Heart_Disease_Prediction_Using_Deep_Learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "id": "UvRbMznhUMv5"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import r2_score,accuracy_score,confusion_matrix\n",
    "from sklearn import metrics\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "from sklearn.preprocessing import LabelEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "id": "aRRNP-ADUi5o"
   },
   "outputs": [],
   "source": [
    "data=pd.read_csv('Heart_disease.csv',encoding='latin1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "0m26_nR5VKML",
    "outputId": "38ce3ed9-0961-40f0-a7aa-85e40bb6e993",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Heart rate</th>\n",
       "      <th>Systolic blood pressure</th>\n",
       "      <th>Diastolic blood pressure</th>\n",
       "      <th>Blood sugar</th>\n",
       "      <th>CK-MB</th>\n",
       "      <th>Troponin</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>160</td>\n",
       "      <td>83</td>\n",
       "      <td>160.0</td>\n",
       "      <td>1.80</td>\n",
       "      <td>0.012</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>94</td>\n",
       "      <td>98</td>\n",
       "      <td>46</td>\n",
       "      <td>296.0</td>\n",
       "      <td>6.75</td>\n",
       "      <td>1.060</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>160</td>\n",
       "      <td>77</td>\n",
       "      <td>270.0</td>\n",
       "      <td>1.99</td>\n",
       "      <td>0.003</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>66</td>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "      <td>120</td>\n",
       "      <td>55</td>\n",
       "      <td>270.0</td>\n",
       "      <td>13.87</td>\n",
       "      <td>0.122</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>112</td>\n",
       "      <td>65</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.08</td>\n",
       "      <td>0.003</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Gender  Heart rate  Systolic blood pressure  Diastolic blood pressure  \\\n",
       "0   63       1          66                      160                        83   \n",
       "1   20       1          94                       98                        46   \n",
       "2   56       1          64                      160                        77   \n",
       "3   66       1          70                      120                        55   \n",
       "4   54       1          64                      112                        65   \n",
       "\n",
       "   Blood sugar  CK-MB  Troponin    Result  \n",
       "0        160.0   1.80     0.012  negative  \n",
       "1        296.0   6.75     1.060  positive  \n",
       "2        270.0   1.99     0.003  negative  \n",
       "3        270.0  13.87     0.122  positive  \n",
       "4        300.0   1.08     0.003  negative  "
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Dos_2LadUi74",
    "outputId": "614d1b95-b785-4f5e-ec52-146a373bacff"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1319, 9)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "AtSCUwENUi-R",
    "outputId": "87e87919-7188-4120-b58c-b6bcfef56420"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age                         0\n",
       "Gender                      0\n",
       "Heart rate                  0\n",
       "Systolic blood pressure     0\n",
       "Diastolic blood pressure    0\n",
       "Blood sugar                 0\n",
       "CK-MB                       0\n",
       "Troponin                    0\n",
       "Result                      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "id": "XedQP7WGZpTT"
   },
   "outputs": [],
   "source": [
    "lb = LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "id": "ujYVC5_1Zvp0"
   },
   "outputs": [],
   "source": [
    "data['Result'] = lb.fit_transform(data['Result'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aJxtDtKlZvtQ",
    "outputId": "f153c3f4-b2f5-4472-94c1-5f27f751811f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1])"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Result'] .unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dPUVC9b1n3On",
    "outputId": "8961eefb-d1f4-4853-cacb-71572aa0fdcb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Result\n",
       "1    810\n",
       "0    509\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Result'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KauHKmZ3WB3g"
   },
   "source": [
    "# **Analyzing The Data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "gIuWcpWJWTHD",
    "outputId": "8700708f-0559-4db6-b4c8-ed888f4e29a1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Gender', 'Result']"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Discrete_features=[feature for feature in data.columns if (data[feature].dtypes !='O') and (len(data[feature].unique()) < 25 ) ]\n",
    "Discrete_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Wf9vvCr1X-qj",
    "outputId": "9deb6450-37b6-40ec-a6ad-bc7d15d7dc7d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Age',\n",
       " 'Gender',\n",
       " 'Heart rate',\n",
       " 'Systolic blood pressure',\n",
       " 'Diastolic blood pressure',\n",
       " 'Blood sugar',\n",
       " 'CK-MB',\n",
       " 'Troponin',\n",
       " 'Result']"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numerical_features=[features for features in data.columns if data[features].dtypes != \"O\"]\n",
    "numerical_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_abX8kbOWTJC",
    "outputId": "8066f9bb-dd4b-4cae-9654-70cdf4d72225"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Age',\n",
       " 'Heart rate',\n",
       " 'Systolic blood pressure',\n",
       " 'Diastolic blood pressure',\n",
       " 'Blood sugar',\n",
       " 'CK-MB',\n",
       " 'Troponin']"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "continous_features=[feature for feature in numerical_features if feature not in Discrete_features]\n",
    "continous_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZkgWP3dmaSLX"
   },
   "source": [
    "So, from this observation most of the features have the continous values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Oxhy4PXDY5-O"
   },
   "source": [
    "### Checking The Relationship Between the Discrete Feature and Dependent Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 467
    },
    "id": "JkSNOb1yWTMk",
    "outputId": "84d85f36-a23b-438b-ad75-f6ff49a0319c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data=data.copy()\n",
    "data.groupby('Gender')['Result'].median().plot.bar()\n",
    "plt.xlabel('Gender')\n",
    "plt.ylabel('Result')\n",
    "plt.title('Gender')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tRaVhbghakAv"
   },
   "source": [
    "### Plotting the continous_features in Histogram to check if it is skewed data or not"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "zecdXxqWUjAn",
    "outputId": "fe22478a-09d7-4a26-eee7-b13597e24950"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHFCAYAAAAUpjivAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAvC0lEQVR4nO3de1TVdb7/8dcWNptLyHhlg5LiZEsLKtNuZmmp2MWpjjNZqWW/bAbzSth4GadEK1DPyTxHy7LVqCdzdE1lWXkasJIyLI3UvJTdyEvBUImiQoDw+f3Rcjc7vGy2X9n08flYi7Xan+/7+/2+v+8wX333zWWMMQIAALBUs1A3AAAAcDoRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AARt8eLFcrlc+vDDD4+5feDAgerYsWPjNvVvVq9eraysLOvPCeDECDsArLV69WpNnz7d+nMCODHCDgDrVFRUNMljAQgNwg6ARmWM0ZNPPqmLLrpIUVFRatGihf7whz/oq6++8qvLy8vTzTffrPbt2ysyMlLnnHOO0tPT9f333/vVZWVlyeVy6aOPPtIf/vAHtWjRQr/97W91991364knnpAkuVwu38/XX3993N769OmjlJQUvfPOO+rZs6eio6N1zz33SJJWrFihtLQ0JSQkKCoqSl27dtXkyZN1+PBh3/4nO2eg1w7AWeGhbgDAr19tba2OHDlSb90YU28tPT1dixcv1rhx4zRr1izt27dPM2bMUM+ePbVlyxbFx8dLkr788ktdccUVuvfeexUXF6evv/5ac+bMUa9evbR161a53W6/4w4aNEi33367Ro4cqcOHDyslJUWHDx/WCy+8oPXr1/vqEhISTngtxcXFGjZsmCZOnKjs7Gw1a/bT/xN+/vnnuuGGG5SRkaGYmBh9+umnmjVrljZs2KC33npLkvTggw+e8JyBXjsAhxkACNKiRYuMpBP+dOjQwVe/fv16I8k89thjfsfZs2ePiYqKMhMnTjzmeerq6kxNTY3ZtWuXkWReeeUV37Zp06YZSeahhx6qt9/o0aNNQ/4z17t3byPJvPnmmyesO9pPfn6+kWS2bNly0nMGe+0ATh1PYwE4Zf/7v/+rjRs31vvp1auXX91rr70ml8ulYcOG6ciRI74fr9erCy+8UGvXrvXVlpaWauTIkUpKSlJ4eLjcbrc6dOggSfrkk0/q9fD73//ekWtp0aKFrr322nrrX331lYYMGSKv16uwsDC53W717t37uP38UkOuHYCzeBoLwCnr2rWrevToUW89Li5Oe/bs8T3+17/+JWPMcZ+u6dSpkySprq5OaWlp+vbbb/Xggw8qNTVVMTExqqur0+WXX67Kysp6+57s6alAHes4hw4d0lVXXaXIyEg98sgjOvfccxUdHa09e/Zo0KBBx+znlwK9dgDOI+wAaDStW7eWy+XSu+++K4/HU2/70bVt27Zpy5YtWrx4sYYPH+7b/sUXXxz32C6Xy5Eej3Wct956S99++63Wrl3ru5sjSfv37w/4uIFeOwDnEXYANJqBAwdq5syZ+uabbzR48ODj1h0NHL8MAE8//XSDznd0/8rKSkVFRTWw2+D6Od45A712AM4j7ABoNFdeeaX+9Kc/6f/9v/+nDz/8UFdffbViYmJUXFysdevWKTU1Vffdd5+6dOmi3/72t5o8ebKMMWrZsqVeffVV5eXlNeh8qampkqRZs2bp+uuvV1hYmC644AJFREQ06Dg9e/ZUixYtNHLkSE2bNk1ut1vPP/+8tmzZEvA5A712AM4j7ABoVE8//bQuv/xyPf3003ryySdVV1enxMREXXnllbr00kslSW63W6+++qrGjx+v9PR0hYeHq1+/flqzZo3OPvvsgM81ZMgQvffee3ryySc1Y8YMGWNUVFTU4K+waNWqlV5//XVNmDBBw4YNU0xMjG6++WatWLFCF198ccDnDOTaATjPZcwxPggDAADAErz1HAAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAanzOjn76Hp5vv/1WsbGxjn3kPAAAOL2MMTp48KASExPVrNnx798QdiR9++23SkpKCnUbAAAgCHv27FH79u2Pu52wIyk2NlbST8Nq3rx5wPvV1NQoNzdXaWlpcrvdp6u9MwKzdA6zdAZzdA6zdAZzrK+8vFxJSUm+v8ePh7Cjn7/kr3nz5g0OO9HR0WrevDm/eKeIWTqHWTqDOTqHWTqDOR7fyV6CwguUAQCA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYLD3UDCEzHya87dqyvZ97o2LEAAGjquLMDAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwWkjDzpEjR/TXv/5VycnJioqKUqdOnTRjxgzV1dX5aowxysrKUmJioqKiotSnTx9t377d7zhVVVUaO3asWrdurZiYGN10003au3dvY18OAABogkIadmbNmqWnnnpK8+fP1yeffKLZs2frP//zPzVv3jxfzezZszVnzhzNnz9fGzdulNfrVf/+/XXw4EFfTUZGhlauXKnly5dr3bp1OnTokAYOHKja2tpQXBYAAGhCwkN58vXr1+vmm2/WjTfeKEnq2LGj/v73v+vDDz+U9NNdnblz52rq1KkaNGiQJGnJkiWKj4/XsmXLlJ6ergMHDujZZ5/Vc889p379+kmSli5dqqSkJK1Zs0YDBgwIzcUBAIAmIaRhp1evXnrqqaf02Wef6dxzz9WWLVu0bt06zZ07V5JUVFSkkpISpaWl+fbxeDzq3bu3CgoKlJ6ersLCQtXU1PjVJCYmKiUlRQUFBccMO1VVVaqqqvI9Li8vlyTV1NSopqYm4P6P1jZkn2B5woxjx2qMfhuqMWdpO2bpDOboHGbpDOZYX6CzCGnYmTRpkg4cOKAuXbooLCxMtbW1evTRR3XHHXdIkkpKSiRJ8fHxfvvFx8dr165dvpqIiAi1aNGiXs3R/X8pJydH06dPr7eem5ur6OjoBl9HXl5eg/dpqNmXOnes1atXO3cwhzXGLM8UzNIZzNE5zNIZzPFnFRUVAdWFNOysWLFCS5cu1bJly3T++edr8+bNysjIUGJiooYPH+6rc7lcfvsZY+qt/dKJaqZMmaLMzEzf4/LyciUlJSktLU3NmzcPuP+amhrl5eWpf//+crvdAe8XjJSsfzp2rG1ZTe+pvcacpe2YpTOYo3OYpTOYY31Hn5k5mZCGnT//+c+aPHmybr/9dklSamqqdu3apZycHA0fPlxer1fST3dvEhISfPuVlpb67vZ4vV5VV1errKzM7+5OaWmpevbseczzejweeTyeeututzuoX6Bg92uIqtoTh7uGaMp/SBpjlmcKZukM5ugcZukM5vizQOcQ0ndjVVRUqFkz/xbCwsJ8bz1PTk6W1+v1u2VXXV2t/Px8X5Dp3r273G63X01xcbG2bdt23LADAADOHCG9s/O73/1Ojz76qM4++2ydf/752rRpk+bMmaN77rlH0k9PX2VkZCg7O1udO3dW586dlZ2drejoaA0ZMkSSFBcXpxEjRmjChAlq1aqVWrZsqQceeECpqam+d2cBAIAzV0jDzrx58/Tggw9q1KhRKi0tVWJiotLT0/XQQw/5aiZOnKjKykqNGjVKZWVluuyyy5Sbm6vY2FhfzeOPP67w8HANHjxYlZWV6tu3rxYvXqywsLBQXBYAAGhCQhp2YmNjNXfuXN9bzY/F5XIpKytLWVlZx62JjIzUvHnz/D6MEAAAQOK7sQAAgOUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrhTzsfPPNNxo2bJhatWql6OhoXXTRRSosLPRtN8YoKytLiYmJioqKUp8+fbR9+3a/Y1RVVWns2LFq3bq1YmJidNNNN2nv3r2NfSkAAKAJCmnYKSsr05VXXim3263/+7//044dO/TYY4/pN7/5ja9m9uzZmjNnjubPn6+NGzfK6/Wqf//+OnjwoK8mIyNDK1eu1PLly7Vu3TodOnRIAwcOVG1tbQiuCgAANCXhoTz5rFmzlJSUpEWLFvnWOnbs6PtnY4zmzp2rqVOnatCgQZKkJUuWKD4+XsuWLVN6eroOHDigZ599Vs8995z69esnSVq6dKmSkpK0Zs0aDRgwoFGvCQAANC0hDTurVq3SgAEDdOuttyo/P1/t2rXTqFGj9Mc//lGSVFRUpJKSEqWlpfn28Xg86t27twoKCpSenq7CwkLV1NT41SQmJiolJUUFBQXHDDtVVVWqqqryPS4vL5ck1dTUqKamJuD+j9Y2ZJ9gecKMY8dqjH4bqjFnaTtm6Qzm6Bxm6QzmWF+gswhp2Pnqq6+0YMECZWZm6i9/+Ys2bNigcePGyePx6K677lJJSYkkKT4+3m+/+Ph47dq1S5JUUlKiiIgItWjRol7N0f1/KScnR9OnT6+3npubq+jo6AZfR15eXoP3aajZlzp3rNWrVzt3MIc1xizPFMzSGczROczSGczxZxUVFQHVhTTs1NXVqUePHsrOzpYkdevWTdu3b9eCBQt01113+epcLpfffsaYemu/dKKaKVOmKDMz0/e4vLxcSUlJSktLU/PmzQPuv6amRnl5eerfv7/cbnfA+wUjJeufjh1rW1bTe2qvMWdpO2bpDOboHGbpDOZY39FnZk4mpGEnISFB5513nt9a165d9eKLL0qSvF6vpJ/u3iQkJPhqSktLfXd7vF6vqqurVVZW5nd3p7S0VD179jzmeT0ejzweT711t9sd1C9QsPs1RFXticNdQzTlPySNMcszBbN0BnN0DrN0BnP8WaBzCOm7sa688krt3LnTb+2zzz5Thw4dJEnJycnyer1+t+yqq6uVn5/vCzLdu3eX2+32qykuLta2bduOG3YAAMCZI6R3du6//3717NlT2dnZGjx4sDZs2KCFCxdq4cKFkn56+iojI0PZ2dnq3LmzOnfurOzsbEVHR2vIkCGSpLi4OI0YMUITJkxQq1at1LJlSz3wwANKTU31vTsLAACcuUIadi655BKtXLlSU6ZM0YwZM5ScnKy5c+dq6NChvpqJEyeqsrJSo0aNUllZmS677DLl5uYqNjbWV/P4448rPDxcgwcPVmVlpfr27avFixcrLCwsFJcFAACakJCGHUkaOHCgBg4ceNztLpdLWVlZysrKOm5NZGSk5s2bp3nz5p2GDgEAwK9ZyL8uAgAA4HQi7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsFlTYufbaa7V///566+Xl5br22mtPtScAAADHBBV21q5dq+rq6nrrP/74o959991TbgoAAMAp4Q0p/vjjj33/vGPHDpWUlPge19bW6o033lC7du2c6w4AAOAUNSjsXHTRRXK5XHK5XMd8uioqKkrz5s1zrDkAAIBT1aCwU1RUJGOMOnXqpA0bNqhNmza+bREREWrbtq3CwsIcbxIAACBYDQo7HTp0kCTV1dWdlmYAAACc1qCw8+8+++wzrV27VqWlpfXCz0MPPXTKjQEAADghqLDzzDPP6L777lPr1q3l9Xrlcrl821wuF2EHAAA0GUGFnUceeUSPPvqoJk2a5HQ/AAAAjgrqc3bKysp06623Ot0LAACA44IKO7feeqtyc3Od7gUAAMBxQT2Ndc455+jBBx/U+++/r9TUVLndbr/t48aNc6Q5AACAUxVU2Fm4cKHOOuss5efnKz8/32+by+Ui7AAAgCYjqLBTVFTkdB8AAACnRVCv2QEAAPi1COrOzj333HPC7X/729+CagYAAMBpQYWdsrIyv8c1NTXatm2b9u/ff8wvCAUAAAiVoMLOypUr663V1dVp1KhR6tSp0yk3BQAA4BTHXrPTrFkz3X///Xr88cedOiQAAMApc/QFyl9++aWOHDni5CEBAABOSVBPY2VmZvo9NsaouLhYr7/+uoYPH+5IYwAAAE4IKuxs2rTJ73GzZs3Upk0bPfbYYyd9pxYAAEBjCirsvP322073AQAAcFoEFXaO+u6777Rz5065XC6de+65atOmjVN9AQAAOCKoFygfPnxY99xzjxISEnT11VfrqquuUmJiokaMGKGKigqnewQAAAhaUGEnMzNT+fn5evXVV7V//37t379fr7zyivLz8zVhwgSnewQAAAhaUE9jvfjii3rhhRfUp08f39oNN9ygqKgoDR48WAsWLHCqPwAAgFMS1J2diooKxcfH11tv27YtT2MBAIAmJaiwc8UVV2jatGn68ccffWuVlZWaPn26rrjiCseaAwAAOFVBPY01d+5cXX/99Wrfvr0uvPBCuVwubd68WR6PR7m5uU73CAAAELSgwk5qaqo+//xzLV26VJ9++qmMMbr99ts1dOhQRUVFOd0jAABA0IIKOzk5OYqPj9cf//hHv/W//e1v+u677zRp0iRHmgMAADhVQb1m5+mnn1aXLl3qrZ9//vl66qmnTrkpAAAApwQVdkpKSpSQkFBvvU2bNiouLj7lpgAAAJwSVNhJSkrSe++9V2/9vffeU2Ji4ik3BQAA4JSgXrNz7733KiMjQzU1Nbr22mslSW+++aYmTpzIJygDAIAmJaiwM3HiRO3bt0+jRo1SdXW1JCkyMlKTJk3SlClTHG0QAADgVAQVdlwul2bNmqUHH3xQn3zyiaKiotS5c2d5PB6n+wMAADglQYWdo8466yxdcsklTvUCAADguKBeoAwAAPBrQdgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKzWZMJOTk6OXC6XMjIyfGvGGGVlZSkxMVFRUVHq06ePtm/f7rdfVVWVxo4dq9atWysmJkY33XST9u7d28jdAwCApqpJhJ2NGzdq4cKFuuCCC/zWZ8+erTlz5mj+/PnauHGjvF6v+vfvr4MHD/pqMjIytHLlSi1fvlzr1q3ToUOHNHDgQNXW1jb2ZQAAgCYo5GHn0KFDGjp0qJ555hm1aNHCt26M0dy5czV16lQNGjRIKSkpWrJkiSoqKrRs2TJJ0oEDB/Tss8/qscceU79+/dStWzctXbpUW7du1Zo1a0J1SQAAoAk5pS8CdcLo0aN14403ql+/fnrkkUd860VFRSopKVFaWppvzePxqHfv3iooKFB6eroKCwtVU1PjV5OYmKiUlBQVFBRowIABxzxnVVWVqqqqfI/Ly8slSTU1NaqpqQm496O1DdknWJ4w49ixGqPfhmrMWdqOWTqDOTqHWTqDOdYX6CxCGnaWL1+ujz76SBs3bqy3raSkRJIUHx/vtx4fH69du3b5aiIiIvzuCB2tObr/seTk5Gj69On11nNzcxUdHd3g68jLy2vwPg01+1LnjrV69WrnDuawxpjlmYJZOoM5OodZOoM5/qyioiKgupCFnT179mj8+PHKzc1VZGTkcetcLpffY2NMvbVfOlnNlClTlJmZ6XtcXl6upKQkpaWlqXnz5gFewU+JMi8vT/3795fb7Q54v2CkZP3TsWNtyzr2Ha9QasxZ2o5ZOoM5OodZOoM51nf0mZmTCVnYKSwsVGlpqbp37+5bq62t1TvvvKP58+dr586dkn66e5OQkOCrKS0t9d3t8Xq9qq6uVllZmd/dndLSUvXs2fO45/Z4PPJ4PPXW3W53UL9Awe7XEFW1Jw54DdGU/5A0xizPFMzSGczROczSGczxZ4HOIWQvUO7bt6+2bt2qzZs3+3569OihoUOHavPmzerUqZO8Xq/f7brq6mrl5+f7gkz37t3ldrv9aoqLi7Vt27YThh0AAHDmCNmdndjYWKWkpPitxcTEqFWrVr71jIwMZWdnq3PnzurcubOys7MVHR2tIUOGSJLi4uI0YsQITZgwQa1atVLLli31wAMPKDU1Vf369Wv0awIAAE1PyN+NdSITJ05UZWWlRo0apbKyMl122WXKzc1VbGysr+bxxx9XeHi4Bg8erMrKSvXt21eLFy9WWFhYCDsHAABNRZMKO2vXrvV77HK5lJWVpaysrOPuExkZqXnz5mnevHmntzkAAPCrFPIPFQQAADidCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYLTzUDdiu4+TXQ90CAABnNO7sAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNVCGnZycnJ0ySWXKDY2Vm3bttUtt9yinTt3+tUYY5SVlaXExERFRUWpT58+2r59u19NVVWVxo4dq9atWysmJkY33XST9u7d25iXAgAAmqiQhp38/HyNHj1a77//vvLy8nTkyBGlpaXp8OHDvprZs2drzpw5mj9/vjZu3Civ16v+/fvr4MGDvpqMjAytXLlSy5cv17p163To0CENHDhQtbW1obgsAADQhISH8uRvvPGG3+NFixapbdu2Kiws1NVXXy1jjObOnaupU6dq0KBBkqQlS5YoPj5ey5YtU3p6ug4cOKBnn31Wzz33nPr16ydJWrp0qZKSkrRmzRoNGDCg0a8LAAA0HSENO7904MABSVLLli0lSUVFRSopKVFaWpqvxuPxqHfv3iooKFB6eroKCwtVU1PjV5OYmKiUlBQVFBQcM+xUVVWpqqrK97i8vFySVFNTo5qamoD7PVp7on08YSbg4zWWhlxjYwlklggMs3QGc3QOs3QGc6wv0Fk0mbBjjFFmZqZ69eqllJQUSVJJSYkkKT4+3q82Pj5eu3bt8tVERESoRYsW9WqO7v9LOTk5mj59er313NxcRUdHN7j3vLy8426bfWmDD3farV69OtQtHNeJZomGYZbOYI7OYZbOYI4/q6ioCKiuyYSdMWPG6OOPP9a6devqbXO5XH6PjTH11n7pRDVTpkxRZmam73F5ebmSkpKUlpam5s2bB9xzTU2N8vLy1L9/f7nd7mPWpGT9M+DjNZZtWU3vqb1AZonAMEtnMEfnMEtnMMf6jj4zczJNIuyMHTtWq1at0jvvvKP27dv71r1er6Sf7t4kJCT41ktLS313e7xer6qrq1VWVuZ3d6e0tFQ9e/Y85vk8Ho88Hk+9dbfbHdQv0In2q6o9cSgLhab8hyTYfweoj1k6gzk6h1k6gzn+LNA5hPTdWMYYjRkzRi+99JLeeustJScn+21PTk6W1+v1u2VXXV2t/Px8X5Dp3r273G63X01xcbG2bdt23LADAADOHCG9szN69GgtW7ZMr7zyimJjY32vsYmLi1NUVJRcLpcyMjKUnZ2tzp07q3PnzsrOzlZ0dLSGDBniqx0xYoQmTJigVq1aqWXLlnrggQeUmprqe3cWAAA4c4U07CxYsECS1KdPH7/1RYsW6e6775YkTZw4UZWVlRo1apTKysp02WWXKTc3V7Gxsb76xx9/XOHh4Ro8eLAqKyvVt29fLV68WGFhYY11KQAAoIkKadgx5uRvy3a5XMrKylJWVtZxayIjIzVv3jzNmzfPwe4AAIAN+G4sAