Classificacao.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b5d107ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Classificação "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a40226d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# imports\n",
    "import numpy as np\n",
    "from sklearn.datasets import fetch_openml\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.base import BaseEstimator\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score, precision_recall_curve, roc_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9209a1f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\datasets\\_openml.py:1002: FutureWarning: The default value of `parser` will change from `'liac-arff'` to `'auto'` in 1.4. You can set `parser='auto'` to silence this warning. Therefore, an `ImportError` will be raised from 1.4 if the dataset is dense and pandas is not installed. Note that the pandas parser may return different data types. See the Notes Section in fetch_openml's API doc for details.\n",
      "  warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<function Bunch.keys>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# baixar o dataset\n",
    "mnist = fetch_openml('mnist_784', version = 1)\n",
    "mnist.keys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e8c0b306",
   "metadata": {},
   "outputs": [],
   "source": [
    "X , y = mnist['data'], mnist['target']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "50b42e5c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000, 784)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f4e92a47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000,)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e46ed18d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# mostra um numero \n",
    "\n",
    "some_digit = X.iloc[0]\n",
    "some_digit_image = some_digit.values.reshape(28, 28)\n",
    "\n",
    "plt.imshow(some_digit_image, cmap='binary')\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cd052f1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "y[0]\n",
    "y = y.astype(np.uint8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3438ee59",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dividir o conjunto, test e train\n",
    "X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ca8621cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Treinando o classificador binario \n",
    "y_train_5 = y_train == 5\n",
    "y_test_5 = y_test == 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0a354468",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SGDClassifier(random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SGDClassifier</label><div class=\"sk-toggleable__content\"><pre>SGDClassifier(random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "SGDClassifier(random_state=42)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# usar o SGDClassifier, vantagem é lidar bem com um volume grandes de dados\n",
    "# pois ele lida com as instancias de forma independente, uma de cada vez.\n",
    "sgd_clf = SGDClassifier(random_state=42)\n",
    "sgd_clf.fit(X_train, y_train_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "accf9307",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SGDClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# detecta imagens do numero 5\n",
    "sgd_clf.predict([some_digit])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e7d150df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.95035, 0.96035, 0.9604 ])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cáculo do desempenho\n",
