrO3Nyc6tWr59neysqKV199VX3/66+/AnDt2jWSk5PV9a1bt36u7b527RoffPABu3btonLlyur6KVOmUKlSJb799lumTJmirt+zZw9Xr14tstx58+ap+21iYsLkyZPZvXs3gwcP5syZM8VqW2RkJB9++CF79+7liy++wNHRERsbG6ZNm8a6devYs2cPBw8eZPfu3YwZMwaA7OxsJk2aBPzftZm5b/JUo0YNjhw5oi6FuXv3Lv379yctLQ2Apk2b8u2337JixQo8PT0BiI2NJTg4ON/tk5OTKV26NOvWrWP58uXY2toCT667Xbp0aZH736JFC/V1zjVyWVlZ/PTTT+r6I0eOkJ2dbZDn6W3ffPNNli9fzs6dOzl48CD79+9n0aJFWFpaAk+uC7tz5w7w5Pq7nGl0MTExfP/992o5q1evVl8HBQVRqlQp9XXOdO86deqwYcMGwsPD6dmzp9qPvXr1Utu5efNmFEUBoHv37oSHh7Nnzx6++uorBg0ahIeHR5F982+wceNG9fWoUaPYv38/O3bsYMmSJfTu3RtHR0cABg0axJEjR3B1dVXzT5o0ST3HJ0+enKfsmJgY7O3t+fLLL9m3bx8ffvghAK+88gqffvop33zzDfv27ePgwYNs3bqVNm3aABAfH8+cOXOA/7s2M/d1hAMHDlTrXbhwYaH7N3ToUPX1qlWr1NeJiYns3r0beHIztpy/39ztNDMzY+rUqezbt48vvvgCJycnAJYsWcKGDRsKrRfg66+/5n//+5/6vnXr1qxfv57du3cze/Zs/Pz88t3u/v37ZGRksGDBAvbt28dnn32GnZ0dgwcPVvN8+eWX6usjR45w69YtACpWrEijRo3ylJmUlISNjQ0bNmxg48aNvPLKK8CTv/URI0bw+PHjYh/jN998k0uXLgFgZ2fHf//7X3bu3Ennzp2BJ5+Po0eP5sKFC0X2UXEU5zNOq9XSq1cvdZvcx3rv3r3ExcUBULlyZbV/pk2bxt69ewFwdXVlyZIl7N+/n9DQUExMTMjMzGTw4MFERkbm265r167x9ttvs2vXLrp166au//LLL3nw4EGx/zfs3r1bPd8Bhg0bxu7du5k1a5b6+Xn48GE++OCDfNvxxx9/UKlSJVatWsXevXt5++23qV+/PnXq1AGefDZu27ZNzZ/7b753797qZ+xLraQjXiGE+Dsq6lf9nOWVV15RHj16pCiKorRo0UJd7+rqWmDZuUeoLCwsFEVRlB9//NGg3C+//PIv70PuX4J79Oihrs89NQxQLl26pCjKk1/pdTqduj5nBFFRCv4VPPfUy+DgYIP669Spo6YVNsJZ0C/xOdOb3N3dFXNz83z7P/fU4sKmLCpKwSOcCxcuVNdptVolKSlJ3Wbfvn0G21y4cEFRlLznx4kTJ9Rtck+N7tKlSwFH5/9kZ2crDg4O6jYRERHK8ePHFXgyQpUzIn769Gnlt99+M/jVPPcUxN9++00ZNGiQ4ufnl++oNaDs2LFDzZ/fyGpWVpbi4eGhrs8Zpb9w4YK6TqPRKFu3blWOHDmiHDlyRDl06JDBKNqhQ4cURVGUyZMnq+v+85//KBEREQYjqcWRlJSk1vNnl8ePHxdZT+4+etEjnEFBQer6OXPmKDExMYW29ekphU/LXY9Go1HOnTuXJ09mZqayePFipWnTpoqDg4PBpQA5S+3atQ22Kagvitr3R48eGYwM/v7774qiKMrSpUvVdR07dlTzd+vWzWB97mOZ+2+rOLNA6tatW+hnQkHtf/rzL8eNGzfUkSy9Xq+OeI0aNUrd7pNPPimwzJzPD0Ux/JsClD179qhphR3jxMREg+O1bNkyNS0jI8Ng29yzGgpSnBHO4n7G5Xxu5ZwDOZ9PvXr1UtfPmzdPUZQnn3vOzs4G51TuY12vXj01bdq0afn2TWBgoLr+9u3bBu399ddf1bSi/jd07dpVTW/SpIlB2qRJk9Q0JycndZ9y91upUqWU27dv5yk39+yFVq1aqetzj04fP3680OPzspARTiGE+BNKlSrFoEGDOHjwoHpHRr1er6bfu3dPHel5Ws4oE/zfnVLt7OwM8ty7d++5trdhw4bqawcHB/W1ra2tOuJpYmKCvb29mpaQkFBkudeuXVNfN2nSxCCtuDdi6dq1a551YWFhtG3blu3btxMbG0tGRka+2yYmJharjsLkjBTAkzuU5vx6D0/2Kfdjb3LnzaHT6ahbt676Pnf/FqcPNRoNzZs3V98fPnxYHcV87bXXaNq0KfBkZDP3MyJff/11tW0XLlygbt26fP3111y/fp3Hjx/nW1fu9gwYMEC9YdWWLVt48OABP/zwAzExMQDUrl1bHb2/ePGiup2iKHTu3JkmTZrQpEkTXnvtNR49eqSmnz9/HoB+/fqh0+kAmDt3Lr6+vpQuXZoaNWrwzjvvFGsE/cyZM2o9f3bJGXn6uxoxYoT6iKP33nsPDw8PtFot/v7+TJw4kdjY2D9dtp+fH9WqVcuzfvDgwYwaNYrDhw9z7949srKy8uQpzrlbHFZWVvTt21d9nzPylXskPfcoaO5zbfv27QbHMveMgZzzrDC5y8o9AlYUS0tL2rdvn2e9r6+veoO4pKQkNm3aRHZ2Nt988w3wZES2X79++ZZpZ2dHlSpV1PdVqlQx+J9RnL+HnHy5j1fuz1kzMzMaNGigvs/v8+rPKO5nnL+/v3q+3bx5k8OHD5OSksL27duBJ/2a0z9xcXHcvXtX3XbatGkGx/r48eNqWkHH+vXXX8+3TU+3qyi5++np/1s5n785bY6Pj8+zfaNGjfK9w2/Pnj3VG2nt37+fiIgILl68qNb36quv4u/vX+x2/pNJwCmEEMWwadMmjhw5wtGjR7l06RJJSUksW7ZMneIFGHyxy8jIMPiykyMtLc1gmlPOtKry5ctjY2Ojrt+3b99zbX/uICr3nRZzf+F5mvL/p0MaW36PJ/n444/V1/7+/mzevJkjR44YPIIGKDCof5FyB+mAwfNRi9uHT0+rzR1wvvbaa+r63AFn7m0+//xzUlNTgSePNli2bBmHDh3iyJEj6pRMMOwvZ2dn3njjDeDJXRW3bt1aYBCQ37NmC5LTjgoVKnD+/HmmTp1Ky5Yt8fX1JT09nXPnzrFw4ULq169PdHR0scv9J8jdT5mZmQZpOVMKn9a4cWPOnj3L2LFjCQgIwMPDgwcPHvDLL78wa9YsmjRpQkpKyp9qT35/W7GxsaxYsUJ9HxQUxJ49ezhy5Ajjxo1T1z/Pv61hw4apr1evXk1ERIQ6ZdzT05O2bduq6cU913LOM2NwcXEpsB1vv/22+vrLL7/k0KFD6o+I7dq1M5gO+7J4ls+43Md61apVbNmyhYcPHwLQuXNnNTD8M58phbXr6edSv6j/X1DwI7YsLS3V/lAUhWXLlhlMp809RftlJwGnEEIUQ506dWjcuDENGjSgcuXK6shQbt27dzcI5mbOnJknz5IlSwx+Ic35Um9ubk5QUJC6/vvvv2f9+vX5tuXmzZt/+gvo81ahQgX19c8//2yQ9uOPPxarjPy+eOS+Zuf999+na9euNG7cOM+X+Nxy9/2zfFnOfU3rhQsXDK7P/fHHHw2+uOTO+zzlDh5/+OEHte9ee+019XmHuUc+n94md38FBwczaNAgmjZtire3d6Gj5bmDyi+++IItW7YAULp0afr06aOm5d5vU1NToqOjUZ7ceNBgSU1NZfz48cCTL1g+Pj5MmzaNffv2cePGDZKTk9XrzBITE9m1a1eh/RIQEJBvPc+yvMjn1eaeqRAfH69eF5yZmUl4eHi+2yiKQpUqVZg9ezY//PAD0dHRJCQkUL9+feDJ9WFHjx5V8z/LeV7U3xbA0qVLadOmDY0bNy70XPmzf1/wZDQvZ+Tt+vXrvP322+rf1aBBgzA1NVXz5j7Xhg8fXuBxzW+k6Wk5P+gB6ihkbgUFJYUFQ6+//rpa7o8//siMGTPUtEGDBhW4XWJiIr/99pv6PueHyxy5H01SWF9XqFDBoL9yX+udmZlp8DlsrM+rwgQFBWFtbQ08uY572bJlalruYNTR0dHgB9t169ble5zT0tLUEdI/q6hzN3c/5e5PMHxGr5OTk8EPeDkKO19yz2D4+uuv1WuPLSwsCrwvwMtInsMphBDPSaVKlXjrrbf4/PPPgSf/QLOystR/wHv37uWzzz5T87du3VodYYInU4r27Nmj3pglKCiI/fv30759e/R6Pbdu3WL//v2sX7+e3377TZ2uWJJ69uyp3vho1apV+Pj4UK9ePbZs2cLJkyf/dLlly5ZVv5zNnz8fc3Nzfv/9d4MbGz0t95SqX3/9lS1btuDs7Ixerze4UVN++zBx4kQePHhASkoKnTp1YsyYMSQnJ6s3JoInI62FlfNXlC1bljJlynDjxg3++OMP4MkoS8WKFVEUBXt7e4MpYuXLl8fb29tg+xybN2+mQYMGZGdnM23atEJ/6c8ZeYyIiDD4YtWrVy+D8+vVV1+lXr16HD9+nKysLFq2bElISAhly5YlISGBmzdvcuTIES5fvszly5eBJ9Nod+zYQWBgIGXKlMHJyYn4+HiD6Wu5p+K+DCpWrKi+TktLo3v37rRp04ZNmzapx/Vpo0eP5urVq7Rq1Qpvb28cHByIiopSPwfAsJ8cHBy4ceMG8GTmha+vLxYWFlSsWNHgC3xBcp8r8OSmNB06dOD7779n+fLlBW6X++9r9+7dNG7cmFKlSuHj44OXl1eR9Q4dOlQNhnbu3Ak8CQSeDtKGDBmiBodffPEFFhYWtGrVCgsLC6Kiorh06RI7duxg0qRJDBgwoNA6hw0bpn4OHTx4kHbt2tG/f39sbGz47bff2LVrFwcOHCiy7U976623GDVqFIB6wy1XV1cCAwML3a5bt26Ehoai0WgIDQ1V1zs7O9OsWTP1fVHHuHPnzmzevBl4Mg07LS0Nb29vvv76a27evAk86dv+/fs/8779VXq9nu7du7Ny5Uru37+vfq74+fmpP57BkyBt0KBBfPLJJwCMHDmSmzdvUrNmTdLS0oiKiuLkyZNs376dkydP/qUfjor63zBo0CD1x7bDhw8zcuRI3njjDc6fP8/cuXPVbQcMGPBMI7MAHh4edOnShY0bNxpMj+/YsWO+wetLyyhXhgohxD9cUY9FKUhGRkaht+HPWdq2baukpKTk2f73339XatWqVeT2xWlPQTeeKOzGJgVt82cei5J7fWE3DcpP7me65V6aN29eYD9cvnzZ4Nb0Ocvrr7+uKErhj0XZvHnzn34syrM8b7EwQ4cONagz942ecj8+AvI+FuLChQuKhYVFnnZXqVLF4MYc+d1k5unH+QDKsWPH8uS7fv264uXlVeh5mbsvcj9HNL/F1tZWiYyMLHb/GFPudv2VmwYpiuHjbHL/PVSpUiXfbYr6vPD29jb4rMh9M6bcS86zVotz/uV+RmDupVmzZgWe119++WW+20yfPl1RlML/JhRFUVJTUw0e/QQobdq0ybd9uW/UUtCS37n8tOzsbGXw4MHFOl+Lav/T+5L70ViAMn78+Dz5cpdpb2+f72OFTExM1Gci5yjqGL/ox6I862fckSNH8rRp1qxZefI9evRIef3114s81rk/qwu7oVJBf8dF/W9QFEX5z3/+U2gbCnssytPPpn7a0zcFBMObRP0byJRaIYR4jszMzFi6dCk///wzgwYNokKFCpQuXdogz6RJk9i9ezdarTbP9mXLluX48eOsX7+eLl264OXlhZWVFVqtlgoVKtCzZ0+++eYb9XEdJc3a2poffviBESNG4OTkhJWVFXXq1GHbtm0Gv2Y/3QdFGTp0KF9++SVVqlTBysoKX19fPvzwQ4NHHDytYsWKrFmzhmrVqmFhYfFM9XXt2pVTp04xaNAgdTTB2tqaV199lQkTJnDu3DmDKW/GkHuKLBjerCLnOs6C8lapUoXvv/+eRo0aodVqcXR0ZODAgfzwww/q9LaCPD2lsWrVqtSrVy9PvnLlyvHrr78ybdo0ateujU6nw8LCAg8PD+rXr8+UKVPUUQKAtm3b8s477+Dv74+rqysWFhZYWlpStmxZBg4cyIkTJ4o1MvZPs3LlSnr37o1er8fKyopGjRoRHh5e4E1r+vTpw/Dhw6lRowZOTk6YmZlhbW1NpUqVePvtt/n5558NPismT57MqFGjcHV1NZgq+Cy+/PJL3nvvPXx8fLCysqJGjRps2LChwBvewJPzZMqUKXh7exucL8X19DRtMJzSndtHH33E999/T7du3fDw8MDc3BxbW1sqVapEjx49WLVqFV26dCmyTo1Gw1dffcWOHTt44403cHNzU8uqWbOmweMynnVfnh6ZLWw6LTy58c7Ro0fp168fDg4OWFpaUq9ePXbt2pVnX4o6xk5OTpw4cYKZM2dSp04dtFotZmZmuLu706NHD3788UeDa01ftMaNGxtMUzU3N893NNrKyop9+/bx9ddf07x5cxwcHDAzM8PJyYnq1aszYsQIdu/e/Zc/J4rzv2HOnDns3r2bDh064OzsjJmZGba2tjRs2JD//ve/fPfdd0V+lhakUaNG6g3Y4Ml1y61atfpTZf1TaRTlBV5VK4QQ/1JXrlyhQYMGJCYmYmtry+HDh/O9e+Q/kaIoeaYZKYpCnTp1OH36NAAhISEGU5OEEOKf7MiRI+qPQo0bN873eb9hYWFqUOvj42MwTVr8u0yfPp2pU6cCT55/PX369BJu0YslI5xCCPECVKxYkW+++QZzc3Pu379PmzZtXpovH7169WL27NkcO3aM6OhoTpw4Qd++fdVg08TEpNCREyGE+KdITU0lJibG4Hr8kSNHlmCLxN9VVlYWqampXLp0Sb0DuJmZWYEj+y8zuWmQEEK8IM2aNWPr1q388ssvABw9evSF3kHTWGJiYgxu9Z6biYkJ8+fPp3r16i+4VUII8fw9fbO22rVr07NnzxJqjfg7O3LkiMHNoODJDady3/Dt30ICTiGEeIHatWtHu3btSroZz1Xfvn2xtLTkt99+4969e5iamuLp6Unjxo158803qV27dkk3UQghnitHR0dat27NnDlz/tQ1reLfw8TEBE9PT/r3769Oq/23kWs4hRBCCCGEEEIYhVzDKYQQQgghhBDCKCTgFEIIIYQQQghhFBJwCiGEEEIIIYQwCgk4hRBGkZmZSXR0NJmZmSXdFCGEEEIIUUIk4BRCGMXt27fx8vLi9u3bJd0UIYQQQghRQiTgFEIIIYQQQghhFBJwCiGEEEIIIYQwCgk4hRBCCCGEEEIYhQScQgghhBBCCCGMQgJOIYQQQgghhBBGIQGnEEIIIYQQQgijkIBTCCGEEEIIIYRRSMAphBBCCCGEEMIoJOAUQgghhBBCCGEUEnAKIYQQQgghhDAKCTiFEEIIIYQQQhiFBJxCCCGEEEIIIYzCrKQbIIR4uV1a8Ar3bOW3LSGEEEKI5636+0kl3YQiybdAIYQQQgghhBBGIQGnEEIIIYQQQgijkIBTCCGEEEIIIYRRSMAphBBCCCGEEMIoJOAUQgghhBBCCGEUL0XAGRAQwNKlS41W/oABA5gwYcI/ruyXTfPmzfnpp58ACAsLo379+iXcohfr4MGDuLq6vvCyY2JiqFy5MmlpaUapWwghhBBCvLyea8AZEBCARqNRg4Ico0ePRqPRGDUoFH8PZ8+eRaPRMHr06Dxpvr6+WFtbo9VqcXZ2pnv37ty+fRuA0NBQevXqVWC5+/btIysri0aNGhmr6aIAHh4eNGnShC+//LKkmyKEEEIIIf5hnvsIZ4UKFVixYoX6PiMjgw0bNuDn5/e8qzKKzMzMkm7C38Kf7Yfly5djb2/P2rVrycjIyJO+detWUlNTuXTpEvHx8YwZM6ZY5S5evJi+ffv+qTYVJTs7G0VRjFL2y6Jfv37yg5EQQgghhHhmzz3gDAoKYsuWLTx69AiAXbt2Ua1aNTw8PAzy7d27lzp16qDX66lVqxZHjhxR0wICApg8eTKvvfYapUuXJiAggLt37/Kf//wHe3t7ypYty8GDBw3Ku3nzJg0bNkSn0/H6668TFRWlpl27do22bdvi6OhIuXLlWLx4sZoWGhpKly5dGDRoEHq9nk8//TTf/UpMTCQwMBCdToe/vz8XLlxQ0z799FP8/PzQ6XRUrlyZb775Rk3Lmfo5efJkHBwc8PDwYM2aNfnW8fjxYzp37kzv3r3zDdYSExPp0qULtra2VKtWjVmzZuHr66umazQaLl++rL6fMGECAwYMUN//8ssvNG3aFDs7OypXrsyWLVvUtAEDBjBixAg6d+6MVqtlzZo1WFlZcevWLTXPrVu3sLKyUkcln5aens6aNWuYN28e6enp7Ny5M998AI6OjnTr1o3z588XmCdHRkYG+/bto3nz5nnSQkJC0Ov1lCtXjo0bN6rrn55mHR4ebtBXvr6+zJ49m9q1a1OqVClu375d6Hly8uRJGjZsiF6vx9XVlVGjRqlTTCMiItBoNDx+/FjN36tXL0JDQwvdr6eD3OIGvZ9++inOzs54enry2WefFauN8OT8+OKLL6hUqRK2trYEBweTnp6ebx2rV6+mTJky/PbbbwDUr1+fP/74gxs3bhSrjUIIIYQQQoARAk5nZ2caNmzIt99+CzwJuHIHPQDnzp2jT58+LFiwgISEBD788EM6depEfHy8mmfNmjX873//Iy4ujqysLOrXr0+VKlWIi4tj1KhRDBs2zKDMr7/+mkWLFhEXF4efnx/BwcEAPHz4kNdff52OHTty69Ytdu/ezaxZs9i/f7+67Y4dO2jVqhUJCQn5TgUFWLVqFSEhISQkJNCxY0c6deqkjgKWKVOGQ4cOcf/+faZPn07fvn2Jjo5Wtz116hSurq7cuXOHhQsXMnz4cJKTkw3Kv3//Pm3atMHNzY01a9Zgbm6epw1vvfUWWVlZREdHs3XrVr7++uvCD0Yut27dok2bNoSEhBAfH09YWBhDhgxRAwp4EmS88847pKSk0KNHD1q1amUQHK9evZoWLVoUeK3fjh07yMjIoEePHnTr1o3ly5cX2J67d++yadMmatWqVWTbr127RlZWFmXLljVYf/LkSRwdHYmLi2PJkiX079+fa9euFVlejpUrV7Jp0yZSUlIoXbp0oeeJqakpc+bMIT4+nuPHj3Po0CEWLlxY7Lqe9uuvv+Lv728QpI4cOZL//ve/hW4XHx/PjRs3iIyMZNu2bUybNo3vv/++2G3cvHkzP/30E9euXePnn39m9erVeeqYP38+M2bM4ODBg1SuXBkAMzMz/Pz8OHv2bIFtS05OJjo6Wl1y/1ghhBBCCCH+nYxy06ABAwawYsUK4uLiOHr0KJ07dzZI/9///seQIUNo3LgxJiYmtG/fnho1arB7926DMipVqkSpUqXo3LkzFhYWDBo0CFNTU/r06cO1a9dITU1V8wcHB1OrVi2srKz45JNPOHLkCNHR0ezcuRM3NzdGjhyJubk5FStWZOjQoaxbt07dtnbt2vTq1QsTExOsra3z3ae2bdvSokULzM3NmThxIvfv3+fYsWMAdOvWDQ8PD0xMTOjWrRsVK1bk+PHj6rYeHh68/fbbmJmZ0aVLF0xMTLh69aqafvv2bQICAmjSpAmLFy/GxCTvYcnKymLTpk3MmDEDnU5HuXLlePPNN4t9TFatWkWLFi3o1KkTpqam1KtXj86dO7Np0yY1T/v27WnWrBkajQZra2sGDhzIqlWrDMro379/gXUsX76cLl26YG1tTXBwMHv27OHOnTsGebp164adnR21a9fGx8eH+fPnF9n2xMREdDpdnvX29vZMmDABc3NzWrVqRZs2bdiwYUNxugN4EsCXLVsWc3NzwsPDCz1PatasScOGDTEzM8PHx4dhw4Zx6NChYteV2+nTp2ndujUTJkzAyspKXT9hwgTmzp3LggULCtw2OzubmTNnYmVlRe3atenfvz9r164tdhsnTZqEg4MDzs7OtGvXjtOnTxukT5w4kdWrV3PkyBF8fHwM0nQ6HYmJiQW2bd68eXh5eamLv79/cbtECCGEEEK8pIwScLZv354zZ84we/ZsOnfubPClGp5MQVy4cCF6vV5djh07RmxsrJon9yhaqVKl8rwHDAJOb29v9bVer8fGxoaYmBgiIiI4c+aMQV2zZ882mBaa+4v1iBEj0Gq1aLVaRowYkW/5pqameHh4EBMTAzwZKatRo4Za/vnz5w1Ga58eESxVqpRB2/fs2cPdu3cLHF0FiIuLIyMjw6AdTwcEhYmIiGDbtm0G/bBhwwaDUainy2vfvj2xsbGcPXuWM2fOEB0dTceOHfMt/9atW4SHh6vXWb722mu4u7sbBKzwZIQtMTGRqKgoVqxYgYODQ5Ftt7OzIyUlJc/6nCA/d/tzjklx5N7fos6Tq1ev0r59e1xdXbGxsWHixIkGx/hZDBw4kNu3b9OtWzc0Go26lClThhs3bjBmzBj++OOPfLe1tbXF1tY2330uThuf/jvKfR7ev3+f//73v4wdOxYnJ6c8daekpGBnZ1fgfoWEhBAVFaUuJ06cKF6HCCGEEEKIl5ZRAk4LCwt69uzJ3Llz80ynhSfB27hx40hKSlKXBw8e/KXHg0RGRqqvk5KSSE5OxsPDA29vbxo2bGhQV0pKisFoqkajUV8vXbqU1NRUUlNTDa4BzF1+VlYWMTExeHh4cPPmTYYMGcLnn3/OvXv3SEpKomrVqs90E5qBAwfSq1cvWrRoQUJCQr55nJycMDc3N2hH7tcApUuX5uHDh+r73EG1t7c3vXr1MuiH1NRUlixZkm8/AJibmxMcHMzKlStZuXIlPXv2xNLSMt/2rVq1iqysLHr37o2rqytubm7cuXOn0Gm1xeXn54epqWmeICwmJobs7Gz1fWRkpHqtsFarLbAvcuTe36LOk5EjR+Ln58fVq1dJTk5m5syZ6jHWarUARdaXY/ny5bi6urJ582YURVGXGzduUKZMGebPn59n+nCO+/fvG0zHzr3PhbWxOGxtbQkPD+ftt99mz549BmmZmZlcv36dGjVqFLi9jY0Nnp6e6uLm5lbsuoUQQgghxMvJaM/hnDJlCgcOHKBBgwZ50oYNG8YXX3zBjz/+SHZ2No8ePeKHH34wuO7xWa1Zs4azZ8/y+PFjJk6cSKNGjfD09KR9+/ZERESwbNky0tLSyMzM5Pz58/zyyy/PVH54eDjff/89GRkZfPLJJ+h0OurVq8eDBw8A1BGhlStXGtxQqLjmzp1Ls2bNCgw6TU1N6dq1K1OnTiUlJYU//vgjz/V+NWvWVAO/o0ePsm3bNjUtZ4rrjh07yMzMJD09nePHjxtcw5mfgQMHsnbtWtatW5fvjwc5li9fztixYzl37hxnz57l7NmzHD16lMuXL//lkS4LCwtatmyZ50ZRCQkJzJ49m4yMDA4cOMCePXvo3r078KQvNm/eTGpqKlFRUUVeb1nUeZKSkoKNjQ06nY6rV68a/Bjh6OiIp6cnK1asICsri23btvHzzz8XWFetWrXYu3cvs2bNMriGc9asWfznP/8pdKTbxMSESZMmkZaWxpkzZ1ixYgW9e/cuso3FlXP9db9+/QgPD1fXHz9+HF9fX8qUKfPMZQohhBBCiH8vowWczs7O+d5VFJ584V6xYgVjx47FwcEBHx8f5s6dazBa9awGDBjAqFGjcHJy4sqVK+rNbrRaLfv372f79u14eXnh5OTEsGHD8ty0pyjBwcHMmTMHOzs7tm7dytatWzE3N+eVV15h7NixNGrUCBcXF86ePUvDhg3/1D7MmzeP1157rcCgc9GiRSiKgqenJ506dWLgwIEG6Z9//jnfffcder2euXPnqoEIgKenJ7t372bBggW4uLjg7u7OxIkTDe5imp+cOwzr9Xrq1auXb55jx45x/fp13n33XVxdXdWlVq1aBAYGPpdRzlGjRrFy5UqDdXXq1OHu3bvqMf3666+pWLEiAGPGjMHW1hY3Nzc6d+5MUFBQoeUXdZ7MmTOHjRs3otPpGDx4sBrY5li2bBmLFi3C3t6eXbt20aFDh0Lrq1atGidOnDCYbr5kyZIir8t1dHTEx8cHLy8v2rdvz5QpU3j99deL1cbiatSoEVu3bqVv375q0Llq1SpGjhz5p8oTQgghhBD/XhpFHkD4jxUeHs6IESOIiIgwaj1dunShbt26TJw40aj1FKVZs2bMmDGDRo0alWg7/m1iYmJo0aIFZ8+eLXBKdX6io6Px8vJi7390uNga7bctIYQQQoh/rervJ5V0E4pkVtINEH9vv/32G/v37ze41rOk/PDDDyXdhH8lDw+PIqdeCyGEEEIIkR8ZdhAFGjx4MPXq1ePjjz/GxcWlpJsjhBBCCCGE+IeRKbVCCKOQKbVCCCGEEMb1T5hSK98ChRBCCCGEEEIYhQScQgghhBBCCCGMQm4aJIQwqldGX8LT07OkmyGEEEIIIUqAjHAKIYQQQgghhDAKCTiFEEIIIYQQQhiFBJxCCCGEEEIIIYxCAk4hhBBCCCGEEEYhAacQQgghhBBCCKOQu9QKIYzq0oJXuGcrv20JIYQQ4sWr/n5SSTfhX0++BQohhBBCCCGEMAoJOIUQQgghhBBCGIUEnEIIIYQQQgghjEICTiGEEEIIIYQQRiEBpxBCCCGEEEIIo5CAU5QYrVbL1atXjVZ+lSpVOHDggNHKF0IIIYQQQhROAs4SEhAQgJWVFVqtFgcHB9q0acOVK1cACAsLo379+mrex48f06FDB5o2bUpycnKedGOKiIhAo9Hw+PHjQvNFR0fTsmVLbGxscHd3Z+bMmUWWnZqaSoUKFZ5XU/O4ePEiLVq0MErZGRkZfPTRR1SqVInSpUvj5eVFjx49OH/+vEG+BQsWoNFo+Pbbb59b3b6+voSHhxusO3jwIK6urur73377jbZt2+Lg4ICtrS1Vq1YlLCxMTddoNNja2vLo0SODcmrUqIFGo+Hy5cvPrb1CCCGEEOLfSwLOErRgwQJSU1O5efMmdnZ2DBgwIE+e1NRU2rVrR3p6OuHh4djY2DxTHZmZmc+ptYWbNGkSdnZ2xMXFceHCBZo0afJC6i0pPXv2ZOPGjSxfvpykpCQuX75M+/bt2blzp0G+5cuXY29vbxDsvQjt27encePGxMTEEB8fT1hYGC4uLgZ5XF1d2bp1q/r+7NmzRf6wIIQQQgghxLOQgPNvQKvVEhwcnGd0LDExUR013L59O6VKlSpWeRqNhsWLF1OpUiX0ej0Ae/bsoXr16tja2lKvXj1+/vlnNX9ycjKDBw/G1dUVT09PQkJCSEtLA6Bhw4YAODo6otVq84ys5TAzM8PT0xNLS0vs7e1p3LhxsdqZM5IWHh5O1apV0el0uLq6MnbsWODJ6O6AAQPUkbrq1atz6dIl4Mko8dKlS9XywsPD8fX1Vd/nHgkMDQ2la9euDB06FFtbW/z8/Aym20ZGRtK8eXN0Oh2NGzdm0qRJBAQE5Nvu77//nt27d7N9+3YaNGiAubk5pUuXpl+/fkycOFHNd/r0aS5cuMCXX37Jrl27uHv3rpp279493njjDezs7LCzs6NevXrEx8cX2WfFER8fzx9//MGwYcOwsrLC3Nyc2rVr07ZtW4N8/fv3NwiEw8LC6N+/v0GekydPUq9ePWxsbHByciIoKOi5tFEIIYQQQvw7SMD5N5CcnMyqVauoVauWui4xMZGAgADKlSvHpk2bsLS0fKYyN27cyOHDh4mLi+PatWt07dqVjz/+mHv37vHmm2/Stm1bNcB55513iImJ4fLly5w8eZKffvqJDz/8EICjR48CT4KY1NRU2rRpk299/v7+LFy4kE2bNv2ZLmDgwIGMGzeOlJQUrl+/Trdu3QBYsWIF58+f5/fffycpKYn169djb2//p+rYuXMngYGBJCQk8OabbzJo0CA1rXfv3rzyyivExcWxaNEili9fXmA5+/btw9/fHx8fn0LrW758Oc2bN6dLly74+vqyevVqNW3OnDlkZ2erI5CLFy/GysrqT+3X0xwcHKhQoQJ9+/Zl8+bNxMTE5Juvffv2nDt3jpiYGDIyMtiwYQP9+vUzyPP222/TsWNHkpKSiIqKYtSoUQXWm5ycTHR0tLrcunXrueyPEEIIIYT455KAswSFhIRgZ2dHpUqVSEtLY8WKFWpabGwsFy5cYODAgZiZmT1z2RMmTMDZ2Rlra2s2bNhA69atadeuHWZmZvTr149KlSqxbds2srKyWLduHZ988gl6vR5XV1emTZvGypUri13XTz/9xMcff8zBgwcZO3Ysa9asUdN0Op06IlkYCwsLrl+/Tnx8PFqtlnr16qnrU1JSuHz5MoqiULlyZYNrFZ9FgwYN6Ny5M6ampvTv35+oqCji4+OJjIzk559/5uOPP8bKyooaNWoUOpJ379493N3dC60rLS2NtWvXEhwcDEBQUJBBEGthYcG9e/e4fv06pqam1K5dG61WW+x96datG3q9Xl3at2+vpmk0Gn744QcqVqzI5MmT8fLyonbt2vzyyy8GZVhYWNCzZ09WrVrFrl27qFatGh4eHnny3Lx5k9jYWKysrGjUqFGBbZo3bx5eXl7q4u/vX+z9EUIIIYQQLycJOEvQvHnzSExMJDY2lq1bt1KmTBk1rUqVKixevJhOnTpx6NChZy479+hbTEyMwVRTeDLdNGd0LT09Pc9U1Fu3bqEoSrHqWrRoEUOHDqVRo0aEh4fz3nvv8fXXX3PmzBns7e2pXLlykWVs3bqV8+fPU758eerWrateC9m3b1/69+/PsGHDcHZ2ZtiwYSQnJxerXU/LHajmTE9OTU0lNjYWW1tbg+tjvby8CizHwcGB2NjYQuvavn07jx8/pmvXrgAEBwdz4cIFTp48CcDYsWNp3LgxXbt2xd3dnXHjxpGRkVHsfdm8eTNJSUnq8vS1o+7u7nz22WdcuXKF27dvU6VKFTp16pTnmA4YMIAVK1YQFhaW7zXEX3/9NQ8fPqRWrVpUqVKFr7/+usA2hYSEEBUVpS4nTpwo9v4IIYQQQoiXkwScf2PDhw9n9uzZtG/fniNHjjzTthqNRn3t4eFBRESEQXpERAQeHh44OjpiYWFhkB4REYGbmxsajcagnIJkZmaq13xWqlSJvXv3Mm7cOHr16sX7779frDJq1arFli1biI+PZ8yYMXTr1o0HDx5gZmbG5MmT+fXXXzl//jyXLl3i008/BZ5c+/rw4UO1jNu3bxdZT37c3d25f/8+KSkp6rqoqKgC87dq1YoTJ04QGRlZYJ7ly5eTnp6On58frq6uNG7cGI1Go45yarVaZs+ezbVr1zhy5AjffvvtM40qPwtnZ2fGjRtHbGwsCQkJBmm1atXC3Nycw4cP07lz5zzblitXjtWrV3Pnzh3++9//MmLECK5fv55vPTY2Nnh6eqqLm5ubUfZHCCGEEEL8c0jA+Tc3cuRIPvnkE9q1a8dPP/30p8ro0aMHe/fuZc+ePWRmZrJ69WouX77MG2+8gampKb169WLixIkkJSVx584dpk2bRt++fQFwcnLCxMSE33//vcDye/fuzaJFi9i/fz/Z2dnY2tpSq1Ytrl69Wqxpounp6axatYrExERMTU3R6/VoNBpMTU354YcfOHfuHFlZWWi1WiwtLTE1NQWgZs2abN68mdTUVKKioli4cOGf6h9vb2/q16/P5MmTSUtL49dff2Xt2rUF5m/evDlt27alU6dOHD9+nIyMDB4+fMiaNWuYNWsWsbGx7Nu3j02bNnH27Fl1WbRoEevWrSMtLY2dO3dy9epVsrOzsbGxwdzcXN2v0NDQAm9YVByJiYm8//77avn3799n0aJFlC9fHgcHhzz5N23axIEDB/K9hnTlypXcvXsXjUZjcFyEEEIIIYQoDgk4/wFGjRrFxx9/TGBgoHoTn2dRoUIFNm7cyPjx43FwcOCzzz5j165dODo6AvD555/j4uJCxYoVqVmzJv7+/kydOhV4MvV0ypQpNGvWDL1ez969e/OU36VLFz7//HPGjBmDTqcjICCAxo0bs2vXLoYMGcLBgweLbOPatWspV64cOp2O8ePHs3HjRqysrLh9+zY9e/bE1taWcuXK4eHhod7BdsyYMdja2uLm5kbnzp3/0h1U165dy6+//oqjoyMjR44kKCio0Bs1bdy4kS5dutCvXz/0ej0VKlRgx44ddOjQgZUrV1KxYkU6deqEq6urugwaNAhTU1O+/fZbrl+/Tps2bdDpdFSrVo1WrVqpQX5UVFSh10oWxcLCgqioKFq3bo2NjQ3ly5fn1q1b7NixI9/8FStWNLhhVW779++natWqaLVaunfvzuLFiw2mfgshhBBCCFEYjVLcC/WE+Bd59913efDgAV999dULr7tGjRp89913+Y5G/pNER0fj5eXF3v/ocLGV37aEEEII8eJVfz+ppJvwryffAoUATp06xbVr11AUhR9//JEVK1bQpUuXEmnL2bNn//HBphBCCCGEEADP/rwNIV5Cd+7coUuXLsTFxeHq6srUqVMJDAws6WYJIYQQQgjxjyYBpxBAYGAgN2/eLOlmCCGEEEII8VKRKbVCCCGEEEIIIYxCRjiFEEb1yuhLeHp6lnQzhBBCCCFECZARTiGEEEIIIYQQRiEBpxBCCCGEEEIIo5CAUwghhBBCCCGEUUjAKYQQQgghhBDCKCTgFEIIIYQQQghhFHKXWiGEUV1a8Ar3bOW3LSGEEEK8eNXfTyrpJvzrybdAIYQQQgghhBBGIQGnEEIIIYQQQgijkIBTCCGEEEIIIYRRSMAphBBCCCGEEMIoJOAUQgghhBBCCGEUEnCWIF9fX8LDw/NNi4yMRKvVkpaW9oJb9de5urpy8ODBkm7G35JGo+Hy5cv/mHKFEEIIIYT4KyTg/BM6deqEhYUFcXFxRqvD29ub1NRULC0tjVZHfg4dOkS1atXQ6XRUrFiRPXv2GLW+v2ugFBERgUaj4fHjxyXdFCGEEEIIIf6xJOB8Rnfv3mXXrl3odDrWrFlTaN7MzMwX1Krnp2/fvgwaNIjk5GT27NmDl5dXSTfpb+ufeHyFEEIIIYR4kSTgfEarV6/Gz8+PsWPHsnz5coO00NBQunTpwqBBg9Dr9Xz66aekpaUxceJEypQpg06no06dOkRFRanbXLx4kdq1a2NjY0NgYCCJiYmA4Qjbxo0bqVKlikFdGzdu5JVXXgFAURTmzZtHhQoVsLe3JzAwkOjoaDVt7NixuLi4YGNjQ6VKlQqd7mpmZkbZsmXRaDSULVuWV199tcg+mT9/Ph4eHjg7OzN37lyDtJMnT9KwYUP0ej2urq6MGjVKnSbcsGFDAGrXro1Wq+WLL74AoHfv3ri5uWFra0uTJk04f/58ofWfOHGCpk2bYmdnh6urKzNnzlTTVq9ezauvvoper6dJkyZcvHhRTfP19WXu3Ln59n9O2xwdHdFqtYSHh3Pw4EFcXV1ZsGABHh4etG/fvsg68rN//378/Pywt7dn6NChan8kJyfToUMHnJ2dsbOzIzAwkMjISHW7+/fvM2LECDw9PdW+efToUZ7yz507h4+PD5s3byYsLIz69esbpOee8pxzzgYHB6PT6ahSpQqHDh1S865atQo/Pz90Oh1eXl7Mnz+/0H0TQgghhBAiNwk4n9Hy5csJDg4mKCiI8+fPc/r0aYP0HTt20KpVKxISEhg9ejTjx4/n8OHDHDx4kPv377Ns2TJKlSql5l+9ejVbt24lNjaWpKSkfL/Qd+zYkejoaM6ePauuW7NmDcHBwQAsWrSINWvWsG/fPu7cuUOtWrXo1asXAPv27WP9+vWcPXtWHbX09vbOd98URaF+/foMGzaM3377rVj9sX//fmbMmMGOHTuIjIzk2rVrxMfHq+mmpqbMmTOH+Ph4jh8/zqFDh1i4cCEAR48eBeDUqVOkpqYybNgwAFq1asWVK1e4e/cu/v7+9O7du8D6o6OjadmyJYMGDeLOnTtcvXqV5s2bA0+OxZQpU1i3bh337t0jODiYDh06kJ6erm5fUP/ntC0+Pp7U1FTatGmjvr9x4wa///47W7duLVYdT1u/fj1Hjx7l8uXLnD17lo8++giA7Oxs+vfvT0REBFFRUdjY2DBq1Ch1u/79+3Pnzh3OnDlDQkICs2bNwsTE8E/48OHDtGvXjmXLltGtW7fCDp1q+/bttG7dmsTERMaNG8cbb7xBYmIiDx48YNCgQSxfvpyUlBTOnTtHQEBAgeUkJycTHR2tLrdu3SpW/UIIIYQQ4uUlAeczOHnyJBcvXiQoKAgvLy+aNm2aZ5Szdu3a9OrVCxMTEywtLfniiy/47LPP8PHxwcTEhOrVq+Pg4KDmf/fdd/H29kar1dKtW7c8ASyAlZUVXbt2VafwJiQksHfvXoKCggBYsmQJM2bMwNfXF3Nzc0JDQ/nll1+IjIzEwsKCx48fc/HiRTIyMihTpgxly5bNd/9mz55NXFwcS5cupWXLlvz666/qfjs7O+e7zbp16xgwYAC1atXCysqKWbNmkZ2drabXrFmThg0bYmZmho+PD8OGDTMYQcvPwIEDsbGxwdLSkqlTp3Lx4kXu3buXb97Vq1fTpEkTBgwYgIWFBTY2NtSrV0/tl/Hjx1O1alVMTU0ZPnw4Go2GY8eOPVP/56YoCjNnzsTKygpra+ti1fG0CRMm4OzsjLOzM1OmTGHt2rUA6PV6unXrRqlSpdBqtUycOFHtq9u3b7Nt2za++OILnJycMDU1pVGjRgbX+G7bto0+ffqwZcsWWrRoUeh+5Fa9enX69u2LmZkZ/fv3p0yZMuzatQsAc3NzLl26RHJyMvb29tSsWbPAcubNm4eXl5e6+Pv7F7sNQgghhBDi5SQB5zNYvnw5jRs3xtfXF3hyvePatWsN7iTr4+Ojvo6Pj+fRo0f4+fkVWKarq6v6ulSpUqSmpuabLzg4mHXr1pGdnc2mTZvw9/dX64qIiKBnz57o9Xr0ej2Ojo6YmJgQHR1Ns2bNmDZtGpMmTcLJyYlevXoRGxubbx0LFixg6tSpdOrUifnz59OyZUtOnz7NkSNH1BG+p8XGxhqMmOr1emxsbNT3V69epX379ri6umJjY8PEiRMNRkCflpWVxYQJEyhXrhw2NjbqPsbHx6t37s1ZIiMjiYyMLLB/IyIiGDt2rNover2eW7duERMTo+Ypbv/ncHR0NBihLk4dT8vdXz4+Pmrehw8fMnz4cHx8fLCxsaFJkyakpqaSlpZGZGQktra2ODk5FVjuvHnzaN++/TMHek+PeOe0qXTp0mzfvp2tW7fi5eXFa6+9xs8//1xgOSEhIURFRanLiRMnnqkdQgghhBDi5SMBZzGlpaWxbt06Tp06haurK66urowfP56EhAS2b9+u5tNoNOprR0dHrK2tuX79+l+uPyAgAI1Gw8GDB1m9erU6nRaeBAzbt28nKSlJXR49eqRehzhq1ChOnDjBjRs3yMzMZPz48fnWkZmZqQbP3bt359NPP6VFixbMnz+fCRMm5LuNu7u7wXWGSUlJJCcnq+9HjhyJn58fV69eJTk5mZkzZ6IoSoH7uXbtWrZs2cL+/fu5f/8+N2/eBJ6MLObcuTdn8fb2xtvbu8D+9fb2ZtGiRQb98vDhw0Kn6ObIfRwLW/9n6sjdX5GRkXh4eAAwd+5cLl26xLFjx0hOTubIkSMG+37//v1Cg/WNGzdy+PBhpk6dqq7TarU8fPhQfZ+RkUFCQkKB7Xm6TS1atCA8PJy4uDjatWtHjx49CqzfxsYGT09PdXFzcyswrxBCCCGE+HeQgLOYvv32W9LT0zl//jxnz57l7NmzXLhwgaCgoDzTanOYmJgwZMgQxowZQ2RkJIqicO7cuQKnhxbGxMSE3r17M3PmTE6ePEn37t3VtJEjRzJ58mR+//13ABITE9m4cSMAv/zyCz///DPp6emUKlWKUqVKYWpqmm8dvXv3ZuzYsVy+fJns7GwqVaqETqcjJSWlwG169uzJihUrOHv2LI8fP2bSpEkG1xWmpKRgY2ODTqfj6tWrLF261GB7FxcXtd05+S0tLXFwcODRo0dMmTKl0H4JCgri8OHDrFq1ioyMDJKTkzl+/LjaL7NmzeLcuXMoikJqaio7duwgJSWl0DIBnJycMDExMWhbfv5MHTlTl+Pi4vjoo4/U4DQlJQVra2v0ej2JiYlMnz5d3cbV1ZUOHTowYsQI4uPjycrK4ujRowaj6y4uLvzwww9s3ryZDz74AHgyXfby5cucOHGC9PR0pk6dajDlGZ7cZGjt2rVkZmayatUqfv/9dwIDA7lz5w7ffvstqampmJubo9PpCjwPhBBCCCGEyI8EnMW0fPly+vXrR9myZdURTldXV8aMGcO+ffsKnKY6e/Zs6tWrR+PGjbG1tWXIkCH53lm0OIKDgzlw4ABt27bFzs5OXf/222/Tq1cv2rdvj42NDdWrV2fv3r3Akxu5jBgxAgcHB9zd3bl//z6zZs3Kt/xPP/2UwMBAWrdujU6nY/jw4XzyySeMHj2aNm3a5HsTmNatWzNx4kTatWuHt7c35cqVw9HRUU2fM2cOGzduRKfTMXjwYINAGWDatGkMGTIEvV7Pl19+qfaxh4cHlStXpm7duoX2iZeXF+Hh4SxduhQnJycqVqyo3oH1jTfe4IMPPqB///7o9XrKly/PqlWritXXpUqVYsqUKTRr1gy9Xq/259P+TB09evSgQYMGVKhQgVdffZXJkycDMHr0aNLT03FycqJevXq0bNnSYLsVK1Zga2tL1apVcXBwYOLEiXmCRxcXF77//ns2btxIaGgo5cuXZ8aMGbRr1w4fHx+8vLwMjg88uSnV7t27sbOzY+bMmWzduhV7e3uys7NZsGABnp6e6PV6wsLC1OtNhRBCCCGEKA6NUtj8RiHESy00NJTLly+zfv365152dHQ0Xl5e7P2PDhdb+W1LCCGEEC9e9feTSroJ/3ryLVAIIYQQQgghhFFIwCmEEEIIIYQQwijMSroBQoiSExoaWtJNEEIIIYQQLzEZ4RRCCCGEEEIIYRQScAohhBBCCCGEMAqZUiuEMKpXRl/C09OzpJshhBBCCCFKgIxwCiGEEEIIIYQwCgk4hRBCCCGEEEIYhQScQgghhBBCCCGMQgJOIYQQQgghhBBGITcNEkIYVdcxl7Asfa+kmyGEEEKIv6HDYdVLugnCyGSEUwghhBBCCCGEUUjAKYQQQgghhBDCKCTgFEIIIYQQQghhFBJwCiGEEEIIIYQwCgk4hRBCCCGEEEIYhQSc/1KhoaH06tWrwPSAgACWLl0KwJo1a2jevPmLalqhRowYwQcffPCvq/t5qVKlCgcOHCjpZgghhBBCiH+JZw44AwICsLKyQqvVqkvt2rWN0baXWu6ArjCXLl2iQoUKL6BFBQsKCuL7778v0TbkWLp0KdOmTSsyX3H71xh1Q9EB/YuQXx9cvHiRFi1alFCLhBBCCCHEv82fGuFcsGABqamp6nLq1Knn3a7nLjMzs6Sb8Kfs3LmT9u3bl3QzngtFUcjKyirpZgCQnZ2Noigl3Yw/7Z96PgshhBBCiH8Xo0+pTUhIoHPnztja2lKtWjU++eQTfH191fSrV6/SsmVL7OzsqFChAsuWLQMgLS0NOzs7fvnlFzVvWloa9vb2nDhxAoBffvmFpk2bYmdnR+XKldmyZYuad8CAAYwYMYLOnTuj1WrZtGkTAQEBvP/++zRr1gydTkejRo24efOmuo1Go2Hp0qVUqlSJ0qVLM2LECOLi4mjXrh06nY6GDRsSHR2t5r927Rpt27bF0dGRcuXKsXjxYjUtNDSUrl27MnToUGxtbfHz81OnMo4fP54jR44wevRotFotffv2LbD/Cgs4NRoNCxcuxM/PD3t7e4YOHUpaWhoAYWFh1K9f3yC/q6srBw8eVN+np6cTHByMTqejSpUqHDp0KN96ni7rypUrBAYG4ujoiKOjI2+99VaB7Q8ICGDSpEkEBARQunRpTpw4we3bt+nZsycuLi54eXkRGhpKdnY28CQQnDBhAk5OTnh7e/P111+j0WiIiIgAnhzXCRMmAHDv3j3eeOMN7OzssLOzo169esTHxxfYv76+vsyePZvatWtTqlQp1q5dyyuvvGLQ3g0bNlC1atV89yV33QcPHsTV1ZWFCxfi5uaGs7Mzn376KfDkmH388cd88803aLVa9XxPT09n0qRJlClTBkdHR/r06UNiYqJa/rp16yhbtix2dna899571K9fn7CwMINjMH78eJydnXnzzTdJTk6mQ4cOODs7Y2dnR2BgIJGRkUDB55ivry/h4eFqe8aOHYunpycuLi4MGDCA+/fvq+3RaDR88cUXVKpUCVtbW4KDg0lPTy/wWAshhBBCCPE0owecb7/9NgAxMTFs27aNFStWqGkZGRm0b9+eBg0acOfOHdauXcv48eM5cOAAlpaWdO/enTVr1qj5d+/ejZOTE/7+/ty6dYs2bdoQEhJCfHw8YWFhDBkyhN9++03Nv3r1at555x1SUlLo1KkTACtXrmThwoXcu3cPNzc33n//fYP2btu2jZ9//plLly6xceNGWrduTWhoKAkJCTg7O/Phhx8C8PDhQ15//XU6duzIrVu32L17N7NmzWL//v1qWTt37iQwMJCEhATefPNNBg0aBMAnn3xCkyZN1JHiVatW5dt3iYmJXLx4kSZNmhTYv+vXr+fo0aNcvnyZs2fP8tFHHxXnsACwfft2WrduTWJiIuPGjeONN94wCIDyk5qaSosWLWjUqBFRUVFERUUVOXU0LCxM3ddatWrRsWNH/Pz8uHnzJsePH2fbtm3qDw3Lli3jm2++4ZdffuHSpUvs2bOnwHLnzJlDdnY2MTExxMfHs3jxYqysrArt35UrV7Jp0yZSUlLo3r07d+/e5eTJkwbp/fv3L073ER8fT1RUFDdv3mTnzp1MnjyZ69ev0759eyZNmkTXrl1JTU1Vg+WJEydy+vRpjh8/TlRUFBYWFmqwfvnyZQYPHsxXX33F3bt3cXJyyjNz4OTJkzg6OhITE8OCBQvIzs6mf//+REREEBUVhY2NDaNGjQKKd459/PHHfPfdd5w4cYKrV6+SkJCgbp9j8+bN/PTTT1y7do2ff/6Z1atXF9gfycnJREdHq8utW7eK1Y9CCCGEEOLl9acCzpCQEPR6vboU9AU9KyuLTZs2MX36dLRaLWXKlDH4Qnv8+HESExP54IMPsLCwoE6dOgwZMkQNSoODg1m/fr06DXPNmjUEBQUBsGrVKlq0aEGnTp0wNTWlXr16dO7cmU2bNqnlt2/fnmbNmqHRaLC2tgZg4MCBvPrqq1hYWNCnTx9Onz5t0Obx48djZ2eHj48PjRs3xt/fn7p162Jubk6PHj3U/Dt37sTNzY2RI0dibm5OxYoVGTp0KOvWrVPLatCgAZ07d8bU1JT+/fsTFRVFfHx8sft5z549vP7665ibmxeYZ8KECTg7O+Ps7MyUKVNYu3ZtscuvXr06ffv2xczMjP79+1OmTBl27dpV6DY7d+7E1taWyZMnY21tjbW1NY0bNy50m379+lGjRg1MTEw4d+4cUVFRzJgxAysrK9zd3QkJCVH7bd26dbz77rv4+vqi1WoLvUmPhYUF9+7d4/r165iamlK7dm20Wm2hbXnrrbcoW7Ys5ubmWFhYEBQUxMqVKwG4c+cO3333nXqOFcXExIQZM2ZgYWGBv78/lSpV4uzZs/nmVRSFpUuXMn/+fJydnbG2tmb69Ols2rSJrKwsNm7cSGBgIM2bN8fc3JyxY8diZ2dnUIazszPvvfce5ubmWFtbo9fr6datG6VKlUKr1TJx4sQCR6nzs3r1aj744APc3d2xtbVl9uzZbNiwwWAUc9KkSTg4OODs7Ey7du3y/L3kNm/ePLy8vNTF39+/2G0RQgghhBAvpz8VcM6bN4+kpCR1yQkQR4wYod5IKGc6akZGBl5eXuq2uV/HxMTg6emJqampus7X15eYmBgAmjRpgpWVFd999x33799n165dBAcHAxAREcG2bdsMAt8NGzYYjKr4+Pjkaburq6v6ulSpUqSmphaaXlD+iIgIzpw5Y1D/7NmzuX37doFlAXnqK8yuXbuKvH7T29tbfe3j46P2XXHk3ra420dGRuLn55dvWtu2bdXj//HHHxuUmyMiIoK4uDjs7OzUfnvzzTe5c+cOALGxsQWeL08bO3YsjRs3pmvXrri7uzNu3DgyMjIKbf/T58TAgQNZv349GRkZrF27lubNm+Pm5lZoGTns7e2xsLBQ3+d3PuWIi4vj4cOHNGjQQN3vqlWrYmJiwu3bt/Pst4mJCR4eHgZleHl5odFo1PcPHz5k+PDh+Pj4YGNjQ5MmTUhNTVWnVRclJibGYHq7r68vWVlZhZ7DhZ2/ISEh6qh3VFSUOvVdCCGEEEL8e5k9z8KWLl1qcFfMrKwszM3NiYqKwtbWFoCoqCg13cPDg+joaLKystSgMyIiQv2irdFo6NOnD6tXryYqKoqaNWtStmxZ4Emw1KtXL/Uat/zk/nL+vHl7e9OwYUODayKfRVFty8rKYt++fSxYsKDQfJGRkVSvXl19ndN3Wq2Whw8fqvkyMjJISEjIs+3T77t161Zofd7e3gVOAS5o+mvuffX29sbT01OdZvo0d3d3g3Mk9+unabVaZs+ezezZs/n9999p27YtFStWZPDgwQX279Pra9SogYeHB3v27GHlypXqNZp/1dP1ODo6Ym1tzdmzZw2CvBzu7u78+uuv6vucqcKFlTl37lwuXbrEsWPHcHNz49y5c9SoUUO9GVJR55iHhwcRERHq+RMREYGJiYlBkPksbGxssLGx+VPbCiGEEEKIl5NRr+E0NTWla9eufPDBB6SmpnLz5k2WLFmipterVw+9Xs/06dNJT0/n9OnTLFu2zOAmOsHBwWzdupWvvvpKHd3MWb9nzx527NhBZmYm6enpHD9+3OAaTmNq3749ERERLFu2jLS0NDIzMzl//rzBTY4K4+Liwu+//15g+tGjR/Hz88PJyanQcmbPnk1cXBxxcXF89NFH9O7dG3gyXfby5cucOHGC9PR0pk6dqt6YJ8e5c+dYu3YtmZmZrFq1it9//53AwMBC62vXrh0JCQnMmjWLR48e8ejRI3788cdi7TNA3bp1cXJyYvr06Tx48IDs7GyuXbumTgXt2bMnCxcu5ObNm6SmpjJ9+vQCy9q5cydXr14lOzsbGxsbzM3N1R8uiurf3AYOHMj7779PREQEb7zxRrH3pTAuLi5ERESofW5iYsLw4cMZM2aMOgp/9+5dtm3bBkD37t3ZvXs3Bw8eJDMzk7lz5xZ5PW1KSoo6tTYxMTFPXxXVB0FBQUyfPp1bt26RnJzMhAkT6Nmzp8GorRBCCCGEEH/Fnwo4c+58mbMUNiKyaNEiMjMz8fDwoEOHDvTu3RtLS0sAzM3N2bFjB0eOHMHZ2ZmePXvy0Ucf0apVK3X7V155hfLly3Pq1Cl69uyprvf09GT37t0sWLAAFxcX3N3dmThxYrGnE/5VWq2W/fv3s337dry8vHBycmLYsGEkJycXa/t3332X7du3Y2dnl+81sMV9HEqPHj1o0KABFSpU4NVXX2Xy5MkAlC9fnhkzZtCuXTt8fHzw8vLC0dHRYNuOHTuye/du7OzsmDlzJlu3bsXe3r7Q+nQ6Hfv37+f777/Hw8MDb29vNm7cWKx9hic/QuzYsYNr165Rvnx57Ozs6NGjhxqEDRkyhI4dO1KnTh0qV65M8+bNAdRzJrfr16/Tpk0bdDod1apVo1WrVuqPFUX1b25BQUFcvnyZnj17YmVlVex9KUz37t0xNzfHwcGBcuXKATBr1iyqV69OkyZN1Lse50w7rVy5Ml9++SUDBw7EycmJO3fuUK1atXz3O8fo0aNJT0/HycmJevXq0bJlS4P0ovpg0qRJvPbaa9SpUwc/Pz9sbGwM7rQshBBCCCHEX6VRXvDDCOfPn8+uXbvUR4SI/FWpUoW1a9eq0x3zo9Fo+O2336hUqdILbNmLdebMGerVq8fjx48xMTHOgHxmZibu7u5s27aNBg0aGKWOZ5WZmYmrqyvffvttkTdl+ruKjo5+cvOgbnuxLO1S0s0RQgghxN/Q4bCCv+uKl4PRH4ty5coVzpw5g6IonD9/ns8++4wuXboYu9p/tPT0dIKCggoNNl9W6enp6jTpuLg4Jk6cSKdOnYwWbAKsWLECd3f3Eg82d+zYQWpqKo8fP+aDDz7A2tpa7vQqhBBCCCH+0Z7rTYPy8+DBA3r16kV0dDQODg4EBwczbNgwY1f7j2ZhYcGkSZNKuhklQlEUZsyYQZ8+fbCwsKB58+YsWrTIaPX5+fnx6NEjNmzYYLQ6imv37t3069ePrKwsqlatyrfffivXUwohhBBCiH+0Fz6lVgjx7yBTaoUQQghRFJlS+/Iz+pRaIYQQQgghhBD/ThJwCiGEEEIIIYQwCqNfwymE+Hf7Zv4reHp6lnQzhBBCCCFECZARTiGEEEIIIYQQRiEBpxBCCCGEEEIIo5CAUwghhBBCCCGEUUjAKYQQQgghhBDCKOSmQUIIo+o65hKWpe+VdDOEEEII8Tckz+F8+ckIpxBCCCGEEEIIo5CAUwghhBBCCCGEUUjAKYQQQgghhBDCKCTgFEIIIYQQQghhFBJwCiGEEEIIIYQwin91wDlixAg++OCDkm7Gv8rjx4/RaDRERESUdFP+Vnx9fQkPDwfg448/ZsCAAUapR6vVcvXqVaOULYQQQgghxNNeqoAzICAAKysrtFotDg4OtGnThitXrhSYf+nSpUybNu0FtlC8aDnnhE6nw9bWlmrVqvH++++TkpJikO/OnTsMHz4cDw8PtFotfn5+jBkzhvj4+Bfe5kmTJhEWFvaXy8kdxOZITU2lQoUKf7lsIYQQQgghiuOlCjgBFixYQGpqKjdv3sTOzi7fkSJFUcjKynphbXrR9b2sMjMz/9R2CxYsICUlhcTERFasWMHx48dp1KgRDx8+BCAxMZGGDRuSmJjI4cOHSUlJ4ccff8TW1pbjx48brV1CCCGEEEK87F66gDOHVqslODiY8+fPA09GuiZNmkRAQAClS5fmxIkTDBgwgAkTJgBw8OBBXF1d+eyzz3Bzc8PBwYGvvvqKU6dOUaNGDWxtbenbt68aXCQnJ9OhQwecnZ2xs7MjMDCQyMhItf6n6zt27Bh2dnakpaWpeY4fP46DgwPp6ekF7oeiKIW+f9off/xB8+bNsbW1xcHBgaZNm6ppt2/fpmfPnri4uODl5UVoaCjZ2dlq+sqVK3n11VfR6XT4+fmpo2PJyckMHjwYV1dXPD09CQkJUfcjIiICjUbDqlWrKFOmDHZ2dowZM0YtMzs7m4kTJ+Lk5IS3tzdr1qwxaO+ePXuoVasWNjY2eHl58f7776tpOWWHhYVRpkwZqlWrxttvv82oUaMMyhg5ciRvvfVWof0CYGJiQs2aNdm0aRN3795l+fLlAMyfPx9ra2vWrVtHuXLl0Gg0uLq6EhoaSrt27fItS6PRsHjxYipVqoRerwfgl19+oWnTptjZ2VG5cmW2bNmi5h8wYADDhw8nMDAQnU6Hv78/Fy5cyLfs0NBQevXqpb4/ceKEWq6rqyszZ84E4MaNG7z++us4ODjg6OhI7969SUxMBKB3795ERkbSuXNntFotkydPVtt9+fJl4K8dVyGEEEIIIYrjpQ04k5OTWbVqFbVq1VLXhYWFqSOgudfniI+P586dO9y8eZOwsDDeeustpk2bxp49e7h+/TpHjhxh48aNwJNAqn///kRERBAVFYWNjU2eQCh3fXXq1MHLy4sdO3ao6StXrqR3795YWFjkuw/ff/89r7/+ukGQ2bFjR7799tsC93vy5MlUqFCB+Ph4bt++zUcffaS2t2PHjvj5+XHz5k2OHz/Otm3bWLZsGQDffvst48aN44svviA5OZkffvgBHx8fAN555x1iYmK4fPkyJ0+e5KeffuLDDz80qPfAgQNcuHCBU6dOsXz5cr7//nsAli1bxqZNmzh+/DgXL1402H+A0qVLExYWRlJSErt37+bLL79k8+bNBnnCw8M5d+4cp06dYuDAgWzYsEEN0tPT09mwYcMzXfNoa2tLy5YtOXz4MAD79u2jS5cumJqaFrsMgI0bN3L48GHi4uK4desWbdq0ISQkhPj4eMLCwhgyZAi//fabmn/VqlWEhISQkJBAx44d6dSpU5Gjo9HR0bRs2ZJBgwZx584drl69SvPmzYEnPz6MHz+e2NhYLl++zK1bt9SAfd26dXh7e7N161ZSU1PV8yC3v3Jc85OcnEx0dLS63Lp1q9h9KYQQQgghXk4vXcAZEhKCnZ0dlSpVIi0tjRUrVqhp/fr1o0aNGpiYmGBpaZlnWxMTE6ZNm4aFhQUdOnTAwsKCPn364ObmhpOTE61ateL06dMA6PV6unXrRqlSpdBqtUycOJFDhw4ZlPd0fQMHDmTlypXA/wVK/fv3z3c/9u7dS+/evfnwww/RaDTq+mnTpjFy5Mg8QVkOCwsLbt26xc2bNzE3N6dJkyYAnDx5kqioKGbMmIGVlRXu7u6EhISwbt064Mn1rO+99x4NGzZEo9Hg5eVF5cqVycrKYt26dXzyySfo9XpcXV2ZNm2auh+521W6dGnKli1L06ZN1X5at24d7777LmXLlkWn0+W5ZrZp06ZUq1YNExMTqlatSu/evfP0Y2hoKDY2NlhbW1OrVi08PT3ZuXMnADt27MDNzY06derk2x8FcXd3JyEhAYB79+7h7u7+TNsDTJgwAWdnZ6ytrVm1ahUtWrSgU6dOmJqaUq9ePTp37symTZvU/G3btqVFixaYm5szceJE7t+/z7FjxwqtY/Xq1TRp0oQBAwZgYWGBjY0N9erVA6Bs2bK0atUKS0tLHB0dGTNmTJ6+K8hfPa75mTdvHl5eXuri7+9frLYIIYQQQoiX10sXcM6bN4/ExERiY2PZunUrZcqUUdNyRuwKYm9vj7m5ufq+VKlSuLq6GrxPTU0F4OHDhwwfPhwfHx9sbGxo0qQJqampBlNmn64vODiY7777jri4OHbt2oWzszN169bNty09evTg7t27NGnSBI1Goy61a9fm9u3bBAcH8+jRozzbffrpp7i7u9OsWTPKli3LrFmzgCdTJOPi4rCzs0Ov16PX63nzzTe5c+cOAJGRkfj5+eUpLz4+nvT0dHx9fdV1vr6+3Lp1y2DktaB+io2Nxdvbu8A+OX78OM2aNcPJyQlbW1uWLFmS50Y9T2+TO3BfuXJlgUF7YWJjY7G3twfAwcGB2NjYZy4jd7siIiLYtm2b2rd6vZ4NGzYYjPLl7gdTU1M8PDyIiYkptI6Cjgs8udFRr1698PDwwMbGhqCgoGLf5OivHtf8hISEEBUVpS4nTpwoVluEEEIIIcTL66ULOAuTe6Twr5o7dy6XLl3i2LFjJCcnc+TIEcDwGsun68sZJV23bl2RgdKGDRtwdnbmxx9/RFEUdTl9+jSurq6sXr0aa2vrPNs5OzuzZMkSoqKi2LJlC59++infffcd3t7eeHp6kpSUpC7JyclcvHgReBIMXb9+PU95jo6OWFhYGDzGJCIiAjc3t2L1p7u7u8G1rblfA/Tp04d27doRGRnJ/fv3GTlyZJ7rVJ+uJygoiAMHDnD58mX27dtHcHBwke3ILTk5mQMHDqjXt7Zq1YqtW7caXM9aHLnb5e3tTa9evQz6NzU1lSVLlqh5cu97VlYWMTExeHh4FFpHQccFntzNNjs7m19//ZXk5GTWrFlT6PmX2189rvmxsbHB09NTXdzc3P5UOUIIIYQQ4uXxrwo4n6eUlBSsra3R6/UkJiYyffr0Ym03cOBAlixZwt69e+nbt2+B+dq0acPatWsNbqIDMHXqVJYsWUK3bt3y3W7jxo1ERUUBT6b9mpqaYmpqSt26dXFycmL69Ok8ePCA7Oxsrl27pk7BHDZsGHPnzuXYsWMoikJ0dDSXL1/G1NSUXr16MXHiRJKSkrhz5w7Tpk0rtO259ezZk88//5wbN26QkpJCaGioQXpKSgp2dnZYW1tz8uRJ1q5dW2SZOYF7z549CQgIKPZ0WEVROHfuHD179sTBwYGBAwcCMGbMGB48eEBQUBB//PEHiqJw9+5dpk+fzu7du4tVdnBwMHv27GHHjh1kZmaSnp7O8ePHDa7hDA8P5/vvvycjI4NPPvkEnU6nTo8tSFBQEIcPH2bVqlVkZGSQnJys3jk3JSWF0qVLY2trS2xsLHPmzDHY1sXFhd9//z3fcv/qcRVCCCGEEKI4JOD8k0aPHk16ejpOTk7Uq1ePli1bFmu7wMBAEhMTadq0aZGB0uuvv853331nsG779u106tSpwG1OnTpFgwYNKF26NE2bNmX06NEEBARgamrKjh07uHbtGuXLl8fOzo4ePXqoUz67dOnCjBkzGDRoEDY2NjRr1oybN28C8Pnnn+Pi4kLFihWpWbMm/v7+TJ06tVj7O2TIEDp37oy/vz+vvPJKnru+Ll68mA8//BCdTkdoaCjdu3cvVrkDBw7k119/LdZ02tGjR6PT6dDr9fTt25datWrx008/UapUKQDs7Oz4+eef0el0NG7cGJ1OR8OGDUlKSioyIMzh6enJ7t27WbBgAS4uLri7uzNx4kSDKdbBwcHMmTMHOzs7tm7dytatWw2mcOfHy8uL8PBwli5dipOTExUrVuTgwYMAfPDBB/z666/o9XoCAwPznBcTJ05k9uzZ6PX6PD9cwF87rkIIIYQQQhSHRinqORviuatVqxZjx46ld+/eJd2Uf6xff/2Vpk2bcvv2baysrEq6OUUaMGAArq6u6jW1/wbR0dFPbh7UbS+WpV1KujlCCCGE+Bs6HFa9pJsgjExGOF+w7777jpiYGLp06VLSTfnHyszMZP78+fTv3/8fEWwKIYQQQgjxb2VW0g34N2nRogVnzpzhf//7X76PZRFFu3btGjVq1KBixYrs3bu3pJsjhBBCCCGEKIRMqRVCGIVMqRVCCCFEUWRK7ctPptQKIYQQQgghhDAKCTiFEEIIIYQQQhiFXMMphDCqb+a/gqenZ0k3QwghhBBClAAZ4RRCCCGEEEIIYRQScAohhBBCCCGEMAoJOIUQQgghhBBCGIUEnEIIIYQQQgghjEJuGiSEMKquYy5hWfpeSTdDCCGEEEWQZ2IKY5ARTiGEEEIIIYQQRiEBpxBCCCGEEEIIo5CAUwghhBBCCCGEUUjAKYQQQgghhBDCKCTgFEIIIYQQQghhFBJwipfOgAEDmDBhQkk3o0ChoaH06tWrwPS2bduybNkyAMLCwqhfv/6LapoQQgghhBDPlQScz9GmTZuoX78+Wq0WZ2dnAgIC2LFjR0k36x9Po9Fw+fLl51ZedHQ0LVu2xMbGBnd3d2bOnPncyn4e9uzZw+DBg0u6GUIIIYQQQvxlEnA+J5999hkjR44kJCSEW7ducevWLd5//32+/fbbkm6a0WRmZv4jy580aRJ2dnbExcVx4cIFmjRpYpR6hBBCCCGE+LeTgPM5SE5OZvLkyfz3v/+lR48e6HQ6TE1Nef3119WpkYqiMHv2bMqWLYuDgwOdOnUiNjZWLcPX15e5c+dSu3ZtbGxsCAwMJDExEYDHjx8zYMAAHBwcsLW1pXr16ly6dAmAgIAAli5dqpYTHh6Or6+vQbmffvoptWrVonTp0nTp0oXExET69euHjY0NVatW5cKFC2r+27dv07NnT1xcXPDy8iI0NJTs7Gzg/6Z3jh8/HmdnZ9588808fZGTJyQkBL1eT7ly5di4caNBXw0ePBhXV1c8PT0JCQkhLS0NgIiICDQaDWFhYZQpU4Zq1arRsGFDAGrXro1Wq+WLL74AoHfv3ri5uWFra0uTJk04f/58sY+XmZkZnp6eWFpaYm9vT+PGjYvcRqPRsHTpUipVqkTp0qUZMWIEcXFxtGvXDp1OR8OGDYmOjlbznzhxggYNGmBra0vVqlXZuXOnQXnp6ekEBwej0+moUqUKhw4dUtOePqa5Xbt2jbZt2+Lo6Ei5cuVYvHhxgW3es2cPtWrVwsbGBi8vL95//301Laevv/rqK7y8vHBycmLSpEl5jnVBx1EIIYQQQojikIDzOTh69CiPHz+mS5cuBeZZsWIFixcvZvfu3URHR+Pm5kaPHj0M8qxevZqtW7cSGxtLUlIS8+fPV7c9f/48v//+O0lJSaxfvx57e/tit2/9+vXs2LGDyMhILly4QL169ejTpw+JiYk0a9aMkJAQALKzs+nYsSN+fn7cvHmT48ePs23bNjVoBjh58iSOjo7ExMSwYMGCfOvLyRMXF8eSJUvo378/165dA+Cdd94hJiaGy5cvc/LkSX766Sc+/PBDg+3Dw8M5d+4cp06d4ujRowCcOnWK1NRUhg0bBkCrVq24cuUKd+/exd/fn969exe7P/z9/Vm4cCGbNm0q9jYA27Zt4+eff+bSpUts3LiR1q1bExoaSkJCAs7Ozup+JCYm0qZNGwYNGsS9e/eYO3cuPXv25LffflPL2r59O61btyYxMZFx48bxxhtvqD8wFOThw4e8/vrrdOzYkVu3brF7925mzZrF/v37881funRpwsLCSEpKYvfu3Xz55Zds3rzZIM+ePXu4dOkSx44dY/369YSFhalphR3H/CQnJxMdHa0ut27dKqpLhRBCCCHES04Czufg3r17ODo6Ym5uXmCe1atXM2bMGCpVqoS1tTWffvopx48f5/fff1fzvPvuu3h7e6PVaunWrRunT58GwMLCgpSUFC5fvoyiKFSuXBlXV9dit+/tt9/Gw8MDBwcH2rRpQ9myZWnTpg2mpqb06tVLrefkyZNERUUxY8YMrKyscHd3JyQkhHXr1qllOTs7895772Fubo61tXW+9dnb2zNhwgTMzc1p1aoVbdq0YcOGDWRlZbFu3To++eQT9Ho9rq6uTJs2jZUrVxpsHxoaio2NTYHlAwwcOBAbGxssLS2ZOnUqFy9e5N69e0X2xU8//cTHH3/MwYMHGTt2LGvWrFHTdDqdOnKcn/Hjx2NnZ4ePjw+NGzfG39+funXrYm5uTo8ePdR+3LVrFz4+PgwdOhQzMzNatWpFhw4dWLt2rVpW9erV6du3L2ZmZvTv358yZcqwa9euQtu+c+dO3NzcGDlyJObm5lSsWJGhQ4caHJ/cmjZtSrVq1TAxMaFq1ar07t3bYCQVnvS1TqejXLlyvPvuuwZtLOg4FmTevHl4eXmpi7+/f6H7I4QQQgghXn5mJd2Al4GDgwPx8fFkZGQUGHTGxMQYTHXVarU4ODgQExNDuXLlAAyCyFKlSpGamgpA3759iY2NZdiwYcTGxtKlSxfmzJmDjY1Nsdr3dLkF1RMREUFcXBx2dnZqenZ2Nl5eXup7Ly8vNBpNofV5eHhgYvJ/v2X4+PgQExNDfHw86enpeab83rp1C0VRDPIXJisri8mTJ7Np0ybi4uLUuuLj43FwcCh020WLFjF06FAaNWpEeHg4zZo1Iy0tjZo1a2Jvb0/lypUL3La4/fj0sc7Zz5iYGPW9t7e3QXpOHxUmIiKCM2fOoNfr1XVZWVkFXoN6/PhxJkyYwIULF0hPTyctLY3OnTsb5MndjqfbUNBxLEhISAhDhgxR39+6dUuCTiGEEEKIfzkZ4XwOGjZsiJWVFVu3bi0wj4eHBxEREer71NRU7t27h4eHR5Hlm5mZMXnyZH799VfOnz/PpUuX+PTTT4EngevDhw/VvLdv3/7T++Ht7Y2npydJSUnqkpyczMWLF9U8RQWb8CTgyrkWECAyMhIPDw8cHR2xsLAw6IeIiAjc3NwMyi2qjrVr17Jlyxb279/P/fv3uXnzJoBB0FqQzMxM9ZrRSpUqsXfvXsaNG0evXr14//33i7V/RXn6WMOT/cx9rCMjIw3Sc/qoMN7e3jRs2NDg+KSkpLB79+588/fp04d27doRGRnJ/fv3GTlyZJ4+yt2Op9tQ0HEsiI2NDZ6enuri5uZW6P4IIYQQQoiXnwScz4GNjQ0fffQRb731Fps3byY1NZWsrCwOHTrE0KFDAQgKCmLBggVcvXqVx48fM378ePz9/dXRzcL88MMPnDt3jqysLLRaLZaWlpiamgJQs2ZNtc6oqCgWLlz4p/ejbt26ODk5MX36dB48eEB2djbXrl3LMw2zKAkJCcyePZuMjAwOHDjAnj176N69uzqFd+LEiSQlJXHnzh2mTZtG3759Cy3PxcXFYOpxSkoKlpaWODg48OjRI6ZMmVLstvXu3ZtFixaxf/9+srOzsbW1pVatWly9ehWtVvtM+1mQwMBAIiIi+Prrr8nMzOTAgQPs2LGDPn36qHnOnTvH2rVryczMZNWqVfz+++8EBgYWWm779u2JiIhg2bJlpKWlkZmZyfnz5/nll1/yzZ+SkoKdnR3W1tacPHnSYLpsjmnTppGSksIff/zBZ599ZnAtbEHHUQghhBBCiOKSgPM5effdd1m0aBFz5szBxcUFNzc3QkND6dSpEwD9+/dn2LBhtG7dGnd3d6Kiogq9Hi63nDvH2traUq5cOTw8PBg7diwAY8aMwdbWFjc3Nzp37kxQUNCf3gdTU1N27NjBtWvXKF++PHZ2dvTo0eOZb/5Sp04d7t69i5OTE8OGDePrr7+mYsWKAHz++ee4uLhQsWJFatasib+/P1OnTi20vGnTpjFkyBD0ej1ffvkl/fr1o2zZsnh4eFC5cmXq1q1b7LZ16dKFzz//nDFjxqDT6QgICKBx48bs2rWLIUOGcPDgwWfa1/zY29uze/du/ve//+Hg4MDo0aNZu3atwXTdjh07snv3buzs7Jg5cyZbt24t8kZQWq2W/fv3s337dvXOssOGDSM5OTnf/IsXL+bDDz9Ep9MRGhqab7DYpk0bXnnlFerVq0f37t0ZOHCgmlbYcRRCCCGEEKI4NEpx5iEKUUxhYWEsXbqUY8eOlXRTRCEiIiIoU6YMjx49wsrKKk/68ziO0dHRT24e1G0vlqVd/kpzhRBCCPECHA6rXtJNEC8hGeEUQgghhBBCCGEUEnAKIYQQQgghhDAKmVIrhDAKmVIrhBBC/LPIlFphDDLCKYQQQgghhBDCKCTgFEIIIYQQQghhFGYl3QAhxMvtm/mv4OnpWdLNEEIIIYQQJUBGOIUQQgghhBBCGIUEnEIIIYQQQgghjEICTiGEEEIIIYQQRiEBpxBCCCGEEEIIo5CAUwghhBBCCCGEUchdaoUQRtV1zCUsS98r6WYIIYQQJe5wWPWSboIQL5yMcAohhBBCCCGEMAoJOIUQQgghhBBCGIUEnEIIIYQQQgghjEICTiGEEEIIIYQQRiEB5z9Y27ZtWbZs2QuvNzQ0lF69er3weoUQQgghhBD/LBJw/gMEBARgZWWFVqtVl2+//ZY9e/YwePDgYpUxYMAAJkyYYOSWFuz+/fvY29uTnp5u9LpKIiDOyMhg3LhxeHt7o9Vq8fT0pGfPns+l7N9++422bdvi4OCAra0tVatWJSwsrNBt4uLisLCwoFOnTnnScp9PDg4OtGnThitXrgAQFhZG/fr1n0u7hRBCCCGEkIDzH2LBggWkpqaqS36BxN/Z3r17CQgIwMLCwqj1ZGZmGrX8guqYOXMmP/30Ez/99BOpqamcOHGC5s2bP5f62rdvT+PGjYmJiSE+Pp6wsDBcXFwK3Wb16tXodDp2795NXFxcnvSc8+nmzZvY2dkxYMCA59JWIYQQQgghcpOA8x8sICCApUuXAv83MjV58mQcHBzw8PBgzZo1ACxevJg1a9Ywb948tFotTZs2BSA5OZkRI0bg6emJq6srb731Fo8fP1bL37NnD3Xq1MHW1hYvLy+DUbWMjAyGDh2Kra0tfn5+HDhwoNC27ty5k/bt2+eb5uvry8yZM6latSq2trZ07dqVpKQkNb137964ublha2tLkyZNOH/+vJo2YMAARowYQefOndFqtaxbt46PP/6Yb775Bq1Wi6+vr5pv1KhRdOnSBZ1OR7Vq1Th79qxazu3bt+nZsycuLi54eXkRGhpKdna2Qd+OHz8eZ2dn3nzzzTz7cOLECTp16oSXlxcA7u7uDB8+XE1fuXIlVapUQafTUbZsWf773/+qacnJyXTo0AFnZ2fs7OwIDAwkMjISgPj4eP744w+GDRuGlZUV5ubm1K5dm7Zt2xba38uXL2fcuHGULVuW1atXF5hPq9USHBxs0Ke5zZkzBy8vL7Xd69evL7ReIYQQQgghcpOA8yVy6tQpXF1duXPnDgsXLmT48OEkJyczatQogoKCCAkJITU1lcOHDwMwcOBAHj9+zKVLl7h8+TLXrl1j+vTpalk9evRg2rRpJCQkcObMGapX/7+HFe/cuZPAwEASEhJ48803GTRoUIHtys7OZu/evbRr167APGFhYWzbto3o6GjS0tJ455131LRWrVpx5coV7t69i7+/P7179zbYdvXq1bzzzjukpKTQrVs3Jk2aRNeuXUlNTSUiIkLNt3btWkJCQkhKSqJ58+ZqHdnZ2XTs2BE/Pz9u3rzJ8ePH2bZtm8H1sSdPnsTR0ZGYmBgWLFiQp/0NGzZkwYIFfP7555w5c0YNVnM4Ojqybds2kpOTWbFiBWPHjuWXX35R6+/fvz8RERFERUVhY2PDqFGjAHBwcKBChQr07duXzZs3ExMTU2Af5jh16hQXLlygT58+BAcHs3z58gLzJicns2rVKmrVqpUn7cqVK0ydOpUDBw6QkpLCTz/9RLVq1QotKzo6Wl1u3bpVZFuFEEIIIcTLTQLOf4iQkBD0ej16vR5PT89883h4ePD2229jZmZGly5dMDEx4erVq/nmvXv3Ltu3b2fhwoXY2Nig1+uZMmUK69atA+DLL7+kf//+tGvXDlNTUxwdHalZs6a6fYMGDejcuTOmpqb079+fqKgo4uPj863r2LFj+Pj4FDoN9K233qJs2bLodDo++ugjNmzYoAZtAwcOxMbGBktLS6ZOncrFixe5d++eum379u1p1qwZGo0Ga2vrAuvo1KkTjRs3xtTUlH79+nH69GngSTAZFRXFjBkzsLKywt3dnZCQELUvAJydnXnvvfcwNzfPt47x48fz/vvv880339CoUSOcnZ2ZPXu2mh4YGIifnx8ajYYmTZrQunVrNfDX6/V069aNUqVKodVqmThxIocOHQJAo9Hwww8/ULFiRSZPnoyXlxe1a9dWg9X8LF++nNdeew0vLy+CgoK4cOECp06dMsgTEhKCnZ0dlSpVIi0tjRUrVuQpx8zMDEVRuHDhAo8ePcLNzY1XXnmlwHrnzZuHl5eXuvj7+xeYVwghhBBC/DtIwPkPMW/ePJKSkkhKSiI6OjrfPK6urgbvS5UqRWpqar55IyIiyMrKwsvLSw1k27dvz927dwGIjIzEz8+vwPbkrqtUqVIABda1a9euAqfT5vD29lZf+/j4kJ6eTlxcHFlZWUyYMIFy5cphY2ODj48PgEFwm7OuKE+3+cGDB8CTvoiLi8POzk7tizfffJM7d+6o+b28vNBoNAWWbWpqyogRIzh06BBJSUksWrSIKVOmsHfvXuDJ9OT69etjb2+PXq9n165d6j48fPiQ4cOH4+Pjg42NDU2aNCE1NZW0tDTgyfTczz77jCtXrnD79m2qVKlCp06dUBQlTzvS0tJYu3Ytffv2BaBMmTI0atQozyjnvHnzSExMJDY2lq1bt1KmTJk8ZZUrV44VK1bw+eef4+LiQrt27bh8+XKBfRASEkJUVJS6nDhxosC8QgghhBDi30ECzn+Jp4Mlb29vzMzMuHv3rhrI3r9/Xw0avb29uX79+nOpe+fOnYVOpwXUaxZzXpubm+Pk5MTatWvZsmUL+/fv5/79+9y8eRPAINh6et8KCwzz4+3tjaenp9oPSUlJJCcnc/HixT9VpoWFBb169aJatWpcuHCBtLQ0unbtyujRo7lz5w5JSUm0a9dO3Ye5c+dy6dIljh07RnJyMkeOHMmzjzmcnZ0ZN24csbGxJCQk5Enftm0biYmJjBs3DldXV1xdXTl9+jRr165VA9hn0aNHDw4dOsTt27cpV64cQ4cOLTCvjY0Nnp6e6uLm5vbM9QkhhBBCiJeLBJz/Ei4uLvzxxx/qe1dXV9q1a8e7775LYmIiiqIQFRVFeHg4AEOGDGHlypXs2bOHrKws4uPjDW6yU1w5U23zu0Ywt8WLF3Pjxg1SUlKYMmUKPXv2xMTEhJSUFCwtLXFwcODRo0dMmTKlWPsaERGR5zrKgtStWxcnJyemT5/OgwcPyM7O5tq1a+q01uJYsGAB33//vbr9rl27uHjxIvXr1yc9PZ20tDScnJwwMzNj37597Nu3T902JSUFa2tr9Ho9iYmJ6nW0AImJibz//vtcvXqV7Oxs7t+/z6JFiyhfvjwODg552rF8+XJ69erFhQsXOHv2LGfPnuXChQtkZmaybdu2Yu8PPLmG88CBAzx+/BhLS0u0Wi2mpqbPVIYQQgghhPh3k4DzX2Lw4MFcu3YNOzs7mjVrBsCKFSswNzenRo0a2Nra0rp1a/Wazzp16rBmzRomTZqEXq+nVq1anDt37pnrzbm5UFEjhP369aNjx454enpiamrKZ599pq4vW7YsHh4eVK5cmbp16xZZZ/fu3TE3N8fBwYFy5coVmd/U1JQdO3Zw7do1ypcvj52dHT169Himm96UKlWKcePG4eHhgb29PZMnT+aLL76gUaNG6HQ6Pv/8c/r06YOdnR2rVq2iQ4cO6rajR48mPT0dJycn6tWrR8uWLdU0CwsLoqKiaN26NTY2NpQvX55bt26xY8eOPG2IiYlh3759jBkzRh3ddHV1pUyZMvTv37/QmwflJy0tjcmTJ+Pk5ISDgwPHjh1T74oshBBCCCFEcWiU/ObtCfGctGvXjmHDhvHGG28UmMfX15elS5fSpk2bF9gyYWzR0dFPbh7UbS+WpQt/bqgQQgjxb3A4rHrRmYR4ycgIpzCq1157jRYtWpR0M4QQQgghhBAlwKykGyBebuPGjSvpJgghhBBCCCFKiAScosRFRESUdBOEEEIIIYQQRiBTaoUQQgghhBBCGIUEnEIIIYQQQgghjEKm1AohjOqb+a/g6elZ0s0QQgghhBAlQEY4hRBCCCGEEEIYhQScQgghhBBCCCGMQgJOIYQQQgghhBBGIQGnEEIIIYQQQgijkIBTCCGEEEIIIYRRyF1qhRBG1XXMJSxL3yvpZgghhBAl7nBY9ZJughAvnIxwCiGEEEIIIYQwCgk4hRBCCCGEEEIYhQScQgghhBBCCCGMQgJOIYQQQgghhBBGIQGnEEIIIYQQQgijkIBTiOdMq9Vy9erVkm6GEEIIIYQQJU4CziIEBARgZWWFTqfD1taWatWq8f7775OSklLSTXuuIiIi0Gg0PH78uNB80dHRtGzZEhsbG9zd3Zk5c+YLauGfd/DgQVxdXV9YfampqVSoUMHo9QwYMIAJEyYYvR4hhBBCCCH+LAk4i2HBggWkpKSQmJjIihUrOH78OI0aNeLhw4cl3bQXbtKkSdjZ2REXF8eFCxdo0qTJM22fmZlppJb9Per7J5G+EUIIIYQQxiYB5zMwMTGhZs2abNq0ibt377J8+XIAFEVh9uzZlC1bFgcHBzp16kRsbKy6XWxsLL169cLFxQU7Ozs6deoE5D/yVr9+fcLCwgAICwujfv36TJgwAXt7ezw9Pdm5cyd79+6lYsWK6PV63nvvPYPtV69ezauvvoper6dJkyZcvHhRTfP19WXu3LnUrl0bGxsbAgMDSUxMBKBhw4YAODo6otVqCQ8Pz7cPzMzM8PT0xNLSEnt7exo3blxkv2k0GhYvXkylSpXQ6/UA/PLLLzRt2hQ7OzsqV67Mli1b1PwDBgxg+PDhBAYGotPp8Pf358KFC2r61atXadmyJXZ2dlSoUIFly5apaTl9Nn78eJydnRk6dCht27bl7t27aLVatFotv/32W542KorCvHnzqFChAvb29gQGBhIdHV3gPp08eZJ69ephY2ODk5MTQUFBBvt7+fJlABISEujcubM6Ov7JJ5/g6+ur5i3smAD07t0bNzc3bG1tadKkCefPnwdg8eLFrFmzhnnz5qHVamnatKlaXu5jt3TpUgICAv70sQgPD6dq1arodDpcXV0ZO3ZsgX0ihBBCCCHE0yTg/BNsbW1p2bIlhw8fBmDFihUsXryY3bt3Ex0djZubGz169AAgKyuLDh06YG9vz7Vr17h79y5jxowpdl2nTp3Cx8eHu3fvMn78eAYOHMjy5cs5ceIEp0+f5osvvuDo0aMA7NixgylTprBu3Tru3btHcHAwHTp0ID09XS1v9erVbN26ldjYWJKSkpg/fz6AWkZ8fDypqam0adMm3/b4+/uzcOFCNm3a9Ex9tnHjRg4fPkxcXBy3bt2iTZs2hISEEB8fT1hYGEOGDDEIBFetWkVISAgJCQl07NiRTp06kZmZSUZGBu3bt6dBgwbcuXOHtWvXMn78eA4cOKBue/LkSRwdHYmJiWHx4sXs2bMHZ2dnUlNTSU1NpXLlynnat2jRItasWcO+ffu4c+cOtWrVolevXgXuz9tvv03Hjh1JSkoiKiqKUaNGFZgPICYmhm3btrFixYo8eQo6JgCtWrXiypUr3L17F39/f3r37g3AqFGjCAoKIiQkhNTUVPVcLI5nORYDBw5k3LhxpKSkcP36dbp161ZgucnJyURHR6vLrVu3it0mIYQQQgjxcpKA809yd3cnISEBeBIwjBkzhkqVKmFtbc2nn37K8ePH+f333/nll1+4ceMGCxYswMbGBnNzc1577bVi1+Ph4cHIkSMxMzMjKCiI+Ph4Ro8eja2tLWXLlqV+/fqcPn0agCVLljB+/HiqVq2Kqakpw4cPR6PRcOzYMbW8d999F29vb7RaLd26dVO3LY6ffvqJjz/+mIMHDzJ27FjWrFmjpul0Oi5dulTgthMmTMDZ2Rlra2tWrVpFixYt6NSpE6amptSrV4/OnTsbBLFt27alRYsWmJubM3HiRO7fv8+xY8c4fvw4iYmJfPDBB1hYWFCnTh2GDBliEMg5Ozvz3nvvYW5ujrW1dbH2bcmSJcyYMQNfX1/Mzc0JDQ3ll19+ITIyMt/8FhYW3Lx5k9jYWKysrGjUqFGePFlZWWzatInp06ej1WopU6ZMvoFpYcdk4MCB2NjYYGlpydSpU7l48SL37t0r1j4V5FmOhYWFBdevXyc+Ph6tVku9evUKLHfevHl4eXmpi7+//19qpxBCCCGE+OeTgPNPio2Nxd7eHngyepV7mqRWq8XBwYGYmBgiIyPx9vbGwsLiT9WTe8ptqVKl8l2XmpoKPLnxz9ixY9Hr9epy69YtYmJiCiwvZ9viWLRoEUOHDqVRo0aEh4fz3nvv8fXXX3PmzBns7e3zHTnM4ePjo76OiIhg27ZtBu3csGGDwYiYt7e3+trU1BQPDw9iYmKIiYnB09MTU1NTNd3X19dgH728vNBoNMXer5w29ezZU22Po6MjJiYmREdH8//au/e4nO//8eOPTirV1ZVK6UzJcckhpk1yPpTDcp4+mDCnMQ3DDpihTcaYYcaYc/ZB5DQ+aM60zZmVQ6Qi0RHpdP3+8Ov9dekgVtrheb/drttc7/fr/Xo93+/rYj17nWbNmqUMx+3UqRMAK1as4OHDhzRq1Ih69eqxYsWKQnXevXuXnJwcHB0dtWJ7VnGfSV5eHpMmTcLV1RWVSqU8w+Tk5Be6t2e9yGexZcsWzp07R82aNfHy8iIiIqLYeoODg4mLi1NeJ0+e/FNxCiGEEEKIvz/9ig7g7yg9PZ19+/bx8ccfA096IWNjY5XzmZmZ3Lt3D3t7eypVqsTNmzfJycnBwMBAqx5TU9NCCw/dvn37peNycnJi4sSJDBo06IWvLU2Clpuby+PHjwGoXbs2e/bsoXXr1lhaWvLJJ5+UWMfT55ycnOjbt68yV7UoT/cs5uXlER8fj729PfBkpdy8vDwl6YyNjVXOFXUvpbk3JyenQvMdC3h7ezNlyhStY66urqxZswaNRkNkZCTt27fHx8cHNzc3pYy1tTUGBgbExcVhbm4OQFxc3HNjKbBu3To2b97M3r17qV69Ounp6ajVajQaTbH39ex3qqjv04t8Fo0aNWLz5s3k5eWxceNGevbsyb179zAxMSlUVqVSoVKpSn1/QgghhBDin096OF+ARqPhzJkz9OnTB0tLS9555x0A+vfvz/z584mOjiYrK4sPP/yQpk2b4urqipeXF87OzgQHB5ORkUFOTg6RkZEAuLu7k5eXx+bNm8nNzWXRokVaPXUvasSIEYSEhHDmzBk0Gg2ZmZls3769VFu4WFtbo6ury9WrV4st069fP7755hv27t1Lfn4+5ubmNGrUiOjoaExNTUsdZ2BgILt27WL79u3k5uaSnZ3NiRMntOZw7t69m/3795OTk8MXX3yBmZkZzZo1o1mzZqjVambMmEF2dja//fYby5cv5z//+U+x7dnY2JCSkqK1GM+zRowYwUcffaTcf0pKCmFhYcWW//HHH0lKSkJHRwe1Wo2Ojo5Wrys86Znt0aMHU6dOJTMzkxs3brB48eLSPiYyMjIwNDTE0tKSR48eKb/gePq+rl27pnWsYcOGrFu3juzsbC5evFhiUg8lfxbZ2dmsXr2alJQU9PT0ir1PIYQQQgghiiMJZym8//77mJmZoVar+c9//kOjRo04cuSIMsR14MCBDBs2jA4dOmBnZ0dcXBwbN24EniQd27dvJzExkRo1amBjY8OCBQuAJz1CS5Ys4b333sPGxobbt2/TuHHjl46zW7duTJ06lYEDB6JWq6lZsyarV68u1bWVK1fm448/plWrVqjVavbs2VOoTEBAAAsWLGDcuHGYmZnh6+vLm2++yY4dOxgyZAgHDx4sVVsODg7s3LmT+fPnY2Njg52dHZMnT1Z6T+FJIhQaGoqFhQVbtmxhy5YtGBgYYGBgwPbt2zl06BBVq1alT58+zJw5k/bt2xfbXu3atQkMDMTNzQ21Wl3kKrXvvfceffv2xd/fH5VKRYMGDYp8BgX27t3La6+9hqmpKb169eLbb7+levXqhcp988035ObmYm9vT5cuXejXrx+Ghoalek4DBgygRo0a2NvbU6dOHby8vLTOBwUFERMTg4WFBa1atQJgxowZynDv0aNHM2DAgBLbeN5nsW7dOlxdXTEzM+PDDz8kLCwMIyOjUsUvhBBCCCGEjqZgfJ4QfxGDBg3C1taWkJCQig6lzM2bN48dO3Zorar7T3Xr1q0niwf13IOhiU1FhyOEEEJUuF9WNqjoEIR45aSHU4hy9Mcff/D777+j0Wg4d+4cX3/9NQEBARUdlhBCCCGEEK+ELBokRDl68OABffv25datW1haWhIYGMiwYcMqOiwhhBBCCCFeCUk4xV/O8xa6+TspWFRJCCGEEEKIfyMZUiuEEEIIIYQQolxIwimEEEIIIYQQolzIkFohRLn677y6ODg4VHQYQgghhBCiAkgPpxBCCCGEEEKIciEJpxBCCCGEEEKIciEJpxBCCCGEEEKIciEJpxBCCCGEEEKIciGLBgkhytUXi9ajMq9S0WEIIYT4F5g5KaiiQ3jlbu0M5lHSxXJvx7hqXRw6f1Xu7fyVrFq1isOHD7Ns2bKKDqVcREVFMXbsWI4cOVKu7UgPpxBCCCGEEH9Tj5Iu8jDueLm/SpvUhoSE0KpVK61jvr6+1KxZU+tYYGAgw4cPL7Pn8DyJiYl07doVOzs7dHR0uHz5conlc3NzmTp1KpMmTQIgKSmJt99+G3t7e1QqFU2aNOHnn3/WuiYyMpL69etTuXJlvLy8OHPmjHJu1apVeHl5YW5ujp2dHSNGjCAzM7PQ9U2bNsXU1JRq1arxzTfflBjj48ePGTJkCNWrV8fMzIx69eqxbt06rTL79++nSZMmqFQqnJycmDVrlnKuSZMmmJiYEBERUWI7f5YknEIIIYQQQogy0bJlS44fP87jx4+BJ0nR2bNnycnJISEhQSkXGRmJr6/vK4tLV1eXjh07snXr1lKVj4iIwMHBAVdXVwAyMzNp1KgRJ0+eJDU1lSlTphAQEMCNGzcAuHfvHt26dWPixImkpKTQr18/unbtqjyHhw8fEhoaSlJSEmfPniUmJoYJEyYo7f3xxx/06tWLTz/9lJSUFC5fvkybNm1KjDE3Nxc7Ozv+97//kZ6eztKlSxkxYgTHjh0Dnjz77t27M3jwYFJTU9m3bx8LFizQegYDBgxgyZIlpX2ML0USTiGEEEIIIUSZ8PLyQk9Pj5MnTwJw4sQJPD09adWqFZGRkQBcu3aNW7du4evrS1RUFN7e3qjVamxtbRk5cqSSpI0cOZJRo0Zp1T9y5EhGjhwJQHp6OsOHD8fBwQFbW1tGjx5NVlZWkXHZ2NgwcuRImjZtWqr7iIiIoHXr1sr7GjVqMH78eOzt7dHV1SUgIAAnJyeioqIA2Lx5M25ubgwYMABDQ0PGjRtHfn4++/btA2DEiBG0bNkSQ0NDrKysGDZsGIcPH1bqnzFjBkOHDsXf3x8DAwPMzc2pU6dOiTGamJjw2WefUaNGDXR0dHjzzTd54403OHr0KADJyclkZGQwaNAgdHV1cXd3p0WLFpw7d06po1WrVuzbt6/Y51YWJOEUQgghhBBClAl9fX28vb2V5DIyMhIfHx98fHy0jtWqVQtbW1v09PQIDQ0lOTmZEydOEBkZycKFC4Enw27DwsLIzc0FICcnh7CwMAIDAwF45513yMrK4uLFi1y+fJmYmBhmzJhRJvdx9uxZateuXez5hIQErly5Qr169QA4f/48np6eynkdHR08PDw4f/58kdcXDL8tcOzYMfT09GjQoAE2NjYEBAQQHx//QjE/ePCAqKgopV57e3t69+7N999/T25uLhcvXuTo0aO0a9dOucbe3p5KlSpx6dKlF2rrRUjCKYQQQgghhCgzvr6+z004C4bTNmzYEG9vb/T19XF2dmbYsGFKOW9vb1QqFXv27AFgz549mJub4+3tTVJSEtu2bWPhwoWoVCrUajUff/wx69evL5N7SElJQaVSFXkuKyuL3r17M2TIECUpzczMRK1Wa5VTq9VkZGQUuj48PJz169drJcdxcXGsXLmSDRs2cP36dSwsLJTEujTy8/MZNGgQXl5etG/fXjnev39/Zs6ciZGREfXq1WPo0KG8/vrrWteamZmRkpJS6rZelCSc5WzSpEkMGjSoTOs8dOiQMp78z1q7dq3WcIG/s23btuHs7IypqSkHDx6s6HBemb59+zJt2rSKDkMIIYQQAngyj/Po0aM8ePCAX3/9lebNm+Pq6kpGRgZ37tzRSjijo6Px9/fH1tYWlUrF5MmTSU5OVurq378/a9euBZ783Nq/f38AYmNjycvLw9HREbVajVqtxt/fn6SkpDK5BwsLC9LT0wsdz87OpkePHtjY2LBgwQLluKmpKWlpaVpl09LSMDMz0zq2b98+goKCCA8Px83NTTleuXJlBg4cSJ06dahcuTKfffYZkZGRPHjwgKlTp2Jqaqr1Wr58uXKtRqNh+PDhJCQksHHjRnR0dAC4dOkSffr0Yfny5Tx+/Jjr168TERHB4sWLtWLKyMjAwsLi5R/Wc0jCWYRNmzbx+uuvY2pqStWqVfH19WX79u0VHZaiRYsWXL16tUzq6t+/P/v37y+TuooSGRmJh4cHZmZm1KpVi127dpVbWx988AGhoaFkZmbi4uKCjo5OuY5H/zvy9fUt94nhQgghhPh3K5gnuXjxYurUqYOxsTHw5GfYNWvWEBsbqyScI0aMwM3NjejoaNLT05k9ezYajUapKzAwkPDwcBITE9m2bZvS6+fk5IS+vj5JSUmkpqaSmppKWlpaoZVfX5aHh0ehlWyzs7Pp1asXenp6bNiwAX39/9thsn79+pw+fVp5r9FoOHv2rNaw2f3799O3b182bdpEixYtCrVXkCg+TaPRMH36dDIzM7VeQUFByvlRo0Zx+vRpdu3ahampqXLthQsXcHV1xd/fHz09PVxcXOjTpw87duxQysTHx5Odnf3c+aJ/hiScz/j6668ZMWIEwcHBJCYmkpiYyCeffFLqFa2Etv/85z8MHjyY9PR0du3ahaOjY7m1df36dV577bVyq78oGo2GvLy8V9qmEEIIIcRfmYGBAc2bNyc0NBQfHx/luI+PD3PmzFHmb8KT3jWVSoWZmRnR0dGFfjHu7u5OvXr1GDhwIPXq1cPd3R0AW1tb/Pz8GDt2LCkpKWg0GuLi4ti9e3excWVlZSmdEdnZ2WRlZWklt0/z8/PTGjGXk5ND7969efz4MZs2bcLAwECrfEBAADExMaxZs4bs7Gy+/vprANq2bQvAwYMH6dmzJ2vXri20bQzAkCFDWLVqFVeuXOHx48d89tlntGrVSiuBLMro0aM5fvw4e/bsKTQEuGHDhsTGxrJ79240Gg3x8fGEhYXRoEEDpczBgwdp3bo1RkZGJbbzZ0jC+ZT09HQ++ugjFi1aRO/evTEzM0NPT482bdoo3dYajYYvv/ySGjVqYGlpSffu3bWWeD579ixeXl6YmZnh5+dXaDz0qVOn8PHxwcLCgjp16rB58+Zi41m9ejVubm6YmZnh6OjIvHnzgCdfjIK/pM+26e/vz/Dhw5VhvLGxsejo6LB69WqqV6+OhYUF48aNU65duXKl1jhuHR0dvvvuO2rXro25uTmBgYFkZ2cr5+fNm4e9vT02NjZ89dVX2Nraljh8VV9fX1k5q0aNGlq/5SlKwZLSFhYWWFhY0KxZM2VYxe3bt+nRowdWVlZUr16dWbNmkZ+fz4MHDzA1NSUvL4/GjRtjZWWFt7c3AFZWVpiamrJ7927atm3LmjVrALhy5Qo6OjrKP2q3b9/G2NiYrKws0tPT6dKlC1WrVsXCwoLOnTtz8+ZNJUZfX1+mTJmCr68vJiYmnDx5ktu3b9OnTx9sbGxwdHRk2rRp5OfnF3mPgwYNUvZ0Arh8+bLWb7R8fX2ZPHky3t7emJmZ0aZNG+Li4pTzBw4coF69epiZmTFgwACtz6ek2D/88EMOHTrE+++/j6mpKf/5z3+Uey8u9mvXrtG6dWvMzc2xtLTU+p+GEEIIIURxfH19uXPnTqGE886dO1rboYSGhhIWFoaZmRlBQUH06tWrUF2BgYHs3bu30JzGVatWYWBggKenJ+bm5nTo0IHo6OhiYzI2NlZ6Wxs0aICxsbGyrcmzunTpwo0bN7h+/ToAR48eJTw8nF9++QVLS0tlaGvBcF9LS0u2bt1KSEgI5ubmrF27lm3btmFoaAjA9OnTSU9Pp0ePHsq1BQsOwZPtSYYOHcobb7xBtWrVSEpKYvXq1SU9Ym7cuMG3337LxYsXcXR0VOot2GvT1dWVVatWMXHiRMzNzWnSpAmNGzfm448/VupYvXo1I0aMKLGdP0v/+UX+PY4ePUpWVhYBAQHFllm1ahXffvstu3fvxtnZmeDgYHr37s3hw4fJycmhW7duDB06lAkTJnDgwAG6d+9O7969gScbznbs2JHly5fTpUsXoqKi6NSpE3Xq1CnUjf3gwQMGDx7M/v37adGiBffv3y/yL0RBm8OHDyc4OJhDhw7RtWtXevbsqVVu3759nD9/njt37tCoUSO6dOlS7NzNn376iSNHjpCXl0fz5s1Zs2YNgwcPZu/evcyaNYu9e/dSu3ZtJkyYoDXG/lkajYbXX3+dYcOGUbNmzVJ11YeGhpKfn098fDyGhoacPn1a+Y1Lv379qF69Ojdv3iQhIYGOHTtibW3N0KFDyczMREdHh19//ZXatWsTGxtL9erVSU5OVq6Piori4MGDBAYGcuDAAWrUqMHBgwcZPnw4Bw4coFmzZhgZGZGVlcXAgQPZuHEj+fn5DBkyhJEjR2ptirty5Up27tyJh4cHOTk5tGjRgnbt2rFq1Sru37+Pn58f9vb2DB069Ln3XJQVK1awa9cu6taty9ixYwkMDCQyMpL79+/TrVs3Fi1aRL9+/Vi7di1BQUF4eHgATyaMFxf7F198wYkTJ+jbt6+y0XJ+fj5du3YtNvaPPvoId3d3ZbL+8ePHi405PT1da65DYmLiS927EEIIIUrPuGrdv2Q7H3/8sVZiA1CvXr1CPYo+Pj6Fhq5+9tlnWu/HjBnDmDFjCrWhUqlYsGCB1lzKkhTXm1kUfX19pk+fzqxZs1i2bBktW7Z87vW+vr7Frkp74MCB57ZZ1DMribOz83Nj6tGjBz169CjyXFRUFBkZGXTp0qXUbb4MSTifcu/ePaysrAp1kT9tzZo1jBs3TlmRas6cOVhYWHD16lXi4+N58OABkyZNQldXl/bt29OhQwfl2tWrV9O2bVu6d+8OQLNmzXjrrbfYtGkTn376aaG2DAwMuHjxIg0aNKBKlSpUqVKlUJljx47x8OFDJkyYgK6uLq1bt6Zjx46Fyk2fPh0TExNq1KiBj48Pv/32W7EJ55QpU7C0tASeDCf47bffGDx4MOvXr2fgwIHKks8zZ85k0aJFxT6rL7/8krt377JkyRLatWunJGhRUVF07ty5yEndlSpV4t69e1y5cgUPDw8aN24MwK1bt4iMjGTz5s1UrlwZNzc3xo8fz48//ljqpM7X11fp+T148CBTpkxRnvvBgweV37ap1WqthH3y5Mm8+eabWnUNGDBAeQ5nzpwhLi6Ozz//HB0dHezs7AgODuaHH3546YQzMDCQRo0aAfDFF19QpUoVbt26xf79+6lZs6bSOzlw4EDmz5+vXFea2J8WFRVVYuyVKlUiMTGRGzdu4ObmVmi+wdO++uorpk+f/lL3K4QQQoiX49D5q4oO4R9r0KBBZb74519JkyZNOHLkSLm3I0Nqn2JpaUlycjI5OTnFlomPj8fFxUV5b2pqiqWlJfHx8SQkJCibwRZwdnZW/hwbG0t4eLiykpZarWbjxo1F9gSZmJiwbds2tmzZgqOjIy1btuTYsWOFyiUkJGBnZ6fVZlHzJJ8eglu5cuUSJ1QXVzYhIUGrbpVKhbm5ebH1zJ8/n08//ZTu3bszb9482rVrx2+//cahQ4eKTIoBJkyYwJtvvkmPHj2ws7Nj4sSJ5OTkEB8fj7m5udYKWi4uLi+0P1HTpk1JTEwkLi6OyMhIunXrhq2tLZcuXeLAgQNKwvnw4UPeffddnJ2dUalUtGjRgszMTGUTYij8ud69excLCwvlcx01ahR37twpdWzPcnJyUv6sVqtRqVTKd+zpc8/GUprYn/a82OfMmYOdnR2tWrWiRo0ahISEFBtzcHAwcXFxyqtgw2chhBBCCPHvJT2cT/H29sbIyIgtW7Yow2CfZW9vT2xsrPI+MzOTe/fuYW9vDzxJSPPz85UE8ObNm8oEXicnJ/r27cvKlStLFU/btm1p27Yt2dnZzJ8/n969e2vN5QOws7MjISFBq824uLjnTjB+GXZ2dlrtp6enF1r++Wm5ublKotOrVy8ePXpE27ZtlTmVRTE1NeXLL7/kyy+/5OrVq3Tq1IlatWrRoUMH0tLSSE1NVfY4io2NVZ77s4pa5atSpUo0b96c7777DktLS6ysrPD19WXdunXExcUpc1nnzp3LxYsXOX78ONWqVePMmTN4enpqDVl4un4nJyccHBy0vhclMTU15eHDh8r727dvFyrz9JzR1NRU0tPTsbe3x87OTutcQdmC3tbnxf7sc3le7FWrVmXx4sUsXryY06dP06ZNG7y8vGjTpk2hsiqVqtj9qoQQQgghxL+T9HA+RaVSMXPmTEaPHs1PP/1EZmYmeXl5REZGKkMj+/fvz/z584mOjiYrK4sPP/yQpk2b4urqSvPmzTE2NubLL78kJyeHffv2aSVWgYGB7Nq1i+3bt5Obm0t2djYnTpzg0qVLhWK5c+cOW7duJTMzEwMDA2UBo2c1b94cIyMj5s6dS25uLgcPHixxda4/o0+fPvz444+cPXuWx48f88knn2j1rD6rX79+TJgwgcuXL5Ofn0/t2rUxMzMjIyOjyHsBiIiIIDo6mvz8fFQqFQYGBujp6eHg4ECLFi0YP348jx494urVq8ydO1cZWvosa2trdHV1C20f4+vry4IFC5TVwQreF8zfhCerpRkbG6NWq0lJSdHalLcoXl5eWFtbM2PGDB48eEB+fj4xMTHKpsXPatiwITt37uTu3bvcv3+fL774olCZtWvXcvr0abKyspg8eTJvvPEGDg4O+Pn5ER0dzbp168jNzWX16tWcO3dOue55sdvY2Gg9k+fFHhYWpvySQa1Wo6enV+xnJ4QQQgghxLMk4XzG2LFj+eabbwgNDcXGxoZq1aoxbdo0Zd7lwIEDGTZsGB06dFB6/DZu3Ag8mXMZHh7OTz/9hIWFBfPmzdNKiBwcHNi5cyfz58/HxsYGOzs7Jk+eXORwx/z8fObPn4+DgwNqtZqVK1eybt26QuUK2ty4cSNqtZo5c+bQp08fZUWsstShQwcmTpxIx44dcXR0xMnJCXNz82LbmjNnDp07d6ZDhw6YmZnx7rvv8sUXX/D+++/TsWPHIocSX7lyhY4dO2JmZoaHhwft27dXnuH69eu5d+8eDg4OtG7dmgEDBjBkyJAi265cuTIff/wxrVq1Qq1WK4ve+Pr6kp6ergyf9fHx4cGDB1qrpb3//vtkZ2djbW1Ns2bNaNeuXYnPRU9Pj+3btxMTE0PNmjWxsLCgd+/exS6aExgYSLNmzXBzc+ONN94ocpGqQYMGMXLkSKytrfnjjz+0VkDbsmULn3/+ORYWFuzdu1drovfzYh87dizbtm3DwsKCgQMHPjf2gs2aTUxM8PHx4f3339d6VkIIIYQQQpRER/MiyzWJv4Vu3brh5eX1QqtcvYyUlBSqVKlCXFwcDg4O5drWv4mvr6/WSrJ/V7du3cLR0ZHRk75EZV54wSshhBCirM2cFFTRIQghniE9nP8Av/zyC/Hx8eTl5bFt2zZ2797NW2+9VS5tbd68mcePH5ORkcG4ceNo2rSpJJtCCCGEEEKIIknC+Q9w5coVmjRpgkql4sMPP2TFihVaG8mWpRUrVlC1alUcHR1JTEwscpivEEIIIYQQQoAMqRVClBMZUiuEEOJVkyG1oiytWrWKw4cPs2zZsooOpVxERUUxduzYct+LU7ZFEUIIIYQQ4m9q7qpbXLv1qNzbqeFgzAcDnz+NKiQkhD179nDgwAHlmK+vL/Hx8cTExCjHAgMDMTU1ZcmSJeUS77N27NhBSEgI586do1KlSrRp00ZZyLMoubm5TJ06lf/9738AJCUl8f777xMZGUlGRgbu7u7MmjWL9u3bK9dERkYyatQorl27Rr169fj+++9p0KAB8CR5/eabb4iOjsbExIRu3boxZ84cZSvD+/fvM2rUKPbt20d+fj4tWrTg22+/xc7OrsT7mjZtGj/88AMpKSmYmZnRp08fvvjiCwwMDIAn2yUGBQVx5MgRqlatyuzZs+nbty8ATZo0wcTEhIiICPz9/f/cAy6BDKkVQgghhBDib+rarUeci3lY7q/SJrUtW7bk+PHjyi4Mjx8/5uzZs+Tk5JCQkKCUi4yMfKUr36elpTFp0iTi4+O5du0aenp6DBo0qNjyERERODg44OrqCkBmZiaNGjXi5MmTpKamMmXKFAICArhx4wYA9+7do1u3bkycOJGUlBT69etH165dlefw8OFDQkNDSUpK4uzZs8TExDBhwgSlvUmTJnH37l1iYmK4desWJiYmjBw58rn31b9/f86fP096ejpnzpzh999/Z968ecr5fv364ebmRnJyMj/88ANDhw7l/PnzyvkBAwaUe9IvQ2qFEOWiYEitrGIshBBClJ9RM2M4F/Ow3Nt5rWZlFn1U87nlcnNzUavV7Nq1ixYtWvDLL78wbdo0nJ2dad++Pf369ePatWu4urqSmJjIrVu3GDNmDBcvXsTIyIiAgADmzZuHoaEhI0eOREdHh0WLFin1FyRh3377Lenp6UycOJGIiAhyc3Pp2bMnoaGhyt7qJTl58iRt2rQhIyOjyPNDhgzBzs6Ozz77rNg66taty4wZM+jRowfLli1j6dKlREVFAaDRaHBycmLJkiX4+fkVujYsLIwZM2Yo+6m3a9eOt956S7m/HTt2MGbMmEJ7ypckOTmZPn364OrqynfffUdMTAx169YlKSkJCwsL4EmCamdnx5w5cwCIj4/H1dWV1NTUUj23lyE9nEIIIYQQQogyoa+vj7e3N5GRkcCTnkwfHx98fHy0jtWqVQtbW1v09PQIDQ0lOTmZEydOEBkZycKFC4Enw27DwsLIzc0FICcnh7CwMAIDAwF45513yMrK4uLFi1y+fJmYmBhmzJhRqjgjIyOpX79+sefPnj1L7dq1iz2fkJDAlStXlIU6z58/j6enp3JeR0cHDw8Prd7EktofPXo027Zt4969e2RmZrJ69Wo6depUqntZvHgxZmZmWFtbc+bMGUaMGKHE5OzsrCSbAJ6enlox2dvbU6lSJS5dulSqtl6GJJxCCCGEEEKIMuPr6/vchLNgOG3Dhg3x9vZGX18fZ2dnhg0bppTz9vZGpVKxZ88eAPbs2YO5uTne3t4kJSWxbds2Fi5ciEqlQq1W8/HHH7N+/frnxnfy5Ek+//xzQkNDiy2TkpKCSqUq8lxWVha9e/dmyJAhSlKamZmJWq3WKqdWq4vsQQ0PD2f9+vVayXGTJk3Iy8vDysoKc3NzoqOjS508jxgxgoyMDC5evMjQoUOpVq3aC8VkZmZGSkpKqdp6GZJwCiGEEEIIIcpMy5YtOXr0KA8ePODXX3+lefPmuLq6kpGRwZ07d7QSzujoaPz9/bG1tUWlUjF58mSSk5OVuvr378/atWsBWLt2Lf379wcgNjaWvLw8HB0dUavVqNVq/P39SUpKKjG2s2fP0qVLF77//nveeOONYstZWFiQnp5e6Hh2djY9evTAxsaGBQsWKMdNTU1JS0vTKpuWloaZmZnWsX379hEUFER4eDhubm7K8V69euHs7ExaWhoPHjzAz89P6eFcvnw5pqamWq+pU6cWiq1OnTrUrVtXGZZb2pgyMjK0ekHLmiScQgghhBBCiDLTtGlT4MlQzzp16mBsbAxAixYtWLNmDbGxsUrCOWLECNzc3IiOjiY9PZ3Zs2fz9BIzgYGBhIeHk5iYyLZt25ThtE5OTujr65OUlERqaiqpqamkpaWRmZlZbFznzp2jffv2fPXVV/Tq1avEe/Dw8ODy5ctax7Kzs+nVqxd6enps2LABff3/2/Cjfv36nD59Wnmv0Wg4e/as1rDZ/fv307dvXzZt2kSLFi206j579izDhw9HpVJhZGTEmDFjOHHiBMnJyQQFBZGZman1mj59epFx5+bmKvM+69evz40bN0hNTVXOnz59Wium+Ph4srOzqVOnTonP48+QbVGEEOXqi0XrZR9OISqA7EcohKgoBgYGNG/enNDQUK2VYH18fJg+fboyfxOe9K6pVCrMzMyIjo5myZIlmJubK9e4u7tTr149Bg4cSL169XB3dwfA1tYWPz8/xo4dy6xZs1Cr1dy6dYsLFy7QsWPHQjFduHCBdu3aMXv2bKWXtCR+fn5aq73m5OTQu3dvHj9+THh4uLLtSIGAgAAmTJjAmjVr6N27N99++y0Abdu2BeDgwYP07NmT9evX06pVq0LtNWvWjGXLllGnTh0MDAxYtGgR9vb2WFlZlRjn4sWL6d27N1WqVOHcuXOEhIQoW5zUrFkTLy8vPv74Y+bMmcOpU6fYtm0bR48eVa4/ePAgrVu3LrcFg0B6OIUQQgghhBBlzNfXlzt37uDj46Mc8/Hx4c6dO1rboYSGhhIWFoaZmRlBQUFF9jwGBgayd+9epXezwKpVqzAwMMDT0xNzc3M6dOhAdHR0kfEUbEny3nvvaQ1NLU6XLl24ceMG169fB+Do0aOEh4fzyy+/YGlpqVxfMNzX0tKSrVu3EhISgrm5OWvXrmXbtm0YGhoCMH36dNLT0+nRo4dybcGCQwArVqzg9u3bODs7Y2Njw4EDB9i6dWvJDxnYtWsXtWvXxszMjO7du/PWW28xa9Ys5fyGDRv4448/sLS0ZMCAASxdulSrh3P16tXKIkPlRbZFEUKUi4JtUUZP+lJ6OIWoANLDKcS/w9xVt0q9R+afUcPBmA8G/ru2OVu5ciVHjhxh2bJlFR1KuYiKimLs2LEcOXKkXNuRhFMIUS4k4RSiYknCKYQQ4q9AhtQKIYQQQgghhCgXknAKIYQQQgghhCgXknCKfyQdHZ1CS1m/amvXrqV169YVGsOf4eLiwu7duys6DCGEEEII8TcmCacoFV9fX5YsWVKqstOmTaNv377lHNFfX//+/dm/f39FhyGEEEIIIUSFkYRT/OXk5uZWdAjPpdFoyMvLq+gwhBBCCCGE+EuThFO8sIMHD2Jra8vChQupVq0aVatWZc6cOQBEREQwa9Ys/vvf/2JqaoqLiwsA2dnZTJkyherVq2NlZcXbb79NSkoKALGxsejo6LBy5UqqV6+Oh4dHiW3Ak2Wcvb29UavV2NraMnLkSB4/flyq+FevXo2bmxtmZmY4Ojpqbeq7Z88emjRpglqtplGjRhw6dEg55+vry5QpU/D19cXExITDhw9TpUoVsrOzlTInT57EwsKCx48fs3LlSl5//XXl3B9//EHnzp2xsrLCysqK0aNHl6rdZ7m4uDBnzhwaNWqEiYkJAQEBpKSkMGDAAFQqFa+99hrnz59Xyt++fZs+ffpgY2ODo6Mj06ZNIz8/H4Br167RunVrzM3NsbS01NorC55skty4cWNUKhWdO3dWPjMhhBBCCCFKQxJO8VKSk5OJi4vjxo0bRERE8NFHH3HlyhX8/f2ZMmUKPXr0IDMzk9jYWAAmT57Mb7/9xokTJ4iLi6NSpUpaCRfA7t27OXPmDL/++muJbQDo6ekRGhpKcnIyJ06cIDIykoULFz437gcPHjB48GB++OEHMjIyOHPmjLL58JkzZ3j77beZP38+9+/f57PPPqN79+4kJycr169cuZL58+eTmZnJ66+/jrW1NTt37lTOr127ll69eimb/BbIzMykbdu2vPHGG8TFxREXF6cMOy5Nu8/asGED27dv5+bNm5w/f55mzZopSXyrVq0IDg4GID8/n65du+Lm5saNGzc4ceIE4eHhLF++HICPPvoId3d3kpOTuX37NjNnztRqZ82aNWzZsoWEhARSU1O1kvNnpaenc+vWLeWVmJj43M9DCCGEEEL8s0nCKV6Krq4un3/+OZUqVaJp06bUrl2b06dPF1lWo9GwZMkS5s2bR9WqVTE2NmbGjBls2rRJa1jqtGnTUKlUGBsbP7eNhg0b4u3tjb6+Ps7OzgwbNozIyMhSxW5gYMDFixdJT0+nSpUqNGzYEIClS5cyZMgQ3nzzTXR1dfH398fT01MroRwwYACenp7o6upiaGhI//79Wbt2LQB5eXls3LiRwMDAQm1GRERgbm7ORx99hLGxMcbGxrz55pulbvdZ7733Hvb29lhaWtKxY0dq1KhBx44d0dPTo2/fvvz222/Ak57guLg4Pv/8c4yMjLCzsyM4OJj169cDUKlSJRITE7lx4wYGBga0aNFCq52xY8fi5OSEqakpPXv2VOotyldffYWjo6Pyatq0aWk+DiGEEEKIv6RVq1YxdOjQig6j3ERFRfHGG2+Uezv65d6C+EeqUqUKlSpVUt5XrlyZzMzMIsvevXuXhw8f0rx5c63jurq63L59W3nv7Oxc6jaio6MJDg4mKiqKhw8fkpubS4MGDZ4bt4mJCdu2bSM0NJSJEyfi6elJSEgIzZs3JzY2loMHD7J06VKlfE5ODu3atSs2xsDAQOrXr096ejrHjh3DyMioUNIGcPPmTdzc3IqMqTTtPsvW1lb5c+XKlQu9L3hOsbGx3L17FwsLC+V8fn4+jo6OAMyZM4epU6fSqlUrDAwMGDZsGJMmTSq2neI+Y4Dg4GCGDBmivE9MTJSkUwghhChn4XuOcOdu+U95sbG2oFuH5ycnISEh7NmzhwMHDijHfH19iY+PJyYmRjkWGBiIqalpqRel/LOOHz/O8OHDuXHjBgBNmjRh/vz51KtXr8jyubm5TJ06lf/9738AJCUl8f777xMZGUlGRgbu7u7MmjWL9u3bK9dERkYyatQorl27Rr169fj++++Vn0/Dw8OZPHkyCQkJ6Ovr4+Pjw8KFC7G3tweeTD8bM2YMGzZsQF9fn6FDhzJr1ix0dHRKvK9hw4axY8cOMjIyqFKlCsOGDWPKlCnKeRcXF+7cuYOenh7w5GfZCxcuKM/AxMSEiIgI/P39X+axlooknKLMPfsXw8rKCmNjY06fPq3M6XxawbDb5/2FetqIESN47bXXWLduHSqViq+//lrptXuetm3b0rZtW7Kzs5k/fz69e/cmLi4OJycnJk6cyLRp04q99tkYa9SogaenJz/99BMHDhzg7bffLvI+nJycWL16dZF1lqbdl+Xk5ISDg4PyjJ9VtWpVFi9ezOLFizl9+jRt2rTBy8uLNm3avHBbKpUKlUr1JyMWQgghxIu4czeFG7fuVHQYipYtWzJ9+nQeP36MoaEhjx8/5uzZs6hUKhISErCzswOeJGdPr89R3tzc3Ni+fTsODg7k5eWxaNEievXqxcWLF4ssHxERgYODA66ursCT6VGNGjVizpw5VKtWja1btxIQEMCFCxdwdnbm3r17dOvWjQULFtCnTx8WLVpE165diY6OxtDQkCZNmnDgwAFsbGzIysrik08+YejQocqIts8++4zffvuN6OhosrKyaNeuHU5OTowYMaLE+3r//ff5+uuvMTY2Ji4ujg4dOuDm5kbv3r2VMlu2bKFjx45FXj9gwACWLFlSrgmnDKkVZc7GxobY2FhlYRpdXV3effddxo0bp8zrS0pKIjw8/KXbyMjIQKVSYWZmRnR0dKl/O3bnzh22bt1KZmYmBgYGmJmZKb/xGTZsGN999x2HDx8mPz+fR48eceDAAW7dulVinYGBgXz//fds3bq1yOG0AH5+fty/f5+QkBAePXrEo0ePOHz48J9qtzS8vLywtrZmxowZPHjwgPz8fGJiYpThx2FhYcTFxQGgVqvR09NTnocQQgghxIvy8vJCT0+PkydPAnDixAk8PT1p1aqV8vPHtWvXuHXrFr6+viUuBDly5EhGjRqlVf/IkSMZOXIk8GT9iOHDh+Pg4ICtrS2jR48mKyuryLisrKxwdHRER0cHjUaDrq4uV69eRaPRFFk+IiJCaz/1GjVqMH78eOzt7dHV1SUgIAAnJyeioqIA2Lx5M25ubgwYMABDQ0PGjRtHfn4++/btA8De3h4bGxulPl1dXWVtEoAffviBTz/9lKpVq+Lk5MT48eNZsWLFc5933bp1leloRdX7PK1atWLfvn3FPreyIAmnKHO9evXCwMAAS0tL5bdCISEhNGjQgBYtWmBmZoa3t7fyD9HLCA0NJSwsDDMzM4KCgujVq1eprsvPz2f+/Pk4ODigVqtZuXIl69atA6BRo0asWrWKCRMmYGlpibOzM3PnzlUS5+L06dOHqKgoatasSd26dYssY2Zmxt69e9m/fz/29vY4OTkRFhb2p9otDT09PbZv305MTAw1a9bEwsKC3r17K4n/r7/+SvPmzTExMcHHx4f3339fWURJCCGEEOJF6evr4+3trSSXkZGR+Pj44OPjo3WsVq1a2NralrgQZGBgIGFhYcqWeTk5OYSFhSm/4H/nnXfIysri4sWLXL58mZiYGGbMmFFsbGlpaajVaoyMjBg7dixTpkwpdoTd2bNnqV27drF1JSQkcOXKFWVI7vnz5/H09FTO6+jo4OHhobVzwLlz51Cr1RgbG/PVV1/x4YcfApCSkkJCQoLW9Z6enlrXlmTy5MmYmJjg5OTEgwcPCnWADBw4EGtra3x9fTly5IjWOXt7eypVqsSlS5dK1dbL0NEUl9YLIcSfcOvWLRwdHRk96UtU5lUqOhwh/nVmTgqq6BCEEK/Ad2siXsmQWmcHG4YFlm7Y5axZszhw4AB79+6lbdu2TJkyBWdnZ/z9/bl06RKDBg3CyMioyBFqX3/9Nfv27WP79u0AuLq6smDBAvz8/IiIiGDs2LFcvXqVpKQk7O3tuX//PmZmZgAcOnSIgQMHcu3atRLjy8jIYPny5dSsWRM/P78iy9SsWZN58+YVOdQ0KyuLtm3b4uHhwbfffgtAUFAQFhYWhIaGKuX69+9P9erV+fzzz7WuT05OZvHixbRr147XX39dmdqVkZGBqakpADExMbi7u5OTk4O+/vNnQWo0GqKioti6dSuTJk1Snsnhw4dp3Lgx8GS3hQ8//JBz585prUtib2/P6tWrtXp0y5L0cAohhBBCCCHKTMuWLTl69CgPHjxQRlO5urqSkZHBnTt3iIyMVEZURUdH4+/vj62tLSqVismTJ2ttDff0jgBr166lf//+wJM1QPLy8nB0dEStVqNWq/H39ycpKem58ZmZmTF69Gj69+9fbHkLCwvS09MLHc/OzqZHjx7Y2NiwYMEC5bipqSlpaWlaZdPS0pTE72lWVlYMGjSIbt26kZubqySZT1+flpaGkZER+vr6TJ06FVNTU61XwRZ3BXR0dPDy8sLIyIipU6cqx998801lh4QRI0bQsGFDdu3apXVtRkaG1gKTZU0STiGEEEIIIUSZKVilfvHixdSpU0eZY9iiRQvWrFlDbGysknCOGDECNzc3oqOjSU9PZ/bs2VrzKgMDAwkPDycxMZFt27Ypw0WdnJzQ19cnKSmJ1NRUUlNTSUtLK3FF/afl5+fz8OFD4uPjizzv4eHB5cuXtY5lZ2fTq1cv9PT0lNVkC9SvX19ri0CNRsPZs2epX79+kfXn5uaSlJREeno6FhYW2NnZaV1/+vRp5drp06eTmZmp9QoKKnoUS25uLlevXi32vnV1dbWeb3x8PNnZ2dSpU6fYa/4sSTiFEEIIIYQQZcbAwIDmzZsTGhqKj4+PctzHx4c5c+Yo8zfh+QtBuru7U69ePQYOHEi9evVwd3cHnmzd5ufnx9ixY0lJSUGj0RAXF8fu3buLjCk8PJyLFy+Sn59PamoqY8aMwdrauthEy8/Pj4MHDyrvc3Jy6N27N48fP2bTpk0YGBholQ8ICCAmJoY1a9aQnZ3N119/DTzZHQFg/fr1XL9+HY1Gw+3btxk3bhyNGjWiSpUn044GDRrEjBkzuHv3LnFxccydO5fBgweX+JxTUlJYvXo16enp5Ofnc+TIERYvXqy0efPmTQ4dOkR2djbZ2dksW7aMU6dOaW3lcvDgQVq3bo2RkVGJbf0ZknAKIYQQQgghypSvry937twplHDeuXNHa4HC0iwEGRgYyN69ewsthrNq1SoMDAzw9PTE3NycDh06EB0dXWQ8d+7coWvXrpiZmeHu7k58fDx79uwpNtHq0qULN27c4Pr16wAcPXqU8PBwfvnlFywtLZWhrQXDfS0tLdm6dSshISGYm5uzdu1atm3bhqGhIfBk6HDLli0xNTWlYcOGVK5cmS1btijtTZ06lQYNGlCzZk08PT3p1q0bw4cPL/EZ6+jo8MMPP+Ds7Iy5uTlBQUF88MEHjB49Gniylct7771HlSpVsLW1ZdWqVURERCiLegKsXr36uVuv/FmyaJAQolzIokFCVCxZNEiIf4fwPUe4czel3NuxsbagW4c3yr2dv5KVK1dy5MgRli1bVtGhlIuoqCjGjh1baOXasiYJpxCiXBQknHFxcTg4OFR0OEIIIYQQogLIkFohhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCEk4hhBBCCCGEEOVCv6IDEEL8M+Xm5gKQmJhYwZEIIYQQQoiyZmtri77+89NJSTiFEOXi+vXrADRt2rSCIxFCCCGEEGUtLi4OBweH55aThFMIUS4cHR0BOHr0qPJnIRITE2natCknT56kWrVqFR2O+AuQ74R4lnwnRFHke/HXY2trW6pyknAKIcqFkZER8CTxLM1vv8S/S7Vq1eR7IbTId0I8S74Toijyvfj7kUWDhBBCCCGEEEKUC0k4hRBCCCGEEEKUC0k4hRDlQqVSMXXqVFQqVUWHIv5C5HshniXfCfEs+U6Iosj34u9LR6PRaCo6CCGEEEIIIYQQ/zzSwymEEEIIIYQQolxIwimEEEIIIYQQolxIwimEEEIIIYQQolxIwimEEEIIIYQQolxIwimEKHOXL1+mXbt2mJiYYGtry8SJE8nOzq7osEQFunLlCsOHD8fT0xN9fX3q169f0SGJCrRp0ya6deuGg4MDJiYmeHp6smLFCmQdw3+3nTt30rJlS6ytrTE0NKRGjRoEBweTlpZW0aGJv4DMzEwcHBzQ0dEhKiqqosMRL0C/ogMQQvyzpKSk0Lp1a2rWrMnmzZuJj48nODiYhw8f8s0331R0eKKCXLhwgR07dtCsWTPy8/PJz8+v6JBEBfrqq69wcXFh7ty5WFtbs3fvXoYOHUpcXBxTp06t6PBEBbl//z7NmjVjzJgxWFpacv78eaZNm8b58+f5+eefKzo8UcFmzJhBbm5uRYchXoJsiyKEKFOzZ89m5syZ3Lx5kypVqgDw3XffMXLkSG7evImdnV0FRygqQn5+Prq6TwbVDBo0iKioKM6fP1/BUYmKkpycjJWVldaxYcOGsXHjRlJSUpTvihDLli1j2LBhxMfHy/8//sUuX75MkyZNmDt3LsOHD+fUqVM0adKkosMSpST/ogshytSuXbto27atkmwC9O7dm/z8fPkN9b+YJBDiac8mmwANGzYkPT2dBw8eVEBE4q/K0tISQKZl/Mu99957DB8+nFq1alV0KOIlyE8AQogydfnyZWrXrq11TK1WU61aNS5fvlxBUQkh/uoOHz6Mvb09ZmZmFR2KqGB5eXlkZWXx22+/8dlnn9G1a1dcXFwqOixRQX766SfOnTvHp59+WtGhiJckCacQokylpKSgVqsLHbewsOD+/fuvPiAhxF/e4cOH2bBhA+PHj6/oUMRfgLOzM8bGxjRu3Jhq1aqxbt26ig5JVJCHDx8SHBzMrFmzUKlUFR2OeEmScAohhBCiwty6dYs+ffrQqlUrxowZU9HhiL+AnTt3cvToUZYtW8alS5fo0qULeXl5FR2WqACff/45NjY2vPPOOxUdivgTZJVaIUSZsrCwKHIJ+5SUFK15nUIIkZqaSqdOnbC0tOS///2vzPUVAHh4eADQvHlzvLy88PT0ZMuWLfTs2bOCIxOv0o0bN5g7dy5btmxRfq7IzMxU/puZmYmpqWlFhihKSRJOIUSZql27dqG5mmlpaSQmJhaa2ymE+Pd69OgR/v7+pKWlcezYMczNzSs6JPEX5OHhgYGBAVeuXKnoUMQrdv36dbKzs/Hz8yt0rlWrVjRr1ozjx49XQGTiRUnCKYQoU506dWLWrFmkpqYqczk3bdqErq4u7du3r9jghBB/Cbm5ufTu3ZtLly5x6NAh7O3tKzok8Rd14sQJcnJyqFGjRkWHIl4xT09PDhw4oHXs9OnTjBs3jiVLluDl5VVBkYkXJQmnEKJMDR8+nIULF9K9e3emTJlCfHw8EyZMYPjw4bKH2r/Yw4cP2blzJ/BkmFR6ejo//fQTAC1btsTa2roiwxOv2MiRI4mIiGDu3Lmkp6dr9VI0bNgQQ0PDCoxOVJSAgACaNGmCh4cHxsbGnDlzhjlz5uDh4UH37t0rOjzxiqnVanx9fYs817hxYxo1avRqAxIvTUej0WgqOgghxD/LpUuXeO+99zh69ChmZmYMGDCAmTNnUqlSpYoOTVSQ2NhYqlevXuS5AwcOFPtDhfhncnFx4caNG0Weu379umyB8S8VEhLCxo0buXr1Kvn5+bi4uBAQEMD48eNlhVIBwMGDB2nVqhWnTp2iSZMmFR2OKCVJOIUQQgghhBBClAtZDk4IIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYQQQgghRLmQhFMIIYT4G5s2bRo6OjrKy9ramtatW3Po0KHnXnvw4EF0dHSIiooqk1iKqm/atGkcPXq0TOr/u0tKSsLMzIzz588rx3R0dAgNDVXe+/r64u/vr7yfNm0apqamZR7L1q1b+fbbb8u83vnz57Nz585Cx11cXBg9enSZtxcbG4uJiQmxsbFlXrcQomxIwimEEEL8zRkbG3Ps2DGOHTvG4sWLuXfvHm3atNFKbIrSqFEjjh07Rp06dcokjqLqmz59uiSc/9/MmTPx9fWlfv36yrFjx47Rv3//Vx7Lq044t2zZwvjx48u8PRcXF3r27MnUqVPLvG4hRNnQr+gAhBBCCPHn6Orq8vrrryvvmzZtiouLC0uWLOGbb74pVF6j0ZCdnY1KpdK67mWVdX1/RY8ePcLY2Pilr8/MzGT58uWsXr1a6/hf/XkVfLaGhoZ/qp6GDRuWUUSFBQUF0bZtW0JDQ7G2ti63doQQL0d6OIUQQoh/GCcnJ6ytrbl+/ToAgwYNon79+uzcuZMGDRpgaGjI9u3bixwCm5WVRXBwMHZ2dhgZGeHp6cmWLVu06i9tfTo6OgBMmDBBGfJ78OBBevTowRtvvFEo7sWLF2NkZMT9+/eLvbeQkBDc3NwwMjLC2tqatm3bKvcJ8PjxYz7++GNq1KiBoaEhDg4ODBo0SKuOzZs34+npiZGREXZ2dgQHB5OVlaWcL7iPHTt20LNnT1QqFb169QIgNTWVkSNHUq1aNQwNDWncuDE///zzcz+Tn376CYBOnTppHX92SO3zrFy5Eh0dHZKTk7WOe3p6at3nhQsX6Ny5M5aWllSuXJlatWrx5ZdfAk8+v1WrVnHhwgXlcym4trjP9sGDB4wePZpatWpRuXJlXFxcGD58OGlpaUqbLi4u3Lhxg0WLFin1rly5UjlXMKR25cqV6Ovrc+fOHa17uH//PpUqVWLp0qXKsWPHjtG6dWtMTEwwNzfn7bffJikpSeu6N998E0tLS9atW1fq5yiEeHUk4RRCCCH+YdLT07l37x52dnbKsYSEBMaMGcO4cePYvXs3np6eRV7bv39/li5dysSJE9m6dSt169alR48ebNu2Tatcaeo7duwYAO+9954y5LdRo0YMHTqUo0eP8scff2iVX7FiBW+99RZVqlQpMrYff/yRTz75hKCgIHbv3s3333+Pp6cn6enpSpkePXrw1VdfMXjwYHbs2MGcOXN48OCBcn7btm307NmTunXrsnXrViZOnMiSJUsIDAws1N6wYcNwdXVVhoNmZ2fTrl07IiIimDlzJtu2baNu3br4+flx7ty5ImMusG/fPho1aoSRkVGJ5cpKly5dSElJYfny5ezYsYPx48crz+GTTz6hc+fO1KhRQ/lcPvnkE+Xaoj7bhw8fkpeXx8yZM9m1axeff/45kZGRdO/eXbluy5Yt2Nra0rNnT6VePz+/QrG99dZb6Ovrs2nTJq3j//3vfwGU5P7YsWP4+vpibm7Oxo0b+e677zh16hTdunXTuq6gh3/v3r1l8uyEEGVMI4QQQoi/ralTp2pMTEw0OTk5mpycHM3169c1AQEBGkCze/dujUaj0QwcOFADaI4fP6517YEDBzSA5tSpUxqNRqM5c+aMBtAsWbJEq1zz5s01jRo1Ut6Xtj6NRqMBNHPmzNEql5eXp3FyctJMnDhROXbu3DkNoPn555+LvddRo0ZpxfGsn3/+WQNo1q1bV2yZhg0bapo3b651bOnSpRpAc/bsWa37GD58uFa5FStWaPT19TUXLlzQOt6sWTNNr169im1To9Fo3N3dNaNGjSp0/Nnn07JlS42fn5/yvuDzLfDDDz9oAM3du3e16mnQoIFm4MCBGo1Go7l7964G0Gzbtq3YeAYOHKipV69ekceL+myflZOTozl8+LAG0Pzxxx/KcWdn5yLv89njb731lsbb21urTKtWrbTu3cfHR+Pt7a3Jz89Xjl24cEGjo6Oj2bFjh9a1U6dO1VhZWZUYsxCiYkgPpxBCCPE39+DBAwwMDDAwMKB69eocOHCAb775hg4dOihlLC0tadasWYn1FKxsW9DDVKBPnz78/vvvWj2FpamvOLq6ugQFBfHjjz+Sm5sLPOnddHZ2pk2bNsVe16hRI37//XeCg4M5fPgwOTk5Wuf/97//UblyZfr27Vvk9ZmZmZw+fZqePXsWuj+Aw4cPax1/tnfu559/5rXXXsPd3Z3c3Fzl1a5dO06dOlXiPScmJr6y+YWWlpY4OzszefJkVq1axa1bt174+qI+29WrV9OwYUNMTU0xMDDgzTffBCA6OvqFY+zXrx/Hjh3j5s2bwJPnExkZSb9+/QB4+PAhR44coVevXuTl5SnP2t3dHUdHx0LP28rKiuTk5ELfCSFExZOEUwghhPibMzY25tSpU0RFRREbG0tycjKjRo3SKmNjY/PcelJSUjAwMCg0pNXGxgaNRkNqauoL1VeSwYMHc/fuXXbu3ElOTg5r1qxh0KBB6OoW/6PJoEGDmDdvHnv27KFFixZYW1szduxYHj16BMC9e/eoVq2aMnf0WampqWg0mkKxm5ubY2hoWGju6LPlkpOT+f3335XkvuD1+eefExcXV+L9ZmVl/emFd0pLR0eHn3/+mTp16jBq1CgcHR1p0qQJv/zyS6muL+qz3bJlCwMGDKBp06aEhYVx/PhxZW7v0/NfS8vf3x8TExM2bNgAQFhYGEZGRsoQ3ZSUFPLy8hg3blyh533z5s1Cz7vg2b5MLEKI8iWr1AohhBB/c7q6ujRp0qTEMsUlYU+rUqUKOTk5pKSkYGFhoRy/c+cOOjo6qNXqF6qvJA4ODnTs2JEVK1aQm5tLcnIy77zzTonX6OrqMnbsWMaOHUt8fDwbNmxg0qRJWFlZ8cknn2BpaUliYiIajabI+NRqNTo6OoUWnUlLS+Px48eFEu1n66hSpQoeHh4sX778he+3SpUqWgn7yyqYA5qdna11PCUlReu9u7s7mzZtIicnh6NHjzJlyhS6dOlCfHz8c/f1LOrZbdq0CU9PT60FfSIjI1/2NjA2NqZ79+5s2LCBiRMnsmHDBrp06YKJiQnwf5/VlClTtOaJFrCystJ6n5qaSqVKlTAzM3vpmIQQ5UN6OIUQQggBoAyRfHYxl02bNtGwYUMlGXgRBgYGxfY6DR06lB07dhAaGkqbNm1wdnYudb329vZ88MEHeHh4cOnSJQDatm3Lw4cPCQsLK/IaU1NTPD09lRVjCxSUL7j/4rRt25Zr165hZ2dHkyZNCr1KUqtWLa3VdF+Wg4MDgHLPBX8urofVwMCAli1bMmnSJNLT00lISACgUqVKL9Qb+OjRIypVqqR1bO3atYXKvUi9/fr14/fff2fPnj0cP35cGU4LYGJiQvPmzbl06VKRz9rFxUWrrtjYWNzd3Ut9P0KIV0d6OIUQQggBgIeHBwEBAQQHB/Po0SNq1arFmjVrOHr0KOHh4S9VZ506dQgPD6dFixaYmJhQq1YtpRfKz88Pa2trjh07xvr1659b17vvvouFhQWvv/46FhYWHDlyhDNnzjBy5EjgSULYuXNnBg8ezNWrV2nWrBn379/np59+YuPGjQBMmzaN7t27ExgYSGBgIH/88QdTpkyhR48evPbaayW2P2DAAJYuXYqvry/jx4/H3d2d1NRUfv/9d7Kzs5k9e3ax177xxhvFJsIvolmzZjg6OjJu3Dhmz55Neno6ISEhWFpaKmXOnj3LBx98QJ8+fXB1dSUtLY3Zs2fj4uKCq6sr8ORzWbFiBevXr6dmzZpYWVkVSuKe1q5dO0aNGsWMGTNo3rw5O3fu5H//+1+hcnXq1GH//v3s3bsXCwsLqlevrhXbs3VaWloyePBg1Gp1oS1j5syZQ+vWrenTpw99+/bFwsKCW7dusXfvXt555x18fX2VslFRUbRo0eIFnqQQ4lWRHk4hhBBCKNasWcPQoUMJCQmhW7dunDt3jp9++okuXbq8VH2LFi0iPz+fTp064eXlxa+//qqc09fXp0uXLlhYWPDWW289ty5vb28OHz5MUFAQHTt2ZO3atcybN4+goCClzH//+1/GjBnD0qVL6dSpE8HBwVpDSLt27cqmTZs4d+4c3bp1IyQkhGHDhrFmzZrntm9oaMj+/fvx9/dn5syZtG/fnpEjRxIVFfXc3tGePXty9epVYmJinttOSQwMDNiyZQtGRkb06tWL2bNn89VXX2Fvb6+UsbW1xdbWltmzZ9OpUyfeffddHB0d+fnnn9HT0wMgKCiIXr168d577+Hl5cW0adNKbPfdd9/lgw8+YOHChQQEBBAXF1fkvpezZs3CwcGBHj164OXlxfbt20u8l549e5KQkECPHj0K9aAWfN6ZmZm88847dO7cmc8++4zKlSvj5uamlEtKSuLXX38ttBiUEOKvQUej0WgqOgghhBBC/Pvk5+fj6uqKv78/CxcurOhwyl3jxo3p1q0bn376aUWH8o+yaNEi5s2bR0xMzJ+eWyyEKHvSwymEEEKIVyo7O5tTp04xefJk4uLiGD16dEWH9Ep8+umnLFmyhMePH1d0KP8Y+fn5fP3113z66aeSbArxFyVzOIUQQgjxSiUkJNC0aVOsra355ptvqFWrVkWH9Ep069aNmJgY4uLitIaEipeXkJDAoEGDCAwMrOhQhBDFkCG1QgghhBBCCCHKhQypFUIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLiThFEIIIYQQQghRLv4fkYRlcoY0vbwAAAAASUVORK5CYII=\n"
     },
     "metadata": {}
    }
   ],
   "source": "CBOM = pd.DataFrame([\n    # system, current crypto (typical), data horizon (yrs), external exposure 1-5, criticality 1-5\n    ('Member-bank query API (bureau ↔ bank)',      'TLS 1.2/1.3, ECDHE + RSA certs', 15, 5, 5),\n    ('Consumer portal & mobile app',               'TLS 1.3, ECDHE + ECDSA',          10, 5, 4),\n    ('E-government / public-registry integration', 'TLS + RSA signatures',            15, 4, 4),\n    ('PKI root & issuing CAs, HSMs',               'RSA-4096 / ECDSA roots',          20, 2, 5),\n    ('Archives & data-centre backups',             'AES-256 + RSA key wrap',          20, 2, 5),\n    ('Primary ↔ secondary DC replication',         'IPsec IKEv2 (ECDH)',              15, 3, 5),\n    ('Fintech / partner SaaS APIs',                'TLS + OAuth2 JWT (RS256)',        10, 5, 3),\n    ('Internal service mesh',                      'mTLS (ECDSA)',                     5, 1, 3),\n    ('Code signing & software updates',            'RSA-3072 signatures',             10, 3, 4),\n    ('Document & report e-signatures',             'RSA / ECDSA e-signature',         10, 4, 4),\n], columns=['System (illustrative)', 'Typical current crypto', 'Data horizon yrs', 'Exposure', 'Criticality'])\nCBOM['HNDL risk'] = np.clip(CBOM['Data horizon yrs'] / 4, 1, 5).round(1)\nCBOM['Priority score'] = (0.4 * CBOM['HNDL risk'] + 0.3 * CBOM['Exposure'] + 0.3 * CBOM['Criticality']).round(2)\nCBOM['Wave'] = pd.cut(CBOM['Priority score'], [0, 3.2, 4.0, 5], labels=['Wave 3 (2030–35)', 'Wave 2 (2028–30)', 'Wave 1 (2026–28)'])\nCBOM = CBOM.sort_values('Priority score', ascending=False).reset_index(drop=True); display(CBOM)\n\nfig, ax = plt.subplots(figsize=(8.5, 3.6))\nwc = {'Wave 1 (2026–28)': AMBER, 'Wave 2 (2028–30)': BLUE, 'Wave 3 (2030–35)': GRAY}\nax.barh(CBOM['System (illustrative)'][::-1], CBOM['Priority score'][::-1], color=[wc[w] for w in CBOM['Wave'][::-1]])\nfrom matplotlib.patches import Patch\nax.legend(handles=[Patch(color=c, label=w) for w, c in wc.items()], frameon=False, fontsize=8, loc='lower right')\nax.set_xlabel('Priority score (illustrative)'); ax.set_title('PQC migration waves — illustrative crypto inventory'); ax.tick_params(axis='y', labelsize=8)\nplt.tight_layout(); FIG_CBOM = savefig('08_cbom_waves'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "588e53fc",
   "metadata": {
    "id": "588e53fc"
   },
   "source": "## Stage 15 — Scale and hardware realities"
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8b19f65a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:55.419472Z",
     "iopub.status.busy": "2026-10-03T09:47:55.418963Z",
     "iopub.status.idle": "2026-10-03T09:47:56.354400Z",
     "shell.execute_reply": "2026-10-03T09:47:56.352495Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 380
    },
    "id": "8b19f65a",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022096915,
     "user_tz": -180,
     "elapsed": 1106,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "62126e41-832f-42ec-d683-531d51047747"
   },
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 935x374 with 1 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    }
   ],
   "source": "scale = [('This PoC\\n(feature selection)', N_CAND), ('This PoC\\n(quantum kernel)', N_QUBITS), ('Full attribute set\\n(~200 bureau vars)', 200),\n         ('Segment-level selection\\n(1,000 vars)', 1000), ('Portfolio-level QUBO\\n(100k vars)', 100_000)]\nfig, ax = plt.subplots(figsize=(8.5, 3.4))\nbars = ax.bar([s[0] for s in scale], [s[1] for s in scale], color=[BLUE, BLUE, NAVY, NAVY, AMBER], width=0.55)\nax.set_yscale('log'); ax.set_ylim(3, 1e6); ax.set_ylabel('Qubits / variables (log)')\nfor b, (l, v) in zip(bars, scale): ax.text(b.get_x() + b.get_width() / 2, v * 1.3, f'{v:,}', ha='center', fontsize=9, fontweight='bold')\nfor yv, c, t in [(25, TEAL, 'Laptop / Colab statevector (~25 qubits)'), (156, AMBER, 'IBM Heron (156 qubits)'), (4400, GRAY, 'D-Wave Advantage2 (4,400+) · hybrid solvers 1M+')]:\n    ax.axhline(yv, color=c, ls='--', lw=1.2, label=t)\nax.legend(frameon=False, fontsize=8, loc='upper left')\nax.tick_params(axis='x', labelsize=8); ax.set_title('Where today\\'s hardware fits')\nplt.tight_layout(); FIG_SCALE = savefig('09_scale'); plt.show()"
  },
  {
   "cell_type": "markdown",
   "id": "fb1573d1",
   "metadata": {
    "id": "fb1573d1"
   },
   "source": "## Stage 16 — Sovereign hybrid architecture for KKB\n```\n ┌──────────────── KKB (Türkiye, BDDK-regulated) ────────────────┐        ┌──── Off-shore (optional) ────┐\n │ Bureau data ──► Feature store ──► AI Safe Testing Environment  │        │ Cloud QPUs (IBM, IonQ,       │\n │                                    │  (Kloudeks, GPU-as-a-Svc) │  PQC   │ Quantinuum) / annealers      │\n │                                    ├─ Quantum simulators &     │◄─────►│ ONLY anonymised / synthetic  │\n │                                    │  quantum-inspired solvers  │ links │ features, never personal data │\n │                                    └─ Classical baselines       │        └──────────────────────────────┘\n │ Results ──► Model risk validation ──► Shadow scoring ──► Decision support (human in the loop) │\n └────────────────────────────────────────────────────────────────┘\n```\n- Personal credit data never leaves Türkiye; quantum workloads run on-shore as simulation / quantum-inspired compute (e.g., CUDA-Q on Kloudeks GPUs).\n- Real QPUs are accessed only with anonymised or synthetic feature matrices — the QUBO in Track 1 contains only correlation statistics, no personal data.\n- Every quantum result is benchmarked against a classical baseline before it is used."
  },
  {
   "cell_type": "markdown",
   "id": "ad03c563",
   "metadata": {
    "id": "ad03c563"
   },
   "source": "---\n## Stage 17 — Narrator Engine: executive Word report\nCollects every result above and writes an executive narrative (English by default, Turkish optional) with charts into a `.docx`. No number is typed by hand — change the parameters, re-run, and the report updates itself."