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgtfBQN4DG13Hy644c5+uZNzpyHAAATifu7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwmjVh58knn1RycrIiIyPVvXt3vfvuu6FuCQAANAFWhJ0VK1YoIyNDU6dO1aZNm3TVVVfp+uuv1+7du0PdGgAACDErws6cOXM0YsQI3Xvvveratavmzp2rpKQkLViwINStAQCAEPvVh53q6moVFhYqLS3Nbz0tLU0FBQUh6goAADQV4aFu4FR9//33qq2tVXx8vN96fHy8SkpKjrlPVVWVqqqqfI8PHDggSdq3b59qamoCPndNTY0qKir0ww8/yO12H7Mm/MjhgI/3a/PDDz84dqxAZonAMEtnMEfnMEtnhGKOl+W86chxPpjS15Hj/NLBgwclScaYE9b96sPOUS6Xy++xMabe2lE5OTmaPn16vfXk5OTT0putWj8W6g4AAL8Gp/vvi4MHDyouLu6423/1Yad169YKCwurdxentLS03t2eo6ZMmaLMzEzf47q6Ou3bt0+tWrU6bkA6lvLyciUlJWnPnj1q3rx5cBcASczSSczSGczROczSGcyxPmOMDh48qMTExBPW/erDTkREhLp37668vDz9x3/8h289Ly9PN9988zH38Xg88ng8fmu/+c1vgu6hefPm/OI5hFk6h1k6gzk6h1k6gzn6O9EdnaN+9WFHkjIzM3XnnXeqR48euuKKK7Rw4ULt3r1bI0eODHVrAAAgxKwIO7fddpt++OEHzZgxQ8XFxUpJSdHq1avVoUOHULcGAABCzIqwI0mjRo3SqFGjGvWcHo9H06ZNq/eUGBqOWTqHWTqDOTqHWTqDOQbPZU72fi0AAIBfsV/9hwoCAACcCGEHAABYjbADAACsRtgBAABWI+ycgieffFLJycmKjIxU9+7d9e6774a6pSYlJydHl1xyiWJjY9W2bVvdcsst2rlzp1+NMUZZWVlKTExUVFSU+vTpo+3bt/vVVFVVaezYsWrdurViYmJ00003ae/evY15KU1KTk6OXC6XMjIyfGvMMXDffPONhg0bplatWik6OloXXXSRCgsLfduZZWCOHDmiv/71r0pOTlZUVJQ6deqkGTNmqK6uzlfDLOt755139Lvf/U6JiYlyuVx6+eWX/bY7NbOysjLdeeediouLU1xcnO68807t37//NF9dE2YQlOXLlxu3222eeeYZs2PHDjN+/HgTExNjdu3aFerWmowBAwaYRYsWmW3btpnNmzebG2+80Zx99tnm0KFDvpqZM2ea2NhY8+KLL5qtW7ea2267zSQkJJjy8nJfzciRI027du1MXl6e+eijj8w111xjLrzwQnPkyJFQXFZIbdiwwXTs2NFccMEFZvz48b515hiYffv2mQ4dOpi7777bfPDBB6aoqMisWbPGfPHFF74aZhmYRx55xLRq1cq89tprpqioyPzjH/8wZ511lpk7d66vhlnWt3r1ajN16lTz4osvGklm5cqVftudmtl1111nUlJSTEFBgSkoKDApKSlm4MCBjXWZTQ5hJ0iXXnqpGTlypN9aly5dzOTJk0PUUdNXWlpqJJn8/HxjjDF1dXXG6/WamTNn+mp+/PFHExcXZ5566iljjDH79+83brfbLF++3FfzzTffmGbNmpk33nijcS8gxA4ePGg6d+5s8vLyTO/evX1hhzkGbtKkSaZXr17H3c4sA3fjjTeae+65x29t0KBBZtiwYcYYZhmIX4Ydp2a2Y8cOI8m8//77vpr169cbSebTTz89zVfVNPE0VhCqq6tVWFiotLQ0v/W0tDQVFBSEqKum78CBA5Kkli1bSpKKiopUUlLiN0ePx6PevXv75lhYWKiamhq/msTERKWkpJxxsx49erRuvPFG9evXz2+dOQZu1apV6tGjh2699Va1bdtW3bp10zPPPOPbziwD16tXL7355pv67LPPJElbtmzRunXrdMMNN0hilsFwambr169XXFycLrvsMl/N5Zdfrri4uDNyrpJFn6DcmL7//nvV1tbW+1b1+Pj4et++jp8YY5SZmalevXopJSVFknyzOtYcd+3a5auJiIhQixYt6tWcSbNevny5PvroI23cuLHeNuYYuK+++koLFixQZmam/vKXv2jDhg0aN26cPB6P7rrrLmbZAJMmTdKBAwfUpUsXhYWFqba2Vo8++qjuuOMOSfxeBsOpmZWUlKht27b1jt+2bdszcq4SYeeUuFwuv8fGmHpr+MmYMWP08ccfa926dfW2BTPHM2nWe/bs0fjx45Wbm6vIyMjj1jHHk6urq1OPHj2UnZ0tSerWrZu2b9+uBQsW6K677vLVMcuTW7FihZYuXaply5bp/PPP1+bNm5WRkaHExEQNHz7cV8csG86JmR2r/kyeK09jBaF169YKCwurl5BLS0vrJXJIY8eO1apVq/T222+rffv2vnWv1ytJJ5yj1+tVdXW1ysrKjltju8LCQpWWlqp79+4KDw9XeHi48vPz9T//8z8KDw/3zYE5nlxCQoLOO+88v7WuXbtq9+7dkvidbIg///nPmjx5sm6//Xalpqbqzjvv1P3336+cnBxJzDIYTs3M6/XqX//6V73jf/fdd2fkXCXCTlAiIiLUvXt35eXl+a3n5eWpZ8+eIeqq6THGaMyYMXrppZf01ltvKTk52W97cnKyvF6v3xyrq6uVn5/vm2P37t3ldrv9aoqLi7Vt27YzZtZ9+/bV1q1btXnzZt9Pjx49NHToUG3evFmdOnVijgG68sor6338wWeffaYOHTpI4neyISoqKtSsmf9fIWFhYb63njPLhnNqZldccYUOHDigDRs2+Go++OADHThw4IycqyTeeh6so289f/bZZ82OHTtMRkaGiYmJMV9//XWoW2sy7rvvPhMXF2fWrl1riouLfT8VFRW+mpkzZ5q4uDjz0ksvma1bt5o77rjjmG+zbN++vVmzZo356KOPzLXXXmv1W1MD8e/vxjKGOQZqw4YNJjw83Dz66KPm888/N88//7yJjo42S5cu9dUwy8AMHz7ctGvXzvfW85deesm0bt3aTJw40VfDLOs7ePCg2bRpk9m0aZORZObMmWM2bdrk+9gSp2Z23XXXmQsuuMCsX7/erF+/3qSmpvLWcwTniSeeMB06dDARERHm4osv9r2lGj+RdMyfRYsW+Wrq6urMtGnTjNfrNR6Px1x99dVm69atfseprKw0Y8aMMS1btjRRUVFm4MCBZvfu3Y18NU3LL8MOcwzcq6++alJSUozH4zFdunQxCxcu9NvOLANTXl5uxo8fb84++2wTGRlpOnXqZKZOnWqqqqp8NcyyvrfffvuY/10cPny4Mca5mf3www9m6NChJjY21sTGxpqhQ4easrKyRrrKpsdljDGhuacEAABw+vGaHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAKfF3XffrVtuuaXe+tq1a+VyubR///7T3kNWVpYuuuii03b8xrwWAMEj7ACwjjFGR44cCXr/6upqB7sBEGqEHQAhV1BQoKuvvlpRUVFKSkrSuHHjdPjwYd/2pUuXqkePHoqNjZXX69WQIUNUWlrq2370Dss///lP9ejRQx6PR88995ymT5+uLVu2yOVyyeVyafHixcc8/9G7UDk5OUpMTNS555570vN+/fXXuuaaayRJLVq0kMvl0t133y3pp7A1e/ZsderUSVFRUbrwwgv1wgsvnIbJAQgEYQdASG3dulUDBgzQoEGD9PHHH2vFihVat26dxowZ46uprq7Www8/rC1btujll19WUVGRL1j8u4kTJyonJ0effPKJ0tLSNGHCBJ1//vkqLi5WcXGxbrvttuP28eabb+qTTz5RXl6eXnvttZOeNykpSS+++KIkaefOnSouLtZ///d/S5L++te/atGiRVqwYIG2b9+u+++/X8OGDVN+fr5DUwPQIKH9HlIAtho+fLgJCwszMTExfj+RkZFGku8bmO+8807zpz/9yW/fd9991zRr1sxUVlYe89gbNmwwkszBgweNMT9/k/TLL7/sVzdt2jRz4YUXBtRrfHy83zd2N+S8//5t0ocOHTKRkZGmoKDAb98RI0aYO+6446S9AHBeeEiTFgCrXXPNNVqwYIHf2gcffKBhw4b5HhcWFuqLL77Q888/71szxqiurk5FRUXq2rWrNm3apKysLG3evFn79u1TXV2dJGn37t0677zzfPv16NEj6F5TU1MVERHhtxboef/djh079OOPP6p///5+69XV1erWrVvQ/QEIHmEHwGkTExOjc845x29t7969fo/r6uqUnp6ucePG1dv/7LPP1uHDh5WWlqa0tDQtXbpUbdq00e7duzVgwIB6LySOiYk5pV7/XUPO+8vrkaTXX39d7dq189vm8XiC7g9A8Ag7AELq4osv1vbt2+uFoqO2bt2q77//XjNnzlRSUpIk6cMPPwzo2BEREaqtrQ2qr08//fSk5z16J+jfz3HeeefJ4/Fo9+7d6t27d1DnBuAsXqAMIKQmTZqk9evXa/To0dq8ebM+//xzrVq1SmPHjpX0092diIgIzZs3T1999ZVWrVqlhx9+OKBjd+zYUUVFRdq8ebO+//57VVVVBdxXIOft0KGDXC6XXnvtNX333Xc6dOiQYmNj9cADD+j+++/XkiVL9OWXX2rTpk164okntGTJksAHA8AxhB0AIXXBBRcoPz9fn3/+ua666ip169ZNDz74oBISEiRJbdq00eLFi/WPf/xD5513nmbOnKn/+q//CujYv//973XdddfpmmuuUZs2bfT3v/894L4COW+7du00ffp0TZ48WfHx8b53kD388MN66KGHlJOTo65du2rAgAF69dVXlZycHPD5ATjHZYwxoW4CAADgdOHODgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABW+/8G1tdsg+ZlIQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHFCAYAAAAUpjivAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABFgElEQVR4nO3deXhU5f3//9cQJpOFBAghJJEQAoK0QoOAIqASQKJhVwpS+GiQpVQQpeAXpRaTKApi3YoVbSuLAkJdQBGrBCSILBZZyqJF1ASwJKBoEiAQhuT+/eEvU4esZJvJyfNxXXPVc5/7nHm/MzPh1bNkbMYYIwAAAItq4OkCAAAAahJhBwAAWBphBwAAWBphBwAAWBphBwAAWBphBwAAWBphBwAAWBphBwAAWBphBwAAWBphB6gln376qW677Ta1atVKDodDLVq0UI8ePTRjxowaeb7jx48rOTlZe/furfQ+lixZIpvNpoyMDNfY2LFj1bp16yrXV7SvRo0aVWhu69atNXbs2Gp53stls9mUnJzskecGUHWEHaAWrFu3Tj179lRubq7mz5+v9evX6/nnn1evXr20atWqGnnO48ePKyUlpUphpySzZ8/W6tWrq3WfAFCTGnq6AKA+mD9/vmJiYvThhx+qYcP/fexGjRql+fPne7Cyy9e2bVtPl2BJeXl5CggI8HQZl+XcuXPy9/f3dBlAuTiyA9SCU6dOKTQ01C3oFGnQ4H8fw/HjxyskJER5eXnF5vXt21dXX321a/mNN95Q9+7d1bhxYwUEBKhNmzYaN26cJCktLU3XXnutJOnuu++WzWYrdirm3XffVY8ePRQQEKCgoCD1799f27dvL7eXkk5jFRYWasGCBercubP8/f3VpEkTXX/99Xr33XfL3Z8kHTx4UP369VNgYKCaN2+ue++9t8SfwaWOHj2q//u//1NYWJgcDod+8Ytf6Omnn1ZhYaHbvB9++EGTJ0/WFVdcIV9fX7Vp00YPP/yw8vPz3ebl5uZq4sSJatasmRo1aqRbb71VX375ZYV6SEtLk81m07JlyzR9+nSFh4fL399fvXv31p49e9zmFp2+279/v+Lj4xUUFKR+/fpJki5cuKA5c+aoQ4cOcjgcat68ue6++2599913bvv46KOPFBcXp2bNmsnf31+tWrXS8OHD3X5uCxcuVGxsrBo1aqSgoCB16NBBf/jDH1zrk5OTZbPZivVS0unL1q1ba9CgQXr77bd1zTXXyM/PTykpKZKkrKwsTZo0SS1btpSvr69iYmKUkpKiixcvVuhnB9Q0juwAtaBHjx76+9//rvvuu09jxoxRly5dZLfbi827//77tWjRIq1YsUITJkxwjX/++efatGmT/vKXv0iStm/frjvuuEN33HGHkpOT5efnpyNHjuijjz6SJHXp0kWLFy/W3XffrT/+8Y8aOHCgJKlly5aSpBUrVmjMmDGKj4/X66+/rvz8fM2fP19xcXHauHGjbrjhhsvqb+zYsVq2bJnGjx+vRx99VL6+vtq9e7fbP5alcTqdGjBggCZNmqSHHnpI27Zt05w5c3TkyBGtXbu21O2+++479ezZUxcuXNBjjz2m1q1b67333tMDDzygr7/+Wi+++KIk6fz58+rTp4++/vprpaSk6Fe/+pW2bNmiuXPnau/evVq3bp0kyRijYcOGadu2bXrkkUd07bXXauvWrUpISLisn8Uf/vAHdenSRX//+9+Vk5Oj5ORkxcXFac+ePWrTpo1r3oULFzRkyBBX3xcvXlRhYaGGDh2qLVu2aObMmerZs6eOHDmipKQkxcXF6bPPPpO/v78yMjI0cOBA3XjjjVq0aJGaNGmi//73v/rggw904cIFBQQEaOXKlZo8ebKmTp2qP/3pT2rQoIG++uorff7555fVz8/t3r1bX3zxhf74xz8qJiZGgYGBysrK0nXXXacGDRrokUceUdu2bbV9+3bNmTNHGRkZWrx4caWfD6g2BkCN+/77780NN9xgJBlJxm63m549e5q5c+ea06dPu83t3bu36dy5s9vYPffcY4KDg11z//SnPxlJJjs7u9Tn3Llzp5FkFi9e7DZeUFBgIiMjTadOnUxBQYFr/PTp0yYsLMz07NnTNbZ48WIjyaSnp7vGEhMTTXR0tGv5448/NpLMww8/XNEfh9u+JJnnn3/ebfzxxx83kswnn3ziGouOjjaJiYmu5YceeshIMp9++qnbtvfcc4+x2Wzm0KFDxhhjXnrpJSPJ/OMf/3Cb9+STTxpJZv369cYYY/75z3+WWUtSUlKZvWzatMlIMl26dDGFhYWu8YyMDGO3282ECROK9b1o0SK3fbz++utGknnrrbfcxoteyxdffNEYY8ybb75pJJm9e/eWWs+9995rmjRpUmbNSUlJpqR/Bkp63aOjo42Pj4/r51pk0qRJplGjRubIkSNu40Xv0YMHD5ZZA1AbOI0F1IJmzZppy5Yt2rlzp+bNm6ehQ4fqyy+/1KxZs9SpUyd9//33rrn333+/9u7dq61bt0r66dTKa6+9psTERNedS0WnqEaOHKl//OMf+u9//1vhWg4dOqTjx4/rzjvvdDuF1qhRIw0fPlw7duyo0CmkIv/85z8lSVOmTKnwNpcaM2aM2/Lo0aMlSZs2bSp1m48++ki//OUvdd1117mNjx07VsYY11Gujz76SIGBgfr1r39dbJ4kbdy40e25SqulokaPHu12aig6Olo9e/YssZfhw4e7Lb/33ntq0qSJBg8erIsXL7oenTt3Vnh4uNLS0iRJnTt3lq+vr377299q6dKl+uabb4rt+7rrrlN2drZ+85vf6J133nF7j1XWr371K7Vv375YzX369FFkZKRbzUVHxDZv3lzl5wWqirAD1KJu3brpwQcf1BtvvKHjx4/r97//vTIyMtwuUh46dKhat27tOmW1ZMkSnT171i1M3HTTTVqzZo0uXryou+66Sy1btlTHjh31+uuvl1vDqVOnJEkRERHF1kVGRqqwsFA//vhjhXv67rvv5OPjo/Dw8Apv83MNGzZUs2bN3MaK9lVUa0lOnTpVag8/3/bUqVMKDw8vdm1KWFiYGjZs6DavrFoqqqT54eHhxXoJCAhQcHCw29iJEyeUnZ0tX19f2e12t0dWVpYrsLRt21YbNmxQWFiYpkyZorZt26pt27Z6/vnnXfu68847tWjRIh05ckTDhw9XWFiYunfvrtTU1Mvq5+dK+nmfOHFCa9euLVZv0fVl1RGygKoi7AAeYrfblZSUJEk6cOCAa7xBgwaaMmWK3nzzTWVmZurFF19Uv379dNVVV7ltP3ToUG3cuFE5OTlKS0tTy5YtNXr06HIvMi76xzwzM7PYuuPHj6tBgwZq2rRphfto3ry5CgoKlJWVVeFtfu7ixYvFgkDRvi4NHj/XrFmzUnuQpNDQUNe8EydOyBjjNu/kyZO6ePGi27yyaqmokuZnZWUV66WkC4NDQ0PVrFkz7dy5s8RH0XVIknTjjTdq7dq1ysnJ0Y4dO9SjRw9NmzZNK1eudM25++67tW3bNuXk5GjdunUyxmjQoEE6cuSIJMnPz0+Sil2oXVpAKa3m+Pj4UmseP358aT8qoNYQdoBaUNI/ypL0xRdfSPrf0YgiEyZMkK+vr8aMGaNDhw7p3nvvLXXfDodDvXv31pNPPilJrjt/HA6HpJ9uD/65q666SldccYVWrFjhFgDOnj2rt956y3WHVkUVna5YuHBhhbe51PLly92WV6xYIUmKi4srdZt+/frp888/1+7du93GX331VdlsNvXp08c178yZM1qzZk2xeUXrJbnml1ZLRb3++utuP9cjR45o27ZtZfZSZNCgQTp16pQKCgrUrVu3Yo9LA68k+fj4qHv37q4jgZf+PCQpMDBQCQkJevjhh3XhwgUdPHhQklx31e3bt89tflkXhpdU84EDB9S2bdsSa770vQ14AndjAbXglltuUcuWLTV48GB16NBBhYWF2rt3r55++mk1atRI999/v9v8Jk2a6K677tLChQsVHR2twYMHu61/5JFH9O2336pfv35q2bKlsrOz9fzzz8tut6t3796SfjrV4e/vr+XLl+sXv/iFGjVqpMjISEVGRmr+/PkaM2aMBg0apEmTJik/P19PPfWUsrOzNW/evMvq7cYbb9Sdd96pOXPm6MSJExo0aJAcDof27NmjgIAATZ06tcztfX199fTTT+vMmTO69tprXXdjJSQklHlX2O9//3u9+uqrGjhwoB599FFFR0dr3bp1evHFF3XPPfe4ri2566679Je//EWJiYnKyMhQp06d9Mknn+iJJ57QgAEDdPPNN0uS4uPjddNNN2nmzJk6e/asunXrpq1bt+q11167rJ/HyZMnddttt2nixInKyclRUlKS/Pz8NGvWrHK3HTVqlJYvX64BAwbo/vvv13XXXSe73a5vv/1WmzZt0tChQ3XbbbfppZde0kcffaSBAweqVatWOn/+vBYtWiRJrn4mTpwof39/9erVSxEREcrKytLcuXPVuHFj1zVfAwYMUEhIiOsuuoYNG2rJkiU6duxYhft99NFHlZqaqp49e+q+++7TVVddpfPnzysjI0Pvv/++XnrpJdddgIDHePTyaKCeWLVqlRk9erRp166dadSokbHb7aZVq1bmzjvvNJ9//nmJ26SlpRlJZt68ecXWvffeeyYhIcFcccUVxtfX14SFhZkBAwaYLVu2uM17/fXXTYcOHYzdbi92R9GaNWtM9+7djZ+fnwkMDDT9+vUzW7duddu+IndjGfPTHV7PPvus6dixo/H19TWNGzc2PXr0MGvXri3z55KYmGgCAwPNvn37TFxcnPH39zchISHmnnvuMWfOnHGbe+ndWMYYc+TIETN69GjTrFkzY7fbzVVXXWWeeuopt7vMjDHm1KlT5ne/+52JiIgwDRs2NNHR0WbWrFnm/PnzbvOys7PNuHHjTJMmTUxAQIDp37+/+c9//nNZd2O99tpr5r777jPNmzc3DofD3Hjjjeazzz4rse+SOJ1O86c//cnExsYaPz8/06hRI9OhQwczadIkc/jwYWOMMdu3bze33XabiY6ONg6HwzRr1sz07t3bvPvuu679LF261PTp08e0aNHC+Pr6msjISDNy5Eizb98+t+f717/+ZXr27GkCAwPNFVdcYZKSkszf//73Eu/GGjhwYIk1f/fdd+a+++4zMTExxm63m5CQENO1a1fz8MMPF3sdAU+wGXPJiWwAXmHGjBlauHChjh07Vua1K/AOaWlp6tOnj954441id34B8CxOYwFeZseOHfryyy/14osvatKkSQQdAKgiwg7gZYouEB40aJDmzJnj6XIAoM7jNBYAALA0bj0HAACWRtgBAACWRtgBAACWxgXKkgoLC3X8+HEFBQWV+OfQAQCA9zHG6PTp04qMjHT7YuNLEXb003fpREVFeboMAABQCceOHSvzL3UTdiQFBQVJ+umHdem3EHsTp9Op9evXKz4+Xna73dPl1Kr62jt903d9UV97p++q9Z2bm6uoqCjXv+OlIezof9/kGxwc7PVhJyAgQMHBwfXqQyHV397pm77ri/raO31XT9/lXYLCBcoAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSCDsAAMDSGnq6AAB1S+uH1lXLfjLmDayW/QBAeTiyAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALM2jYefjjz/W4MGDFRkZKZvNpjVr1ritt9lsJT6eeuop15y4uLhi60eNGlXLnQAAAG/l0bBz9uxZxcbG6oUXXihxfWZmpttj0aJFstlsGj58uNu8iRMnus17+eWXa6N8AABQB3j0i0ATEhKUkJBQ6vrw8HC35XfeeUd9+vRRmzZt3MYDAgKKzQUAAJDq0LeenzhxQuvWrdPSpUuLrVu+fLmWLVumFi1aKCEhQUlJSQoKCip1X/n5+crPz3ct5+bmSpKcTqecTmf1F19Nimrz5hprSn3t3Rv7dviYatlPWT15Y9+1ob72LdXf3um7an1XdHubMaZ6fnNVkc1m0+rVqzVs2LAS18+fP1/z5s3T8ePH5efn5xr/29/+ppiYGIWHh+vAgQOaNWuWrrzySqWmppb6XMnJyUpJSSk2vmLFCgUEBFS5FwAAUPPy8vI0evRo5eTkKDg4uNR5dSbsdOjQQf3799eCBQvK3M+uXbvUrVs37dq1S126dClxTklHdqKiovT999+X+cPyNKfTqdTUVPXv3192u93T5dSq+tq7N/bdMfnDatnPgeRbSl3njX3Xhvrat1R/e6fvqvWdm5ur0NDQcsNOnTiNtWXLFh06dEirVq0qd26XLl1kt9t1+PDhUsOOw+GQw+EoNm632+vEm62u1FkT6mvv3tR3foGtWvZTkX68qe/aVF/7lupv7/Rd+e0rok78nZ1XXnlFXbt2VWxsbLlzDx48KKfTqYiIiFqoDAAAeDuPHtk5c+aMvvrqK9dyenq69u7dq5CQELVq1UrST4eo3njjDT399NPFtv/666+1fPlyDRgwQKGhofr88881Y8YMXXPNNerVq1et9QEAALyXR8POZ599pj59+riWp0+fLklKTEzUkiVLJEkrV66UMUa/+c1vim3v6+urjRs36vnnn9eZM2cUFRWlgQMHKikpST4+PrXSAwAA8G4eDTtxcXEq7/ro3/72t/rtb39b4rqoqCht3ry5JkoDAAAWUSeu2QEAAKgswg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0wg4AALA0j4adjz/+WIMHD1ZkZKRsNpvWrFnjtn7s2LGy2Wxuj+uvv95tTn5+vqZOnarQ0FAFBgZqyJAh+vbbb2uxCwAA4M08GnbOnj2r2NhYvfDCC6XOufXWW5WZmel6vP/++27rp02bptWrV2vlypX65JNPdObMGQ0aNEgFBQU1XT4AAKgDGnryyRMSEpSQkFDmHIfDofDw8BLX5eTk6JVXXtFrr72mm2++WZK0bNkyRUVFacOGDbrllluqvWYAAFC3eDTsVERaWprCwsLUpEkT9e7dW48//rjCwsIkSbt27ZLT6VR8fLxrfmRkpDp27Kht27aVGnby8/OVn5/vWs7NzZUkOZ1OOZ3OGuymaopq8+Yaa0p97d0b+3b4mGrZT1k9eWPftaG+9i3V397pu2p9V3R7mzGmen5zVZHNZtPq1as1bNgw19iqVavUqFEjRUdHKz09XbNnz9bFixe1a9cuORwOrVixQnfffbdbcJGk+Ph4xcTE6OWXXy7xuZKTk5WSklJsfMWKFQoICKjWvgAAQM3Iy8vT6NGjlZOTo+Dg4FLnefWRnTvuuMP13x07dlS3bt0UHR2tdevW6fbbby91O2OMbDZbqetnzZql6dOnu5Zzc3MVFRWl+Pj4Mn9YnuZ0OpWamqr+/fvLbrd7upxaVV9798a+OyZ/WC37OZBc+mlmb+y7NtTXvqX62zt9V63vojMz5fHqsHOpiIgIRUdH6/Dhw5Kk8PBwXbhwQT/++KOaNm3qmnfy5En17Nmz1P04HA45HI5i43a7vU682epKnTWhvvbuTX3nF5T+fyQuR0X68aa+a1N97Vuqv73Td+W3r4g69Xd2Tp06pWPHjikiIkKS1LVrV9ntdqWmprrmZGZm6sCBA2WGHQAAUH949MjOmTNn9NVXX7mW09PTtXfvXoWEhCgkJETJyckaPny4IiIilJGRoT/84Q8KDQ3VbbfdJklq3Lixxo8frxkzZqhZs2YKCQnRAw88oE6dOrnuzgIAAPWbR8POZ599pj59+riWi66jSUxM1MKFC7V//369+uqrys7OVkREhPr06aNVq1YpKCjItc2zzz6rhg0bauTIkTp37pz69eunJUuWyMfHp9b7AQAA3sejYScuLk5l3Qz24YflXwjp5+enBQsWaMGCBdVZGgAAsIg6dc0OAADA5apTd2MBqLzWD63zdAkA4BEc2QEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJbW0NMFAChfx+QPlV9g83QZ1ar1Q+tKXefwMZp/XcX6zpg3sLpLA2AxHNkBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWxt1YAOq0su7quhzc1QVYF0d2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApXk07Hz88ccaPHiwIiMjZbPZtGbNGtc6p9OpBx98UJ06dVJgYKAiIyN111136fjx4277iIuLk81mc3uMGjWqljsBAADeyqNh5+zZs4qNjdULL7xQbF1eXp52796t2bNna/fu3Xr77bf15ZdfasiQIcXmTpw4UZmZma7Hyy+/XBvlAwCAOqChJ588ISFBCQkJJa5r3LixUlNT3cYWLFig6667TkePHlWrVq1c4wEBAQoPD6/RWgEAQN3k0bBzuXJycmSz2dSkSRO38eXLl2vZsmVq0aKFEhISlJSUpKCgoFL3k5+fr/z8fNdybm6upJ9OnTmdzhqpvToU1ebNNdaU+tp7Ub+OBsbDldSuon5rs29veG/V1/e5VH97p++q9V3R7W3GGK/4LWqz2bR69WoNGzasxPXnz5/XDTfcoA4dOmjZsmWu8b/97W+KiYlReHi4Dhw4oFmzZunKK68sdlTo55KTk5WSklJsfMWKFQoICKhyLwAAoObl5eVp9OjRysnJUXBwcKnz6kTYcTqdGjFihI4ePaq0tLQyG9q1a5e6deumXbt2qUuXLiXOKenITlRUlL7//vsy9+1pTqdTqamp6t+/v+x2u6fLqVX1tfeivmd/1kD5hTZPl1NrHA2MHutWWKt9H0i+pVaepyz19X0u1d/e6btqfefm5io0NLTcsOP1p7GcTqdGjhyp9PR0ffTRR+WGkS5dushut+vw4cOlhh2HwyGHw1Fs3G6314k3W12psybU197zC23KL6g/YadIbfbtTe+r+vo+l+pv7/Rd+e0rwqvDTlHQOXz4sDZt2qRmzZqVu83BgwfldDoVERFRCxUCAABv59Gwc+bMGX311Veu5fT0dO3du1chISGKjIzUr3/9a+3evVvvvfeeCgoKlJWVJUkKCQmRr6+vvv76ay1fvlwDBgxQaGioPv/8c82YMUPXXHONevXq5am2AACAF/Fo2Pnss8/Up08f1/L06dMlSYmJiUpOTta7774rSercubPbdps2bVJcXJx8fX21ceNGPf/88zpz5oyioqI0cOBAJSUlycfHp9b6AAAA3sujYScuLk5lXR9d3rXTUVFR2rx5c3WXBQAALITvxgIAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJbW0NMFAICVtH5oXaW3dfgYzb9O6pj8oQ49PqgaqwLqN47sAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAAS6tU2Onbt6+ys7OLjefm5qpv375VrQkAAKDaVCrspKWl6cKFC8XGz58/ry1btlS5KAAAgOpyWX9Bed++fa7//vzzz5WVleVaLigo0AcffKArrrii+qoDAACoossKO507d5bNZpPNZivxdJW/v78WLFhQbcUBAABU1WWFnfT0dBlj1KZNG/3rX/9S8+bNXet8fX0VFhYmHx+fai8SAACgsi7rmp3o6Gi1bt1ahYWF6tatm6Kjo12PiIiIyw46H3/8sQYPHqzIyEjZbDatWbPGbb0xRsnJyYqMjJS/v7/i4uJ08OBBtzn5+fmaOnWqQkNDFRgYqCFDhujbb7+9rDoAAIB1Vfpbz7/88kulpaXp5MmTKiwsdFv3yCOPVGgfZ8+eVWxsrO6++24NHz682Pr58+frmWee0ZIlS9S+fXvNmTNH/fv316FDhxQUFCRJmjZtmtauXauVK1eqWbNmmjFjhgYNGqRdu3ZxlAkAAFQu7Pztb3/TPffco9DQUIWHh8tms7nW2Wy2CoedhIQEJSQklLjOGKPnnntODz/8sG6//XZJ0tKlS9WiRQutWLFCkyZNUk5Ojl555RW99tpruvnmmyVJy5YtU1RUlDZs2KBbbrmlMu0BqIdaP7TO0yUAqCGVCjtz5szR448/rgcffLC663FJT09XVlaW4uPjXWMOh0O9e/fWtm3bNGnSJO3atUtOp9NtTmRkpDp27Kht27aVGnby8/OVn5/vWs7NzZUkOZ1OOZ3OGuqo6opq8+Yaa0p97b2oX0cD4+FKaldRv/W57/r6Xqfv+qG6+q7o9pUKOz/++KNGjBhRmU0rrOi29hYtWriNt2jRQkeOHHHN8fX1VdOmTYvN+flt8ZeaO3euUlJSio2vX79eAQEBVS29xqWmpnq6BI+pr70/1q2w/EkWVJ/7fv/99z1dhkfU1884fVdOXl5eheZVKuyMGDFC69ev1+9+97vKbH5Zfn6KTPrp9NalY5cqb86sWbM0ffp013Jubq6ioqIUHx+v4ODgqhVcg5xOp1JTU9W/f3/Z7XZPl1Or6mvvRX3P/qyB8gvLft9biaOB0WPdCut137seudXT5dSq+v4Zp+/KKTozU55KhZ0rr7xSs2fP1o4dO9SpU6dihd53332V2a2b8PBwST8dvYmIiHCNnzx50nW0Jzw8XBcuXNCPP/7odnTn5MmT6tmzZ6n7djgccjgcxcbtdnudeLPVlTprQn3tPb/QpvyC+vOPfpH63Hd9fJ9L9fczTt+V374iKhV2/vrXv6pRo0bavHmzNm/e7LbOZrNVS9iJiYlReHi4UlNTdc0110iSLly4oM2bN+vJJ5+UJHXt2lV2u12pqakaOXKkJCkzM1MHDhzQ/Pnzq1wDAACo+yoVdtLT06vlyc+cOaOvvvrKbb979+5VSEiIWrVqpWnTpumJJ55Qu3bt1K5dOz3xxBMKCAjQ6NGjJUmNGzfW+PHjNWPGDDVr1kwhISF64IEH1KlTJ9fdWQAAoH6r9N/ZqQ6fffaZ+vTp41ouuo4mMTFRS5Ys0cyZM3Xu3DlNnjxZP/74o7p3767169e7/saOJD377LNq2LChRo4cqXPnzqlfv35asmQJf2MHAABIqmTYGTduXJnrFy1aVKH9xMXFyZjSby212WxKTk5WcnJyqXP8/Py0YMECvpMLAACUqNK3nv+c0+nUgQMHlJ2dXeIXhAIAAHhKpcLO6tWri40VFhZq8uTJatOmTZWLAgAAqC6X9UWgZe6oQQP9/ve/17PPPltduwQAAKiyags7kvT111/r4sWL1blLAACAKqnUaayf//Vh6ae/WJyZmal169YpMTGxWgoDAACoDpUKO3v27HFbbtCggZo3b66nn3663Du1AAAAalOlws6mTZuquw4AAIAaUaU/Kvjdd9/p0KFDstlsat++vZo3b15ddQEAAFSLSl2gfPbsWY0bN04RERG66aabdOONNyoyMlLjx4+v8NetAwAA1IZKhZ3p06dr8+bNWrt2rbKzs5Wdna133nlHmzdv1owZM6q7RgAAgEqr1Gmst956S2+++abi4uJcYwMGDJC/v79GjhyphQsXVld9AAAAVVKpIzt5eXlq0aJFsfGwsDBOYwEAAK9SqbDTo0cPJSUl6fz5866xc+fOKSUlRT169Ki24gAAAKqqUqexnnvuOSUkJKhly5aKjY2VzWbT3r175XA4tH79+uquEQAAoNIqFXY6deqkw4cPa9myZfrPf/4jY4xGjRqlMWPGyN/fv7prBAAAqLRKhZ25c+eqRYsWmjhxotv4okWL9N133+nBBx+sluIAAACqqlLX7Lz88svq0KFDsfGrr75aL730UpWLAgAAqC6VCjtZWVmKiIgoNt68eXNlZmZWuSgAAIDqUqmwExUVpa1btxYb37p1qyIjI6tcFAAAQHWp1DU7EyZM0LRp0+R0OtW3b19J0saNGzVz5kz+gjIAAPAqlQo7M2fO1A8//KDJkyfrwoULkiQ/Pz89+OCDmjVrVrUWCAAAUBWVCjs2m01PPvmkZs+erS+++EL+/v5q166dHA5HddcHAABQJZUKO0UaNWqka6+9trpqAQAAqHaVukAZAACgriDsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAAS6vSt54DAGpG64fWVct+MuYNrJb9AHWZ1x/Zad26tWw2W7HHlClTJEljx44ttu7666/3cNUAAMBbeP2RnZ07d6qgoMC1fODAAfXv318jRoxwjd16661avHixa9nX17dWawQAAN7L68NO8+bN3ZbnzZuntm3bqnfv3q4xh8Oh8PDw2i4NAADUAV4fdn7uwoULWrZsmaZPny6bzeYaT0tLU1hYmJo0aaLevXvr8ccfV1hYWKn7yc/PV35+vms5NzdXkuR0OuV0OmuugSoqqs2ba6wp9bX3on4dDYyHK6ldRf3Sd9XVlc9Mff+M03fV9lMemzGmzvw2+cc//qHRo0fr6NGjioyMlCStWrVKjRo1UnR0tNLT0zV79mxdvHhRu3btksPhKHE/ycnJSklJKTa+YsUKBQQE1GgPAACgeuTl5Wn06NHKyclRcHBwqfPqVNi55ZZb5Ovrq7Vr15Y6JzMzU9HR0Vq5cqVuv/32EueUdGQnKipK33//fZk/LE9zOp1KTU1V//79ZbfbPV1OraqvvRf1PfuzBsovtJW/gUU4Ghg91q2QvqvBgeRbqmU/Na2+f8bpu3Jyc3MVGhpabtipM6exjhw5og0bNujtt98uc15ERISio6N1+PDhUuc4HI4Sj/rY7fY68WarK3XWhPrae36hTfkF9ecf/SL0XXV17fNSXz/j9F357SvC6289L7J48WKFhYVp4MCy/2bEqVOndOzYMUVERNRSZQAAwJvVibBTWFioxYsXKzExUQ0b/u9g1JkzZ/TAAw9o+/btysjIUFpamgYPHqzQ0FDddtttHqwYAAB4izpxGmvDhg06evSoxo0b5zbu4+Oj/fv369VXX1V2drYiIiLUp08frVq1SkFBQR6qFgAAeJM6EXbi4+NV0nXU/v7++vDDDz1QEQAAqCvqxGksAACAyiLsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAAS2vo6QIAq2r90Loq78PhYzT/umooBgDqMY7sAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAAS+Nbz4FLVMe3lQMAvAdHdgAAgKURdgAAgKURdgAAgKURdgAAgKV5ddhJTk6WzWZze4SHh7vWG2OUnJysyMhI+fv7Ky4uTgcPHvRgxQAAwNt4ddiRpKuvvlqZmZmux/79+13r5s+fr2eeeUYvvPCCdu7cqfDwcPXv31+nT5/2YMUAAMCbeH3YadiwocLDw12P5s2bS/rpqM5zzz2nhx9+WLfffrs6duyopUuXKi8vTytWrPBw1QAAwFt4fdg5fPiwIiMjFRMTo1GjRumbb76RJKWnpysrK0vx8fGuuQ6HQ71799a2bds8VS4AAPAyXv1HBbt3765XX31V7du314kTJzRnzhz17NlTBw8eVFZWliSpRYsWbtu0aNFCR44cKXO/+fn5ys/Pdy3n5uZKkpxOp5xOZzV3UX2KavPmGmtKbfbu8DE1/hwV5Whg3P63vqDv6uu7rvy+qK+/3+i7an1XdHubMabO/DY5e/as2rZtq5kzZ+r6669Xr169dPz4cUVERLjmTJw4UceOHdMHH3xQ6n6Sk5OVkpJSbHzFihUKCAiokdoBAED1ysvL0+jRo5WTk6Pg4OBS53n1kZ1LBQYGqlOnTjp8+LCGDRsmScrKynILOydPnix2tOdSs2bN0vTp013Lubm5ioqKUnx8fJk/LE9zOp1KTU1V//79ZbfbPV1OrarN3jsmf1ij+78cjgZGj3Ur1OzPGii/0ObpcmoNfVdf3weSb6mW/dS0+vr7jb6r1nfRmZny1Kmwk5+fry+++EI33nijYmJiFB4ertTUVF1zzTWSpAsXLmjz5s168skny9yPw+GQw+EoNm632+vEm62u1FkTaqP3/ALv+8c1v9DmlXXVNPquurr2u6K+/n6j78pvXxFeHXYeeOABDR48WK1atdLJkyc1Z84c5ebmKjExUTabTdOmTdMTTzyhdu3aqV27dnriiScUEBCg0aNHe7p0AADgJbw67Hz77bf6zW9+o++//17NmzfX9ddfrx07dig6OlqSNHPmTJ07d06TJ0/Wjz/+qO7du2v9+vUKCgrycOUAAMBbeHXYWblyZZnrbTabkpOTlZycXDsFAQCAOsfr/84OAABAVRB2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApXn1t54DAKqm9UPrqmU/GfMGVst+AE/gyA4AALA0juwAAMrFESLUZRzZAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlubVYWfu3Lm69tprFRQUpLCwMA0bNkyHDh1ymzN27FjZbDa3x/XXX++higEAgLfx6rCzefNmTZkyRTt27FBqaqouXryo+Ph4nT171m3erbfeqszMTNfj/fff91DFAADA2zT0dAFl+eCDD9yWFy9erLCwMO3atUs33XSTa9zhcCg8PLy2ywMAAHWAVx/ZuVROTo4kKSQkxG08LS1NYWFhat++vSZOnKiTJ096ojwAAOCFvPrIzs8ZYzR9+nTdcMMN6tixo2s8ISFBI0aMUHR0tNLT0zV79mz17dtXu3btksPhKHFf+fn5ys/Pdy3n5uZKkpxOp5xOZ802UgVFtXlzjTWlNnt3+Jgaf46KcjQwbv9bX9C3dfsu7TNcX3+/0XfV+q7o9jZjTJ34VE2ZMkXr1q3TJ598opYtW5Y6LzMzU9HR0Vq5cqVuv/32EuckJycrJSWl2PiKFSsUEBBQbTUDAICak5eXp9GjRysnJ0fBwcGlzqsTYWfq1Klas2aNPv74Y8XExJQ7v127dpowYYIefPDBEteXdGQnKipK33//fZk/LE9zOp1KTU1V//79ZbfbPV1OrarN3jsmf1ij+78cjgZGj3Ur1OzPGii/0ObpcmoNfVu37wPJt5Q4Xl9/v9F31frOzc1VaGhouWHHq09jGWM0depUrV69WmlpaRUKOqdOndKxY8cUERFR6hyHw1HiKS673V4n3mx1pc6aUBu95xd43z8y+YU2r6yrptG39ZT3+a2vv9/ou/LbV4RXX6A8ZcoULVu2TCtWrFBQUJCysrKUlZWlc+fOSZLOnDmjBx54QNu3b1dGRobS0tI0ePBghYaG6rbbbvNw9QAAwBt49ZGdhQsXSpLi4uLcxhcvXqyxY8fKx8dH+/fv16uvvqrs7GxFRESoT58+WrVqlYKCgjxQMQAA8DZeHXbKu5zI399fH37oPddXAAAA7+PVp7EAAACqyquP7AAArKX1Q+tKHHf4GM2/7qe7IStycXbGvIHVXRosjCM7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0rgbC5ZR2l0eAID6jSM7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0hp6ugCg9UPryp3j8DGaf53UMflD5RfYaqEqAIBVcGQHAABYGkd2AACoooocoS7JpUetM+YNrObKIHFkBwAAWBxHdgAA8BKVPUJUU6xypIkjOwAAwNI4soNK87b/BwIAl4vfY/UDR3YAAIClEXYAAIClEXYAAIClcc0OAAAoUXVd0+Tpu7o4sgMAACyNIzsAgDqHu6hwOTiyAwAALI2wAwAALM0yYefFF19UTEyM/Pz81LVrV23ZssXTJQEAAC9gibCzatUqTZs2TQ8//LD27NmjG2+8UQkJCTp69KinSwMAAB5mibDzzDPPaPz48ZowYYJ+8Ytf6LnnnlNUVJQWLlzo6dIAAICH1fm7sS5cuKBdu3bpoYcechuPj4/Xtm3bPFTV/1jlbxQAAFBX1fmw8/3336ugoEAtWrRwG2/RooWysrJK3CY/P1/5+fmu5ZycHEnSDz/8IKfTWa31Nbx4tlr2c+rUKTmdTuXl5enUqVOy2+3Vst+qqK7eKvRchUZ5eYVq6GyggkJbrT2vp9E3fdcX9bX3+tL3qVOn3Jar69+z06dPS5KMMWXOq/Nhp4jN5v4mMcYUGysyd+5cpaSkFBuPiYmpkdqqQ+jTnq7A80Z7ugAPoe/6pb72LdXf3utD3zX9b9jp06fVuHHjUtfX+bATGhoqHx+fYkdxTp48WexoT5FZs2Zp+vTpruXCwkL98MMPatasWakByRvk5uYqKipKx44dU3BwsKfLqVX1tXf6pu/6or72Tt9V69sYo9OnTysyMrLMeXU+7Pj6+qpr165KTU3Vbbfd5hpPTU3V0KFDS9zG4XDI4XC4jTVp0qQmy6xWwcHB9epD8XP1tXf6rl/qa99S/e2dviuvrCM6Rep82JGk6dOn684771S3bt3Uo0cP/fWvf9XRo0f1u9/9ztOlAQAAD7NE2Lnjjjt06tQpPfroo8rMzFTHjh31/vvvKzo62tOlAQAAD7NE2JGkyZMna/LkyZ4uo0Y5HA4lJSUVOwVXH9TX3umbvuuL+to7fddO3zZT3v1aAAAAdZgl/oIyAABAaQg7AADA0gg7AADA0gg7AADA0gg7Xig5OVk2m83tER4e7lpvjFFycrIiIyPl7++vuLg4HTx40IMVV4/WrVsX69tms2nKlCmSpLFjxxZbd/3113u46sv38ccfa/DgwYqMjJTNZtOaNWvc1lfk9c3Pz9fUqVMVGhqqwMBADRkyRN9++20tdnH5yurb6XTqwQcfVKdOnRQYGKjIyEjdddddOn78uNs+4uLiir0HRo0aVcudXL7yXvOKvLet9ppLKvHzbrPZ9NRTT7nm1MXXfO7cubr22msVFBSksLAwDRs2TIcOHXKbY8XPeXl9e/JzTtjxUldffbUyMzNdj/3797vWzZ8/X88884xeeOEF7dy5U+Hh4erfv7/rC9Hqqp07d7r1nJqaKkkaMWKEa86tt97qNuf999/3VLmVdvbsWcXGxuqFF14ocX1FXt9p06Zp9erVWrlypT755BOdOXNGgwYNUkFBQW21cdnK6jsvL0+7d+/W7NmztXv3br399tv68ssvNWTIkGJzJ06c6PYeePnll2uj/Cop7zWXyn9vW+01l+TWb2ZmphYtWiSbzabhw4e7zatrr/nmzZs1ZcoU7dixQ6mpqbp48aLi4+N19uz/vjzZip/z8vr26OfcwOskJSWZ2NjYEtcVFhaa8PBwM2/ePNfY+fPnTePGjc1LL71USxXWjvvvv9+0bdvWFBYWGmOMSUxMNEOHDvVsUdVMklm9erVruSKvb3Z2trHb7WblypWuOf/9739NgwYNzAcffFBrtVfFpX2X5F//+peRZI4cOeIa6927t7n//vtrtrgaVlLv5b2368trPnToUNO3b1+3MSu85idPnjSSzObNm40x9edzfmnfJamtzzlHdrzU4cOHFRkZqZiYGI0aNUrffPONJCk9PV1ZWVmKj493zXU4HOrdu7e2bdvmqXKr3YULF7Rs2TKNGzfO7ctZ09LSFBYWpvbt22vixIk6efKkB6usfhV5fXft2iWn0+k2JzIyUh07drTUeyAnJ0c2m63Y99YtX75coaGhuvrqq/XAAw/U+SOaRcp6b9eH1/zEiRNat26dxo8fX2xdXX/Nc3JyJEkhISGS6s/n/NK+S5tTG59zy/wFZSvp3r27Xn31VbVv314nTpzQnDlz1LNnTx08eND17e6XfqN7ixYtdOTIEU+UWyPWrFmj7OxsjR071jWWkJCgESNGKDo6Wunp6Zo9e7b69u2rXbt2Weavj1bk9c3KypKvr6+aNm1abE7R9nXd+fPn9dBDD2n06NFuXxI4ZswYxcTEKDw8XAcOHNCsWbP073//23XKs64q771dH17zpUuXKigoSLfffrvbeF1/zY0xmj59um644QZ17NhRUv34nJfU96Vq83NO2PFCCQkJrv/u1KmTevToobZt22rp0qWuixZ/frRD+umNdelYXfbKK68oISFBkZGRrrE77rjD9d8dO3ZUt27dFB0drXXr1hX7BVnXVeb1tcp7wOl0atSoUSosLNSLL77otm7ixImu/+7YsaPatWunbt26affu3erSpUttl1ptKvvetsprLkmLFi3SmDFj5Ofn5zZe11/ze++9V/v27dMnn3xSbJ2VP+dl9S3V/uec01h1QGBgoDp16qTDhw+77sq6NNmfPHmy2P9LqKuOHDmiDRs2aMKECWXOi4iIUHR0tA4fPlxLldW8iry+4eHhunDhgn788cdS59RVTqdTI0eOVHp6ulJTU93+315JunTpIrvdbqn3gFT8vW3l11yStmzZokOHDpX7mZfq1ms+depUvfvuu9q0aZNatmzpGrf657y0vot44nNO2KkD8vPz9cUXXygiIsJ1aO/nh/MuXLigzZs3q2fPnh6ssvosXrxYYWFhGjhwYJnzTp06pWPHjikiIqKWKqt5FXl9u3btKrvd7jYnMzNTBw4cqNPvgaJfgIcPH9aGDRvUrFmzcrc5ePCgnE6npd4DUvH3tlVf8yKvvPKKunbtqtjY2HLn1oXX3Bije++9V2+//bY++ugjxcTEuK236ue8vL4lD37Oq/VyZ1SLGTNmmLS0NPPNN9+YHTt2mEGDBpmgoCCTkZFhjDFm3rx5pnHjxubtt982+/fvN7/5zW9MRESEyc3N9XDlVVdQUGBatWplHnzwQbfx06dPmxkzZpht27aZ9PR0s2nTJtOjRw9zxRVX1Lm+T58+bfbs2WP27NljJJlnnnnG7Nmzx3U3QkVe39/97nemZcuWZsOGDWb37t2mb9++JjY21ly8eNFTbZWrrL6dTqcZMmSIadmypdm7d6/JzMx0PfLz840xxnz11VcmJSXF7Ny506Snp5t169aZDh06mGuuucar+zam7N4r+t622mteJCcnxwQEBJiFCxcW276uvub33HOPady4sUlLS3N7L+fl5bnmWPFzXl7fnvycE3a80B133GEiIiKM3W43kZGR5vbbbzcHDx50rS8sLDRJSUkmPDzcOBwOc9NNN5n9+/d7sOLq8+GHHxpJ5tChQ27jeXl5Jj4+3jRv3tzY7XbTqlUrk5iYaI4ePeqhSitv06ZNRlKxR2JiojGmYq/vuXPnzL333mtCQkKMv7+/GTRokNf/LMrqOz09vcR1ksymTZuMMcYcPXrU3HTTTSYkJMT4+vqatm3bmvvuu8+cOnXKs41VQFm9V/S9bbXXvMjLL79s/P39TXZ2drHt6+prXtp7efHixa45Vvycl9e3Jz/ntv+/QAAAAEvimh0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AVTJ27FgNGzbMtRwXF6dp06ZVen+tW7fWc889V+Ycm82mNWvWVPo5KiIjI0M2m0179+6t0ecBUPMIO0AdcfLkSU2aNEmtWrWSw+FQeHi4brnlFm3fvr1a9n9paKmst99+W4899ljVCwKAatLQ0wUAqJjhw4fL6XRq6dKlatOmjU6cOKGNGzfqhx9+8HRpbkJCQjxdgldzOp2y2+2eLsONN9YEVCeO7AB1QHZ2tj755BM9+eST6tOnj6Kjo3Xddddp1qxZrm+HHzdunAYNGuS23cWLFxUeHq5FixZJkt5880116tRJ/v7+atasmW6++WadPXtWycnJWrp0qd555x3ZbDbZbDalpaVJkvbv36++ffu6tvntb3+rM2fOlFrrpaex8vPzNXPmTEVFRcnhcKhdu3Z65ZVXyuz39OnTGj16tBo1aqTIyEgtWLCgzPnl1VhYWKhHH31ULVu2lMPhUOfOnfXBBx+47eNf//qXrrnmGvn5+albt27as2dPmc8p/XTK7bHHHiuzVpvNppdeeklDhw5VYGCg5syZI0lau3atunbtKj8/P7Vp00YpKSm6ePGia7vk5GTXUbzIyEjdd999rnUvvvii2rVrJz8/P7Vo0UK//vWv3Wq69DRg586dlZycXOWagDqrSt+sBaBWOJ1O06hRIzNt2jRz/vz5Euds3brV+Pj4mOPHj7vG3nnnHRMYGGhOnz5tjh8/bho2bGieeeYZk56ebvbt22f+8pe/mNOnT5vTp0+bkSNHmltvvdXtW4jPnj3r+jLa/fv3m40bN5qYmBi3L3JMTEw0Q4cOdS337t3b3H///a7lkSNHmqioKPP222+br7/+2mzYsMGsXLmy1F6jo6NNUFCQmTt3rjl06JD585//bHx8fMz69etdcySZ1atXG2NMhWp85plnTHBwsHn99dfNf/7zHzNz5kxjt9vNl19+aYwx5syZM6Z58+bmjjvuMAcOHDBr1641bdq0MZLMnj17qlxrWFiYeeWVV8zXX39tMjIyzAcffGCCg4PNkiVLzNdff23Wr19vWrdubZKTk40xxrzxxhsmODjYvP/+++bIkSPm008/NX/961+NMcbs3LnT+Pj4mBUrVpiMjAyze/du8/zzz7vV9Oyzz7rVGRsba5KSkqpUE1CXEXaAOuLNN980TZs2NX5+fqZnz55m1qxZ5t///rfbnF/+8pfmySefdC0PGzbMjB071hhjzK5du4wkk5GRUeL+Lw0txhjz17/+1TRt2tScOXPGNbZu3TrToEEDk5WVVeJ2Pw87hw4dMpJMampqhfuMjo42t956q9vYHXfcYRISElzLPw87FakxMjLSPP744277vPbaa83kyZONMT9983ZISIg5e/asa/3ChQsrFHYqUuu0adPc5tx4443miSeecBt77bXXTEREhDHGmKefftq0b9/eXLhwodhzvvXWWyY4ONjk5uaWWlNFws7l1gTUZZzGAuqI4cOH6/jx43r33Xd1yy23KC0tTV26dNGSJUtccyZMmKDFixdL+umC5nXr1mncuHGSpNjYWPXr10+dOnXSiBEj9Le//U0//vhjmc/5xRdfKDY2VoGBga6xXr16qbCwUIcOHSq35r1798rHx0e9e/e+rF579OhRbPmLL76oVI25ubk6fvy4evXq5bZdr169XPss2kdAQECpNVSl1m7durkt79q1S48++qgaNWrkekycOFGZmZnKy8vTiBEjdO7cObVp00YTJ07U6tWrXaeT+vfvr+joaLVp00Z33nmnli9frry8vArVWpWagLqMsAPUIX5+furfv78eeeQRbdu2TWPHjlVSUpJr/V133aVvvvlG27dv17Jly9S6dWvdeOONkiQfHx+lpqbqn//8p375y19qwYIFuuqqq5Senl7q8xljZLPZSlxX2vjP+fv7X2aHpSvt+Spa46Vzfr6dMaaaqiz5uX4exKSfriFKSUnR3r17XY/9+/fr8OHD8vPzU1RUlA4dOqS//OUv8vf31+TJk3XTTTfJ6XQqKChIu3fv1uuvv66IiAg98sgjio2NVXZ2tiSpQYMGxfpxOp3FarzcmoC6jLAD1GG//OUvdfbsWddys2bNNGzYMC1evFiLFy/W3Xff7TbfZrOpV69eSklJ0Z49e+Tr66vVq1dLknx9fVVQUFBs/3v37nV7jq1bt6pBgwZq3759ufV16tRJhYWF2rx582X1tWPHjmLLHTp0KHFueTUGBwcrMjJSn3zyidt227Zt0y9+8QvXPv7973/r3LlzpdZQHbUW6dKliw4dOqQrr7yy2KNBg59+Lfv7+2vIkCH685//rLS0NG3fvl379++XJDVs2FA333yz5s+fr3379ikjI0MfffSRJKl58+bKzMx0PVdubm6ZgfZyagLqKm49B+qAU6dOacSIERo3bpx+9atfKSgoSJ999pnmz5+voUOHus2dMGGCBg0apIKCAiUmJrrGP/30U23cuFHx8fEKCwvTp59+qu+++871D37r1q314Ycf6tChQ2rWrJkaN26sMWPGKCkpSYmJiUpOTtZ3332nqVOn6s4771SLFi3Krbt169ZKTEzUuHHj9Oc//1mxsbE6cuSITp48qZEjR5a63datWzV//nwNGzZMqampeuONN7Ru3boS51akxv/3//6fkpKS1LZtW3Xu3FmLFy/W3r17tXz5cknS6NGj9fDDD2v8+PH64x//qIyMDP3pT38qt7/LrbXII488okGDBikqKkojRoxQgwYNtG/fPu3fv19z5szRkiVLVFBQoO7duysgIECvvfaa/P39FR0drffee0/ffPONbrrpJjVt2lTvv/++CgsLddVVV0mS+vbtqyVLlmjw4MFq2rSpZs+eLR8fn3L7KK8moE7z6BVDACrk/Pnz5qGHHjJdunQxjRs3NgEBAeaqq64yf/zjH01eXp7b3MLCQhMdHW0GDBjgNv7555+bW265xTRv3tw4HA7Tvn17s2DBAtf6kydPmv79+5tGjRoZSWbTpk3GGGP27dtn+vTpY/z8/ExISIiZOHGiOX36tGu78u7GOnfunPn9739vIiIijK+vr7nyyivNokWLSu01OjrapKSkmJEjR5qAgADTokUL89xzz7nN0c8uUK5IjQUFBSYlJcVcccUVxm63m9jYWPPPf/7TbZ/bt283sbGxxtfX13Tu3Nm89dZbFbpA+XJrLfLBBx+Ynj17Gn9/fxMcHGyuu+461x1Xq1evNt27dzfBwcEmMDDQXH/99WbDhg3GGGO2bNlievfubZo2bWr8/f3Nr371K7Nq1SrXfnNycszIkSNNcHCwiYqKMkuWLCnxAuXLrQmoy2zGVPPJagAelZeXp8jISC1atEi33367p8uxtNatW2vatGlV+noMADWP01iARRQWFiorK0tPP/20GjdurCFDhni6JADwCoQdwCKOHj2qmJgYtWzZUkuWLFHDhny8AUCSOI0FAAAsjfsJAQCApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApRF2AACApf1/hp6LiWe7aUYAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHFCAYAAAAUpjivAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABDtUlEQVR4nO3de3iMd/7/8dckJiMhoRE5VcShTkWjWC12i2qoU6meyLZiWz1ri2rRfjXRA6WXVpelh120xbLfLt1ubUuK6AEtcajTkmocWtHUKUE0RvL5/dFf5mvkPEYmuT0f1zWXzuf+3Pf9vt9m5NX7vmdiM8YYAQAAWJSfrwsAAAC4nAg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7AADA0gg7wGUyf/582Ww216NmzZqKjIxUjx49NGXKFGVlZRVZJzk5WTab7bLVNHv2bM2fP/+StmGz2ZScnOx6npqaKpvNptTU1Eva7oXb+vDDD8uce7l7VZrhw4erUaNGPtk3gIoj7ACX2bx587R+/XqlpKToL3/5i9q1a6epU6eqVatW+vzzz93mjhgxQuvXr79stXgj7Fysffv2Wr9+vdq3b+/V7QKAt9TwdQGA1bVp00YdO3Z0Pb/jjjs0evRo/f73v9fgwYOVnp6uiIgISVKDBg3UoEEDX5XqkZCQEN14442+LsNynE6nbDabatSoPv9M5+bmKigoyNdlAEVwZgfwgYYNG2r69Ok6deqU3n77bdd4cZdmlixZol69eikqKkqBgYFq1aqVxo8frzNnzrjN++GHHzRkyBBFR0fL4XAoIiJCPXv21NatWyVJjRo10s6dO7V27VrXpbULL8UcPHhQ9957r8LDw+VwONSqVStNnz5dBQUFpR5LSZexvvnmGw0YMED16tVTzZo11bRpU40aNapc/fn11181ZswYRUZGKjAwUN26ddOWLVvKXK+goEDTpk1Ty5Yt5XA4FB4ermHDhunHH38sMnfu3LmKi4tTzZo1FRoaqttvv127d+8uMm/+/Plq0aKFqyfvv/9+uY5B+q3n/fv317Jly3TdddepZs2aatKkif785z+7zSvs4QcffKCnn35aV199tRwOh77//ntJ0ueff66ePXsqJCREQUFB6tq1q1atWuW2jV9++UUPPfSQYmJi5HA4VL9+fXXt2tXt7OGWLVvUv39/199xdHS0+vXr5+rP/v37ZbPZij37d/Hly8LX6ubNm3XnnXfqqquuUtOmTSVJxhjNnj1b7dq1U2BgoK666irdeeed+uGHH8rdO8Cbqs//MgAW07dvX/n7++uLL74odV56err69u2rUaNGqVatWvrvf/+rqVOn6ttvv9Xq1avdtpefn69p06apYcOGOnr0qNatW6eTJ09KkpYtW6Y777xTderU0ezZsyVJDodD0m8/KLt06aJz587ppZdeUqNGjfTJJ59o7Nix2rdvn2t+ea1YsUIDBgxQq1at9Prrr6thw4bav3+/Vq5cWa71n3vuObVv315//etflZ2dreTkZHXv3l1btmxRkyZNSlzv0Ucf1TvvvKORI0eqf//+2r9/vyZOnKjU1FRt3rxZYWFhkqQpU6boueee09ChQzVlyhQdO3ZMycnJ6ty5szZu3KhmzZpJ+i3o/OlPf9LAgQM1ffp0Vy15eXny8yvf/ytu3bpVo0aNUnJysiIjI7Vw4UI99dRTOnfunMaOHes2d8KECercubPeeust+fn5KTw8XAsWLNCwYcM0cOBAvffee7Lb7Xr77bfVu3dvrVixQj179pQk3Xfffdq8ebNeeeUVNW/eXCdPntTmzZt17NgxSdKZM2cUHx+vxo0b6y9/+YsiIiJ05MgRrVmzRqdOnSrXsRRn8ODBGjJkiB555BFXAH/44Yc1f/58Pfnkk5o6daqOHz+uF198UV26dNG2bdtcZzKBSmMAXBbz5s0zkszGjRtLnBMREWFatWrlep6UlGRKe1sWFBQYp9Np1q5daySZbdu2GWOMOXr0qJFkZsyYUWpNrVu3Nt26dSsyPn78eCPJfPPNN27jjz76qLHZbGbPnj2uMUkmKSnJ9XzNmjVGklmzZo1rrGnTpqZp06bm7NmzpdZzscJttW/f3hQUFLjG9+/fb+x2uxkxYoRr7OJe7d6920gyjz32mNs2v/nmGyPJPPfcc8YYY06cOGECAwNN37593eYdPHjQOBwOk5CQYIwxJj8/30RHR5dYS2xsbJnHExsba2w2m9m6davbeHx8vAkJCTFnzpxxO+6bbrrJbd6ZM2dMaGioGTBggNt4fn6+iYuLM506dXKN1a5d24waNarEWjZt2mQkmY8++qjEORkZGUaSmTdvXpFlF/+9F/b/hRdecJu3fv16I8lMnz7dbfzQoUMmMDDQPPvssyXuH7hcuIwF+JAxpsw5P/zwgxISEhQZGSl/f3/Z7XZ169ZNklyXXUJDQ9W0aVO99tprev3117Vly5YyLz9daPXq1br22mvVqVMnt/Hhw4fLGON2Bqkse/fu1b59+/TAAw+oZs2a5V7vQgkJCW6X82JjY9WlSxetWbOmxHUKlw0fPtxtvFOnTmrVqpXrss/69et19uzZIvNiYmJ08803u+bt2bNHhw8fLrGW8mrdurXi4uKKHF9OTo42b97sNn7HHXe4PV+3bp2OHz+uxMREnT9/3vUoKCjQrbfeqo0bN7rOpnTq1Enz58/Xyy+/rA0bNsjpdLpt65prrtFVV12lcePG6a233tKuXbvKfQylubjmTz75RDabTffee69bzZGRkYqLi/PKp/aAiiLsAD5y5swZHTt2TNHR0SXOOX36tP7whz/om2++0csvv6zU1FRt3LhRS5culSSdPXtW0m/3U6xatUq9e/fWtGnT1L59e9WvX19PPvlkuS5RHDt2TFFRUUXGC2srvBRSHr/88oskXdKN1pGRkcWOlVZH4bKSjqNweUXnlVRLeZW2/sXHc3FNP//8syTpzjvvlN1ud3tMnTpVxhgdP35c0m/3diUmJuqvf/2rOnfurNDQUA0bNkxHjhyRJNWpU0dr165Vu3bt9Nxzz6l169aKjo5WUlJSkWBUEcXVbIxRREREkZo3bNigo0ePerwvwFPcswP4yPLly5Wfn6/u3buXOGf16tU6fPiwUlNTXWdzJLnuw7lQbGys/va3v0n67ezKP/7xDyUnJ+vcuXN66623Sq2lXr16yszMLDJ++PBhSXLd61Ie9evXl6Ribwour8If0BeP1atXr8R1CpdlZmYWCVqHDx92HcOF8y5W3LySaimv0ta/+Hguvjm9sJaZM2eW+Im3wvtfwsLCNGPGDM2YMUMHDx7Uxx9/rPHjxysrK0ufffaZJKlt27ZavHixjDH67rvvNH/+fL344osKDAzU+PHjXWfi8vLy3PZRWsgsrmabzaYvv/zSdU/YhYobAy43zuwAPnDw4EGNHTtWderU0cMPP1zivMIfJBf/gLjwE1zFad68uf7nf/5Hbdu2dbtU4nA4XGeDLtSzZ0/t2rWryGWV999/XzabTT169CjzmC7cd9OmTTV37twiPzTL6+9//7vbJb4DBw5o3bp1pQbDm2++WZK0YMECt/GNGzdq9+7drht5O3furMDAwCLzfvzxR61evdo1r0WLFoqKiiqxlvLauXOntm3b5ja2aNEiBQcHl/ndRF27dlXdunW1a9cudezYsdhHQEBAkfUaNmyokSNHKj4+vsjfqfTb6youLk5vvPGG6tat65oTERGhmjVr6rvvvnOb/69//avcx9u/f38ZY/TTTz8VW2/btm3LvS3AWzizA1xmO3bscN23kJWVpS+//FLz5s2Tv7+/li1b5joTUpwuXbroqquu0iOPPKKkpCTZ7XYtXLiwyA/P7777TiNHjtRdd92lZs2aKSAgQKtXr9Z3332n8ePHu+YV/p/9kiVL1KRJE9WsWVNt27bV6NGj9f7776tfv3568cUXFRsbq+XLl2v27Nl69NFH1bx58wod81/+8hcNGDBAN954o0aPHq2GDRvq4MGDWrFihRYuXFjm+llZWbr99tv14IMPKjs7W0lJSapZs6YmTJhQ4jotWrTQQw89pJkzZ8rPz099+vRxfRorJiZGo0ePliTVrVtXEydO1HPPPadhw4Zp6NChOnbsmCZNmqSaNWsqKSlJkuTn56eXXnpJI0aMcNVy8uRJ16eqyis6Olq33XabkpOTFRUVpQULFiglJUVTp04t8ztpateurZkzZyoxMVHHjx/XnXfeqfDwcP3yyy/atm2bfvnlF82ZM0fZ2dnq0aOHEhIS1LJlSwUHB2vjxo367LPPNHjwYEm/3Usze/ZsDRo0SE2aNJExRkuXLtXJkycVHx8vSa57bebOnaumTZsqLi5O3377rRYtWlTu4+3ataseeugh/elPf9KmTZt00003qVatWsrMzNRXX32ltm3b6tFHHy339gCv8Nmt0YDFFX4aq/AREBBgwsPDTbdu3czkyZNNVlZWkXWK+zTWunXrTOfOnU1QUJCpX7++GTFihNm8ebPbp2Z+/vlnM3z4cNOyZUtTq1YtU7t2bXPdddeZN954w5w/f961rf3795tevXqZ4OBgI8ntE0UHDhwwCQkJpl69esZut5sWLVqY1157zeTn57vVo3J8GsuY3z6V06dPH1OnTh3jcDhM06ZNzejRo0vtWeG2PvjgA/Pkk0+a+vXrG4fDYf7whz+YTZs2ldmr/Px8M3XqVNO8eXNjt9tNWFiYuffee82hQ4eK7Ouvf/2rue6660xAQICpU6eOGThwoNm5c2ex85o1a2YCAgJM8+bNzdy5c01iYmK5P43Vr18/8+GHH5rWrVubgIAA06hRI/P6668Xe9z/+7//W+x21q5da/r162dCQ0ON3W43V199tenXr59r/q+//moeeeQRc91115mQkBATGBhoWrRoYZKSklyf+Prvf/9rhg4dapo2bWoCAwNNnTp1TKdOncz8+fPd9pWdnW1GjBhhIiIiTK1atcyAAQPM/v37S/w01i+//FJszXPnzjU33HCDqVWrlgkMDDRNmzY1w4YNK/L3CFQGmzHl+DgIAKDCGjVqpDZt2uiTTz7xdSnAFY17dgAAgKURdgAAgKVxGQsAAFgaZ3YAAIClEXYAAIClEXYAAICl8aWCkgoKCnT48GEFBwcX+epzAABQNRljdOrUKUVHR8vPr+TzN4Qd/fb7cGJiYnxdBgAA8MChQ4dK/eXDhB1JwcHBkn5rVkhISKXt1+l0auXKlerVq5fsdnul7be6o2+eoW8VR888Q988Q98qLicnRzExMa6f4yUh7Oj/ftliSEhIpYedoKAghYSE8MKuAPrmGfpWcfTMM/TNM/TNc2XdgsINygAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNIIOwAAwNJq+LoAoKppNH65V7az/9V+XtkOAODScGYHAABYGmEHAABYGmEHAABYGmEHAABYGmEHAABYGmEHAABYmk/DzhdffKEBAwYoOjpaNptNH330kdtym81W7OO1115zzenevXuR5UOGDKnkIwEAAFWVT8POmTNnFBcXp1mzZhW7PDMz0+0xd+5c2Ww23XHHHW7zHnzwQbd5b7/9dmWUDwAAqgGffqlgnz591KdPnxKXR0ZGuj3/17/+pR49eqhJkyZu40FBQUXmAgAASNXoG5R//vlnLV++XO+9916RZQsXLtSCBQsUERGhPn36KCkpScHBwSVuKy8vT3l5ea7nOTk5kiSn0ymn0+n94ktQuK/K3KcVXO6+OfyNV7ZT1f5eeb1VHD3zDH3zDH2ruPL2ymaM8c6/7JfIZrNp2bJlGjRoULHLp02bpldffVWHDx9WzZo1XePvvvuuGjdurMjISO3YsUMTJkzQNddco5SUlBL3lZycrEmTJhUZX7RokYKCgi75WAAAwOWXm5urhIQEZWdnKyQkpMR51SbstGzZUvHx8Zo5c2ap20lLS1PHjh2Vlpam9u3bFzunuDM7MTExOnr0aKnN8jan06mUlBTFx8fLbrdX2n6ru8vdtzbJK7yynR3Jvb2yHW/h9VZx9Mwz9M0z9K3icnJyFBYWVmbYqRaXsb788kvt2bNHS5YsKXNu+/btZbfblZ6eXmLYcTgccjgcRcbtdrtPXmC+2m91d7n6lpdv88p2qurfKa+3iqNnnqFvnqFv5VfePlWL79n529/+pg4dOiguLq7MuTt37pTT6VRUVFQlVAYAAKo6n57ZOX36tL7//nvX84yMDG3dulWhoaFq2LChpN9OUf3v//6vpk+fXmT9ffv2aeHCherbt6/CwsK0a9cuPf3007r++uvVtWvXSjsOAABQdfk07GzatEk9evRwPR8zZowkKTExUfPnz5ckLV68WMYYDR06tMj6AQEBWrVqld58802dPn1aMTEx6tevn5KSkuTv718pxwAAAKo2n4ad7t27q6z7ox966CE99NBDxS6LiYnR2rVrL0dpAADAIqrFPTsAAACeIuwAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABL8+nvxgKsrNH45V7Zzv5X+3llO22SV2hap9/+zMu3+bweAKgsnNkBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWRtgBAACWVsPXBQCNxi+v0HyHv9G0TlKb5BXKy7e5xve/2s/bpQEALIAzOwAAwNIIOwAAwNIIOwAAwNJ8Gna++OILDRgwQNHR0bLZbProo4/clg8fPlw2m83tceONN7rNycvL0xNPPKGwsDDVqlVLt912m3788cdKPAoAAFCV+TTsnDlzRnFxcZo1a1aJc2699VZlZma6Hv/5z3/clo8aNUrLli3T4sWL9dVXX+n06dPq37+/8vPzL3f5AACgGvDpp7H69OmjPn36lDrH4XAoMjKy2GXZ2dn629/+pg8++EC33HKLJGnBggWKiYnR559/rt69e3u9ZgAAUL1U+Y+ep6amKjw8XHXr1lW3bt30yiuvKDw8XJKUlpYmp9OpXr16ueZHR0erTZs2WrduXYlhJy8vT3l5ea7nOTk5kiSn0ymn03kZj8Zd4b4qc59VkcPfVGy+n3H7s5C3+ljRei43rx1XCX2rqCvp9cp71DP0zTP0reLK2yubMaZK/Mtus9m0bNkyDRo0yDW2ZMkS1a5dW7GxscrIyNDEiRN1/vx5paWlyeFwaNGiRfrTn/7kFlwkqVevXmrcuLHefvvtYveVnJysSZMmFRlftGiRgoKCvHpcAADg8sjNzVVCQoKys7MVEhJS4rwqfWbnnnvucf13mzZt1LFjR8XGxmr58uUaPHhwiesZY2Sz2UpcPmHCBI0ZM8b1PCcnRzExMerVq1epzfI2p9OplJQUxcfHy263V9p+q5o2ySsqNN/hZ/RSxwJN3OSnvIL/+3vekeydy5YVredy89ZxdXjxs2L75qt6qgPeo56hb56hbxVXeGWmLFU67FwsKipKsbGxSk9PlyRFRkbq3LlzOnHihK666irXvKysLHXp0qXE7TgcDjkcjiLjdrvdJy8wX+23qrjwW5ArtF6BzW1db/XQ03ouF68d1/8POBf3zVf1VCdX+nvUU/TNM/St/Mrbp2r1PTvHjh3ToUOHFBUVJUnq0KGD7Ha7UlJSXHMyMzO1Y8eOUsMOAAC4cvj0zM7p06f1/fffu55nZGRo69atCg0NVWhoqJKTk3XHHXcoKipK+/fv13PPPaewsDDdfvvtkqQ6derogQce0NNPP6169eopNDRUY8eOVdu2bV2fzgIAAFc2n4adTZs2qUePHq7nhffRJCYmas6cOdq+fbvef/99nTx5UlFRUerRo4eWLFmi4OBg1zpvvPGGatSoobvvvltnz55Vz549NX/+fPn7+1f68QAAgKrHp2Gne/fuKu3DYCtWlH2jaM2aNTVz5kzNnDnTm6UBAACLqFb37AAAAFQUYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFgaYQcAAFhaDV8XAKB0jcYv98p2HP5e2QwAVDuc2QEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJbm07DzxRdfaMCAAYqOjpbNZtNHH33kWuZ0OjVu3Di1bdtWtWrVUnR0tIYNG6bDhw+7baN79+6y2WxujyFDhlTykQAAgKrKp2HnzJkziouL06xZs4osy83N1ebNmzVx4kRt3rxZS5cu1d69e3XbbbcVmfvggw8qMzPT9Xj77bcro3wAAFAN1PDlzvv06aM+ffoUu6xOnTpKSUlxG5s5c6Y6deqkgwcPqmHDhq7xoKAgRUZGXtZaAQBA9VSt7tnJzs6WzWZT3bp13cYXLlyosLAwtW7dWmPHjtWpU6d8UyAAAKhyfHpmpyJ+/fVXjR8/XgkJCQoJCXGN//GPf1Tjxo0VGRmpHTt2aMKECdq2bVuRs0IXysvLU15enut5Tk6OpN/uE3I6nZfvIC5SuK/K3GdV5PA3FZvvZ9z+LOStPla0nuqipL5V1JX0euU96hn65hn6VnHl7ZXNGFMl/mW32WxatmyZBg0aVGSZ0+nUXXfdpYMHDyo1NdUt7FwsLS1NHTt2VFpamtq3b1/snOTkZE2aNKnI+KJFixQUFOTxMQAAgMqTm5urhIQEZWdnl5oNqnzYcTqduvvuu/XDDz9o9erVqlevXqnbMcbI4XDogw8+0D333FPsnOLO7MTExOjo0aOlNsvbnE6nUlJSFB8fL7vdXmn7rWraJK+o0HyHn9FLHQs0cZOf8gpsrvEdyb19Uk91UVLfKspbfa4OeI96hr55hr5VXE5OjsLCwsoMO1X6MlZh0ElPT9eaNWvKDDqStHPnTjmdTkVFRZU4x+FwyOFwFBm32+0+eYH5ar9VRV6+Zz948wpsbut6q4ee1lNdXNy3iroSX6tX+nvUU/TNM/St/MrbJ5+GndOnT+v77793Pc/IyNDWrVsVGhqq6Oho3Xnnndq8ebM++eQT5efn68iRI5Kk0NBQBQQEaN++fVq4cKH69u2rsLAw7dq1S08//bSuv/56de3a1VeHBQAAqhCfhp1NmzapR48erudjxoyRJCUmJio5OVkff/yxJKldu3Zu661Zs0bdu3dXQECAVq1apTfffFOnT59WTEyM+vXrp6SkJPn7+1facQAAgKrLp2Gne/fuKu2WobJuJ4qJidHatWu9XRYAALCQavU9OwAAABVF2AEAAJZG2AEAAJZG2AEAAJZG2AEAAJZWpb9UEKiIRuOX+7oEAEAVxJkdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaR6FnZtvvlknT54sMp6Tk6Obb775UmsCAADwGo/CTmpqqs6dO1dk/Ndff9WXX355yUUBAAB4S42KTP7uu+9c/71r1y4dOXLE9Tw/P1+fffaZrr76au9VBwAAcIkqFHbatWsnm80mm81W7OWqwMBAzZw5s9zb++KLL/Taa68pLS1NmZmZWrZsmQYNGuRabozRpEmT9M477+jEiRO64YYb9Je//EWtW7d2zcnLy9PYsWP197//XWfPnlXPnj01e/ZsNWjQoCKHBg80Gr/c1yUAAFCmCl3GysjI0L59+2SM0bfffquMjAzX46efflJOTo7uv//+cm/vzJkziouL06xZs4pdPm3aNL3++uuaNWuWNm7cqMjISMXHx+vUqVOuOaNGjdKyZcu0ePFiffXVVzp9+rT69++v/Pz8ihwaAACwqAqd2YmNjZUkFRQUeGXnffr0UZ8+fYpdZozRjBkz9Pzzz2vw4MGSpPfee08RERFatGiRHn74YWVnZ+tvf/ubPvjgA91yyy2SpAULFigmJkaff/65evfu7ZU6AQBA9VWhsHOhvXv3KjU1VVlZWUXCzwsvvHDJhWVkZOjIkSPq1auXa8zhcKhbt25at26dHn74YaWlpcnpdLrNiY6OVps2bbRu3boSw05eXp7y8vJcz3NyciRJTqdTTqfzkmsvr8J9VeY+vcnhb3yzXz/j9ifKx1t9q66vV09U9/eor9A3z9C3iitvrzwKO++++64effRRhYWFKTIyUjabzbXMZrN5JewU3vwcERHhNh4REaEDBw645gQEBOiqq64qMufCm6cvNmXKFE2aNKnI+MqVKxUUFHSppVdYSkpKpe/TG6Z18u3+X+ronTOMV5pL7dt//vMfL1VSfVTX96iv0TfP0Lfyy83NLdc8j8LOyy+/rFdeeUXjxo3zZPUKuTBISb9d3rp47GJlzZkwYYLGjBnjep6Tk6OYmBj16tVLISEhl1ZwBTidTqWkpCg+Pl52u73S9ustbZJX+GS/Dj+jlzoWaOImP+UVlP5awP/xVt92JF85l4er+3vUV+ibZ+hbxRVemSmLR2HnxIkTuuuuuzxZtdwiIyMl/Xb2JioqyjWelZXlOtsTGRmpc+fO6cSJE25nd7KystSlS5cSt+1wOORwOIqM2+12n7zAfLXfS5WX79ugkVdg83kN1dGl9q06vlYvVXV9j/oaffMMfSu/8vbJoy8VvOuuu7Ry5UpPVi23xo0bKzIy0u103rlz57R27VpXkOnQoYPsdrvbnMzMTO3YsaPUsAMAAK4cHp3ZueaaazRx4kRt2LBBbdu2LZKsnnzyyXJt5/Tp0/r+++9dzzMyMrR161aFhoaqYcOGGjVqlCZPnqxmzZqpWbNmmjx5soKCgpSQkCBJqlOnjh544AE9/fTTqlevnkJDQzV27Fi1bdvW9eksAABwZfMo7LzzzjuqXbu21q5dq7Vr17ots9ls5Q47mzZtUo8ePVzPC++jSUxM1Pz58/Xss8/q7Nmzeuyxx1xfKrhy5UoFBwe71nnjjTdUo0YN3X333a4vFZw/f778/f09OTQAAGAxHoWdjIwMr+y8e/fuMqbkj8HabDYlJycrOTm5xDk1a9bUzJkzK/TNzQAA4Mrh0T07AAAA1YVHZ3bK+pUQc+fO9agYAAAAb/P4o+cXcjqd2rFjh06ePFnsLwgFAADwFY/CzrJly4qMFRQU6LHHHlOTJk0uuSgAAABv8do9O35+fho9erTeeOMNb20SAADgknn1BuV9+/bp/Pnz3twkAADAJfHoMtaFv1dK+u13UWVmZmr58uVKTEz0SmEAAADe4FHY2bJli9tzPz8/1a9fX9OnTy/zk1oAAACVyaOws2bNGm/XAQAAcFl4FHYK/fLLL9qzZ49sNpuaN2+u+vXre6suAAAAr/DoBuUzZ87o/vvvV1RUlG666Sb94Q9/UHR0tB544AHl5uZ6u0YAAACPeRR2xowZo7Vr1+rf//63Tp48qZMnT+pf//qX1q5dq6efftrbNQIAAHjMo8tY//znP/Xhhx+qe/furrG+ffsqMDBQd999t+bMmeOt+gAAAC6JR2d2cnNzFRERUWQ8PDycy1gAAKBK8SjsdO7cWUlJSfr1119dY2fPntWkSZPUuXNnrxUHAABwqTy6jDVjxgz16dNHDRo0UFxcnGw2m7Zu3SqHw6GVK1d6u0YAAACPeRR22rZtq/T0dC1YsED//e9/ZYzRkCFD9Mc//lGBgYHerhEAAMBjHoWdKVOmKCIiQg8++KDb+Ny5c/XLL79o3LhxXikOAADgUnl0z87bb7+tli1bFhlv3bq13nrrrUsuCgAAwFs8CjtHjhxRVFRUkfH69esrMzPzkosCAADwFo/CTkxMjL7++usi419//bWio6MvuSgAAABv8eienREjRmjUqFFyOp26+eabJUmrVq3Ss88+yzcoAwCAKsWjsPPss8/q+PHjeuyxx3Tu3DlJUs2aNTVu3DhNmDDBqwUCAABcCo/Cjs1m09SpUzVx4kTt3r1bgYGBatasmRwOh7frAwAAuCQehZ1CtWvX1u9+9ztv1QIAAOB1Ht2gDAAAUF0QdgAAgKURdgAAgKURdgAAgKURdgAAgKURdgAAgKURdgAAgKURdgAAgKURdgAAgKVd0jcoA7jyNBq/3Cvb2f9qP69sBwDKwpkdAABgaYQdAABgaYQdAABgaVU+7DRq1Eg2m63I4/HHH5ckDR8+vMiyG2+80cdVAwCAqqLK36C8ceNG5efnu57v2LFD8fHxuuuuu1xjt956q+bNm+d6HhAQUKk1AgCAqqvKh5369eu7PX/11VfVtGlTdevWzTXmcDgUGRlZ2aUBAIBqoMqHnQudO3dOCxYs0JgxY2Sz2VzjqampCg8PV926ddWtWze98sorCg8PL3E7eXl5ysvLcz3PycmRJDmdTjmdzst3ABcp3Fdl7tObHP7GN/v1M25/onyqWt+qw+u+ur9HfYW+eYa+VVx5e2UzxlSNf/nK4R//+IcSEhJ08OBBRUdHS5KWLFmi2rVrKzY2VhkZGZo4caLOnz+vtLQ0ORyOYreTnJysSZMmFRlftGiRgoKCLusxAAAA78jNzVVCQoKys7MVEhJS4rxqFXZ69+6tgIAA/fvf/y5xTmZmpmJjY7V48WINHjy42DnFndmJiYnR0aNHS22WtzmdTqWkpCg+Pl52u73S9ustbZJX+GS/Dj+jlzoWaOImP+UV2MpeAZKqXt92JPf2dQllqu7vUV+hb56hbxWXk5OjsLCwMsNOtbmMdeDAAX3++edaunRpqfOioqIUGxur9PT0Euc4HI5iz/rY7XafvMB8td9LlZfv2x+YeQU2n9dQHVWVvlWn13x1fY/6Gn3zDH0rv/L2qcp/9LzQvHnzFB4ern79Sv+K+WPHjunQoUOKioqqpMoAAEBVVi3CTkFBgebNm6fExETVqPF/J6NOnz6tsWPHav369dq/f79SU1M1YMAAhYWF6fbbb/dhxQAAoKqoFpexPv/8cx08eFD333+/27i/v7+2b9+u999/XydPnlRUVJR69OihJUuWKDg42EfVAgCAqqRahJ1evXqpuPuoAwMDtWKFb26SBQAA1UO1uIwFAADgKcIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwtBq+LgDAlanR+OVe2c7+V/t5ZTsArKtKn9lJTk6WzWZze0RGRrqWG2OUnJys6OhoBQYGqnv37tq5c6cPKwYAAFVNlQ47ktS6dWtlZma6Htu3b3ctmzZtml5//XXNmjVLGzduVGRkpOLj43Xq1CkfVgwAAKqSKh92atSoocjISNejfv36kn47qzNjxgw9//zzGjx4sNq0aaP33ntPubm5WrRokY+rBgAAVUWVv2cnPT1d0dHRcjgcuuGGGzR58mQ1adJEGRkZOnLkiHr16uWa63A41K1bN61bt04PP/xwidvMy8tTXl6e63lOTo4kyel0yul0Xr6DuUjhvipzn97k8De+2a+fcfsT5WPVvl3O9091f4/6Cn3zDH2ruPL2ymaMqbL/8n366afKzc1V8+bN9fPPP+vll1/Wf//7X+3cuVN79uxR165d9dNPPyk6Otq1zkMPPaQDBw5oxYoVJW43OTlZkyZNKjK+aNEiBQUFXZZjAQAA3pWbm6uEhARlZ2crJCSkxHlVOuxc7MyZM2ratKmeffZZ3XjjjeratasOHz6sqKgo15wHH3xQhw4d0meffVbidoo7sxMTE6OjR4+W2ixvczqdSklJUXx8vOx2e6Xt11vaJJccKC8nh5/RSx0LNHGTn/IKbD6poTqyat92JPe+bNuu7u9RX6FvnqFvFZeTk6OwsLAyw06Vv4x1oVq1aqlt27ZKT0/XoEGDJElHjhxxCztZWVmKiIgodTsOh0MOh6PIuN1u98kLzFf7vVR5+b79gZlXYPN5DdWR1fpWGe+d6voe9TX65hn6Vn7l7VOVv0H5Qnl5edq9e7eioqLUuHFjRUZGKiUlxbX83LlzWrt2rbp06eLDKgEAQFVSpc/sjB07VgMGDFDDhg2VlZWll19+WTk5OUpMTJTNZtOoUaM0efJkNWvWTM2aNdPkyZMVFBSkhIQEX5depXnry9wAAKgOqnTY+fHHHzV06FAdPXpU9evX14033qgNGzYoNjZWkvTss8/q7Nmzeuyxx3TixAndcMMNWrlypYKDg31cOQAAqCqqdNhZvHhxqcttNpuSk5OVnJxcOQUBAIBqp1rdswMAAFBRhB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBphB0AAGBpNXxdAABcikbjl3tlO/tf7eeV7QCoejizAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALI2wAwAALK1Kh50pU6bod7/7nYKDgxUeHq5BgwZpz549bnOGDx8um83m9rjxxht9VDEAAKhqqnTYWbt2rR5//HFt2LBBKSkpOn/+vHr16qUzZ864zbv11luVmZnpevznP//xUcUAAKCqqeHrAkrz2WefuT2fN2+ewsPDlZaWpptuusk17nA4FBkZWdnlAQCAaqBKh52LZWdnS5JCQ0PdxlNTUxUeHq66deuqW7dueuWVVxQeHl7idvLy8pSXl+d6npOTI0lyOp1yOp2XofLiFe6rMvcpSQ5/U6n78zaHn3H7E+VD30pX3PvQV+/R6o6+eYa+VVx5e2UzxlSLf/mMMRo4cKBOnDihL7/80jW+ZMkS1a5dW7GxscrIyNDEiRN1/vx5paWlyeFwFLut5ORkTZo0qcj4okWLFBQUdNmOAQAAeE9ubq4SEhKUnZ2tkJCQEudVm7Dz+OOPa/ny5frqq6/UoEGDEudlZmYqNjZWixcv1uDBg4udU9yZnZiYGB09erTUZnmb0+lUSkqK4uPjZbfbK22/bZJXVNq+LgeHn9FLHQs0cZOf8gpsvi6n2qBvpduR3LvImK/eo9UdffMMfau4nJwchYWFlRl2qsVlrCeeeEIff/yxvvjii1KDjiRFRUUpNjZW6enpJc5xOBzFnvWx2+0+eYFV9n7z8q3xgy6vwGaZY6lM9K14pb0HK/IebTR+uVfq2f9qP69sx5d89W9qdUffyq+8farSYccYoyeeeELLli1TamqqGjduXOY6x44d06FDhxQVFVUJFQIAgKquSn/0/PHHH9eCBQu0aNEiBQcH68iRIzpy5IjOnj0rSTp9+rTGjh2r9evXa//+/UpNTdWAAQMUFham22+/3cfVAwCAqqBKn9mZM2eOJKl79+5u4/PmzdPw4cPl7++v7du36/3339fJkycVFRWlHj16aMmSJQoODvZBxQAAoKqp0mGnrHunAwMDtWJF9b7ZFgAAXF5V+jIWAADApSLsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAASyPsAAAAS6vS37MDAJWluN9p5fA3mtbpt1+ey+8TA6ovzuwAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABLI+wAAABL46Pn1URxH4sFAABl48wOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNL5B+TIr7ZuPHf5G0zpJbZJXKC/fVolVAQBw5eDMDgAAsDTCDgAAsDTCDgAAsDTu2QGAKqi0+/0qYv+r/byyHaA648wOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNMIOAACwNL5nBwBQpop8709pv/eP7/2BL1jmzM7s2bPVuHFj1axZUx06dNCXX37p65IAAEAVYIkzO0uWLNGoUaM0e/Zsde3aVW+//bb69OmjXbt2qWHDhr4uDwDgZXzDNCrCEmd2Xn/9dT3wwAMaMWKEWrVqpRkzZigmJkZz5szxdWkAAMDHqv2ZnXPnziktLU3jx493G+/Vq5fWrVvno6oAAKj+rHIGrdqHnaNHjyo/P18RERFu4xERETpy5Eix6+Tl5SkvL8/1PDs7W5J0/PhxOZ1Or9ZX4/yZkpcVGOXmFqiG00/5BbYS58EdffMMfas4K/Ts2LFjXtlOaf+WFZlbSt98UU9pvFWPNzidTuXm5urYsWOy2+2+LkdS1e/zqVOnJEnGmFLnVfuwU8hmc39DGWOKjBWaMmWKJk2aVGS8cePGl6W20iRU+h6tgb55hr5VXHXvWdh03+y3pL75qp6SVLV6rOpy9/nUqVOqU6dOicurfdgJCwuTv79/kbM4WVlZRc72FJowYYLGjBnjel5QUKDjx4+rXr16JQakyyEnJ0cxMTE6dOiQQkJCKm2/1R198wx9qzh65hn65hn6VnHGGJ06dUrR0dGlzqv2YScgIEAdOnRQSkqKbr/9dtd4SkqKBg4cWOw6DodDDofDbaxu3bqXs8xShYSE8ML2AH3zDH2rOHrmGfrmGfpWMaWd0SlU7cOOJI0ZM0b33XefOnbsqM6dO+udd97RwYMH9cgjj/i6NAAA4GOWCDv33HOPjh07phdffFGZmZlq06aN/vOf/yg2NtbXpQEAAB+zRNiRpMcee0yPPfaYr8uoEIfDoaSkpCKX1FA6+uYZ+lZx9Mwz9M0z9O3ysZmyPq8FAABQjVniG5QBAABKQtgBAACWRtgBAACWRtgBAACWRtipZFOmTJHNZtOoUaNcY8YYJScnKzo6WoGBgerevbt27tzpuyKriJ9++kn33nuv6tWrp6CgILVr105paWmu5fStqPPnz+t//ud/1LhxYwUGBqpJkyZ68cUXVVBQ4JpD36QvvvhCAwYMUHR0tGw2mz766CO35eXpUV5enp544gmFhYWpVq1auu222/Tjjz9W4lFUrtJ65nQ6NW7cOLVt21a1atVSdHS0hg0bpsOHD7tt40rrmVT2a+1CDz/8sGw2m2bMmOE2fiX2zdsIO5Vo48aNeuedd3Tddde5jU+bNk2vv/66Zs2apY0bNyoyMlLx8fGuX3B2JTpx4oS6du0qu92uTz/9VLt27dL06dPdvumavhU1depUvfXWW5o1a5Z2796tadOm6bXXXtPMmTNdc+ibdObMGcXFxWnWrFnFLi9Pj0aNGqVly5Zp8eLF+uqrr3T69Gn1799f+fn5lXUYlaq0nuXm5mrz5s2aOHGiNm/erKVLl2rv3r267bbb3OZdaT2Tyn6tFfroo4/0zTffFPtrD67EvnmdQaU4deqUadasmUlJSTHdunUzTz31lDHGmIKCAhMZGWleffVV19xff/3V1KlTx7z11ls+qtb3xo0bZ37/+9+XuJy+Fa9fv37m/vvvdxsbPHiwuffee40x9K04ksyyZctcz8vTo5MnTxq73W4WL17smvPTTz8ZPz8/89lnn1Va7b5ycc+K8+233xpJ5sCBA8YYemZMyX378ccfzdVXX2127NhhYmNjzRtvvOFaRt+8gzM7leTxxx9Xv379dMstt7iNZ2Rk6MiRI+rVq5drzOFwqFu3blq3bl1ll1llfPzxx+rYsaPuuusuhYeH6/rrr9e7777rWk7fivf73/9eq1at0t69eyVJ27Zt01dffaW+fftKom/lUZ4epaWlyel0us2Jjo5WmzZt6OP/l52dLZvN5jobS8+KV1BQoPvuu0/PPPOMWrduXWQ5ffMOy3yDclW2ePFibd68WRs3biyyrPC3tV/8G9ojIiJ04MCBSqmvKvrhhx80Z84cjRkzRs8995y+/fZbPfnkk3I4HBo2bBh9K8G4ceOUnZ2tli1byt/fX/n5+XrllVc0dOhQSbzeyqM8PTpy5IgCAgJ01VVXFZlTuP6V7Ndff9X48eOVkJDg+oWW9Kx4U6dOVY0aNfTkk08Wu5y+eQdh5zI7dOiQnnrqKa1cuVI1a9YscZ7NZnN7bowpMnYlKSgoUMeOHTV58mRJ0vXXX6+dO3dqzpw5GjZsmGsefXO3ZMkSLViwQIsWLVLr1q21detWjRo1StHR0UpMTHTNo29l86RH9PG3m5WHDBmigoICzZ49u8z5V3LP0tLS9Oabb2rz5s0V7sGV3DdPcBnrMktLS1NWVpY6dOigGjVqqEaNGlq7dq3+/Oc/q0aNGq7/e7w4oWdlZRX5P8srSVRUlK699lq3sVatWungwYOSpMjISEn07WLPPPOMxo8fryFDhqht27a67777NHr0aE2ZMkUSfSuP8vQoMjJS586d04kTJ0qccyVyOp26++67lZGRoZSUFNdZHYmeFefLL79UVlaWGjZs6Pr5cODAAT399NNq1KiRJPrmLYSdy6xnz57avn27tm7d6np07NhRf/zjH7V161Y1adJEkZGRSklJca1z7tw5rV27Vl26dPFh5b7VtWtX7dmzx21s7969rt9k37hxY/pWjNzcXPn5ub+t/f39XR89p29lK0+POnToILvd7jYnMzNTO3bsuGL7WBh00tPT9fnnn6tevXpuy+lZUffdd5++++47t58P0dHReuaZZ7RixQpJ9M1rfHhz9BXrwk9jGWPMq6++aurUqWOWLl1qtm/fboYOHWqioqJMTk6O74r0sW+//dbUqFHDvPLKKyY9Pd0sXLjQBAUFmQULFrjm0LeiEhMTzdVXX20++eQTk5GRYZYuXWrCwsLMs88+65pD3377dOSWLVvMli1bjCTz+uuvmy1btrg+OVSeHj3yyCOmQYMG5vPPPzebN282N998s4mLizPnz5/31WFdVqX1zOl0mttuu800aNDAbN261WRmZroeeXl5rm1caT0zpuzX2sUu/jSWMVdm37yNsOMDF4edgoICk5SUZCIjI43D4TA33XST2b59u+8KrCL+/e9/mzZt2hiHw2Fatmxp3nnnHbfl9K2onJwc89RTT5mGDRuamjVrmiZNmpjnn3/e7QcOfTNmzZo1RlKRR2JiojGmfD06e/asGTlypAkNDTWBgYGmf//+5uDBgz44mspRWs8yMjKKXSbJrFmzxrWNK61nxpT9WrtYcWHnSuybt9mMMabyziMBAABULu7ZAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYAQAAlkbYASzAZrPpo48+8mkN8+fPV926dV3Pk5OT1a5dO4+3N3z4cA0aNKjUOd27d9eoUaM83kd5NWrUSDNmzLjs+wFweRB2gCpq+PDhstlsstlsstvtioiIUHx8vObOnev6XVeFMjMz1adPH6/s9+LQ4qmxY8dq1apVl14QAFwiwg5Qhd16663KzMzU/v379emnn6pHjx566qmn1L9/f50/f941LzIyUg6Hw4eVFlW7du0ivwwS/+fcuXO+LqGIqlgT4A2EHaAKczgcioyM1NVXX6327dvrueee07/+9S99+umnmj9/vmvexZexxo0bp+bNmysoKEhNmjTRxIkT5XQ6Xcu3bdumHj16KDg4WCEhIerQoYM2bdqk1NRU/elPf1J2drbrrFJycrIk6cSJExo2bJiuuuoqBQUFqU+fPkpPTy+x9uIuY82dO1etW7eWw+FQVFSURo4cWWYPJk2apPDwcIWEhOjhhx8u9QdyeWr85z//6aqhUaNGmj59utvyrKwsDRgwQIGBgWrcuLEWLlxYZo2Fl9xKq7V79+4aOXKkxowZo7CwMMXHx0uSdu3apb59+6p27dqKiIjQfffdp6NHj7rW+/DDD9W2bVsFBgaqXr16uuWWW3TmzBlJUmpqqjp16qRatWqpbt266tq1qw4cOOBW04VGjRql7t27X3JNQHVD2AGqmZtvvllxcXFaunRpiXOCg4M1f/587dq1S2+++abeffddvfHGG67lf/zjH9WgQQNt3LhRaWlpGj9+vOx2u7p06aIZM2YoJCREmZmZyszM1NixYyX99sNz06ZN+vjjj7V+/XoZY9S3b1+3EFWaOXPm6PHHH9dDDz2k7du36+OPP9Y111xT6jqrVq3S7t27tWbNGv3973/XsmXLNGnSpBLnl1VjWlqa7r77bg0ZMkTbt29XcnKyJk6c6BYchw8frv3792v16tX68MMPNXv2bGVlZZV5fOWp9b333lONGjX09ddf6+2331ZmZqa6deumdu3aadOmTfrss8/0888/6+6775b02+XJoUOH6v7779fu3buVmpqqwYMHyxij8+fPa9CgQerWrZu+++47rV+/Xg899JBsNluZtV5KTUC15NNfQwqgRImJiWbgwIHFLrvnnntMq1atXM8lmWXLlpW4rWnTppkOHTq4ngcHB5v58+cXO3fevHmmTp06bmN79+41kszXX3/tGjt69KgJDAw0//jHP4pdLykpycTFxbmeR0dHm+eff77EGi+WmJhoQkNDzZkzZ1xjc+bMMbVr1zb5+fnGGGO6detmnnrqqXLXmJCQYOLj493288wzz5hrr73WGGPMnj17jCSzYcMG1/Ldu3cbSUV+E7UntbZr185tvYkTJ5pevXq5jR06dMhIMnv27DFpaWlGktm/f3+RfR47dsxIMqmpqSXWdPHr56mnnjLdunVzPfekJqA64swOUA0ZY0r9P/gPP/xQv//97xUZGanatWtr4sSJOnjwoGv5mDFjNGLECN1yyy169dVXtW/fvlL3t3v3btWoUUM33HCDa6xevXpq0aKFdu/eXWa9WVlZOnz4sHr27FmOo/s/cXFxCgoKcj3v3LmzTp8+rUOHDnlU4+7du9W1a1e39bp27ar09HTl5+e7ttGxY0fX8pYtW5brhu3y1HrhdqXfzjStWbNGtWvXdj1atmwpSdq3b5/i4uLUs2dPtW3bVnfddZfeffddnThxQpIUGhqq4cOHq3fv3howYIDefPNNZWZmllnnxSpaE1AdEXaAamj37t1q3Lhxscs2bNigIUOGqE+fPvrkk0+0ZcsWPf/88273jyQnJ2vnzp3q16+fVq9erWuvvVbLli0rcX/GmBLHy3PZJDAwsMw5FVHcPstTY3H1Xrhe4X9X9FJQaS7cVq1atdyWFRQUaMCAAdq6davbIz09XTfddJP8/f2VkpKiTz/9VNdee61mzpypFi1aKCMjQ5I0b948rV+/Xl26dNGSJUvUvHlzbdiwQZLk5+dXpCfFXXKsaE1AdUTYAaqZ1atXa/v27brjjjuKXf71118rNjZWzz//vDp27KhmzZq5blq9UPPmzTV69GitXLlSgwcP1rx58yRJAQEBys/Pd5t77bXX6vz58/rmm29cY8eOHdPevXvVqlWrMmsODg5Wo0aNKvxR9G3btuns2bOu5xs2bFDt2rXVoEGDInPLU+O1116rr776ym29devWqXnz5vL391erVq10/vx5bdq0ybV8z549OnnypFdrLdS+fXvt3LlTjRo10jXXXOP2KAwhNptNXbt21aRJk7RlyxYFBAS4BdPrr79eEyZM0Lp169SmTRstWrRIklS/fv0iZ3q2bt1a5nGUpyaguiHsAFVYXl6ejhw5op9++kmbN2/W5MmTNXDgQPXv31/Dhg0rdp1rrrlGBw8e1OLFi7Vv3z79+c9/dvvhePbsWY0cOVKpqak6cOCAvv76a23cuNEVCBo1aqTTp09r1apVOnr0qHJzc9WsWTMNHDhQDz74oL766itt27ZN9957r66++moNHDiwXMeSnJys6dOn689//rPS09O1efNmzZw5s9R1zp07pwceeEC7du3Sp59+qqSkJI0cOVJ+fkX/6SpPjU8//bRWrVqll156SXv37tV7772nWbNmuW7CbtGihW699VY9+OCD+uabb5SWlqYRI0aU68xURWot9Pjjj+v48eMaOnSovv32W/3www9auXKl7r//fuXn5+ubb77R5MmTtWnTJh08eFBLly7VL7/8olatWikjI0MTJkzQ+vXrdeDAAa1cudIt2N18883atGmT3n//faWnpyspKUk7duwo8zjKqgmolnx2txCAUiUmJhpJRpKpUaOGqV+/vrnlllvM3LlzXTe9FtJFNyg/88wzpl69eqZ27drmnnvuMW+88Ybr5uG8vDwzZMgQExMTYwICAkx0dLQZOXKkOXv2rGv9Rx55xNSrV89IMklJScYYY44fP27uu+8+U6dOHRMYGGh69+5t9u7d61qnrBuUjTHmrbfeMi1atDB2u91ERUWZJ554otTjHzhwoHnhhRdcxzJixAjz66+/uuZceINyeWo0xpgPP/zQXHvttcZut5uGDRua1157zW15Zmam6devn3E4HKZhw4bm/fffN7GxsWXeoFzRWgvt3bvX3H777aZu3bomMDDQtGzZ0owaNcoUFBSYXbt2md69e5v69esbh8NhmjdvbmbOnGmMMebIkSNm0KBBJioqygQEBJjY2FjzwgsvuL02XnjhBRMREWHq1KljRo8ebUaOHFnkBuWK1gRURzZjSrjQDQAol+HDh+vkyZM+/5UdAIrHZSwAAGBphB0AAGBpXMYCAACWxpkdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaYQdAABgaf8PtXMrw/I89ckAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for feature in continous_features:\n",
    "    data=data.copy()\n",
    "    data[feature].hist(bins=25)\n",
    "    plt.xlabel(feature)\n",
    "    plt.ylabel('count')\n",
    "    plt.title(feature)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BM_6cvY-bqJw"
   },
   "source": [
    "## Checking for outliers\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "5E8XEawiUjFg",
    "outputId": "5172ae13-199b-48b3-b345-538d1d48c61f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGxCAYAAACEFXd4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAor0lEQVR4nO3df1iUdb7/8dcI4wSImHbkh2KyymoFbVZmqYmeBNPUOGTZqle5tXtRVi5pWqh70i6F/JknOf4oLTmVP1ojt68rHdijgaau6B5NzUw7ZP4iWlNBYGGA+/uHxzkRWqgj9/Dx+bguL5r7vuf2fXtdwzy7554Zh2VZlgAAAAzVzO4BAAAAriZiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgdAk/P666/L4XAoJibG7lEANAHEDoAm56233pIk7du3T3/9619tngaAryN2ADQpO3bs0O7du/XAAw9IkpYtW2bzRAB8HbEDoEk5HzevvvqqevbsqVWrVqm8vLzONkePHtWwYcMUHBysVq1aaeTIkSooKJDD4dDy5cvrbLtjxw4NHTpUrVu31nXXXadu3brp/fffb6zDAdAIiB0ATUZFRYVWrlyp7t27KyYmRk888YRKS0v1xz/+0bNNWVmZ+vXrp40bN2rmzJl6//33FRoaquHDh9fb38aNG9WrVy+dPn1aixcv1p/+9CfddtttGj58eL0oAtB0OSzLsuweAgAa4p133tFjjz2mxYsXKzk5WWfPnlV4eLi6deum/Px8SdLChQv1zDPPKDs7W/fff7/nvk899ZSWLFmit99+W6NHj5Yk3XTTTQoICND27dvl7+/v2XbIkCHauXOnjh49qmbN+H9CoKnjUQygyVi2bJkCAgL06KOPSpJatGihhx9+WJs2bdLBgwclSXl5eQoODq4TOpL061//us7tQ4cO6YsvvtDIkSMlSdXV1Z4/gwYN0okTJ3TgwIFGOCoAVxuxA6BJOHTokPLz8/XAAw/IsiydPn1ap0+f1rBhwyT93zu0Tp48qdDQ0Hr3//Gyb7/9VpL0wgsvyOl01vkzZswYSdLf//73q3lIABqJ/89vAgD2e+utt2RZltasWaM1a9bUW5+Zmanp06erTZs22r59e731RUVFdW7fcMMNkqTU1FQlJSVd8O/s0qWLFyYHYDdiB4DPq6mpUWZmpjp16qSlS5fWW79u3TrNnTtX2dnZiouL0/vvv6/s7GwNHDjQs82qVavq3KdLly6Kjo7W7t27lZaWdtWPAYB9iB0APi87O1vHjx/XzJkz1bdv33rrY2JilJGRoWXLlundd9/Va6+9plGjRmn69Onq3LmzsrOz9Z//+Z+SVOeC4yVLlmjgwIEaMGCARo8erXbt2un777/X/v379be//a3Ou7wANF1cswPA5y1btkzNmzfXb37zmwuuv+GGG/Qv//IvWrdunc6ePasNGzaob9++mjhxoh566CF98803WrhwoSSpVatWnvv169dP27dvV6tWrZSSkqL+/fvr6aef1l/+8hf179+/MQ4NQCPgrecArglpaWmaMmWKvvnmG7Vv397ucQA0Il7GAmCcjIwMSVLXrl3ldru1YcMGvf766xo1ahShA1yDiB0AxgkMDNRrr72mr7/+WpWVlerQoYNefPFFTZkyxe7RANiAl7EAAIDRuEAZAAAYjdgBAABGI3YAAIDRuEBZUm1trY4fP67g4GA5HA67xwEAAA1gWZZKS0sVERFR5wNDf4zYkXT8+HFFRkbaPQYAALgMR44c+cmPlSB2JAUHB0s694/VsmVLm6cB4E1ut1s5OTlKSEiQ0+m0exwAXlRSUqLIyEjP8/jFEDuS56Wrli1bEjuAYdxutwIDA9WyZUtiBzDUz12CwgXKAADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKPxoYIAjFVVVaUFCxZow4YNOnTokJ577jk1b97c7rEANDLO7AAw0sSJExUUFKQXXnhB69ev1wsvvKCgoCBNnDjR7tEANDLO7AAwzsSJEzV79myFhoZq2rRpcrlcqqys1Msvv6zZs2dLkmbNmmXzlAAai8OyLMvuIexWUlKikJAQnTlzhu/GApq4qqoqBQUFqU2bNjp69Kgsy9L69es1aNAgORwOtW/fXidPnlRZWRkvaQFNXEOfv3kZC4BRFi5cqOrqak2fPl3+/nVPXvv7++uVV15RdXW1Fi5caNOEABobsQPAKF999ZUkafDgwRdcf375+e0AmI/YAWCUTp06SZLWrVt3wfXnl5/fDoD5uGZHXLMDmIRrdoBrB9fsALgmNW/eXM8//7y+/fZbtW/fXkuXLtX333+vpUuXqn379vr222/1/PPPEzrANYS3ngMwzvm3lb/22msaM2aMZ7m/v78mTJjA286BawwvY4mXsQBT/fATlP/5n/+ZT1AGDNPQ52/O7AAwVvPmzTV27Fh17txZgwYNktPptHskADbgmh0AxqqpqVFeXp7y8/OVl5enmpoau0cCYANbYyc/P19DhgxRRESEHA6H1q5dW2e9ZVmaOnWqIiIiFBAQoL59+2rfvn11tqmsrNRzzz2nG264QUFBQRo6dKiOHj3aiEcBwBdlZWWpc+fOio+P17x58xQfH6/OnTsrKyvL7tEANDJbY6esrEy/+tWvlJGRccH1s2bN0rx585SRkaGCggKFhYUpPj5epaWlnm1SUlL04YcfatWqVdq8ebPOnj2rwYMH839wwDUsKytLw4YNU2xsrDZt2qSVK1dq06ZNio2N1bBhwwge4Fpj+QhJ1ocffui5XVtba4WFhVmvvvqqZ9k//vEPKyQkxFq8eLFlWZZ1+vRpy+l0WqtWrfJsc+zYMatZs2bWxx9/3OC/+8yZM5Yk68yZM1d+IABsVV1dbXXs2NEaMmSIVVNTY1VVVVlr1661qqqqrJqaGmvIkCFWVFSUVV1dbfeoAK5QQ5+/ffYC5cLCQhUVFSkhIcGzzOVyKS4uTlu2bFFycrJ27twpt9tdZ5uIiAjFxMRoy5YtGjBgwAX3XVlZqcrKSs/tkpISSZLb7Zbb7b5KRwSgMeTl5enrr7/WO++8o5qaGs9j+vzPCRMmqE+fPtq4caPi4uLsHBXAFWroc7bPxk5RUZEkKTQ0tM7y0NBQHT582LNN8+bNdf3119fb5vz9LyQ9PV3Tpk2rtzwnJ0eBgYFXOjoAG+Xn50uSjh49qpMnT3qW5+bmSpIqKiokSdnZ2SorK2v8AQF4TXl5eYO289nYOc/hcNS5bVlWvWU/9nPbpKamaty4cZ7bJSUlioyMVEJCAp+zAzRxQUFBmjdvntq3b68ePXrI7XYrNzdX8fHxcjqd2rZtmyRp4MCBnNkBmrjzr8z8HJ+NnbCwMEnnzt6Eh4d7lhcXF3vO9oSFhamqqkqnTp2qc3anuLhYPXv2vOi+XS6XXC5XveVOp5PP4QCauH79+qljx46aNWtWnXd4Op1O+fn5afbs2YqKilK/fv3k5+dn36AArlhDn7N99nN2oqKiFBYW5jn1LJ37NNS8vDxPyNxxxx1yOp11tjlx4oT27t37k7EDwFx+fn6aO3eu1q1bp8TERG3btk0VFRXatm2bEhMTtW7dOs2ZM4fQAa4htp7ZOXv2rA4dOuS5XVhYqF27dql169bq0KGDUlJSlJaWpujoaEVHRystLU2BgYEaMWKEJCkkJERPPvmkxo8frzZt2qh169Z64YUXFBsbq/79+9t1WABslpSUpDVr1mj8+PHq06ePZ3lUVJTWrFmjpKQkG6cD0Nhs/W6sTz75RP369au3/PHHH9fy5ctlWZamTZumJUuW6NSpU+rRo4f+/d//XTExMZ5t//GPf2jChAlasWKFKioqdN9992nhwoWKjIxs8Bx8NxZgppqaGm3cuFHZ2dkaOHAgL10Bhmno8zdfBCpiBzCZ2+3W+vXr+W4swEANff722Wt2AAAAvIHYAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARvO3ewAA+LHy8nJ98cUXXtnX2YpKbdnzla6/YYdaBLi8ss+uXbsqMDDQK/sCcPUROwB8zhdffKE77rjDq/uc5cV97dy5U7fffrsX9wjgaiJ2APicrl27aufOnV7Z14ETpzXuj3s07+FYdQlv5ZV9du3a1Sv7AdA4iB0APicwMNBrZ06aHT4p16YK3RTzK912Yxuv7BNA08IFygAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADCaT8dOdXW1pkyZoqioKAUEBOgXv/iFXnnlFdXW1nq2sSxLU6dOVUREhAICAtS3b1/t27fPxqkBAIAv8enYmTlzphYvXqyMjAzt379fs2bN0uzZs7VgwQLPNrNmzdK8efOUkZGhgoIChYWFKT4+XqWlpTZODgAAfIVPx87WrVv14IMP6oEHHlDHjh01bNgwJSQkaMeOHZLOndWZP3++Jk+erKSkJMXExCgzM1Pl5eVasWKFzdMDAABf4G/3AD+ld+/eWrx4sb788kv98pe/1O7du7V582bNnz9fklRYWKiioiIlJCR47uNyuRQXF6ctW7YoOTn5gvutrKxUZWWl53ZJSYkkye12y+12X70DAtDoqqurPT95fANmaehj2qdj58UXX9SZM2fUtWtX+fn5qaamRjNmzNCvf/1rSVJRUZEkKTQ0tM79QkNDdfjw4YvuNz09XdOmTau3PCcnR4GBgV48AgB2O3JWkvy1bds2Hdtr9zQAvKm8vLxB2/l07KxevVrvvvuuVqxYoVtuuUW7du1SSkqKIiIi9Pjjj3u2czgcde5nWVa9ZT+UmpqqcePGeW6XlJQoMjJSCQkJatmypfcPBIBtdn/zvbRnh+6++279qkNru8cB4EXnX5n5OT4dOxMmTNBLL72kRx99VJIUGxurw4cPKz09XY8//rjCwsIknTvDEx4e7rlfcXFxvbM9P+RyueRyueotdzqdcjqdXj4KAHby9/f3/OTxDZiloY9pn75Auby8XM2a1R3Rz8/P89bzqKgohYWFKTc317O+qqpKeXl56tmzZ6POCgAAfJNPn9kZMmSIZsyYoQ4dOuiWW27Rf//3f2vevHl64oknJJ17+SolJUVpaWmKjo5WdHS00tLSFBgYqBEjRtg8PQAA8AU+HTsLFizQH/7wB40ZM0bFxcWKiIhQcnKy/vVf/9WzzcSJE1VRUaExY8bo1KlT6tGjh3JychQcHGzj5AAAwFc4LMuy7B7CbiUlJQoJCdGZM2e4QBkwzK7DJ5W4aJvWPn23bruxjd3jAPCihj5/+/Q1OwAAAFeK2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABjN3+4BAJil8O9lKqustnsMj6++K/P89Pf3nV95QS5/Rd0QZPcYwDXBdx75AJq8wr+Xqd+cT+we44LGr9lj9wj1bHyhL8EDNAJiB4DXnD+jM3/4berctoXN05xTVlGpdZ9s1eC+9ygowGX3OJKkQ8VnlbJ6l0+dAQNMRuwA8LrObVsopl2I3WNIktxut4r+Sbr9xuvldDrtHgeADbhAGQAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNEuO3aqqqp04MABVVfzRXYAAMB3XXLslJeX68knn1RgYKBuueUWffPNN5KksWPH6tVXX/X6gMeOHdOoUaPUpk0bBQYG6rbbbtPOnTs96y3L0tSpUxUREaGAgAD17dtX+/bt8/ocAACgabrk2ElNTdXu3bv1ySef6LrrrvMs79+/v1avXu3V4U6dOqVevXrJ6XQqOztbn3/+uebOnatWrVp5tpk1a5bmzZunjIwMFRQUKCwsTPHx8SotLfXqLAAAoGnyv9Q7rF27VqtXr9bdd98th8PhWX7zzTfrq6++8upwM2fOVGRkpN5++23Pso4dO3r+27IszZ8/X5MnT1ZSUpIkKTMzU6GhoVqxYoWSk5O9Og8AAGh6Ljl2vvvuO7Vt27be8rKysjrx4w0fffSRBgwYoIcfflh5eXlq166dxowZo9/97neSpMLCQhUVFSkhIcFzH5fLpbi4OG3ZsuWisVNZWanKykrP7ZKSEkmS2+2W2+326jEA15Lz1/BVV1f7zGPp/By+Mo/km/9OQFPU0MfPJcdO9+7d9ec//1nPPfecJHkC580339Q999xzqbv7Sf/zP/+jRYsWady4cZo0aZK2b9+usWPHyuVy6bHHHlNRUZEkKTQ0tM79QkNDdfjw4YvuNz09XdOmTau3PCcnR4GBgV49BuBacuSsJPlr8+bNOtzC7mnqys3NtXsED1/+dwKakvLy8gZtd8mxk56ervvvv1+ff/65qqur9W//9m/at2+ftm7dqry8vEse9KfU1tbqzjvvVFpamiSpW7du2rdvnxYtWqTHHnvMs92PzyhZlvWTZ5lSU1M1btw4z+2SkhJFRkYqISFBLVu29OoxANeSfcdLNGfPNvXu3Vu3RPjGY8ntdis3N1fx8fFyOp12jyPJN/+dgKbo/CszP+eSY6dnz5769NNPNWfOHHXq1Ek5OTm6/fbbtXXrVsXGxl7yoD8lPDxcN998c51lN910kz744ANJUlhYmCSpqKhI4eHhnm2Ki4vrne35IZfLJZfLVW+50+n0mV+GQFPk7+/v+elrjyVfenz78r8T0JQ09PFzybEjSbGxscrMzLycu16SXr166cCBA3WWffnll7rxxhslSVFRUQoLC1Nubq66desm6dzn/+Tl5WnmzJlXfT4AAOD7Ljl2LnbKyOFwyOVyqXnz5lc81HnPP/+8evbsqbS0ND3yyCPavn273njjDb3xxhuevzMlJUVpaWmKjo5WdHS00tLSFBgYqBEjRnhtDgAA0HRdcuy0atXqJ6+Had++vUaPHq2XX35ZzZpd2bdRdO/eXR9++KFSU1P1yiuvKCoqSvPnz9fIkSM920ycOFEVFRUaM2aMTp06pR49eignJ0fBwcFX9HcDAAAzXHLsLF++XJMnT9bo0aN11113ybIsFRQUKDMzU1OmTNF3332nOXPmyOVyadKkSVc84ODBgzV48OCLrnc4HJo6daqmTp16xX8XgCvn8C9RYckBNbvON95mVF1drePVx7X/+/2ea2XsVlhyVg7/hl1YCeDKXfIjPzMzU3PnztUjjzziWTZ06FDFxsZqyZIl+q//+i916NBBM2bM8ErsAGhanK3+qknb0+weo56FHy+0e4Q6nK3ukzTI7jGAa8Ilx87WrVu1ePHiesu7deumrVu3SpJ69+7t+c4sANcW9+kemvvACHVq6ztndj7d/Kl69e7lM2d2vio+q7HvefcT5wFc3CU/8tu3b69ly5bV+9LPZcuWKTIyUpJ08uRJXX/99d6ZEECTYlW3VFTLLrq5TYjdo0g69zk7hf6Fuqn1TT7zNu/af5yRVf2d3WMA14xLjp05c+bo4YcfVnZ2trp37y6Hw6GCggLt37/f8/k3BQUFGj58uNeHBQAAuFSXHDtDhw7Vl19+qUWLFunLL7+UZVkaOHCg1q5dq9OnT0uSnn76aW/PCQAAcFku6wXsG2+80fMy1unTp/Xee+/poYce0q5du1RTU+PVAQEAAK7EZX8QzoYNGzRq1ChFREQoIyNDAwcO1I4dO7w5GwAAwBW7pDM7R48e1fLly/XWW2+prKxMjzzyiNxutz744IN632EFAADgCxp8ZmfQoEG6+eab9fnnn2vBggU6fvy4FixYcDVnAwAAuGINPrOTk5OjsWPH6umnn1Z0dPTVnAkAAMBrGnxmZ9OmTSotLdWdd96pHj16KCMjQ999x+dEAAAA39bg2Lnnnnv05ptv6sSJE0pOTtaqVavUrl071dbWKjc3V6WlpVdzTgAAgMtyye/GCgwM1BNPPKHNmzdrz549Gj9+vF599VW1bdtWQ4cOvRozAgAAXLbLfuu5JHXp0kWzZs3S0aNHtXLlSm/NBAAA4DVXFDvn+fn5KTExUR999JE3dgcAAOA1XokdAAAAX0XsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAo/nbPQAAc1S4ayRJe4+dsXmS/1NWUakd30lhh08pKMBl9ziSpEPFZ+0eAbimEDsAvOar/30Sfylrj82T/Ji/3jlUYPcQ9QS5+BUMNAYeaQC8JuGWMElSp7YtFOD0s3macw6cOKPxa/Zo7rBYdQkPsXscjyCXv6JuCLJ7DOCaQOwA8JrWQc316F0d7B6jjurqaklSp38KUkw734kdAI2HC5QBAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGK1JxU56erocDodSUlI8yyzL0tSpUxUREaGAgAD17dtX+/bts29IAADgU5pM7BQUFOiNN97QrbfeWmf5rFmzNG/ePGVkZKigoEBhYWGKj49XaWmpTZMCAABf0iRi5+zZsxo5cqTefPNNXX/99Z7llmVp/vz5mjx5spKSkhQTE6PMzEyVl5drxYoVNk4MAAB8hb/dAzTEM888owceeED9+/fX9OnTPcsLCwtVVFSkhIQEzzKXy6W4uDht2bJFycnJF9xfZWWlKisrPbdLSkokSW63W263+yodBQA7VFdXe37y+AbM0tDHtM/HzqpVq/S3v/1NBQUF9dYVFRVJkkJDQ+ssDw0N1eHDhy+6z/T0dE2bNq3e8pycHAUGBl7hxAB8yZGzkuSvbdu26dheu6cB4E3l5eUN2s6nY+fIkSP6/e9/r5ycHF133XUX3c7hcNS5bVlWvWU/lJqaqnHjxnlul5SUKDIyUgkJCWrZsuWVDw7AZ+z+5ntpzw7dfffd+lWH1naPA8CLzr8y83N8OnZ27typ4uJi3XHHHZ5lNTU1ys/PV0ZGhg4cOCDp3Bme8PBwzzbFxcX1zvb8kMvlksvlqrfc6XTK6XR68QgA2M3f39/zk8c3YJaGPqZ9+gLl++67T3v27NGuXbs8f+68806NHDlSu3bt0i9+8QuFhYUpNzfXc5+qqirl5eWpZ8+eNk4OAAB8hU+f2QkODlZMTEydZUFBQWrTpo1neUpKitLS0hQdHa3o6GilpaUpMDBQI0aMsGNkAADgY3w6dhpi4sSJqqio0JgxY3Tq1Cn16NFDOTk5Cg4Otns0AADgA5pc7HzyySd1bjscDk2dOlVTp061ZR4AAODbfPqaHQAAgCtF7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjObTsZOenq7u3bsrODhYbdu2VWJiog4cOFBnG8uyNHXqVEVERCggIEB9+/bVvn37bJoYAAD4Gp+Onby8PD3zzDPatm2bcnNzVV1drYSEBJWVlXm2mTVrlubNm6eMjAwVFBQoLCxM8fHxKi0ttXFyAADgK/ztHuCnfPzxx3Vuv/3222rbtq127typPn36yLIszZ8/X5MnT1ZSUpIkKTMzU6GhoVqxYoWSk5PtGBsAAPgQn46dHztz5owkqXXr1pKkwsJCFRUVKSEhwbONy+VSXFyctmzZctHYqaysVGVlped2SUmJJMntdsvtdl+t8QE0UHl5eb2XrC9HRUWFps+cq+O7vtSEz36pKS+OV0BAwBXvt0uXLgoMDLzi/QC4Mg19zm4ysWNZlsaNG6fevXsrJiZGklRUVCRJCg0NrbNtaGioDh8+fNF9paena9q0afWW5+Tk8AsM8AFfffWVxo8f79V9/uXEAf0l+/95ZV9z585Vp06dvLIvAJevvLy8Qds1mdh59tln9dlnn2nz5s311jkcjjq3Lcuqt+yHUlNTNW7cOM/tkpISRUZGKiEhQS1btvTe0AAuS3l5uXr37n3Z958wYYLy8/PldDr10MOPKLhtB5UWf6MP/vi+3G63+vTpo9mzZ1/2/jmzA/iG86/M/JwmETvPPfecPvroI+Xn56t9+/ae5WFhYZLOneEJDw/3LC8uLq53tueHXC6XXC5XveVOp1NOp9OLkwO4HCEhIbrrrrsu674VFRXKz89X8+bNVVpaKofDofXr12vQoGnKfPstBQcHKz8/X7GxsV55SQuAfRr6nO3T78ayLEvPPvussrKytGHDBkVFRdVZHxUVpbCwMOXm5nqWVVVVKS8vTz179mzscQH4gAkTJkiSxo0bp+bNm9dZ17x5c6WkpNTZDoD5fDp2nnnmGb377rtasWKFgoODVVRUpKKiIlVUVEg69/JVSkqK0tLS9OGHH2rv3r0aPXq0AgMDNWLECJunB2CHgwcPSpJ++9vfXnD9k08+WWc7AObz6dhZtGiRzpw5o759+yo8PNzzZ/Xq1Z5tJk6cqJSUFI0ZM0Z33nmnjh07ppycHAUHB9s4OQC7REdHS5KWLl16wfXLli2rsx0A8zksy7LsHsJuJSUlCgkJ0ZkzZ7hAGWjiKioqFBgYeIFrdgbJsiwFBwerqqpK5eXlXLMDNHENff726TM7AHCpAgIC9OCDD6qqqkrBwcGaNGmSjh07pkmTJnlC58EHHyR0gGsIZ3bEmR3ARImJifrTn/5Ub/mDDz6otWvXNv5AALyOMzsArmlr165VeXm5nnrqKd1222166qmnVF5eTugA16Am8Tk7AHA5AgIC9Prrr3uu2eFztIBrE2d2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgeAsWpqapSXl6f8/Hzl5eWppqbG7pEA2IDYAWCkrKwsde7cWfHx8Zo3b57i4+PVuXNnZWVl2T0agEZG7AAwTlZWloYNG6bY2Fht2rRJK1eu1KZNmxQbG6thw4YRPMA1hu/GEt+NBZikpqZGnTt3VmxsrNauXauamhrPJyj7+fkpMTFRe/fu1cGDB+Xn52f3uACuAN+NBeCatGnTJn399deaNGmSmjWr+yuuWbNmSk1NVWFhoTZt2mTThAAaG7EDwCgnTpyQJMXExFxw/fnl57cDYD5iB4BRwsPDJUl79+694Przy89vB8B8xA4Ao9x7773q2LGj0tLSVFtbW2ddbW2t0tPTFRUVpXvvvdemCQE0NmIHgFH8/Pw0d+5crVu3TomJidq2bZsqKiq0bds2JSYmat26dZozZw4XJwPXEH+7BwAAb0tKStKaNWs0fvx49enTx7M8KipKa9asUVJSko3TAWhsvPVcvPUcMFVNTY02btyo7OxsDRw4UP369eOMDmCQhj5/c2YHgLH8/PwUFxensrIyxcXFETrANYprdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDR+ARlSee/MaOkpMTmSQB4m9vtVnl5uUpKSuR0Ou0eB4AXnX/e/rlvviJ2JJWWlkqSIiMjbZ4EAABcqtLSUoWEhFx0PV8EKqm2tlbHjx9XcHCwHA6H3eMA8KKSkhJFRkbqyJEjfNEvYBjLslRaWqqIiAg1a3bxK3OIHQBGa+i3IgMwFxcoAwAAoxE7AADAaMQOAKO5XC69/PLLcrlcdo8CwCZcswMAAIzGmR0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiB0CTtWXLFvn5+en++++3exQAPoy3ngNosn7729+qRYsWWrp0qT7//HN16NDB7pEA+CDO7ABoksrKyvT+++/r6aef1uDBg7V8+fI66z/66CNFR0crICBA/fr1U2ZmphwOh06fPu3ZZsuWLerTp48CAgIUGRmpsWPHqqysrHEPBMBVR+wAaJJWr16tLl26qEuXLho1apTefvttnT9R/fXXX2vYsGFKTEzUrl27lJycrMmTJ9e5/549ezRgwAAlJSXps88+0+rVq7V582Y9++yzdhwOgKuIl7EANEm9evXSI488ot///veqrq5WeHi4Vq5cqf79++ull17Sn//8Z+3Zs8ez/ZQpUzRjxgydOnVKrVq10mOPPaaAgAAtWbLEs83mzZsVFxensrIyXXfddXYcFoCrgDM7AJqcAwcOaPv27Xr00UclSf7+/ho+fLjeeustz/ru3bvXuc9dd91V5/bOnTu1fPlytWjRwvNnwIABqq2tVWFhYeMcCIBG4W/3AABwqZYtW6bq6mq1a9fOs8yyLDmdTp06dUqWZcnhcNS5z49PYtfW1io5OVljx46tt38udAbMQuwAaFKqq6v1H//xH5o7d64SEhLqrHvooYf03nvvqWvXrlq/fn2ddTt27Khz+/bbb9e+ffvUuXPnqz4zAHtxzQ6AJmXt2rUaPny4iouLFRISUmfd5MmTtX79emVlZalLly56/vnn9eSTT2rXrl0aP368jh49qtOnTyskJESfffaZ7r77bv3mN7/R7373OwUFBWn//v3Kzc3VggULbDo6AFcD1+wAaFKWLVum/v371wsd6dyZnV27dunUqVNas2aNsrKydOutt2rRokWed2O5XC5J0q233qq8vDwdPHhQ9957r7p166Y//OEPCg8Pb9TjAXD1cWYHwDVhxowZWrx4sY4cOWL3KAAaGdfsADDSwoUL1b17d7Vp00affvqpZs+ezWfoANcoYgeAkQ4ePKjp06fr+++/V4cOHTR+/HilpqbaPRYAG/AyFgAAMBoXKAMAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACM9v8By5o67G4gfTAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGxCAYAAACEFXd4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABSH0lEQVR4nO3dd1gU1/4/8PcCy9JRUFpERCyYiNhrFIyCYu/x6o016o1GxXJN0ChgUBITNdFEvbnGXq+K3RhsIBZMRIlii1FAjWCHpQkLnN8f/tivK6C7srgwvl/Ps4/OmTOznyFZ9u2ZMzMyIYQAERERkUQZGboAIiIiovLEsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ/SGnDlzBn379kXNmjWhUCjg6OiINm3aYNq0aeXyfnfv3kVISAji4+Nfex9r1qyBTCZDUlKSum3EiBGoVatWmesr2peVlZVWfWvVqoURI0bo5X11JZPJEBISYpD3JqKyY9ghegP279+Ptm3bQqlUYsGCBYiMjMT333+Pdu3aYevWreXynnfv3kVoaGiZwk5JZs+ejZ07d+p1n0RE5cnE0AUQvQ0WLFgAd3d3/PrrrzAx+b+P3eDBg7FgwQIDVqY7Dw8PQ5cgSdnZ2bCwsDB0GTrJycmBubm5ocsgeiWO7BC9AY8ePUK1atU0gk4RI6P/+xiOHj0adnZ2yM7OLtbvgw8+wHvvvade3rZtG1q1agVbW1tYWFigdu3aGDVqFAAgKioKLVq0AACMHDkSMpms2KmYPXv2oE2bNrCwsIC1tTX8/Pxw+vTpVx5LSaexCgsLsXTpUjRu3Bjm5uaoUqUKWrdujT179rxyfwBw6dIldOrUCZaWlqhevTo+/fTTEn8GL7p16xb++c9/wsHBAQqFAg0aNMDChQtRWFio0e/x48cYP3483nnnHZiamqJ27dqYNWsWcnNzNfoplUqMGTMG9vb2sLKyQteuXfHnn39qdQxRUVGQyWTYsGEDpk6dCicnJ5ibm8PHxwfnz5/X6Ft0+u7ixYvw9/eHtbU1OnXqBADIy8tDWFgYPD09oVAoUL16dYwcORIPHjzQ2MfRo0fh6+sLe3t7mJubo2bNmujfv7/Gz2358uXw9vaGlZUVrK2t4enpiZkzZ6rXh4SEQCaTFTuWkk5f1qpVCz169EBERASaNGkCMzMzhIaGAgBSU1Mxbtw41KhRA6ampnB3d0doaCjy8/O1+tkRlTeO7BC9AW3atMHKlSsxadIkDB06FE2bNoVcLi/Wb/LkyVi1ahU2bdqEjz/+WN1++fJlHDt2DD/++CMA4PTp0/jwww/x4YcfIiQkBGZmZkhOTsbRo0cBAE2bNsXq1asxcuRIfPHFF+jevTsAoEaNGgCATZs2YejQofD398fmzZuRm5uLBQsWwNfXF0eOHMH777+v0/GNGDECGzZswOjRozF37lyYmpri3LlzGl+WpVGpVOjWrRvGjRuHzz//HKdOnUJYWBiSk5Oxd+/eUrd78OAB2rZti7y8PHz55ZeoVasW9u3bh+nTp+PGjRtYtmwZAODp06fo2LEjbty4gdDQUDRq1AgxMTEIDw9HfHw89u/fDwAQQqBPnz44deoU5syZgxYtWuDkyZMICAjQ6Wcxc+ZMNG3aFCtXrkR6ejpCQkLg6+uL8+fPo3bt2up+eXl56NWrl/q48/PzUVhYiN69eyMmJgYzZsxA27ZtkZycjODgYPj6+uLs2bMwNzdHUlISunfvjvbt22PVqlWoUqUK/v77bxw8eBB5eXmwsLDAli1bMH78eEycOBHffvstjIyM8Ndff+Hy5cs6Hc/zzp07hytXruCLL76Au7s7LC0tkZqaipYtW8LIyAhz5syBh4cHTp8+jbCwMCQlJWH16tWv/X5EeiOIqNw9fPhQvP/++wKAACDkcrlo27atCA8PFxkZGRp9fXx8ROPGjTXaPvnkE2FjY6Pu++233woAIi0trdT3/P333wUAsXr1ao32goIC4eLiIry8vERBQYG6PSMjQzg4OIi2bduq21avXi0AiMTERHXb8OHDhZubm3r5+PHjAoCYNWuWtj8OjX0BEN9//71G+7x58wQAceLECXWbm5ubGD58uHr5888/FwDEmTNnNLb95JNPhEwmE9euXRNCCLFixQoBQPzvf//T6Pf1118LACIyMlIIIcQvv/zy0lqCg4NfeizHjh0TAETTpk1FYWGhuj0pKUnI5XLx8ccfFzvuVatWaexj8+bNAoDYsWOHRnvRf8tly5YJIYTYvn27ACDi4+NLrefTTz8VVapUeWnNwcHBoqSvgZL+u7u5uQljY2P1z7XIuHHjhJWVlUhOTtZoL/p/9NKlSy+tgehN4GksojfA3t4eMTEx+P333/HVV1+hd+/e+PPPPxEUFAQvLy88fPhQ3Xfy5MmIj4/HyZMnATw7tbJ+/XoMHz5cfeVS0SmqQYMG4X//+x/+/vtvrWu5du0a7t69i48++kjjFJqVlRX69++P2NhYrU4hFfnll18AABMmTNB6mxcNHTpUY3nIkCEAgGPHjpW6zdGjR/Huu++iZcuWGu0jRoyAEEI9ynX06FFYWlpiwIABxfoBwJEjRzTeq7RatDVkyBCNU0Nubm5o27ZticfSv39/jeV9+/ahSpUq6NmzJ/Lz89Wvxo0bw8nJCVFRUQCAxo0bw9TUFGPHjsXatWtx8+bNYvtu2bIl0tLS8I9//AO7d+/W+H/sdTVq1Aj16tUrVnPHjh3h4uKiUXPRiFh0dHSZ35eorBh2iN6g5s2b47PPPsO2bdtw9+5dTJkyBUlJSRqTlHv37o1atWqpT1mtWbMGWVlZGmGiQ4cO2LVrF/Lz8zFs2DDUqFEDDRs2xObNm19Zw6NHjwAAzs7Oxda5uLigsLAQT5480fqYHjx4AGNjYzg5OWm9zfNMTExgb2+v0Va0r6JaS/Lo0aNSj+H5bR89egQnJ6dic1McHBxgYmKi0e9ltWirpP5OTk7FjsXCwgI2NjYabffu3UNaWhpMTU0hl8s1XqmpqerA4uHhgcOHD8PBwQETJkyAh4cHPDw88P3336v39dFHH2HVqlVITk5G//794eDggFatWuHQoUM6Hc/zSvp537t3D3v37i1Wb9H8Mn2ELKKyYtghMhC5XI7g4GAAQEJCgrrdyMgIEyZMwPbt25GSkoJly5ahU6dOqF+/vsb2vXv3xpEjR5Ceno6oqCjUqFEDQ4YMeeUk46Iv85SUlGLr7t69CyMjI1StWlXr46hevToKCgqQmpqq9TbPy8/PLxYEivb1YvB4nr29fanHAADVqlVT97t37x6EEBr97t+/j/z8fI1+L6tFWyX1T01NLXYsJU0MrlatGuzt7fH777+X+CqahwQA7du3x969e5Geno7Y2Fi0adMGgYGB2LJli7rPyJEjcerUKaSnp2P//v0QQqBHjx5ITk4GAJiZmQFAsYnapQWU0mr29/cvtebRo0eX9qMiemMYdojegJK+lAHgypUrAP5vNKLIxx9/DFNTUwwdOhTXrl3Dp59+Wuq+FQoFfHx88PXXXwOA+sofhUIB4Nnlwc+rX78+3nnnHWzatEkjAGRlZWHHjh3qK7S0VXS6Yvny5Vpv86KNGzdqLG/atAkA4OvrW+o2nTp1wuXLl3Hu3DmN9nXr1kEmk6Fjx47qfpmZmdi1a1exfkXrAaj7l1aLtjZv3qzxc01OTsapU6deeixFevTogUePHqGgoADNmzcv9nox8AKAsbExWrVqpR4JfPHnAQCWlpYICAjArFmzkJeXh0uXLgGA+qq6CxcuaPR/2cTwkmpOSEiAh4dHiTW/+P82kSHwaiyiN6BLly6oUaMGevbsCU9PTxQWFiI+Ph4LFy6ElZUVJk+erNG/SpUqGDZsGJYvXw43Nzf07NlTY/2cOXNw584ddOrUCTVq1EBaWhq+//57yOVy+Pj4AHh2qsPc3BwbN25EgwYNYGVlBRcXF7i4uGDBggUYOnQoevTogXHjxiE3NxfffPMN0tLS8NVXX+l0bO3bt8dHH32EsLAw3Lt3Dz169IBCocD58+dhYWGBiRMnvnR7U1NTLFy4EJmZmWjRooX6aqyAgICXXhU2ZcoUrFu3Dt27d8fcuXPh5uaG/fv3Y9myZfjkk0/Uc0uGDRuGH3/8EcOHD0dSUhK8vLxw4sQJzJ8/H926dUPnzp0BAP7+/ujQoQNmzJiBrKwsNG/eHCdPnsT69et1+nncv38fffv2xZgxY5Ceno7g4GCYmZkhKCjoldsOHjwYGzduRLdu3TB58mS0bNkScrkcd+7cwbFjx9C7d2/07dsXK1aswNGjR9G9e3fUrFkTT58+xapVqwBAfTxjxoyBubk52rVrB2dnZ6SmpiI8PBy2trbqOV/dunWDnZ2d+io6ExMTrFmzBrdv39b6eOfOnYtDhw6hbdu2mDRpEurXr4+nT58iKSkJBw4cwIoVK9RXARIZjEGnRxO9JbZu3SqGDBki6tatK6ysrIRcLhc1a9YUH330kbh8+XKJ20RFRQkA4quvviq2bt++fSIgIEC88847wtTUVDg4OIhu3bqJmJgYjX6bN28Wnp6eQi6XF7uiaNeuXaJVq1bCzMxMWFpaik6dOomTJ09qbK/N1VhCPLvCa/HixaJhw4bC1NRU2NraijZt2oi9e/e+9OcyfPhwYWlpKS5cuCB8fX2Fubm5sLOzE5988onIzMzU6Pvi1VhCCJGcnCyGDBki7O3thVwuF/Xr1xfffPONxlVmQgjx6NEj8a9//Us4OzsLExMT4ebmJoKCgsTTp081+qWlpYlRo0aJKlWqCAsLC+Hn5yeuXr2q09VY69evF5MmTRLVq1cXCoVCtG/fXpw9e7bE4y6JSqUS3377rfD29hZmZmbCyspKeHp6inHjxonr168LIYQ4ffq06Nu3r3BzcxMKhULY29sLHx8fsWfPHvV+1q5dKzp27CgcHR2FqampcHFxEYMGDRIXLlzQeL/ffvtNtG3bVlhaWop33nlHBAcHi5UrV5Z4NVb37t1LrPnBgwdi0qRJwt3dXcjlcmFnZyeaNWsmZs2aVey/I5EhyIR44UQ2EVUI06ZNw/Lly3H79u2Xzl2hiiEqKgodO3bEtm3bil35RUSGxdNYRBVMbGws/vzzTyxbtgzjxo1j0CEiKiOGHaIKpmiCcI8ePRAWFmbocoiIKj2exiIiIiJJ46XnREREJGkMO0RERCRpDDtEREQkaZygDKCwsBB3796FtbV1ibdDJyIioopHCIGMjAy4uLhoPNj4RQw7ePYsHVdXV0OXQURERK/h9u3bL71TN8MOAGtrawDPflgvPoWYiCo3lUqFyMhI+Pv7Qy6XG7ocItIjpVIJV1dX9fd4aRh28H9P8rWxsWHYIZIYlUoFCwsL2NjYMOwQSdSrpqBwgjIRERFJGsMOERERSRrDDhEREUkaww4RERFJGsMOERERSRrDDhEREUmaQcNOeHg4WrRoAWtrazg4OKBPnz64du2aer1KpcJnn30GLy8vWFpawsXFBcOGDcPdu3c19uPr6wuZTKbxGjx48Js+HCIiIqqADBp2oqOjMWHCBMTGxuLQoUPIz8+Hv78/srKyAADZ2dk4d+4cZs+ejXPnziEiIgJ//vknevXqVWxfY8aMQUpKivr1n//8500fDhEREVVABr2p4MGDBzWWV69eDQcHB8TFxaFDhw6wtbXFoUOHNPosXboULVu2xK1bt1CzZk11u4WFBZycnN5I3URUOeTk5GDq1KmIjY3FwYMHsWjRIpibmxu6LCJ6wyrUHZTT09MBAHZ2di/tI5PJUKVKFY32jRs3YsOGDXB0dERAQACCg4NLvX10bm4ucnNz1ctKpRLAs9NmKpWqjEdBRBVB//79sXfvXvVyfHw8VqxYgZ49e2LHjh0GrIyI9EXb72yZEEKUcy1aEUKgd+/eePLkCWJiYkrs8/TpU7z//vvw9PTEhg0b1O3//e9/4e7uDicnJyQkJCAoKAh16tQpNipUJCQkBKGhocXaN23aBAsLC/0cEBEZzPz58/Hbb7/BxMQEvXr1gp+fHw4dOoQ9e/YgPz8fLVu2xMyZMw1dJhGVUXZ2NoYMGYL09PSXPu6pwoSdCRMmYP/+/Thx4kSJTy5VqVQYOHAgbt26haioqJceVFxcHJo3b464uDg0bdq02PqSRnZcXV3x8OFDPhuLqJLLycmBra0tTE1N8fjxY8hkMhw6dAh+fn4QQsDOzg55eXlIT0/nKS2iSk6pVKJatWqvDDsV4jTWxIkTsWfPHhw/frzUoDNo0CAkJibi6NGjrwwkTZs2hVwux/Xr10sMOwqFAgqFoli7XC7ngwKJKrkpU6YAAKZOnQpLS0v1MHfR5zswMBALFizAzJkz8cMPPxiyVCIqI22/sw16NZYQAp9++ikiIiJw9OhRuLu7F+tTFHSuX7+Ow4cPw97e/pX7vXTpElQqFZydncujbCKqwK5fvw4A+Pjjj0tcP3r0aI1+RCR9Bg07EyZMwIYNG7Bp0yZYW1sjNTUVqampyMnJAQDk5+djwIABOHv2LDZu3IiCggJ1n7y8PADAjRs3MHfuXJw9exZJSUk4cOAABg4ciCZNmqBdu3aGPDwiMoC6desCAFauXIm8vDwsWbIEP/30E5YsWYK8vDz8/PPPGv2ISPoMOmdHJpOV2L569WqMGDECSUlJJY72AMCxY8fg6+uL27dv45///CcSEhKQmZkJV1dXdO/eHcHBwS+9qut5SqUStra2rzznR0QVX05ODiwsLGBkZASZTIaCggL1OmNjYwghUFhYiOzsbM7ZIarktP3+NuicnVflrFq1ar2yj6urK6Kjo/VZFhFVYubm5vDw8MCNGzcAAP7+/vD19UVUVBQiIyMBAB4eHgw6RG+RCnM1liFxZIdIOvLy8mBpaQkjIyP16e7nmZqaorCwEFlZWTA1NTVAhUSkL9p+f/NBoEQkKcuWLUN+fj5+/PFHZGdn41//+hcaN26Mf/3rX8jOzsbSpUuRn5+PZcuWGbpUInpDKsSl50RE+lJ0+qpHjx4wNzfHkiVLcODAAXTr1g1yuRw9evTQ6EdE0seRHSKSFA8PDwDAvn37Slxf1F7Uj4ikj3N2wDk7RFJSNGfH3t4ed+7cgRBCPbIjk8lQo0YNPHr0iHN2iCSAc3aI6K1kamqKKVOm4N69e6hRowZWrlyJx48fY+XKlahRowbu3buHKVOmMOgQvUU4Z4eIJGfBggUAgMWLF2P8+PHqdhMTE/z73/9WryeitwNPY4GnsYikKi8vD0uXLsXRo0fxwQcfYOLEiRzRIZIQnsYiordeZmYmVq1ahZiYGKxatQqZmZmGLomIDICnsYhIkpycnHDv3j318uXLl2Fvbw9HR0ekpqYasDIietM4skNEkvN80GnVqhVCQ0PRqlUrAMC9e/fg5ORkyPKI6A1j2CEiSXn8+LE66GRkZCAmJgbe3t6IiYlBRkYGgGeB5/Hjx4Ysk4jeIIYdIpIUHx8fAEDr1q1hZWWlsc7KygotW7bU6EdE0sewQ0SScvfuXQDAvHnzSlw/d+5cjX5EJH0MO0QkKS4uLgCAWbNmlbh+zpw5Gv2ISPoYdohIUqKjowEAsbGxxS41z8zMxG+//abRj4ikj2GHiCTFzs4Ojo6OAABra2u0a9cO586dQ7t27WBtbQ0AcHR0hJ2dnSHLJKI3iHdQBu+gTCRFL95npwjvs0MkHbyDMhG91YYNGwZjY2ONNmNjYwwbNsxAFRGRofAOykQkOTNmzMA333yD6tWrw9nZGY8ePYK9vT1SUlLwzTffAAAfBkr0FuFpLPA0FpGU5OXlwdLSEgCQn59fbL2JybN/42VlZfGhoESVHE9jEdFbadmyZcjPz0d+fj5kMhn++c9/YvHixfjnP/8JmUymXrds2TJDl0pEbwjDDhFJypUrVwAAMpkM2dnZWLVqFdzd3bFq1SpkZ2dDJpNp9CMi6WPYISJJOXXqFIBnDwA1MzPTWGdmZoYWLVpo9CMi6WPYISJJ+vPPP4vN2cnPz8dff/1loIqIyFAYdohIUurUqQPg2dPPa9SogZUrV+Lx48dYuXIlatSooX7aeVE/IpI+Xo0FXo1FJCWZmZnqOyUbGRmhsLBQvc7Y2BgFBQUAgIyMjGJPRSeiyoVXYxHRW8nKyko9L0cIgTp16qBu3bqoU6eOOvi0aNGCQYfoLcKbChKR5Pz222+oU6cObty4UWyOjoeHh/phoET0dmDYISLJiYiIwM2bNxEQEAClUolbt26hZs2asLGxwcGDBxEREYF+/foZukwiekN4GouIJKWgoADTpk1Djx49sG3bNnh5ecHe3h5eXl7Ytm0bevTogenTp6vn7hCR9DHsEJGkxMTEICkpCWlpabCyssKKFSsQHx+PFStWwMrKCmlpaUhMTERMTIyhSyWiN4SnsYhIUlJSUgA8Cz2mpqYIDAyEu7s7EhMT8d1336lDTlE/IpI+juwQkaRUqVIFACCXy5GRkYGwsDA4OzsjLCwMGRkZkMvlGv2ISPoYdohIUvbs2QMAqFmzJlQqFSZNmoSQkBBMmjQJKpUKNWvW1OhHRNJn0LATHh6OFi1awNraGg4ODujTpw+uXbum0UcIgZCQELi4uMDc3By+vr64dOmSRp/c3FxMnDgR1apVg6WlJXr16oU7d+68yUMhogri5s2bAIAbN26UOGfnxo0bGv2ISPoMGnaio6MxYcIExMbG4tChQ8jPz4e/vz+ysrLUfRYsWIBFixbhhx9+wO+//w4nJyf4+fkhIyND3ScwMBA7d+7Eli1bcOLECWRmZqJHjx682oLoLVS3bl299iOiyq9CPS7iwYMHcHBwQHR0NDp06AAhBFxcXBAYGIjPPvsMwLNRHEdHR3z99dcYN24c0tPTUb16daxfvx4ffvghAODu3btwdXXFgQMH0KVLl1e+Lx8XQSQdjx8/hr29/Sv7PXr0CHZ2dm+gIiIqL9p+f1eoq7HS09MBQP0LKDExEampqfD391f3USgU8PHxwalTpzBu3DjExcVBpVJp9HFxcUHDhg1x6tSpEsNObm4ucnNz1ctKpRIAoFKpoFKpyuXYiOjNmDVrlsZy9erVYW1tjYyMDDx48ECj35IlS950eUSkR9p+Z1eYsCOEwNSpU/H++++jYcOGAIDU1FQAgKOjo0ZfR0dHJCcnq/uYmpqiatWqxfoUbf+i8PBwhIaGFmuPjIyEhYVFmY+FiAzn5MmTGssPHjzQCDnP9ztw4MCbKouIykF2drZW/SpM2Pn0009x4cIFnDhxotg6mUymsSyEKNb2opf1CQoKwtSpU9XLSqUSrq6u8Pf352ksokpu5syZ6r+bmpqiQYMGyM3NhUKhwJUrV5CXlwfg2Z2Wu3XrZqgyiUgPis7MvEqFCDsTJ07Enj17cPz4cdSoUUPd7uTkBODZ6I2zs7O6/f79++rRHicnJ+Tl5eHJkycaozv3799H27ZtS3w/hUIBhUJRrF0ul6vvwUFElZODgwMuX74M4Nm8HIVCgQMHDqBbt27Izc2FtbW1uh8/70SVm7afYYNejSWEwKeffoqIiAgcPXoU7u7uGuvd3d3h5OSEQ4cOqdvy8vIQHR2tDjLNmjWDXC7X6JOSkoKEhIRSww4RSdfDhw/Vf7e3t8fMmTPx999/Y+bMmRoTl5/vR0TSZtCRnQkTJmDTpk3YvXs3rK2t1XNsbG1tYW5uDplMhsDAQMyfPx9169ZF3bp1MX/+fFhYWGDIkCHqvqNHj8a0adNgb28POzs7TJ8+HV5eXujcubMhD4+IDMDR0REJCQkAnv3j6Ntvvy21HxG9HQwadpYvXw4A8PX11WhfvXo1RowYAQCYMWMGcnJyMH78eDx58gStWrVCZGSkeigaABYvXgwTExMMGjQIOTk56NSpE9asWQNjY+M3dShEVEF4enriyJEjAJ7N2Xn33XeRk5MDc3NzXL58WT1nx9PT05BlEtEbVKHus2MovM8OkXTk5OTAwsICMpkMJf16K2rPzs6Gubm5ASokIn3R9vubz8YiIkkxNzdH7969IYSAqakpGjdujPr166Nx48YwNTWFEAK9e/dm0CF6i3BkBxzZIZKiPn36YPfu3cXae/fujV27dr35gohI7ziyQ0RvtWHDhsHV1VWjzdXVFcOGDTNQRURkKAw7RCQ5ERERGDBgABo3boyYmBhs3rwZMTExaNy4MQYMGICIiAhDl0hEbxBPY4GnsYikpKCgAHXq1IGXlxd27dqFgoIC9U0FjY2N0adPHyQkJOD69eu8YpOokuNpLCJ6K8XExCApKQkzZ86EkZHmrzgjIyMEBQUhMTERMTExBqqQiN40hh0ikpSUlBQAUD9Q+EVF7UX9iEj6KsSzsYiInpednY2rV6++1rYZGRkAgB07dsDLywuZObk4dfEGqlY7CytzBS5cuKDud+7cudd6D09PT1hYWLzWtkT05nHODjhnh6iiOXfuHJo1a2boMkoVFxeHpk2bGroMoreett/fHNkhogrH09MTcXFxr7390aNHMWPGDLRv3x4f9ByAZeezMb6JBY7u3Y6YmBgsWLAAH3zwQZnqI6LKgyM74MgOkRRFRERg2rRpSEpKUre5u7vj22+/Rb9+/QxXGBHpjbbf3ww7YNghkqqCggKs3rYPQZtOInxIO4wc2IOXmxNJCE9jEdFbz9jYGM3bvA/LeBM0b9OaQYfoLcVLz4mIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0hh2iIiISNIYdoiIiEjSGHaIiIhI0gwado4fP46ePXvCxcUFMpkMu3bt0lgvk8lKfH3zzTfqPr6+vsXWDx48+A0fCREREVVUBg07WVlZ8Pb2xg8//FDi+pSUFI3XqlWrIJPJ0L9/f41+Y8aM0ej3n//8502UT0RERJWAiSHfPCAgAAEBAaWud3Jy0ljevXs3OnbsiNq1a2u0W1hYFOtLREREBBg47Oji3r172L9/P9auXVts3caNG7FhwwY4OjoiICAAwcHBsLa2LnVfubm5yM3NVS8rlUoAgEqlgkql0n/xRGQw+fn56j/5+SaSFm0/02UKO0+fPoWZmVlZdqG1tWvXwtraGv369dNoHzp0KNzd3eHk5ISEhAQEBQXhjz/+wKFDh0rdV3h4OEJDQ4u1R0ZGwsLCQu+1E5Hh3M4EABPExsbi7wRDV0NE+pSdna1VP5kQQuiy48LCQsybNw8rVqzAvXv38Oeff6J27dqYPXs2atWqhdGjR79WwTKZDDt37kSfPn1KXO/p6Qk/Pz8sXbr0pfuJi4tD8+bNERcXh6ZNm5bYp6SRHVdXVzx8+BA2NjavVT8RVUx/3HqMAf89i+1jmsO7pp2hyyEiPVIqlahWrRrS09Nf+v2t88hOWFgY1q5diwULFmDMmDHqdi8vLyxevPi1w87LxMTE4Nq1a9i6desr+zZt2hRyuRzXr18vNewoFAooFIpi7XK5HHK5vMz1ElHFYWJiov6Tn28iadH2M63z1Vjr1q3DTz/9hKFDh8LY2Fjd3qhRI1y9elXX3Wnl559/RrNmzeDt7f3KvpcuXYJKpYKzs3O51EJERESVi84jO3///Tfq1KlTrL2wsFDnyX+ZmZn466+/1MuJiYmIj4+HnZ0datasCeDZENW2bduwcOHCYtvfuHEDGzduRLdu3VCtWjVcvnwZ06ZNQ5MmTdCuXTsdj4yIiIikSOeRnffeew8xMTHF2rdt24YmTZrotK+zZ8+iSZMm6u2mTp2KJk2aYM6cOeo+W7ZsgRAC//jHP4ptb2pqiiNHjqBLly6oX78+Jk2aBH9/fxw+fFhj1ImIiIjeXjqP7AQHB+Ojjz7C33//jcLCQkRERODatWtYt24d9u3bp9O+fH198ar50WPHjsXYsWNLXOfq6oro6Gid3pOIiIjeLjqP7PTs2RNbt27FgQMHIJPJMGfOHFy5cgV79+6Fn59fedRIRERE9Np0GtnJz8/HvHnzMGrUKI6oEBERUaWg08iOiYkJvvnmGxQUFJRXPURERER6pfNprM6dOyMqKqocSiEiIiLSP50nKAcEBCAoKAgJCQlo1qwZLC0tNdb36tVLb8URERERlZXOYeeTTz4BACxatKjYOplMxlNcREREVKHoHHYKCwvLow4iIiKicqHznB0iIiKiykTnkZ25c+e+dP3zdz8mIiIiMjSdw87OnTs1llUqFRITE2FiYgIPDw+GHSIiIqpQdA4758+fL9amVCoxYsQI9O3bVy9FEREREemLXubs2NjYYO7cuZg9e7Y+dkdERESkN3qboJyWlob09HR97Y6IiIhIL3Q+jbVkyRKNZSEEUlJSsH79enTt2lVvhRERERHpg85hZ/HixRrLRkZGqF69OoYPH46goCC9FUZERESkDzqHncTExPKog4iIiKhclHnOjlKpxK5du3DlyhV91ENERESkVzqHnUGDBuGHH34AAOTk5KB58+YYNGgQGjVqhB07dui9QCIiIqKy0DnsHD9+HO3btwfw7AaDQgikpaVhyZIlCAsL03uBRERERGWhc9hJT0+HnZ0dAODgwYPo378/LCws0L17d1y/fl3vBRIRERGVhc5hx9XVFadPn0ZWVhYOHjwIf39/AMCTJ09gZmam9wKJiIiIykLnq7ECAwMxdOhQWFlZwc3NDb6+vgCend7y8vLSd31EREREZaJz2Bk/fjxatmyJ27dvw8/PD0ZGzwaHateuzTk7REREVOHoHHYAoHnz5mjevDkAoKCgABcvXkTbtm1RtWpVvRZHREREVFY6z9kJDAzEzz//DOBZ0PHx8UHTpk3h6uqKqKgofddHREREVCY6h53t27fD29sbALB3714kJibi6tWrCAwMxKxZs/ReIBEREVFZ6Bx2Hj58CCcnJwDAgQMHMHDgQNSrVw+jR4/GxYsX9V4gERERUVnoHHYcHR1x+fJlFBQU4ODBg+jcuTMAIDs7G8bGxnovkIiIiKgsdJ6gPHLkSAwaNAjOzs6QyWTw8/MDAJw5cwaenp56L5CIiIioLHQOOyEhIWjYsCFu376NgQMHQqFQAACMjY3x+eef671AIiIiorJ4rUvPBwwYAAB4+vSpum348OH6qYiIiIhIj3Ses1NQUIAvv/wS77zzDqysrHDz5k0AwOzZs9WXpBMRERFVFDqHnXnz5mHNmjVYsGABTE1N1e1eXl5YuXKlXosjIiIiKiudw866devw008/YejQoRpXXzVq1AhXr17Va3FEREREZaVz2Pn7779Rp06dYu2FhYVQqVR6KYqIiIhIX3QOO++99x5iYmKKtW/btg1NmjTRaV/Hjx9Hz5494eLiAplMhl27dmmsHzFiBGQymcardevWGn1yc3MxceJEVKtWDZaWlujVqxfu3Lmj62ERERGRROl8NVZwcDA++ugj/P333ygsLERERASuXbuGdevWYd++fTrtKysrC97e3hg5ciT69+9fYp+uXbti9erV6uXn5wkBz57VtXfvXmzZsgX29vaYNm0aevTogbi4ON7kkIiIiHQPOz179sTWrVsxf/58yGQyzJkzB02bNsXevXvVNxjUVkBAAAICAl7aR6FQqB9P8aL09HT8/PPPWL9+vfpOzhs2bICrqysOHz6MLl266FQPERERSY9OYSc/Px/z5s3DqFGjEB0dXV41aYiKioKDgwOqVKkCHx8fzJs3Dw4ODgCAuLg4qFQq+Pv7q/u7uLigYcOGOHXqVKlhJzc3F7m5ueplpVIJAFCpVJx3RCQx+fn56j/5+SaSFm0/0zqFHRMTE3zzzTdv7AaCAQEBGDhwINzc3JCYmIjZs2fjgw8+QFxcHBQKBVJTU2FqaoqqVatqbOfo6IjU1NRS9xseHo7Q0NBi7ZGRkbCwsND7cRCR4dzOBAATxMbG4u8EQ1dDRPqUnZ2tVT+dT2N17twZUVFRGDFihK6b6uzDDz9U/71hw4Zo3rw53NzcsH//fvTr16/U7YQQkMlkpa4PCgrC1KlT1ctKpRKurq7w9/eHjY2Nfoonogrhj1uPgYtn0bp1a3jXtDN0OUSkR0VnZl5F57ATEBCAoKAgJCQkoFmzZrC0tNRY36tXL113qTVnZ2e4ubnh+vXrAAAnJyfk5eXhyZMnGqM79+/fR9u2bUvdj0KhUD/T63lyuRxyuVz/hRORwZiYmKj/5OebSFq0/UzrHHY++eQTAMCiRYuKrZPJZCgoKNB1l1p79OgRbt++DWdnZwBAs2bNIJfLcejQIQwaNAgAkJKSgoSEBCxYsKDc6iAiIqLKQ+ewU1hYqLc3z8zMxF9//aVeTkxMRHx8POzs7GBnZ4eQkBD0798fzs7OSEpKwsyZM1GtWjX07dsXAGBra4vRo0dj2rRpsLe3h52dHaZPnw4vLy/11VlERET0dnutp57ry9mzZ9GxY0f1ctE8muHDh2P58uW4ePEi1q1bh7S0NDg7O6Njx47YunUrrK2t1dssXrwYJiYmGDRoEHJyctCpUyesWbOG99ghIiIiAIBMCCF03ejIkSNYvHgxrly5AplMBk9PTwQGBlba0RSlUglbW1ukp6dzgjKRxMQnP0Kf5bHY9UlrNHazN3Q5RKRH2n5/6/y4iB9++AFdu3aFtbU1Jk+ejEmTJsHGxgbdunXDDz/8UKaiiYiIiPRN59NY4eHhWLx4MT799FN126RJk9CuXTvMmzdPo52IiIjI0HQe2VEqlejatWuxdn9/f62vdyciIiJ6U3QOO7169cLOnTuLte/evRs9e/bUS1FERERE+qLzaawGDRpg3rx5iIqKQps2bQAAsbGxOHnyJKZNm4YlS5ao+06aNEl/lRIRERG9Bp2vxnJ3d9duxzIZbt68+VpFvWm8GotIfxIfZiErN9/QZahdS0nHtO0XsXCAF+o72xq6HDVLhQncq1m+uiMRlUrb72+dR3YSExPLVBgRSVfiwyx0/DbK0GWUaNr2i4YuoZhj030ZeIjeAIPeVJCIpKVoROe7DxujjoOVgat5JisnF/uiTqOHbxtYmhd/Jp4h/HU/E4Fb4yvUCBiRlDHsEJHe1XGwQsN3KsYpI5VKhdTqQFO3qnwQKNFbSuersYiIiIgqE4YdIiIikjSGHSIiIpI0rebsXLhwQesdNmrU6LWLISIiItI3rcJO48aNIZPJIISATCZ7ad+CggK9FEZERESkD1qdxkpMTMTNmzeRmJiIHTt2wN3dHcuWLcP58+dx/vx5LFu2DB4eHtixY0d510tERESkE61Gdtzc3NR/HzhwIJYsWYJu3bqp2xo1agRXV1fMnj0bffr00XuRRERERK9L5wnKFy9eLPGREe7u7rh8+bJeiiIiIiLSF53DToMGDRAWFoanT5+q23JzcxEWFoYGDRrotTgiIiKistL5DsorVqxAz5494erqCm9vbwDAH3/8AZlMhn379um9QCIiIqKy0DnstGzZEomJidiwYQOuXr0KIQQ+/PBDDBkyBJaWfKAdERERVSyv9WwsCwsLjB07Vt+1EBEREenda4WdGzdu4LvvvsOVK1cgk8nQoEEDTJ48GR4eHvquj4iIiKhMdJ6g/Ouvv+Ldd9/Fb7/9hkaNGqFhw4Y4c+YM3nvvPRw6dKg8aiQiIiJ6bTqP7Hz++eeYMmUKvvrqq2Ltn332Gfz8/PRWHBEREVFZ6Tyyc+XKFYwePbpY+6hRo3ifHSIiIqpwdA471atXR3x8fLH2+Ph4ODg46KMmIiIiIr3R+TTWmDFjMHbsWNy8eRNt27aFTCbDiRMn8PXXX2PatGnlUSMRERHRa9M57MyePRvW1tZYuHAhgoKCAAAuLi4ICQnBpEmT9F4gERERUVnoHHZkMhmmTJmCKVOmICMjAwBgbW2t98KIiIiI9OG17rMDAA8ePMC1a9cgk8lQv359VKtWTZ91EREREemFzhOUs7KyMGrUKDg7O6NDhw5o3749nJ2dMXr0aGRnZ5dHjURERESvTeewM3XqVERHR2Pv3r1IS0tDWloadu/ejejoaE5QJiIiogpH59NYO3bswPbt2+Hr66tu69atG8zNzTFo0CAsX75cn/URERERlYnOIzvZ2dlwdHQs1u7g4MDTWERERFTh6Bx22rRpg+DgYDx9+lTdlpOTg9DQULRp00avxRERERGVlc6nsb7//nt07doVNWrUgLe3N2QyGeLj42FmZoZff/21PGokIiIiem06j+w0bNgQ169fR3h4OBo3boxGjRrhq6++wvXr1/Hee+/ptK/jx4+jZ8+ecHFxgUwmw65du9TrVCoVPvvsM3h5ecHS0hIuLi4YNmwY7t69q7EPX19fyGQyjdfgwYN1PSwiIiKSqNe6z465uTnGjBlT5jfPysqCt7c3Ro4cif79+2usy87Oxrlz5zB79mx4e3vjyZMnCAwMRK9evXD27FmNvmPGjMHcuXM16iMiIiICtAw7e/bs0XqHvXr10rpvQEAAAgICSlxna2uLQ4cOabQtXboULVu2xK1bt1CzZk11u4WFBZycnLR+39zcXOTm5qqXlUolgGejSSqVSuv9EJGm/Px89Z8V5bNUVEdFqQeomD8nospI28+PVmGnT58+Wu1MJpOhoKBAq76vIz09HTKZDFWqVNFo37hxIzZs2ABHR0cEBAQgODj4pY+wCA8PR2hoaLH2yMhIWFhY6LtsorfG7UwAMMGJEyeQbGXoajS9+I8nQ6rIPyeiykTbq8C1CjuFhYVlKkYfnj59is8//xxDhgyBjY2Nun3o0KFwd3eHk5MTEhISEBQUhD/++OOlv9iCgoIwdepU9bJSqYSrqyv8/f019k1Eurl0V4mFVyJRw7sW3KtbGrocAM9GT87EnkGr1q1gYvLaT8jRK/EgC7IrSXj/fX+858LfOUSvq+jMzKtUjE/+K6hUKgwePBiFhYVYtmyZxrrn5w41bNgQdevWRfPmzXHu3Dk0bdq0xP0pFAooFIpi7XK5HHK5XL/FE71FTExMIK9yBrPj5hu6lGKWHV726k5vkLxKJ5iYdOPvHKIy0Pbz81ph58iRI1i8eDGuXLkCmUwGT09PBAYGonPnzq+zu5dSqVQYNGgQEhMTcfTo0VeOvDRt2hRyuRzXr18vNewQUflRpbXCwu5D4OFQMc7P5Ofn4+SJk2j3frsKM7Jz434mJm28YegyiN4aOn/yf/jhB0yZMgUDBgzA5MmTAQCxsbHo1q0bFi1ahE8//VRvxRUFnevXr+PYsWOwt7d/5TaXLl2CSqWCs7Oz3uogIu2JfBu429THu/a2hi4FwLPfI4kmiWhg16DCjKIUPk2HyH9g6DKI3ho6h53w8HAsXrxYI9RMmjQJ7dq1w7x583QKO5mZmfjrr7/Uy4mJiYiPj4ednR1cXFwwYMAAnDt3Dvv27UNBQQFSU1MBAHZ2djA1NcWNGzewceNGdOvWDdWqVcPly5cxbdo0NGnSBO3atdP10IiIiEiCdL6poFKpRNeuXYu1+/v7az1RqMjZs2fRpEkTNGnSBMCzJ6o3adIEc+bMwZ07d7Bnzx7cuXMHjRs3hrOzs/p16tQpAICpqSmOHDmCLl26oH79+pg0aRL8/f1x+PBhGBsb63poREREJEE6j+z06tULO3fuxL///W+N9t27d6Nnz5467cvX1xdCiFLXv2wdALi6uiI6Olqn9yQiIqK3i1ZhZ8mSJeq/N2jQAPPmzUNUVJT6wZ+xsbE4efIkpk2bVj5VEhEREb0mrcLO4sWLNZarVq2Ky5cv4/Lly+q2KlWqYNWqVfjiiy/0WyERERFRGWgVdhITE8u7DiIiIqJyofMEZSIiIqLKhGGHiIiIJI1hh4iIiCSNYYeIiIgkjWGHiIiIJE3nsLN69Wps27atWPu2bduwdu1avRRFREREpC86h52vvvoK1apVK9bu4OCA+fPn66UoIiIiIn3ROewkJyfD3d29WLubmxtu3bqll6KIiIiI9EXnsOPg4IALFy4Ua//jjz9gb2+vl6KIiIiI9EXnsDN48GBMmjQJx44dQ0FBAQoKCnD06FFMnjwZgwcPLo8aiYiIiF6bzk89DwsLQ3JyMjp16gQTk2ebFxYWYtiwYZyzQ0RERBWOzmHH1NQUW7duxZdffok//vgD5ubm8PLygpubW3nUR0RERFQmOoedIvXq1UO9evX0WQsRERGR3mkVdqZOnYovv/wSlpaWmDp16kv7Llq0SC+FEREREemDVmHn/PnzUKlU6r+XRiaT6acqIiIiIj3RKuwcO3asxL8TERERVXSvPWeHiOhFOaoCAEDC3+kGruT/ZOXk4uwDwCn5CSzNFYYuBwDw1/1MQ5dA9FbRKuz069dP6x1GRES8djFEVLnd+P9f4p9HXDRwJS8ywfq/fjd0EcVYKvjvTaI3QatPmq2tbXnXQUQS4P+eEwDAw8EK5nJjA1fzzLWUdEzbfhELB3ihvnPF+V1mqTCBezVLQ5dB9FbQKuysXr26vOsgIgmwszTF4JY1DV2Ghvz8fACAR3VLNHyn4oQdInpzXnsM9cGDB7h27RpkMhnq1auH6tWr67MuIiIiIr3Q+dlYWVlZGDVqFJydndGhQwe0b98eLi4uGD16NLKzs8ujRiIiIqLXpnPYmTp1KqKjo7F3716kpaUhLS0Nu3fvRnR0NKZNm1YeNRIRERG9Np1PY+3YsQPbt2+Hr6+vuq1bt24wNzfHoEGDsHz5cn3WR0RERFQmOo/sZGdnw9HRsVi7g4MDT2MRERFRhaNz2GnTpg2Cg4Px9OlTdVtOTg5CQ0PRpk0bvRZHREREVFY6n8b67rvvEBAQgBo1asDb2xsymQzx8fEwMzPDr7/+Wh41EhEREb02ncOOl5cXrl+/jg0bNuDq1asQQmDw4MEYOnQozM3Ny6NGIiIiotemc9g5fvw42rZtizFjxmi05+fn4/jx4+jQoYPeiiMiIiIqK53n7HTs2BGPHz8u1p6eno6OHTvqpSgiIiIifdE57AghIJPJirU/evQIlpZ8zgsRERFVLFqfxip68rlMJsOIESOgUCjU6woKCnDhwgW0bdtW/xUSERERlYHWIzu2trawtbWFEALW1tbqZVtbWzg5OWHs2LHYsGGDTm9+/Phx9OzZEy4uLpDJZNi1a5fGeiEEQkJC4OLiAnNzc/j6+uLSpUsafXJzczFx4kRUq1YNlpaW6NWrF+7cuaNTHURERCRdWo/sFD35vFatWpg+fbpeTlllZWXB29sbI0eORP/+/YutX7BgARYtWoQ1a9agXr16CAsLg5+fH65duwZra2sAQGBgIPbu3YstW7bA3t4e06ZNQ48ePRAXFwdjY+My10hERESVm85XY82YMQNCCPVycnIydu7ciXfffRf+/v467SsgIAABAQElrhNC4LvvvsOsWbPUp9DWrl0LR0dHbNq0CePGjUN6ejp+/vlnrF+/Hp07dwYAbNiwAa6urjh8+DC6dOmi6+ERERGRxOgcdnr37o1+/frhX//6F9LS0tCyZUuYmpri4cOHWLRoET755BO9FJaYmIjU1FSNAKVQKODj44NTp05h3LhxiIuLg0ql0ujj4uKChg0b4tSpU6WGndzcXOTm5qqXlUolAEClUkGlUumlfiKqGPLz89V/8vNNJC3afqZ1Djvnzp3D4sWLAQDbt2+Hk5MTzp8/jx07dmDOnDl6CzupqakAUOw5XI6OjkhOTlb3MTU1RdWqVYv1Kdq+JOHh4QgNDS3WHhkZCQsLi7KWTkQVyO1MADBBbGws/k4wdDVEpE/aPpNT57CTnZ2tni8TGRmJfv36wcjICK1bt1aHEH168TL30i5916VPUFAQpk6dql5WKpVwdXWFv78/bGxsylYwEVUof9x6DFw8i9atW8O7pp2hyyEiPSo6M/MqOoedOnXqYNeuXejbty9+/fVXTJkyBQBw//59vQYFJycnAM9Gb5ydndXt9+/fV4/2ODk5IS8vD0+ePNEY3bl///5LL4NXKBQal84XkcvlkMvl+joEIqoATExM1H/y800kLdp+pnW+qeCcOXMwffp01KpVC61atVI/6TwyMhJNmjTRdXelcnd3h5OTEw4dOqRuy8vLQ3R0tDrINGvWDHK5XKNPSkoKEhISeM8fIiIiAvAaIzsDBgzA+++/j5SUFHh7e6vbO3XqhL59++q0r8zMTPz111/q5cTERMTHx8POzg41a9ZEYGAg5s+fj7p166Ju3bqYP38+LCwsMGTIEADP7v0zevRoTJs2Dfb29rCzs8P06dPh5eWlvjqLiIiI3m46h501a9bgww8/VJ9mKtKyZUud3/zs2bMaz9MqmkczfPhwrFmzBjNmzEBOTg7Gjx+PJ0+eoFWrVoiMjFTPGQKAxYsXw8TEBIMGDUJOTg46deqENWvW8B47REREBACQiedvmqMFZ2dnZGVlYeDAgRg9erQkThcplUrY2toiPT2dE5SJJCY++RH6LI/Frk9ao7GbvaHLISI90vb7W+c5O3fu3MGGDRvw5MkTdOzYEZ6envj6669feqk3ERERkaHoHHaMjY3Rq1cvRERE4Pbt2xg7diw2btyImjVrolevXti9ezcKCwvLo1YiIiIinekcdp7n4OCAdu3aoU2bNjAyMsLFixcxYsQIeHh4ICoqSk8lEhEREb2+1wo79+7dw7fffov33nsPvr6+UCqV2LdvHxITE3H37l3069cPw4cP13etRERERDrT+Wqsnj174tdff0W9evUwZswYDBs2DHZ2/3dXUnNzc0ybNk39SAkiIiIiQ9I57Dg4OCA6Olp9M8GSODs7IzExsUyFEREREemDzmHn559/fmUfmUwGNze31yqIiIiISJ+0nrNz5swZ/PLLLxpt69atg7u7OxwcHDB27Fjk5ubqvUAiIiKistA67ISEhODChQvq5YsXL2L06NHo3LkzPv/8c+zduxfh4eHlUiQRERHR69I67MTHx6NTp07q5S1btqBVq1b473//i6lTp2LJkiX43//+Vy5FEhEREb0urcPOkydP4OjoqF6Ojo5G165d1cstWrTA7du39VsdERERURlpHXYcHR3VV1jl5eXh3LlzGldkZWRkQC6X679CIiIiojLQOux07doVn3/+OWJiYhAUFAQLCwu0b99evf7ChQvw8PAolyKJiIiIXpfWl56HhYWhX79+8PHxgZWVFdauXQtTU1P1+lWrVsHf379ciiQiIiJ6XVqHnerVqyMmJgbp6emwsrKCsbGxxvpt27bByspK7wUSERERlYXONxW0tbUtsf35R0YQERERVRRleuo5ERERUUXHsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENERESSxrBDREREksawQ0RERJJW4cNOrVq1IJPJir0mTJgAABgxYkSxda1btzZw1URERFRRmBi6gFf5/fffUVBQoF5OSEiAn58fBg4cqG7r2rUrVq9erV42NTV9ozUSERFRxVXhw0716tU1lr/66it4eHjAx8dH3aZQKODk5PSmSyMiIqJKoMKHnefl5eVhw4YNmDp1KmQymbo9KioKDg4OqFKlCnx8fDBv3jw4ODiUup/c3Fzk5uaql5VKJQBApVJBpVKV3wEQ0RuXn5+v/pOfbyJp0fYzXanCzq5du5CWloYRI0ao2wICAjBw4EC4ubkhMTERs2fPxgcffIC4uDgoFIoS9xMeHo7Q0NBi7ZGRkbCwsCiv8onIAG5nAoAJYmNj8XeCoashIn3Kzs7Wqp9MCCHKuRa96dKlC0xNTbF3795S+6SkpMDNzQ1btmxBv379SuxT0siOq6srHj58CBsbG73XTUSG88etxxjw37PYPqY5vGvaGbocItIjpVKJatWqIT09/aXf35VmZCc5ORmHDx9GRETES/s5OzvDzc0N169fL7WPQqEocdRHLpdDLpeXuVYiqjhMTEzUf/LzTSQt2n6mK/yl50VWr14NBwcHdO/e/aX9Hj16hNu3b8PZ2fkNVUZEREQVWaUIO4WFhVi9ejWGDx+u/lcaAGRmZmL69Ok4ffo0kpKSEBUVhZ49e6JatWro27evASsmIiKiiqJSnMY6fPgwbt26hVGjRmm0Gxsb4+LFi1i3bh3S0tLg7OyMjh07YuvWrbC2tjZQtURERFSRVIqw4+/vj5LmUZubm+PXX381QEVERERUWVSK01hEREREr4thh4iIiCSNYYeIiIgkjWGHiIiIJI1hh4iIiCSNYYeIiIgkjWGHiIiIJI1hh4iIiCSNYYeIiIgkjWGHiIiIJI1hh4iIiCStUjwbi4jeLtnZ2bh69ape9nUtJQ25qX/hSoI5Ch9V0cs+PT09YWFhoZd9EVH5Y9ghogrn6tWraNasmV73OWSt/vYVFxeHpk2b6m+HRFSuGHaIqMLx9PREXFycXvaVmZOL/cdOo3vHNrAyV+hln56ennrZDxG9GQw7RFThWFhY6G3kRKVS4cnD+2jTsjnkcrle9klElQsnKBMREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaQx7BAREZGkMewQERGRpDHsEBERkaRV6LATEhICmUym8XJyclKvF0IgJCQELi4uMDc3h6+vLy5dumTAiomIiKiiqdBhBwDee+89pKSkqF8XL15Ur1uwYAEWLVqEH374Ab///jucnJzg5+eHjIwMA1ZMREREFUmFDzsmJiZwcnJSv6pXrw7g2ajOd999h1mzZqFfv35o2LAh1q5di+zsbGzatMnAVRMREVFFYWLoAl7l+vXrcHFxgUKhQKtWrTB//nzUrl0biYmJSE1Nhb+/v7qvQqGAj48PTp06hXHjxpW6z9zcXOTm5qqXlUolAEClUkGlUpXfwRDRG1f0meZnm0h6tP1cV+iw06pVK6xbtw716tXDvXv3EBYWhrZt2+LSpUtITU0FADg6Omps4+joiOTk5JfuNzw8HKGhocXaIyMjYWFhob8DIKIK49ChQ4YugYj0LDs7W6t+MiGEKOda9CYrKwseHh6YMWMGWrdujXbt2uHu3btwdnZW9xkzZgxu376NgwcPlrqfkkZ2XF1d8fDhQ9jY2JTrMRDRm6VSqXDo0CH4+flBLpcbuhwi0iOlUolq1aohPT39pd/fFXpk50WWlpbw8vLC9evX0adPHwBAamqqRti5f/9+sdGeFykUCigUimLtcrmcvwyJJIqfbyLp0fYzXeEnKD8vNzcXV65cgbOzM9zd3eHk5KQxNJ2Xl4fo6Gi0bdvWgFUSERFRRVKhR3amT5+Onj17ombNmrh//z7CwsKgVCoxfPhwyGQyBAYGYv78+ahbty7q1q2L+fPnw8LCAkOGDDF06URERFRBVOiwc+fOHfzjH//Aw4cPUb16dbRu3RqxsbFwc3MDAMyYMQM5OTkYP348njx5glatWiEyMhLW1tYGrpyIiIgqiko1Qbm8KJVK2NravnKCExFVPiqVCgcOHEC3bt04Z4dIYrT9/q5Uc3aIiIiIdMWwQ0RERJLGsENEklVQUIDo6GgcP34c0dHRKCgoMHRJRGQADDtEJEkRERGoU6cO/Pz8sGjRIvj5+aFOnTqIiIgwdGlE9IYx7BCR5ERERGDAgAHw8vJCTEwMNm/ejJiYGHh5eWHAgAEMPERvGV6NBV6NRSQlBQUFqFOnDry8vLBr1y4UFBSor8YyNjZGnz59kJCQgOvXr8PY2NjQ5RJRGfBqLCJ6K8XExCApKQkzZ86EkZHmrzgjIyMEBQUhMTERMTExBqqQiN40hh0ikpSUlBQAQMOGDUtcX9Re1I+IpI9hh4gkpejBwAkJCSWuL2p//gHCRCRtDDtEJCnt27dHrVq1MH/+fBQWFmqsKywsRHh4ONzd3dG+fXsDVUhEbxrDDhFJirGxMRYuXIh9+/ahT58+iI2NRU5ODmJjY9GnTx/s27cP3377LScnE71FKvSDQImIXke/fv2wfft2TJs2DR06dFC3u7u7Y/v27ejXr58BqyOiN42XnoOXnhNJVUFBAY4dO4ZffvkFAQEB6NixI0d0iCRE2+9vjuwQkWQZGxvDx8cHWVlZ8PHxYdAhektxzg4RERFJGsMOERERSRrDDhEREUkaww4RERFJGsMOERERSRrDDhEREUkaww4RERFJGsMOERERSRrDDhEREUka76AMoOiJGUql0sCVEJG+qVQqZGdnQ6lUQi6XG7ocItKjou/tVz35imEHQEZGBgDA1dXVwJUQERGRrjIyMmBra1vqej4IFEBhYSHu3r0La2tryGQyQ5dDRHqkVCrh6uqK27dv80G/RBIjhEBGRgZcXFxgZFT6zByGHSKSNG2fikxE0sUJykRERCRpDDtEREQkaQw7RCRpCoUCwcHBUCgUhi6FiAyEc3aIiIhI0jiyQ0RERJLGsENERESSxrBDREREksawQ0RERJLGsENEZTJixAj06dNHvezr64vAwMDX3l+tWrXw3XffvbSPTCbDrl27Xvs9tJGUlASZTIb4+PhyfR8iKn8MO0SVxP379zFu3DjUrFkTCoUCTk5O6NKlC06fPq2X/b8YWl5XREQEvvzyy7IXRESkJ3wQKFEl0b9/f6hUKqxduxa1a9fGvXv3cOTIETx+/NjQpWmws7MzdAkVmkqlqnBPX6+INRHpE0d2iCqBtLQ0nDhxAl9//TU6duwINzc3tGzZEkFBQejevTsAYNSoUejRo4fGdvn5+XBycsKqVasAANu3b4eXlxfMzc1hb2+Pzp07IysrCyEhIVi7di12794NmUwGmUyGqKgoAMDFixfxwQcfqLcZO3YsMjMzS631xdNYubm5mDFjBlxdXaFQKFC3bl38/PPPLz3ejIwMDBkyBFZWVnBxccHSpUtf2v9VNRYWFmLu3LmoUaMGFAoFGjdujIMHD2rs47fffkOTJk1gZmaG5s2b4/z58y99T+DZKbcvv/zypbXKZDKsWLECvXv3hqWlJcLCwgAAe/fuRbNmzWBmZobatWsjNDQU+fn56u1CQkLUo3guLi6YNGmSet2yZctQt25dmJmZwdHREQMGDNCo6cXTgI0bN0ZISEiZayKqtAQRVXgqlUpYWVmJwMBA8fTp0xL7nDx5UhgbG4u7d++q23bv3i0sLS1FRkaGuHv3rjAxMRGLFi0SiYmJ4sKFC+LHH38UGRkZIiMjQwwaNEh07dpVpKSkiJSUFJGbmyuysrKEi4uL6Nevn7h48aI4cuSIcHd3F8OHD1e/x/Dhw0Xv3r3Vyz4+PmLy5Mnq5UGDBglXV1cREREhbty4IQ4fPiy2bNlS6rG6ubkJa2trER4eLq5duyaWLFkijI2NRWRkpLoPALFz504hhNCqxkWLFgkbGxuxefNmcfXqVTFjxgwhl8vFn3/+KYQQIjMzU1SvXl18+OGHIiEhQezdu1fUrl1bABDnz58vc60ODg7i559/Fjdu3BBJSUni4MGDwsbGRqxZs0bcuHFDREZGilq1aomQkBAhhBDbtm0TNjY24sCBAyI5OVmcOXNG/PTTT0IIIX7//XdhbGwsNm3aJJKSksS5c+fE999/r1HT4sWLNer09vYWwcHBZaqJqDJj2CGqJLZv3y6qVq0qzMzMRNu2bUVQUJD4448/NPq8++674uuvv1Yv9+nTR4wYMUIIIURcXJwAIJKSkkrc/4uhRQghfvrpJ1G1alWRmZmpbtu/f78wMjISqampJW73fNi5du2aACAOHTqk9XG6ubmJrl27arR9+OGHIiAgQL38fNjRpkYXFxcxb948jX22aNFCjB8/XgghxH/+8x9hZ2cnsrKy1OuXL1+uVdjRptbAwECNPu3btxfz58/XaFu/fr1wdnYWQgixcOFCUa9ePZGXl1fsPXfs2CFsbGyEUqkstSZtwo6uNRFVZjyNRVRJ9O/fH3fv3sWePXvQpUsXREVFoWnTplizZo26z8cff4zVq1cDeDahef/+/Rg1ahQAwNvbG506dYKXlxcGDhyI//73v3jy5MlL3/PKlSvw9vaGpaWluq1du3YoLCzEtWvXXllzfHw8jI2N4ePjo9OxtmnTptjylStXXqtGpVKJu3fvol27dhrbtWvXTr3Pon1YWFiUWkNZam3evLnGclxcHObOnQsrKyv1a8yYMUhJSUF2djYGDhyInJwc1K5dG2PGjMHOnTvVp5P8/Pzg5uaG2rVr46OPPsLGjRuRnZ2tVa1lqYmoMmPYIapEzMzM4Ofnhzlz5uDUqVMYMWIEgoOD1euHDRuGmzdv4vTp09iwYQNq1aqF9u3bAwCMjY1x6NAh/PLLL3j33XexdOlS1K9fH4mJiaW+nxACMpmsxHWltT/P3NxcxyMsXWnvp22NL/Z5fjuh50cEvvhezwcx4NkcotDQUMTHx6tfFy9exPXr12FmZgZXV1dcu3YNP/74I8zNzTF+/Hh06NABKpUK1tbWOHfuHDZv3gxnZ2fMmTMH3t7eSEtLAwAYGRkVOx6VSlWsRl1rIqrMGHaIKrF3330XWVlZ6mV7e3v06dMHq1evxurVqzFy5EiN/jKZDO3atUNoaCjOnz8PU1NT7Ny5EwBgamqKgoKCYvuPj4/XeI+TJ0/CyMgI9erVe2V9Xl5eKCwsRHR0tE7HFRsbW2zZ09OzxL6vqtHGxgYuLi44ceKExnanTp1CgwYN1Pv4448/kJOTU2oN+qi1SNOmTXHt2jXUqVOn2MvI6NmvZXNzc/Tq1QtLlixBVFQUTp8+jYsXLwIATExM0LlzZyxYsAAXLlxAUlISjh49CgCoXr06UlJS1O+lVCpfGmh1qYmosuKl50SVwKNHjzBw4ECMGjUKjRo1grW1Nc6ePYsFCxagd+/eGn0//vhj9OjRAwUFBRg+fLi6/cyZMzhy5Aj8/f3h4OCAM2fO4MGDB+ov/Fq1auHXX3/FtWvXYG9vD1tbWwwdOhTBwcEYPnw4QkJC8ODBA0ycOBEfffQRHB0dX1l3rVq1MHz4cIwaNQpLliyBt7c3kpOTcf/+fQwaNKjU7U6ePIkFCxagT58+OHToELZt24b9+/eX2FebGv/9738jODgYHh4eaNy4MVavXo34+Hhs3LgRADBkyBDMmjULo0ePxhdffIGkpCR8++23rzw+XWstMmfOHPTo0QOurq4YOHAgjIyMcOHCBVy8eBFhYWFYs2YNCgoK0KpVK1hYWGD9+vUwNzeHm5sb9u3bh5s3b6JDhw6oWrUqDhw4gMLCQtSvXx8A8MEHH2DNmjXo2bMnqlatitmzZ8PY2PiVx/GqmogqNYPOGCIirTx9+lR8/vnnomnTpsLW1lZYWFiI+vXriy+++EJkZ2dr9C0sLBRubm6iW7duGu2XL18WXbp0EdWrVxcKhULUq1dPLF26VL3+/v37ws/PT1hZWQkA4tixY0IIIS5cuCA6duwozMzMhJ2dnRgzZozIyMhQb/eqq7FycnLElClThLOzszA1NRV16tQRq1atKvVY3dzcRGhoqBg0aJCwsLAQjo6O4rvvvtPog+cmKGtTY0FBgQgNDRXvvPOOkMvlwtvbW/zyyy8a+zx9+rTw9vYWpqamonHjxmLHjh1aTVDWtdYiBw8eFG3bthXm5ubCxsZGtGzZUn3F1c6dO0WrVq2EjY2NsLS0FK1btxaHDx8WQggRExMjfHx8RNWqVYW5ublo1KiR2Lp1q3q/6enpYtCgQcLGxka4urqKNWvWlDhBWdeaiCozmRB6PllNRAaVnZ0NFxcXrFq1Cv369TN0OZJWq1YtBAYGlunxGERU/ngai0giCgsLkZqaioULF8LW1ha9evUydElERBUCww6RRNy6dQvu7u6oUaMG1qxZAxMTfryJiACAp7GIiIhI0ng9IREREUkaww4RERFJGsMOERERSRrDDhEREUkaww4RERFJGsMOERERSRrDDhEREUkaww4RERFJ2v8D9ObKVDho8z0AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for feature in continous_features:\n",
    "    dataset=data.copy()\n",
    "    if 0 in dataset[feature].unique():\n",
    "        pass\n",
    "    else:\n",
    "        dataset[feature]=np.log(dataset[feature])\n",
    "        data.boxplot(column=feature)\n",
    "        plt.ylabel(feature)\n",
    "        plt.title(feature)\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "s0F0xt7beAlv",
    "outputId": "02681ec5-1ab8-47da-e265-0d86ce9ddb74"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Heart rate</th>\n",
       "      <th>Systolic blood pressure</th>\n",
       "      <th>Diastolic blood pressure</th>\n",
       "      <th>Blood sugar</th>\n",
       "      <th>CK-MB</th>\n",
       "      <th>Troponin</th>\n",
       "      <th>Result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>63</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>160</td>\n",
       "      <td>83</td>\n",
       "      <td>160.0</td>\n",
       "      <td>1.80</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>94</td>\n",
       "      <td>98</td>\n",
       "      <td>46</td>\n",
       "      <td>296.0</td>\n",
       "      <td>6.75</td>\n",
       "      <td>1.060</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>56</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>160</td>\n",
       "      <td>77</td>\n",
       "      <td>270.0</td>\n",
       "      <td>1.99</td>\n",
       "      <td>0.003</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>66</td>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "      <td>120</td>\n",
       "      <td>55</td>\n",
       "      <td>270.0</td>\n",
       "      <td>13.87</td>\n",
       "      <td>0.122</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54</td>\n",
       "      <td>1</td>\n",
       "      <td>64</td>\n",
       "      <td>112</td>\n",
       "      <td>65</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.08</td>\n",
       "      <td>0.003</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Gender  Heart rate  Systolic blood pressure  Diastolic blood pressure  \\\n",
       "0   63       1          66                      160                        83   \n",
       "1   20       1          94                       98                        46   \n",
       "2   56       1          64                      160                        77   \n",
       "3   66       1          70                      120                        55   \n",
       "4   54       1          64                      112                        65   \n",
       "\n",
       "   Blood sugar  CK-MB  Troponin  Result  \n",
       "0        160.0   1.80     0.012       0  \n",
       "1        296.0   6.75     1.060       1  \n",
       "2        270.0   1.99     0.003       0  \n",
       "3        270.0  13.87     0.122       1  \n",
       "4        300.0   1.08     0.003       0  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "id": "z3TpBqLceTSg"
   },
   "outputs": [],
   "source": [
    "x=data.drop(['Result'],axis=1).copy()\n",
    "y=data[['Result']].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "id": "EDHfp8Yyicps"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import LeakyReLU,PReLU,ELU\n",
    "from tensorflow.keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "id": "snYZlOpreTad"
   },
   "outputs": [],
   "source": [
    "x_train,x_test,y_train,y_test=train_test_split(x,y,test_size = 0.2, random_state = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "o6t3eBJf_11g"
   },
   "source": [
    "## Scaling the numerical features using StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "id": "_hH9P2h4HEi6"
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "sc=StandardScaler()\n",
    "x_train = sc.fit_transform(x_train)\n",
    "x_test = sc.transform(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "N-wiJKG6fGyS",
    "outputId": "47e34761-2e83-40d6-e313-002f822db80f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1055, 8), (1055, 1))"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape,y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "TerCLuLrfG0h",
    "outputId": "3707f503-832b-4f06-bead-9c6308a679ae"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((264, 8), (264, 1))"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test.shape,y_test.shape\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WP0OZDtio0fo"
   },
   "source": [
    "## Developing the netural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "id": "BDgmtbJcldko"
   },
   "outputs": [],
   "source": [
    "# Initialising the ANN\n",
    "classifier = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "id": "gUvzbQ2RmOfR"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "early_stop= tf.keras.callbacks.EarlyStopping(\n",
    "    monitor=\"val_loss\",\n",
    "    min_delta=0.010,\n",
    "    patience=20,\n",
    "    verbose=1,\n",
    "    mode=\"auto\",\n",
    "    baseline=None,\n",
    "    restore_best_weights=False,\n",
    "    start_from_epoch=0,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "id": "wPfMh8W7lZOA"
   },
   "outputs": [],
   "source": [
    "# Adding the input layer\n",
    "classifier.add(Dense(units=8,activation='relu'))\n",
    "# Adding the first hidden layer\n",
    "classifier.add(Dense(units=7,activation='relu'))\n",
    "# Adding  the second hidden layer\n",
    "classifier.add(Dense(units=6,activation='relu'))\n",
    "# Adding the output layer\n",
    "classifier.add(Dense(units=1,activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "id": "DIZpz8Q2lk_-"
   },
   "outputs": [],
   "source": [
    "classifier.compile(optimizer='adam',loss='binary_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "LMVzDa6blZRf",
    "outputId": "bb1fa2ca-43ec-4cac-db8a-d54d2a001cae"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 12ms/step - accuracy: 0.3489 - loss: 0.7616 - val_accuracy: 0.4527 - val_loss: 0.6946\n",
      "Epoch 2/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6590 - loss: 0.6857 - val_accuracy: 0.8023 - val_loss: 0.6710\n",
      "Epoch 3/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8509 - loss: 0.6666 - val_accuracy: 0.7765 - val_loss: 0.6602\n",
      "Epoch 4/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8606 - loss: 0.6492 - val_accuracy: 0.7794 - val_loss: 0.6473\n",
      "Epoch 5/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8312 - loss: 0.6329 - val_accuracy: 0.7937 - val_loss: 0.6324\n",
      "Epoch 6/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8134 - loss: 0.6209 - val_accuracy: 0.7822 - val_loss: 0.6184\n",
      "Epoch 7/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8391 - loss: 0.5912 - val_accuracy: 0.7966 - val_loss: 0.6040\n",
      "Epoch 8/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8370 - loss: 0.5746 - val_accuracy: 0.7708 - val_loss: 0.5965\n",
      "Epoch 9/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8170 - loss: 0.5563 - val_accuracy: 0.7908 - val_loss: 0.5779\n",
      "Epoch 10/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8517 - loss: 0.5333 - val_accuracy: 0.7937 - val_loss: 0.5666\n",
      "Epoch 11/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8741 - loss: 0.4965 - val_accuracy: 0.7908 - val_loss: 0.5603\n",
      "Epoch 12/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8403 - loss: 0.5159 - val_accuracy: 0.7966 - val_loss: 0.5486\n",
      "Epoch 13/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8568 - loss: 0.4729 - val_accuracy: 0.7994 - val_loss: 0.5488\n",
      "Epoch 14/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8236 - loss: 0.5146 - val_accuracy: 0.7937 - val_loss: 0.5481\n",
      "Epoch 15/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8654 - loss: 0.4632 - val_accuracy: 0.7966 - val_loss: 0.5302\n",
      "Epoch 16/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8458 - loss: 0.4772 - val_accuracy: 0.7851 - val_loss: 0.5386\n",