    "# usando validação cruzada\n",
    "cross_val_score(sgd_clf, X_train, y_train_5, cv= 3, scoring = 'accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ccf21a97",
   "metadata": {},
   "outputs": [],
   "source": [
    "# classificador ruim, que classifica somente os não 5\n",
    "class NeverClassifier(BaseEstimator):\n",
    "    def fit(self, X, y=None):\n",
    "        pass\n",
    "    def predict(self, X):\n",
    "        return np.zeros((len(X), 1),dtype = bool)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "2ae5f27d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.91125, 0.90855, 0.90915])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# estimar a acurácia\n",
    "never_5_clf = NeverClassifier()\n",
    "cross_val_score(never_5_clf, X_train, y_train_5, cv=3,\n",
    "               scoring='accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a369ea86",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Acurácia não é bom para classificador, ainda mais com dados assimétricos, ou seja, \n",
    "# quando algumas classes estão mais presentes do que outras."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "78905223",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Matriz de Confusão\n",
    "# boa maneira de abaliar o desempenho de classificador \n",
    "y_train_pred = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "45479fec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[53892,   687],\n",
       "       [ 1891,  3530]], dtype=int64)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Usar a matriz de confusão com com a classe alvo e as classes previstas\n",
    "confusion_matrix(y_train_5,y_train_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b7e9117b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# A primeira linha considera as imagens não 5(classe negativa): 53,892 foram classificadas corretamente como não 5(True negativos)\n",
    "# enquanto 0.687 foi classificadas de forma errada como 5s(falsos positivos)\n",
    "# A segunda linha considera como imagens 5s(classe positiva):1.891 classificada errada como não-5s falsos negativos\n",
    "# ao passo que 3.530 restante foi classificadas como perfeitamente como 5s(verdadeiros positivos)\n",
    "# um classificador ideal só teria verdadeiros positivos e verdadeiros negativos, ou seja, só teria a diagonal principal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "68f33092",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[54579,     0],\n",
       "       [    0,  5421]], dtype=int64)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_perfect_predictions = y_train_5 # fingindo que alcançamos a perfeição\n",
    "confusion_matrix(y_train_5, y_train_perfect_predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "2bf847b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# A matriz fornece bastante informação mas talvez uma métrica mais concisa seja melhor.\n",
    "# Vamos obsevar a acurácia das predições positivas, precisão do classificador\n",
    "# Precisao = TP/ (TP + FP)\n",
    "# TP são os verdadeiros positivos e FP são os falsos positivos \n",
    "# Outra métrica é usada, chamada de revocação(sensibilidade ou tava de verdadeiros positivos(TPR)). Proporção de instancias \n",
    "# positivas que são detectadas corretamente pelo classificador\n",
    "\n",
    "# revocacao = TP(TP + FN), FN = falsos negativos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b8d33f1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8370879772350012"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Precision e Recall\n",
    "precision_score(y_train_5, y_train_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "449231c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6511713705958311"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a001cf66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7325171197343846"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# não parece tão bom \n",
    "# Vamos encaixar essas duas medidas em só uma, chamada de F1(score), nada mais é do que uma média harmonica\n",
    "# de predição e revocação, media harmonica da mais importancia aos valores mais baixos, logo tera um score F1 alto\n",
    "# se a revocação e a precisão forem altas\n",
    "# basta usar a função abaixo\n",
    "f1_score(y_train_5, y_train_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bf3b10e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SGDClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([2164.22030239])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# trade-off precision/recall\n",
    "# usando a sgd_classifier com exemplo\n",
    "y_scores = sgd_clf.decision_function([some_digit])\n",
    "y_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b0f98f16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold = 0\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "y_some_digit_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "03350804",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Vamos aumentar o limiar\n",
    "threshold = 8000\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "y_some_digit_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "2fee21ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "# O que usar ?\n",
    "# Vamos primeiro usar a func cross_val_predict para obter os scores de tds as instancias no conjunto de treinamento\n",
    "y_scores = cross_val_predict(sgd_clf, X_train,y_train_5,cv=3,\n",
    "                            method='decision_function')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "21d43d3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# com esses scores vamos usar outra função para calcular a precisão e a revocaçãode todos os limiares possíveis\n",
    "precisions, recalls,thresholds = precision_recall_curve(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "63281c2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Função para plotar precisão e recall em relação ao limiar\n",
    "def plot_precision_recall_vs_threshold(precisions, recalls, thresholds):\n",
    "    plt.plot(thresholds, precisions[:-1], 'b--', label='Precision')\n",
    "    plt.plot(thresholds, recalls[:-1], 'g-', label='Recall')\n",
    "\n",
    "    # Adiciona ponto na curva de precisão no limiar 8000\n",
    "    index_precision_8000 = (np.abs(thresholds - 8000)).argmin()\n",
    "    plt.scatter(thresholds[index_precision_8000], precisions[index_precision_8000], color='blue')\n",
    "\n",
    "    # Adiciona ponto na curva de recall no limiar 8000\n",
    "    index_recall_8000 = (np.abs(thresholds - 8000)).argmin()\n",
    "    plt.scatter(thresholds[index_recall_8000], recalls[index_recall_8000], color='green')\n",
    "\n",
    "    # Adiciona linha tracejada até o eixo y a partir do ponto de precisão\n",
    "    plt.plot([thresholds[index_precision_8000], thresholds[index_precision_8000]], [0, precisions[index_precision_8000]], 'k--')\n",