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "97c4bf1f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:56.357514Z",
     "iopub.status.busy": "2026-10-03T09:47:56.357250Z",
     "iopub.status.idle": "2026-10-03T09:47:58.782901Z",
     "shell.execute_reply": "2026-10-03T09:47:58.780710Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 160
    },
    "id": "97c4bf1f",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022102789,
     "user_tz": -180,
     "elapsed": 5869,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "d84fc082-21c6-4419-d593-c388c15c2d43"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "✓ Executive report written: KKB_Quantum_PoC_Executive_Report.docx (451 KB)\n",
      "• BVK Technology ran a three-track quantum proof of concept on a public credit-bureau stand-in dataset (1,000 applicants, 30% default rate), mirroring the decisions KKB makes every day at a scale of ~20 million queries per day.\n",
      "• Quantum feature selection built a 6-attribute scorecard from 20 attributes that retains 98% of the full scorecard's discriminatory power (test Gini 0.588 vs 0.603) with 30% of the attributes; on the hold-out set it beats naive top-IV selection by +0.035 Gini (in cross-validation the two are on par: AUC 0.782 vs 0.784).\n",
      "• A constraint-preserving QAOA (14 qubits, 3 layers) produced only valid scorecards and moved expected scorecard quality 74% of the way from random to optimal; classical local search then reached the exact optimum (gap 0.00%). A standard penalty-based QAOA placed only 12% of its probability on valid selections — circuit design, not qubit count, decides usefulness.\n",
      "• A 6-qubit quantum-kernel classifier is on par with classical models in cross-validation (AUC 0.791 vs best classical 0.782); on the hold-out set it reaches 0.783 versus 0.794 for the logistic scorecard, ahead of RBF-SVM and Gradient boosting. With only 30 training applicants it leads the best classical model by +0.020 AUC — the small-data niche the literature points to.\n",
      "• A credit report was sealed with hybrid X25519 + ML-KEM-768 encryption and an ML-DSA-65 signature (NIST FIPS 203/204); tampering was detected. At KKB scale the post-quantum overhead is at most ~110 GB/day — operationally affordable.\n",
      "• Under the regulatory planning horizon (CRQC by 2035) and a 5-year migration, 4 of 5 credit-data classes are already exposed to 'harvest now, decrypt later' — post-quantum migration should start now.\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ],
      "application/javascript": [
       "\n",
       "    async function download(id, filename, size) {\n",
       "      if (!google.colab.kernel.accessAllowed) {\n",
       "        return;\n",
       "      }\n",
       "      const div = document.createElement('div');\n",
       "      const label = document.createElement('label');\n",
       "      label.textContent = `Downloading \"${filename}\": `;\n",
       "      div.appendChild(label);\n",
       "      const progress = document.createElement('progress');\n",
       "      progress.max = size;\n",
       "      div.appendChild(progress);\n",
       "      document.body.appendChild(div);\n",
       "\n",
       "      const buffers = [];\n",
       "      let downloaded = 0;\n",
       "\n",
       "      const channel = await google.colab.kernel.comms.open(id);\n",
       "      // Send a message to notify the kernel that we're ready.\n",
       "      channel.send({})\n",
       "\n",
       "      for await (const message of channel.messages) {\n",
       "        // Send a message to notify the kernel that we're ready.\n",
       "        channel.send({})\n",
       "        if (message.buffers) {\n",
       "          for (const buffer of message.buffers) {\n",
       "            buffers.push(buffer);\n",
       "            downloaded += buffer.byteLength;\n",
       "            progress.value = downloaded;\n",
       "          }\n",
       "        }\n",
       "      }\n",
       "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
       "      const a = document.createElement('a');\n",
       "      a.href = window.URL.createObjectURL(blob);\n",
       "      a.download = filename;\n",
       "      div.appendChild(a);\n",
       "      a.click();\n",
       "      div.remove();\n",
       "    }\n",
       "  "
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ],
      "application/javascript": [
       "download(\"download_3bd4d82a-badb-4f24-bed8-32576e4d6794\", \"KKB_Quantum_PoC_Executive_Report.docx\", 462332)"
      ]
     },
     "metadata": {}
    }
   ],
   "source": "!pip install -q python-docx\nfrom docx import Document\nfrom docx.shared import Pt, Cm, RGBColor\nfrom docx.enum.text import WD_ALIGN_PARAGRAPH\nfrom docx.enum.table import WD_TABLE_ALIGNMENT\nfrom docx.oxml.ns import qn\nfrom docx.oxml import OxmlElement\n\nNARRATOR_LANG = 'en'      # 'en' | 'tr'\nREPORT_FILE = 'KKB_Quantum_PoC_Executive_Report.docx' if NARRATOR_LANG == 'en' else 'KKB_Kuantum_PoC_Yonetici_Raporu.docx'\n\ndef collect_results():\n    sc = SC.set_index('Scorecard'); qk = QK.set_index('Model')\n    qsvm_row = QK.iloc[0]; best_cls = QK.iloc[1:4].sort_values('AUC', ascending=False).iloc[0]\n    return {\n        'date': datetime.date.today().strftime('%d %B %Y'), 'data_source': DATA_SOURCE, 'n': int(len(raw_df)), 'default_rate': float(y.mean()),\n        'baseline': baseline.round(4).to_dict('records'),\n        'fs': {'n_cand': N_CAND, 'k': K_SELECT, 'selected': SELECTED, 'top_iv': TOP_IV,\n               'exact_obj': float(OBJ[I_OPT]), 'hybrid_obj': objective(x_hybrid), 'sa_obj': objective(x_sa),\n               'hybrid_gap_pct': float(gap(objective(x_hybrid))), 'top32_gap_pct': float(gap(OBJ[i_top_xy])), 'quality_xy': float(QUALITY_XY),\n               'p_feas_std': P_FEAS_STD, 'p_feas_uniform': float(FEAS.mean()), 'qaoa_reps': QAOA_REPS,\n               'qaoa_iters': len(history), 'qaoa_time': t_qaoa,\n               'gini_all': float(sc.iloc[0]['Gini']), 'gini_qubo': float(sc.iloc[1]['Gini']), 'gini_top': float(sc.iloc[2]['Gini']), 'gini_rand': float(sc.iloc[3]['Gini']),\n               'cv_qubo': float(sc.iloc[1]['CV AUC']), 'cv_top': float(sc.iloc[2]['CV AUC']),\n               'red_qubo': mean_red(SELECTED), 'red_top': mean_red(TOP_IV)},\n        'qk': {'qubits': N_QUBITS, 'bw': BW, 'C': C_BEST, 'auc': float(qsvm_row['AUC']), 'gini': float(qsvm_row['Gini']), 'ks': float(qsvm_row['KS']), 'cv': float(qsvm_row['CV AUC']),\n               'best_cls_name': best_cls['Model'], 'best_cls_auc': float(best_cls['AUC']), 'best_cls_cv': float(QK.iloc[1:4]['CV AUC'].max()),\n               'beaten': [m for m, v in zip(QK.iloc[1:4]['Model'], QK.iloc[1:4]['AUC']) if v < float(qsvm_row['AUC'])],\n               'small_gap': SMALL_GAP, 'small_n': min(SMALL_GAP)},\n        'pqc': {'tamper': TAMPER_DETECTED, 'overhead': PQ_OVERHEAD, 'gb_day': GB_DAY, 'extra_per_query': extra_per_query,\n                'at_risk_2035': AT_RISK_2035, 'n_classes': len(DATA_CLASSES), 'migration': MIGRATION_YEARS,\n                'wave1': CBOM[CBOM['Wave'] == 'Wave 1 (2026–28)']['System (illustrative)'].tolist()},\n    }\nR = collect_results()\n\ndef narrate(R, lang='en'):\n    fs, qk, pq = R['fs'], R['qk'], R['pqc']\n    T = (lambda en, tr: en) if lang == 'en' else (lambda en, tr: tr)\n    gini_delta = fs['gini_qubo'] - fs['gini_all']\n    small_best = max(qk['small_gap'].items(), key=lambda kv: kv[1])\n    qk_vs = qk['auc'] - qk['best_cls_auc']\n    S = {}\n    S['title'] = T('Quantum Readiness & Advantage for KKB', 'KKB için Kuantum Hazırlığı ve Avantajı')\n    S['subtitle'] = T('Proof-of-concept results — hybrid quantum analytics and quantum-safe credit data', 'PoC sonuçları — hibrit kuantum analitiği ve kuantum-güvenli kredi verisi')\n    S['summary'] = [\n        T(f\"BVK Technology ran a three-track quantum proof of concept on a public credit-bureau stand-in dataset ({R['n']:,} applicants, {R['default_rate']:.0%} default rate), mirroring the decisions KKB makes every day at a scale of ~20 million queries per day.\",\n          f\"BVK Technology, kamuya açık bir kredi bürosu veri seti üzerinde ({R['n']:,} başvuru, %{R['default_rate']*100:.0f} temerrüt oranı) üç ayaklı bir kuantum PoC çalıştırdı; bu, KKB'nin günde ~20 milyon sorguyla verdiği kararların küçük ölçekli bir yansımasıdır.\"),\n        T(f\"Quantum feature selection built a {fs['k']}-attribute scorecard from 20 attributes that retains {100*fs['gini_qubo']/fs['gini_all']:.0f}% of the full scorecard's discriminatory power (test Gini {fs['gini_qubo']:.3f} vs {fs['gini_all']:.3f}) with {100*fs['k']/20:.0f}% of the attributes; on the hold-out set it beats naive top-IV selection by {fs['gini_qubo']-fs['gini_top']:+.3f} Gini (in cross-validation the two are on par: AUC {fs['cv_qubo']:.3f} vs {fs['cv_top']:.3f}).\",\n          f\"Kuantum öznitelik seçimi 20 değişkenden {fs['k']} değişkenli bir skor kartı kurdu: değişkenlerin %{100*fs['k']/20:.0f}'iyle tam skor kartının ayırt ediciliğinin %{100*fs['gini_qubo']/fs['gini_all']:.0f}'ini koruyor (test Gini {fs['gini_qubo']:.3f} / {fs['gini_all']:.3f}); test setinde naif IV seçimini {fs['gini_qubo']-fs['gini_top']:+.3f} Gini ile geçiyor (çapraz doğrulamada başa baş: AUC {fs['cv_qubo']:.3f} / {fs['cv_top']:.3f}).\"),\n        T(f\"A constraint-preserving QAOA ({fs['n_cand']} qubits, {fs['qaoa_reps']} layers) produced only valid scorecards and moved expected scorecard quality {fs['quality_xy']:.0f}% of the way from random to optimal; classical local search then reached the exact optimum (gap {fs['hybrid_gap_pct']:.2f}%). A standard penalty-based QAOA placed only {fs['p_feas_std']:.0%} of its probability on valid selections — circuit design, not qubit count, decides usefulness.\",\n          f\"Kısıt-koruyan QAOA ({fs['n_cand']} kübit, {fs['qaoa_reps']} katman) yalnızca geçerli skor kartları üretti ve beklenen kaliteyi rastgeleden optimuma giden yolun %{fs['quality_xy']:.0f}'i kadar ilerletti; klasik yerel arama kesin optimuma ulaştı (sapma %{fs['hybrid_gap_pct']:.2f}). Standart cezalı QAOA olasılığının yalnızca %{fs['p_feas_std']*100:.0f}'ini geçerli seçimlere verdi — faydayı kübit sayısı değil devre tasarımı belirliyor.\"),\n        T(f\"A {qk['qubits']}-qubit quantum-kernel classifier is on par with classical models in cross-validation (AUC {qk['cv']:.3f} vs best classical {qk['best_cls_cv']:.3f}); on the hold-out set it reaches {qk['auc']:.3f} versus {qk['best_cls_auc']:.3f} for the {qk['best_cls_name'].lower()}{(', ahead of ' + ' and '.join(qk['beaten'])) if qk['beaten'] else ''}. With only {qk['small_n']} training applicants it {'leads' if qk['small_gap'][qk['small_n']] > 0 else 'trails'} the best classical model by {qk['small_gap'][qk['small_n']]:+.3f} AUC — the small-data niche the literature points to.\",\n          f\"{qk['qubits']} kübitlik kuantum çekirdek sınıflandırıcısı çapraz doğrulamada klasik modellerle başa baş (AUC {qk['cv']:.3f}; en iyi klasik {qk['best_cls_cv']:.3f}); test setinde {qk['auc']:.3f}, lojistik skor kartı {qk['best_cls_auc']:.3f}{(' — ' + ' ve '.join(qk['beaten']) + ' modellerinin önünde') if qk['beaten'] else ''}. Yalnızca {qk['small_n']} başvuruyla eğitildiğinde en iyi klasiğe göre {qk['small_gap'][qk['small_n']]:+.3f} AUC fark — literatürün işaret ettiği az veri nişi.\"),\n        T(f\"A credit report was sealed with hybrid X25519 + ML-KEM-768 encryption and an ML-DSA-65 signature (NIST FIPS 203/204); tampering was detected. At KKB scale the post-quantum overhead is at most ~{pq['gb_day']:,.0f} GB/day — operationally affordable.\",\n          f\"Bir kredi raporu hibrit X25519 + ML-KEM-768 şifreleme ve ML-DSA-65 imzasıyla (NIST FIPS 203/204) korundu; değişiklik girişimi tespit edildi. KKB ölçeğinde kuantum-sonrası ek yük en fazla ~{pq['gb_day']:,.0f} GB/gün — operasyonel olarak karşılanabilir.\"),\n        T(f\"Under the regulatory planning horizon (CRQC by 2035) and a {pq['migration']}-year migration, {pq['at_risk_2035']} of {pq['n_classes']} credit-data classes are already exposed to 'harvest now, decrypt later' — post-quantum migration should start now.\",\n          f\"Düzenleyici planlama ufkunda (2035'te CRQC) ve {pq['migration']} yıllık geçişle {pq['n_classes']} kredi verisi sınıfından {pq['at_risk_2035']}'i 'şimdi topla, sonra çöz' riskine şimdiden açık — kuantum-sonrası geçiş şimdi başlamalı.\"),\n    ]\n    S['sections'] = [\n        (T('Track 1 — Quantum feature selection', 'Ayak 1 — Kuantum öznitelik seçimi'), [\n            T(f\"Problem: choose {fs['k']} of the {fs['n_cand']} strongest attributes maximising Information Value while penalising redundancy — a {fs['n_cand']}-qubit QUBO.\",\n              f\"Problem: en güçlü {fs['n_cand']} değişkenden, Information Value'yu en çoklayıp tekrarı cezalandıran {fs['k']} tanesini seçmek — {fs['n_cand']} kübitlik QUBO.\"),\n            T(f\"Constraint-preserving QAOA ({fs['qaoa_reps']} layers, {fs['qaoa_iters']} optimiser iterations, {fs['qaoa_time']:.0f} s on a CPU simulator) generated only valid selections; the best of its 32 most likely outcomes was within {fs['top32_gap_pct']:.1f}% of the optimum and classical local search closed the gap. Exact enumeration and classical annealing were run on the same objective as ground truth.\",\n              f\"Kısıt-koruyan QAOA ({fs['qaoa_reps']} katman, {fs['qaoa_iters']} optimizasyon iterasyonu, CPU simülatöründe {fs['qaoa_time']:.0f} s) yalnızca geçerli seçimler üretti; en olası 32 sonucun en iyisi optimuma %{fs['top32_gap_pct']:.1f} yakındı ve klasik yerel arama farkı kapattı. Kesin sayım ve klasik tavlama aynı amaç fonksiyonunda referans olarak çalıştırıldı.\"),\n            T(f\"Selected attributes: {', '.join(fs['selected'])}.\", f\"Seçilen değişkenler: {', '.join(fs['selected'])}.\"),\n            T(f\"Business meaning: a smaller scorecard with near-identical discrimination is cheaper to maintain, easier to explain to regulators and members, and less exposed to data-quality drift.\",\n              f\"İş anlamı: neredeyse aynı ayırt edicilikte daha küçük bir skor kartı; bakımı daha ucuz, düzenleyiciye ve üyelere açıklaması daha kolay, veri kalitesi kaymasına daha az açık.\"),\n        ], ['scorecard', 'qaoa']),\n        (T('Track 2 — Quantum-kernel credit scoring', 'Ayak 2 — Kuantum çekirdekli kredi skorlama'), [\n            T(f\"Applicants were encoded into {qk['qubits']}-qubit states (ZZ feature map, bandwidth {qk['bw']}); kernel concentration was measured and controlled — a known failure mode in the literature.\",\n              f\"Başvurular {qk['qubits']} kübitlik durumlara kodlandı (ZZ özellik haritası, bant genişliği {qk['bw']}); literatürde bilinen bir hata modu olan çekirdek yoğunlaşması ölçülüp kontrol edildi.\"),\n            T(f\"Test performance: AUC {qk['auc']:.3f}, Gini {qk['gini']:.3f}, KS {qk['ks']:.3f}. Best classical on the same attributes: {qk['best_cls_name']} (AUC {qk['best_cls_auc']:.3f}).\",\n              f\"Test başarımı: AUC {qk['auc']:.3f}, Gini {qk['gini']:.3f}, KS {qk['ks']:.3f}. Aynı değişkenlerde en iyi klasik: {qk['best_cls_name']} (AUC {qk['best_cls_auc']:.3f}).\"),\n            T(f\"Important caveat: performance peaked at the smallest bandwidth tested ({qk['bw']}), where the quantum kernel behaves almost like a classical one (Shaydulin & Wild, 2022). The result demonstrates methodological readiness and a small-data niche — not a quantum advantage on this data.\",\n              f\"Önemli not: başarım test edilen en küçük bant genişliğinde ({qk['bw']}) zirve yaptı; bu bölgede kuantum çekirdek neredeyse klasik bir çekirdek gibi davranır (Shaydulin & Wild, 2022). Sonuç, metodolojik hazırlığı ve az veri nişini gösterir — bu veride kuantum avantajı değil.\"),\n            T(\"Interpretation: quantum kernels are a credible research direction for thin-file, SME and agricultural segments where data is scarce; they are not yet a replacement for production scorecards.\",\n              \"Yorum: kuantum çekirdekler, verinin kıt olduğu ince dosyalı, KOBİ ve tarım segmentleri için güvenilir bir araştırma yönüdür; henüz üretim skor kartlarının yerini almaz.\"),\n        ], ['roc', 'small']),\n        (T('Track 3 — Quantum-safe credit report', 'Ayak 3 — Kuantum-güvenli kredi raporu'), [\n            T(f\"Hybrid encryption (X25519 + ML-KEM-768 → HKDF → AES-256-GCM) and ML-DSA-65 signatures protected a synthetic credit report end-to-end; per-message overhead {pq['overhead']:,} bytes.\",\n              f\"Hibrit şifreleme (X25519 + ML-KEM-768 → HKDF → AES-256-GCM) ve ML-DSA-65 imzası sentetik bir kredi raporunu uçtan uca korudu; mesaj başına ek yük {pq['overhead']:,} bayt.\"),\n            T(f\"Illustrative inventory: first-wave systems are {', '.join(pq['wave1'])}.\", f\"Örnek envanter: ilk dalga sistemleri {', '.join(pq['wave1'])}.\"),\n            T(\"Regulatory context: NIST (2030/2035), the EU roadmap (critical systems by 2030), UK NCSC (2028/2031/2035) and the G7 financial-sector roadmap all start with a cryptographic inventory.\",\n              \"Düzenleyici bağlam: NIST (2030/2035), AB yol haritası (kritik sistemler 2030), UK NCSC (2028/2031/2035) ve G7 finans yol haritasının hepsi kriptografik envanterle başlar.\"),\n        ], ['mosca', 'cbom']),\n    ]\n    S['next'] = [\n        T('Run Track 1–2 on KKB sandbox data inside the AI Safe Testing Environment (Kloudeks), with KKB model-risk validation.', 'Ayak 1–2’yi KKB sandbox verisiyle, Kloudeks üzerindeki Yapay Zekâ Güvenli Test Ortamı’nda ve KKB model risk doğrulamasıyla çalıştırmak.'),\n        T('Start a cryptographic inventory (CBOM) and a hybrid-TLS pilot on one member-bank query channel.', 'Kriptografik envanteri (CBOM) başlatmak ve bir üye banka sorgu kanalında hibrit TLS pilotu yapmak.'),\n        T('Benchmark the feature-selection QUBO on a real QPU and on quantum-inspired GPU solvers using anonymised statistics only.', 'Öznitelik seçimi QUBO’sunu yalnızca anonim istatistiklerle gerçek QPU ve kuantum-esinli GPU çözücülerde kıyaslamak.'),\n        T('Form a joint KKB–BVK quantum working group with academic partners (İTÜ, TOBB ETÜ) and align with BDDK expectations.', 'Akademik ortaklarla (İTÜ, TOBB ETÜ) ortak bir KKB–BVK kuantum çalışma grubu kurmak ve BDDK beklentileriyle hizalanmak.'),\n    ]\n    S['closing'] = T('BVK Technology designs, builds and integrates quantum, quantum-inspired and quantum-safe solutions — from the first workshop to production. Every number in this report was generated by the accompanying notebook.',\n                     'BVK Technology kuantum, kuantum-esinli ve kuantum-güvenli çözümleri ilk çalıştaydan canlı ortama kadar tasarlar, geliştirir ve entegre eder. Bu rapordaki tüm rakamlar ekteki not defteri tarafından üretilmiştir.')\n    return S\n\nFIGS = {'scorecard': FIG_SC, 'qaoa': FIG_QAOA, 'roc': FIG_ROC, 'small': FIG_SMALL, 'mosca': FIG_MOSCA, 'cbom': FIG_CBOM}\nNAV = RGBColor(0x18, 0x2B, 0x5E); MUT = RGBColor(0x7A, 0x87, 0x9C); AMB = RGBColor(0xC9, 0x8A, 0x1A)\n\ndef build_report(R, S, path, lang='en'):\n    T = (lambda en, tr: en) if lang == 'en' else (lambda en, tr: tr)\n    doc = Document(); st = doc.styles['Normal']; st.font.name = 'Calibri'; st.font.size = Pt(10.5)\n    for sec in doc.sections:\n        sec.top_margin = sec.bottom_margin = Cm(2.0); sec.left_margin = sec.right_margin = Cm(2.2)\n        hp = sec.header.paragraphs[0]; hp.text = 'BVK Technology — Quantum Business Solutions  |  ' + T('CONFIDENTIAL', 'GİZLİ'); hp.alignment = WD_ALIGN_PARAGRAPH.RIGHT\n        hp.runs[0].font.size = Pt(8); hp.runs[0].font.color.rgb = MUT\n    def H(t, lvl=1):\n        p = doc.add_heading(level=lvl); r = p.add_run(t); r.font.color.rgb = NAV; r.font.name = 'Calibri'; r.font.size = Pt(16 if lvl == 1 else 12.5)\n    def Pp(t, it=False, sz=10.5, col=None):\n        p = doc.add_paragraph(); r = p.add_run(t); r.italic = it; r.font.size = Pt(sz)\n        if col: r.font.color.rgb = col\n        p.paragraph_format.space_after = Pt(5)\n    def Bl(t): p = doc.add_paragraph(style='List Bullet'); p.add_run(t); p.paragraph_format.space_after = Pt(3)\n    def IMG(p, w=14):\n        doc.add_picture(p, width=Cm(w)); doc.paragraphs[-1].alignment = WD_ALIGN_PARAGRAPH.CENTER\n    def TABLE(df, widths):\n        t = doc.add_table(rows=len(df) + 1, cols=len(df.columns)); t.style = 'Table Grid'; t.alignment = WD_TABLE_ALIGNMENT.CENTER; t.autofit = False\n        for j, w in enumerate(widths): t.columns[j].width = Cm(w)\n        for j, c in enumerate(df.columns):\n            cell = t.cell(0, j); cell.text = ''; rr = cell.paragraphs[0].add_run(str(c)); rr.bold = True; rr.font.size = Pt(9)\n            sh = OxmlElement('w:shd'); sh.set(qn('w:val'), 'clear'); sh.set(qn('w:color'), 'auto'); sh.set(qn('w:fill'), 'EEF4FB'); cell._tc.get_or_add_tcPr().append(sh)\n        for i, row in enumerate(df.itertuples(index=False), 1):\n            for j, v in enumerate(row):\n                cell = t.cell(i, j); cell.text = ''; rr = cell.paragraphs[0].add_run(f'{v:.3f}' if isinstance(v, float) else str(v)); rr.font.size = Pt(9)\n        doc.add_paragraph()\n\n    p = doc.add_paragraph(); p.paragraph_format.space_before = Pt(80)\n    r = p.add_run(T('QUANTUM SOLUTIONS  |  PROOF-OF-CONCEPT REPORT', 'KUANTUM ÇÖZÜMLERİ  |  PoC RAPORU')); r.bold = True; r.font.size = Pt(10); r.font.color.rgb = AMB\n    p = doc.add_paragraph(); r = p.add_run(S['title']); r.bold = True; r.font.size = Pt(26); r.font.color.rgb = NAV\n    Pp(S['subtitle'], sz=13)\n    Pp(f\"{R['date']} · {T('Data', 'Veri')}: {R['data_source']} · Qiskit {qiskit.__version__}\", sz=9.5, col=MUT)\n    Pp(T('Generated automatically by the Narrator Engine from the results of this notebook run.', 'Bu not defteri çalıştırmasının sonuçlarından Narrator Engine tarafından otomatik üretilmiştir.'), it=True, sz=9, col=MUT)\n    doc.add_page_break()\n    H(T('1. Executive summary', '1. Yönetici özeti'))\n    for t in S['summary']: Bl(t)\n    TABLE(SC[['Scorecard', '# attributes', 'AUC', 'Gini', 'KS']], [6.5, 2.4, 2.2, 2.2, 2.2])\n    for i, (title, paras, figs) in enumerate(S['sections'], 2):\n        H(f'{i}. {title}')\n        for t in paras: Pp(t)\n        for f in figs: IMG(FIGS[f], 13.5)\n        if i == 3: TABLE(QK[['Model', 'AUC', 'Gini', 'KS']], [7.5, 2.4, 2.4, 2.4])\n        if i == 4: TABLE(PQC[['Scheme', 'Quantum-safe', 'Public key B', 'Ciphertext / signature B']], [6, 2.6, 3, 3.6])\n    H(T('5. Recommended next steps', '5. Önerilen sonraki adımlar'))\n    for t in S['next']: Bl(t)\n    Pp(S['closing'], it=True)\n    Pp('BVK Technology · İTÜ ARI Teknokent, Maslak, İstanbul · bvktechnologysales@gmail.com · www.bvktechnology.com', sz=9, col=MUT)\n    doc.save(path); return path\n\nNARRATIVE = narrate(R, NARRATOR_LANG)\nout = build_report(R, NARRATIVE, REPORT_FILE, NARRATOR_LANG)\nprint('✓ Executive report written:', out, f'({os.path.getsize(out)/1024:.0f} KB)')\nfor t in NARRATIVE['summary']: print('•', t)\ntry:\n    from google.colab import files; files.download(out)\nexcept Exception:\n    print('(not running in Colab — file saved to the working directory)')\nwith open('kkb_poc_results.json', 'w') as f: json.dump(R, f, indent=1, default=str, ensure_ascii=False)"
  },
  {
   "cell_type": "markdown",
   "id": "cbd86da9",
   "metadata": {
    "id": "cbd86da9"
   },
   "source": "## Stage 18 — Summary"
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d588dede",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-03T09:47:58.786632Z",
     "iopub.status.busy": "2026-10-03T09:47:58.786049Z",
     "iopub.status.idle": "2026-10-03T09:47:58.796086Z",
     "shell.execute_reply": "2026-10-03T09:47:58.794127Z"
    },
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d588dede",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1791022102827,
     "user_tz": -180,
     "elapsed": 8,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "24f22fba-dc4c-4922-cc3f-973ebc44dd19"
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "\n",
      "══════════════════════════════════════════════════════════════════════════════\n",
      " KKB QUANTUM PoC — SUMMARY\n",
      "══════════════════════════════════════════════════════════════════════════════\n",
      " Data           : UCI Statlog German Credit (OpenML credit-g) · 1,000 applicants\n",
      " Classical bar  : logistic scorecard (20 attrs) Gini 0.603\n",
      " Track 1        : 14-qubit QUBO → 6 attributes · Gini 0.588 · XY-QAOA quality 74% · hybrid gap 0.00%\n",
      " Track 2        : 6-qubit quantum kernel · AUC 0.783 (best classical 0.794)\n",
      " Track 3        : hybrid X25519+ML-KEM-768 / ML-DSA-65 · tamper detected: True · ≤110 GB/day at KKB scale\n",
      " Mosca (2035)   : 4/5 data classes exposed if migration starts 2026\n",
      " Report         : KKB_Quantum_PoC_Executive_Report.docx\n",
      "══════════════════════════════════════════════════════════════════════════════\n",
      "\n"
     ]
    }
   ],
   "source": "print(f'''\n══════════════════════════════════════════════════════════════════════════════\n KKB QUANTUM PoC — SUMMARY\n══════════════════════════════════════════════════════════════════════════════\n Data           : {DATA_SOURCE} · {len(raw_df):,} applicants\n Classical bar  : logistic scorecard (20 attrs) Gini {baseline.iloc[0][\"Gini\"]:.3f}\n Track 1        : {N_CAND}-qubit QUBO → {K_SELECT} attributes · Gini {SC.iloc[1][\"Gini\"]:.3f} · XY-QAOA quality {QUALITY_XY:.0f}% · hybrid gap {gap(objective(x_hybrid)):.2f}%\n Track 2        : {N_QUBITS}-qubit quantum kernel · AUC {QK.iloc[0][\"AUC\"]:.3f} (best classical {QK.iloc[1:4][\"AUC\"].max():.3f})\n Track 3        : hybrid X25519+ML-KEM-768 / ML-DSA-65 · tamper detected: {TAMPER_DETECTED} · ≤{GB_DAY:,.0f} GB/day at KKB scale\n Mosca (2035)   : {AT_RISK_2035}/{len(DATA_CLASSES)} data classes exposed if migration starts {THIS_YEAR}\n Report         : {REPORT_FILE}\n══════════════════════════════════════════════════════════════════════════════\n''')"
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.11.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}