      "Epoch 17/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8542 - loss: 0.4664 - val_accuracy: 0.8080 - val_loss: 0.5211\n",
      "Epoch 18/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8909 - loss: 0.4249 - val_accuracy: 0.8023 - val_loss: 0.5207\n",
      "Epoch 19/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8404 - loss: 0.4518 - val_accuracy: 0.8023 - val_loss: 0.5151\n",
      "Epoch 20/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8604 - loss: 0.4328 - val_accuracy: 0.8052 - val_loss: 0.5138\n",
      "Epoch 21/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8615 - loss: 0.4343 - val_accuracy: 0.7966 - val_loss: 0.5293\n",
      "Epoch 22/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8611 - loss: 0.4304 - val_accuracy: 0.8023 - val_loss: 0.5156\n",
      "Epoch 23/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8561 - loss: 0.4315 - val_accuracy: 0.8080 - val_loss: 0.5075\n",
      "Epoch 24/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8414 - loss: 0.4374 - val_accuracy: 0.8052 - val_loss: 0.5009\n",
      "Epoch 25/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8505 - loss: 0.4235 - val_accuracy: 0.8138 - val_loss: 0.4973\n",
      "Epoch 26/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8723 - loss: 0.3988 - val_accuracy: 0.7994 - val_loss: 0.4956\n",
      "Epoch 27/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8867 - loss: 0.3724 - val_accuracy: 0.8109 - val_loss: 0.4899\n",
      "Epoch 28/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8741 - loss: 0.3885 - val_accuracy: 0.8109 - val_loss: 0.4867\n",
      "Epoch 29/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8709 - loss: 0.3939 - val_accuracy: 0.8080 - val_loss: 0.4876\n",
      "Epoch 30/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8660 - loss: 0.3949 - val_accuracy: 0.8080 - val_loss: 0.4894\n",
      "Epoch 31/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8500 - loss: 0.3935 - val_accuracy: 0.8109 - val_loss: 0.4872\n",
      "Epoch 32/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8839 - loss: 0.3577 - val_accuracy: 0.8052 - val_loss: 0.4848\n",
      "Epoch 33/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8721 - loss: 0.3830 - val_accuracy: 0.8138 - val_loss: 0.4861\n",
      "Epoch 34/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8684 - loss: 0.3664 - val_accuracy: 0.8052 - val_loss: 0.4838\n",
      "Epoch 35/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8771 - loss: 0.3748 - val_accuracy: 0.8166 - val_loss: 0.4861\n",
      "Epoch 36/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8936 - loss: 0.3507 - val_accuracy: 0.8138 - val_loss: 0.4812\n",
      "Epoch 37/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8980 - loss: 0.3463 - val_accuracy: 0.8138 - val_loss: 0.4770\n",
      "Epoch 38/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 5ms/step - accuracy: 0.8698 - loss: 0.3620 - val_accuracy: 0.8080 - val_loss: 0.4817\n",
      "Epoch 39/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8794 - loss: 0.3692 - val_accuracy: 0.8080 - val_loss: 0.4774\n",
      "Epoch 40/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8952 - loss: 0.3379 - val_accuracy: 0.8080 - val_loss: 0.4802\n",
      "Epoch 41/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8728 - loss: 0.3748 - val_accuracy: 0.8052 - val_loss: 0.4742\n",
      "Epoch 42/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8830 - loss: 0.3512 - val_accuracy: 0.8080 - val_loss: 0.4765\n",
      "Epoch 43/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8678 - loss: 0.3661 - val_accuracy: 0.8166 - val_loss: 0.4754\n",
      "Epoch 44/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8717 - loss: 0.3660 - val_accuracy: 0.8195 - val_loss: 0.4745\n",
      "Epoch 45/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8697 - loss: 0.3501 - val_accuracy: 0.8023 - val_loss: 0.4742\n",
      "Epoch 46/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8788 - loss: 0.3396 - val_accuracy: 0.8109 - val_loss: 0.4666\n",
      "Epoch 47/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8864 - loss: 0.3246 - val_accuracy: 0.8223 - val_loss: 0.4621\n",
      "Epoch 48/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8779 - loss: 0.3365 - val_accuracy: 0.8109 - val_loss: 0.4694\n",
      "Epoch 49/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8397 - loss: 0.3799 - val_accuracy: 0.8195 - val_loss: 0.4642\n",
      "Epoch 50/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8739 - loss: 0.3575 - val_accuracy: 0.8281 - val_loss: 0.4740\n",
      "Epoch 51/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8743 - loss: 0.3533 - val_accuracy: 0.8138 - val_loss: 0.4667\n",
      "Epoch 52/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8949 - loss: 0.3303 - val_accuracy: 0.8252 - val_loss: 0.4647\n",
      "Epoch 53/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8941 - loss: 0.3260 - val_accuracy: 0.8195 - val_loss: 0.4768\n",
      "Epoch 54/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8751 - loss: 0.3468 - val_accuracy: 0.8195 - val_loss: 0.4679\n",
      "Epoch 55/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8985 - loss: 0.2976 - val_accuracy: 0.8166 - val_loss: 0.4650\n",
      "Epoch 56/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8881 - loss: 0.3206 - val_accuracy: 0.8052 - val_loss: 0.4754\n",
      "Epoch 57/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8998 - loss: 0.2863 - val_accuracy: 0.8138 - val_loss: 0.4697\n",
      "Epoch 58/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8805 - loss: 0.3339 - val_accuracy: 0.8309 - val_loss: 0.4656\n",
      "Epoch 59/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9027 - loss: 0.2963 - val_accuracy: 0.8223 - val_loss: 0.4682\n",
      "Epoch 60/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9017 - loss: 0.2918 - val_accuracy: 0.8138 - val_loss: 0.4723\n",
      "Epoch 61/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9031 - loss: 0.2855 - val_accuracy: 0.8281 - val_loss: 0.4497\n",
      "Epoch 62/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8955 - loss: 0.3138 - val_accuracy: 0.8309 - val_loss: 0.4545\n",
      "Epoch 63/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8905 - loss: 0.3183 - val_accuracy: 0.8223 - val_loss: 0.4693\n",
      "Epoch 64/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8894 - loss: 0.3041 - val_accuracy: 0.8195 - val_loss: 0.4611\n",
      "Epoch 65/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9167 - loss: 0.2563 - val_accuracy: 0.8252 - val_loss: 0.4615\n",
      "Epoch 66/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8924 - loss: 0.3232 - val_accuracy: 0.8281 - val_loss: 0.4542\n",
      "Epoch 67/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8853 - loss: 0.3248 - val_accuracy: 0.8195 - val_loss: 0.4593\n",
      "Epoch 68/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8987 - loss: 0.3037 - val_accuracy: 0.8052 - val_loss: 0.4662\n",
      "Epoch 69/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8831 - loss: 0.3120 - val_accuracy: 0.8252 - val_loss: 0.4490\n",
      "Epoch 70/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8900 - loss: 0.3107 - val_accuracy: 0.8223 - val_loss: 0.4540\n",
      "Epoch 71/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8889 - loss: 0.3144 - val_accuracy: 0.8309 - val_loss: 0.4601\n",
      "Epoch 72/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8780 - loss: 0.3214 - val_accuracy: 0.8223 - val_loss: 0.4682\n",
      "Epoch 73/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8862 - loss: 0.3109 - val_accuracy: 0.8252 - val_loss: 0.4570\n",
      "Epoch 74/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8869 - loss: 0.3094 - val_accuracy: 0.8281 - val_loss: 0.4719\n",
      "Epoch 75/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8931 - loss: 0.3280 - val_accuracy: 0.8252 - val_loss: 0.4697\n",
      "Epoch 76/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8842 - loss: 0.3231 - val_accuracy: 0.8195 - val_loss: 0.4661\n",
      "Epoch 77/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8705 - loss: 0.3430 - val_accuracy: 0.8252 - val_loss: 0.4682\n",
      "Epoch 78/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8917 - loss: 0.2973 - val_accuracy: 0.8166 - val_loss: 0.4810\n",
      "Epoch 79/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8850 - loss: 0.3025 - val_accuracy: 0.8281 - val_loss: 0.4430\n",
      "Epoch 80/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8907 - loss: 0.3015 - val_accuracy: 0.8338 - val_loss: 0.4287\n",
      "Epoch 81/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8944 - loss: 0.2835 - val_accuracy: 0.8424 - val_loss: 0.4174\n",
      "Epoch 82/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8998 - loss: 0.2853 - val_accuracy: 0.8109 - val_loss: 0.4451\n",
      "Epoch 83/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8728 - loss: 0.3369 - val_accuracy: 0.8281 - val_loss: 0.4516\n",
      "Epoch 84/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9200 - loss: 0.2637 - val_accuracy: 0.8309 - val_loss: 0.4250\n",
      "Epoch 85/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8692 - loss: 0.3162 - val_accuracy: 0.8281 - val_loss: 0.4294\n",
      "Epoch 86/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8910 - loss: 0.3131 - val_accuracy: 0.8424 - val_loss: 0.4242\n",
      "Epoch 87/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8822 - loss: 0.3154 - val_accuracy: 0.8338 - val_loss: 0.4229\n",
      "Epoch 88/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8983 - loss: 0.2997 - val_accuracy: 0.8395 - val_loss: 0.4287\n",
      "Epoch 89/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8978 - loss: 0.2999 - val_accuracy: 0.8338 - val_loss: 0.4229\n",
      "Epoch 90/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9050 - loss: 0.2860 - val_accuracy: 0.8281 - val_loss: 0.4267\n",
      "Epoch 91/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9086 - loss: 0.2640 - val_accuracy: 0.8395 - val_loss: 0.4168\n",
      "Epoch 92/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8831 - loss: 0.2928 - val_accuracy: 0.8338 - val_loss: 0.4279\n",
      "Epoch 93/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9107 - loss: 0.2914 - val_accuracy: 0.8309 - val_loss: 0.4200\n",
      "Epoch 94/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9288 - loss: 0.2471 - val_accuracy: 0.8395 - val_loss: 0.3995\n",
      "Epoch 95/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8787 - loss: 0.2941 - val_accuracy: 0.8195 - val_loss: 0.4116\n",
      "Epoch 96/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9231 - loss: 0.2399 - val_accuracy: 0.8338 - val_loss: 0.4066\n",
      "Epoch 97/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8867 - loss: 0.2824 - val_accuracy: 0.8481 - val_loss: 0.3999\n",
      "Epoch 98/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9193 - loss: 0.2576 - val_accuracy: 0.8195 - val_loss: 0.4098\n",
      "Epoch 99/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8942 - loss: 0.2789 - val_accuracy: 0.8252 - val_loss: 0.4192\n",
      "Epoch 100/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8957 - loss: 0.2643 - val_accuracy: 0.8281 - val_loss: 0.4101\n",
      "Epoch 101/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9065 - loss: 0.2683 - val_accuracy: 0.8309 - val_loss: 0.4025\n",
      "Epoch 102/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9061 - loss: 0.2585 - val_accuracy: 0.8309 - val_loss: 0.4136\n",
      "Epoch 103/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8916 - loss: 0.2945 - val_accuracy: 0.8367 - val_loss: 0.4132\n",
      "Epoch 104/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9114 - loss: 0.2727 - val_accuracy: 0.8338 - val_loss: 0.4022\n",
      "Epoch 105/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9049 - loss: 0.2914 - val_accuracy: 0.8367 - val_loss: 0.4160\n",
      "Epoch 106/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8873 - loss: 0.2968 - val_accuracy: 0.8338 - val_loss: 0.4241\n",
      "Epoch 107/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9012 - loss: 0.2609 - val_accuracy: 0.8338 - val_loss: 0.4061\n",
      "Epoch 108/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9153 - loss: 0.2686 - val_accuracy: 0.8367 - val_loss: 0.3941\n",
      "Epoch 109/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9065 - loss: 0.2717 - val_accuracy: 0.8338 - val_loss: 0.4093\n",
      "Epoch 110/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9069 - loss: 0.2550 - val_accuracy: 0.8252 - val_loss: 0.4162\n",
      "Epoch 111/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.8897 - loss: 0.2900 - val_accuracy: 0.8338 - val_loss: 0.3893\n",
      "Epoch 112/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9024 - loss: 0.2742 - val_accuracy: 0.8252 - val_loss: 0.4008\n",
      "Epoch 113/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9136 - loss: 0.2484 - val_accuracy: 0.8281 - val_loss: 0.3989\n",
      "Epoch 114/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8988 - loss: 0.2712 - val_accuracy: 0.8309 - val_loss: 0.4125\n",
      "Epoch 115/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9095 - loss: 0.2635 - val_accuracy: 0.8338 - val_loss: 0.3849\n",
      "Epoch 116/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.9175 - loss: 0.2485 - val_accuracy: 0.8395 - val_loss: 0.3875\n",
      "Epoch 117/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9141 - loss: 0.2446 - val_accuracy: 0.8367 - val_loss: 0.4194\n",
      "Epoch 118/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9125 - loss: 0.2621 - val_accuracy: 0.8424 - val_loss: 0.3866\n",
      "Epoch 119/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9220 - loss: 0.2306 - val_accuracy: 0.8367 - val_loss: 0.4260\n",
      "Epoch 120/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9060 - loss: 0.2563 - val_accuracy: 0.8395 - val_loss: 0.3958\n",
      "Epoch 121/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8922 - loss: 0.2789 - val_accuracy: 0.8395 - val_loss: 0.3853\n",
      "Epoch 122/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8898 - loss: 0.2804 - val_accuracy: 0.8338 - val_loss: 0.4054\n",
      "Epoch 123/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9092 - loss: 0.2577 - val_accuracy: 0.8080 - val_loss: 0.4161\n",
      "Epoch 124/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8943 - loss: 0.2810 - val_accuracy: 0.8481 - val_loss: 0.3977\n",
      "Epoch 125/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.9059 - loss: 0.2504 - val_accuracy: 0.8367 - val_loss: 0.3806\n",
      "Epoch 126/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9217 - loss: 0.2219 - val_accuracy: 0.8453 - val_loss: 0.3915\n",
      "Epoch 127/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9051 - loss: 0.2430 - val_accuracy: 0.8252 - val_loss: 0.4271\n",
      "Epoch 128/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8988 - loss: 0.2823 - val_accuracy: 0.8453 - val_loss: 0.3912\n",
      "Epoch 129/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9035 - loss: 0.2757 - val_accuracy: 0.8309 - val_loss: 0.4067\n",
      "Epoch 130/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.8794 - loss: 0.3063 - val_accuracy: 0.8338 - val_loss: 0.4012\n",
      "Epoch 131/1000\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.9321 - loss: 0.2189 - val_accuracy: 0.8338 - val_loss: 0.4056\n",
      "Epoch 131: early stopping\n"
     ]
    }
   ],
   "source": [
    "model_history=classifier.fit(x_train,y_train,validation_split=0.33,batch_size=10,epochs=1000,callbacks=early_stop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZXtsAPernQ7T",
    "outputId": "69509280-04b8-46e0-e78f-be60b2f94261",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m9/9\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n"
     ]
    }
   ],
   "source": [
    "y_pred = classifier.predict(x_test)\n",
    "y_pred = (y_pred > 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "id": "lpyqs3zhnQ9d"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "score=accuracy_score(y_pred,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "H45AzTSDnRA2",
    "outputId": "8430dca2-841e-423e-ab0d-0bb3e8765c1d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8257575757575758"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}