    "\n",
    "    # Adiciona linha tracejada até o eixo y a partir do ponto de recall\n",
    "    plt.plot([thresholds[index_recall_8000], thresholds[index_recall_8000]], [0, recalls[index_recall_8000]], 'k--')\n",
    "\n",
    "    # Adiciona linha tracejada horizontal do ponto de precisão até o eixo y\n",
    "    plt.plot([-50000, thresholds[index_precision_8000]], [precisions[index_precision_8000], precisions[index_precision_8000]], 'k--')\n",
    "\n",
    "    # Adiciona linha tracejada horizontal do ponto de recall até o eixo y\n",
    "    plt.plot([-50000, thresholds[index_recall_8000]], [recalls[index_recall_8000], recalls[index_recall_8000]], 'k--')\n",
    "\n",
    "    # Adicionar legenda, rótulos de eixo e grade\n",
    "    plt.legend()\n",
    "    plt.xlabel('Threshold')\n",
    "    plt.ylabel('Score')\n",
    "    \n",
    "    # Definir limites personalizados para o eixo X\n",
    "    plt.xlim(-50000, 50000)\n",
    "    plt.ylim(0, 1)\n",
    "\n",
    "    plt.grid(True)\n",
    "\n",
    "# Chame a função com suas listas de precisões, recalls e thresholds\n",
    "plot_precision_recall_vs_threshold(precisions, recalls, thresholds)\n",
    "\n",
    "# Exiba o gráfico\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "348ed836",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# grafico direto precision and recall\n",
    "plt.plot(recalls,precisions)\n",
    "plt.xlim(0,1)\n",
    "plt.ylim(0,1)\n",
    "plt.grid(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "b50800f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# threshold com 90% de precisão\n",
    "threshold_90_precision = thresholds[np.argmax(precisions >=0.9)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "03ead288",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred_90 = y_scores > threshold_90_precision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c701f86",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "96760547",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "cff8ca7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.47980077476480354"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "01e2304a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# A curva ROC(representa a taxa de verdadeiros positivos por taxa de de positivos falsos(FPR), onde o FPR é a proporção\n",
    "# de instancias negativas classificadas incorretamente como positivas)\n",
    "# É igual a taxa de verdadeiros negativos(TNR) que é a proporção de instancias negativas que são corretamente classificadas como\n",
    "# negativas.Também chamado de especificidade(TNR)\n",
    "fpr,tpr, threshold = roc_curve(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "bc97caa2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotar o gráfico\n",
    "def plot_roc_curve(fpr, tpr, label=None):\n",
    "    plt.plot(fpr,tpr, linewidth=2, label = label)\n",
    "    plt.plot([0,1],[0,1], 'k--')\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate(Recall)')\n",
    "    plt.grid(True)\n",
    "plot_roc_curve(fpr, tpr)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "e217e16e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9604938554008616"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# forma de comparar os classificadores é a área sobre a curva(AUC)\n",
    "from sklearn.metrics import roc_auc_score\n",
    "roc_auc_score(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "d25b2802",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Usando a random e comparar a ROC e score ROC AUC.\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "forest_clf = RandomForestClassifier(random_state =42)\n",
    "y_probas_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3,\n",
    "                                   method='predict_proba')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e42e455e",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores_forest = y_probas_forest[:,1] # score = probabilidade de classe positiva\n",
    "fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5, y_scores_forest)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "d1d82fad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAGwCAYAAABVdURTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAACBMklEQVR4nO3dd3gUVRfA4d+mNxJ6EiD0IiC9SVUgEIqASoePDipNINJbwAJSpChNkaogoGJBEAkISEdKEKVJr0GKkJ5sme+PIZuEBMiG3Uw2Oe/zrO7ezM6cvSlzuFWnKIqCEEIIIUQ24aB1AEIIIYQQ1iTJjRBCCCGyFUluhBBCCJGtSHIjhBBCiGxFkhshhBBCZCuS3AghhBAiW5HkRgghhBDZipPWAWQ2k8nEzZs3yZUrFzqdTutwhBBCCJEOiqIQGRlJoUKFcHB4ettMjktubt68SUBAgNZhCCGEECIDrl27RpEiRZ56TI5LbnLlygWolePt7W3Vc+v1erZt20bz5s1xdna26rlFEqnnzCH1nDmknjOP1HXmsFU9R0REEBAQYL6PP02OS24Su6K8vb1tktx4eHjg7e0tvzg2JPWcOaSeM4fUc+aRus4ctq7n9AwpkQHFQgghhMhWJLkRQgghRLYiyY0QQgghshVJboQQQgiRrUhyI4QQQohsRZIbIYQQQmQrktwIIYQQIluR5EYIIYQQ2YokN0IIIYTIViS5EUIIIUS2omly8/vvv9OmTRsKFSqETqfjhx9+eOZ7du3aRfXq1XF1daV06dKsXLnS5nEKIYQQwn5omtxER0dTpUoVFi5cmK7jL126ROvWrWncuDFhYWEMHz6c/v378+uvv9o4UiGEEELYC003zmzZsiUtW7ZM9/FLliyhRIkSfPzxxwCUL1+evXv3MnfuXIKCgmwVpniMoigYTAoGo4LeZMJgVDAk/v9RmdGkoDcmfk0BFBQFFPM51PMoic9J+mLyspTvUY83GAz8/Z8O1zP/4ujolOI8oGBSUr9fUb+ISXlUlqzc/H/lydc2B534NO1i83WeesxjdZl2HSc/XnlCedrHW4vRaOTsDR3Xfr+Eo6Oj9S8gAKnnzCR1bXuKAiaTWs8N4/Tk1WiDUrvaFfzAgQMEBgamKAsKCmL48OFPfE98fDzx8fHm1xEREYC6a6ler7dqfInns/Z5n5feaCI2wUis3kic3kSsPvG5kVi9+rWoeAORcQYi4vTq/2MNRMYnPY+I0xMZbyBOryYu2nOEM2FaB5EDOLLp6j9aB5EDSD1nHqnrzOHIO5Fx5HKzXnJjyb3VrpKb8PBwfH19U5T5+voSERFBbGws7u7uqd4zffp0pk6dmqp827ZteHh42CTO0NBQm5w3wQixRog1QIwBogw6ovSoj0fPYw0Qa9Q9+r/6OsH07O3hhRBCiIwyxjwERcHRM7e5bO/ePZxytd41YmJi0n2sXSU3GTFu3DiCg4PNryMiIggICKB58+Z4e3tb9Vp6vZ7Q0FCaNWuGcwaa4mITjJy6FcHFuzHceBDLzQexhEfEE/4wjvCIOGL1JqvGm17uzg54uznj5eaEu7MjTo46nBx0ODs64OiQ9NzJQYejgw5nRx1Oj77m/KhMp9OhA3TJ8qzkZTp05q/pHv0nsSz5+3ToMJmMXLx4gdKlS+Pk6Pjo67pH50w6n4PusfLHr5nG9ZOupb52SB4vSS9SfI7klZXsC7o0ilOWPf3Yx5k/yxOO1fGEN2aQwWjgxIkTVKlSBSfHbP+nQjM5pZ4VwGQCxQROjz6m3gARD8HBEfLkTjr2vwcQHwc+ucHdTS2LiYFr18DZBUqWSDr25F/w339Q6UXIk0ctu34DjvyhI19+hYYNko7dulXhzNnbdOpYgEKF1G6pc//Ab785UKSwwquvqq3SN2/CzFkOoMDkEBN5H513x2/w3XcO1Kql0Kd3Ugt2t+4OGAzw2RITefKA0QQhIQ6cOaOjUycTnTqqx0ZFQe8+6nXXrzfi+GjUa0iIA3+f0tG3r4lWLdVjT/wJ77/viLe3wvJlSX/7J4c4cOqUjuBgI/XqqmVHj8H06Y44OSl8vdZk/rsQ/K4DcXEw8G0TlSqpZTt3wZ7fHahZSzFfy2iCJUsciI2Bt9824eWlHnvsGFy4qKNCBYWKFdSyBD18/72OXF5QtZpCIX/46+hBPhodjF9AGT78bA2OTgonTpygbVATvD3cnvpzYYnEnpf0sKvfJD8/P27fvp2i7Pbt23h7e6fZagPg6uqKq2vq1NHZ2TlDCUh6pPfccXojJ288ZNfZf/n93F1O3YqwWpePh4sj3m7OeLs74e3mjIerEx7Ojri7OOLm7Ii7syPuLg64O6uvvVyd8HZ3Nr8nl5sz3m7q/12cstaKAXq9ni1bztOqSRmbfQ+FWs/KtTBaViok9WxD6alng0FNDJycwOHRr+Nff6k3n8qVoWrVpOOWLoV//oE5c5Lev3UrHDwIDRtC06Zq2f37MG4cODrCokVJx65aBb/9Bq+/Dq+9ppbdugW9eoGrK2zalHRsUBBcuQIhIdC1q1p24QI0bgze3mqMidq0gZ9/hq+/hs6d1cR87154rSmUKQPnziUd27KlGvPKldC+l1q2cyd07AwVK6Y87/8aq8nNxo3Q6hW1bMNZWDoFXnkFpg9OOnZsD4WTJ4swoqOBllXU299/f8GmxRAYCK0mqMfFlYNxPeHuXWiwAooXV8vP7YQr+6B+MWhVJem8MedAr4dXykDhwo++P3Vh/Caomj/p2IgIKOqgfvagiuDiopZfagYLzsLrtaDpo2Or+8L9P6FOnZTXchmixlWvHhQtqpbVLQrNK0CBAlCyZNKxrbaTSqsqwLDU5W0+e8KxaXitpvp/k8nE9OnTmTx5MiaTCd8CeXgpwJn8+fOjXAvD28PNqn87LDmXXSU3devWZcuWLSnKQkNDqVu3rkYRWU5vNPHHpfss3XORPf/cfTTY9uk8XRzx83Ejn5crPu7O5kdeTxfzI9+j/+f2cCGXmxPOjlkrIREiq1MUMBrVh4tLUqvYP//Av/9CsWJQpIhaFhcH+/apxzRpknSO0FC4fRtq1oQXXlDLrl+Hvn2hfn01CUjUvz8sWwaHDycdt2oVTJsG7drBN98kHVuypNpq8ccf6rlBPdfGjTB0KHzyiVpmNMKgQerzyZMhd271+S+/qMdMmJCU3MTEwOefg7NzyuTm0CFYvVq9ZmJyExenfrbHe/J9fWHbNvUzJzIa1Vh9fFIemzh+98ABaNVKTX4cHNSEzemxO5GHB3h5pSz38IDSpZNu6Ik6dlTrzdMzqaxECejdO+l7kKhDBxOFC1+lUKEi5rIqVeD991MmBW5usHix+hmSj4To0kVNKvLlS3neS5fUz1KwYFLZuHHqIzlvbzhzhlQGD1Yfyfn5waxZqY99bNgpoLZY1amTutyWbt++TY8ePczDMHr27MnChQvx8vLKEuNONU1uoqKiOH/+vPn1pUuXCAsLI2/evBQtWpRx48Zx48YNVq9eDcDbb7/NggULGD16NH379uW3335jw4YNbN68WauPkG53IuNZsvsCG/64RmS8Ic1jyvp6US0gDxUKeROQ150ieTzw93Gz6oAsIeyJokB8vNpqkfzGevSo+q/XypXB318tu35dvYnnyQMdOiQdO3w43LkDH3yg3vQAPv1UvXF07QozZqhlcXFJ14iIgFy51Ofvvw9ffgkzZ8KoUWrZ3bvqTcbZGRISkmINCVFv3nPmJN1YL19WE4PDh1MmNydOqP8/fNiBAgUc8PdXWzD0evWRXGJrTfJZca++Crt2pUwiHB2hbVs1pitXkpKbBg3Uc9aunXSstzdMnZqUdCR6/XX1Rl+vXlJZgQLw1Vepk5ARI6B1a7VFKFHRomoS9vixq1apiU+uXGq9gXqNtO6D332XuqxOHTXRfNxnn6mP5GrVghUrUh87bpyJLVv+pEKFpOTmxRfVx+OS/wwl8vNTH49LbK3JKX777Te6d+9OeHg4Hh4eLFq0iF69emkdVgqaJjdHjhyhcePG5teJY2N69erFypUruXXrFlevXjV/vUSJEmzevJkRI0Ywf/58ihQpwhdffJHlp4Fv/esWI7/5k6jHkppCPm7UL52fWiXy8kq5AhTMZb2+SSFsRa9XuwF0OvWml2jBAvj9d7VbJPGGGxICJ0+q/1Lv3z/p2BYt1Bvdhg1J4yTGjlW7IUaNgnffVcsuXYJSpdRkpEePpPOOHq12naxZA926qWV//w1vvql20SS/Mf3wg3qjHz48KbmJiFBbF+7dSzou+c3YkOxX1c9PTToS4wS1ZadixaSbNKj10batemMvViypvGRJtXXl8en606dDZCSULWvi/HkTjo5qPV29mrqFJCxMPX/iWAiAPn3UR3JOTvDjj6TSsaP6SM7bW23deVyzZuojOS8v6N499bHVqqmP5NzcklqXknu8JUfYJ4PBwJAhQwgPD6dixYps2LCBChUqaB1WKpomN6+88soT1/kA0lx9+JVXXuH48eM2jMq6Dl68x6A1x0jsfXJ1cqB5RT8CyxekVSV/6T4SNnXxIsTGqjfYxGFply+rYxb8/aFGDbXMYFBvwD4+akuFs7OaWHz0US0GD3Yyt3KAOnajUSMoWxbOnk261uTJatIzerR6c0tIUFsFvv8eChVKGVdoqNoaExeXVBYZqXZvJB8zmNiqMHQo7N6d1FVTurSamCS/Yfr6qt05iQlMogkTIDo6qUsJ1KQgKChlcubkpJ7TySmp1QbUFpuZM1Oes2DBlOM+Eo0dqz6SK1QoqdsoucTuBb0eEhuw8+RJmUQlSmyBEUJrTk5OfP311yxZsoSPP/7YZrOOn5ddjbmxN9HxBkZ+c8Kc2Lxa2Z+QNhUpkMuKc+NEtnbtmto6kfyGOX++mmD06QPNm6tlu3apgziDgtSBmIleflntrjlyJCmR+flnNVno2FFtOQH1pr5kifp87Fj1ZlqkCLi6GrlxQ5dmwvG43r3VBCXxWBcXtYXm4UO1qyO5VavUlojkycm4cfDWWynHOAQEqO93dEw5puLxbghQW2zS2sFlwIDUZYUKpU64dDrImzftzyZETrZt2zauXLnCgEe/TFWqVGHx4sUaR/V0ktzY0IdbTnP9v1gA6pTIy/wu1XBMPrdY5Ej37sGNG+rNNHF65okT8N576sDT1q3VssGDkwZ6Dh+uNveDOpD1m2/UAaqJyU3ihMDHdyLJl09tHUneQOrnp7asJB9A6eurdvvcvp2UvDg7Q9u2F5g2zZ+yZZP+VNStq7a6PD5lPfnsnES1a6cc55Hof/9LXVakSMrWFVDHmlh5xQYhRDoZDAZCQkKYPn06Tk5O1KhRg+rVq2sdVrpIn4iNnAmPYO0hdbyQh4sjMztUlsQmB0jc2iHRwIFqEnDqVFLZihXqDI3Zs5MGU7q4qDNfpk5N6v6ZNQt++km9we/fn/T+Xr1g3ryUgzgrVoTNm9WxGcmFhamDaZOPgejQQR0X8tFHKY9dvVpNjpJ3yZQq9ZBatRTy508qS1xLSAiRfV2/fp3GjRszbdo0FEWhX79+WXJszZNIy42NfH0oaSD0iMCyFMvn+ZSjhT2Jj1dbTlxdkwZpRkUlJQXXriW1QJw8qf5/wYKkVhgfH3XMhpMTnD6tzvgpX16doluhArz9tnqch4c6sNNgSJlMJLbsJOftrQ5GFUKI57VlyxZ69uzJvXv3yJUrF1988QWdOnXSOiyLSMuNDcTrjWw8fgMAN2cHOtUK0DgikR4PH6oDcJOv2REaqiYWybtRIiLULpzevZNm1Tg7Q2Jr7W+/JR377rtqC0m/fkllAwao11i2TE1sEm3frg48Tf6PIzc3aSURQmSeCRMm0Lp1a+7du0f16tU5fvy43SU2IMmNTfz+zz0i49S7XqtK/vi4yzo1WjGZ1EXGkq+b8eef6uDbIkXUlhNQZwaNG6dOO/7gg6RjIyPV///7r7oaLKgtKi++qI5NSRyf4uoKc+eqU6ETu5VAHUg7ZkzSYF4hhMjK8j4aVT906FD2799PqVKlNI4oY6RbygZ2nrtjft6mcqGnHCms6fx5dRaOwaCuIQJqa0jiMkgGg5qMVKqkziC6cUNdpRXU6cOJa5Mknw1Uq5a69kr58kktM56eSd1NyTVqZJOPJYQQNhUdHY3no+mIwcHB1KlThwYNGjzjXVmbtNzYwMnrDwFwdNBRt1S+ZxwtMuKvv/JRoYJTiinSly+rrS5r1yYN1G3ePGlF0cSVZHU6tfvn6FE1aUk0apQ6GHjevKSygAB1AG9aM36EEMKeJSQkMHz4cGrWrElUVBSgbtBr74kNSHJjdQYTnL8TDUCZgl64OT9hURBhkbCwlEmHXu/A+fM6Nm1Su5lAXZ69cmW1yyn5uJmbN9WkJfneqi1bqi0xydefcpDfBiFEDnHx4kXq16/P/PnzOXPmDJuS74iaDUi3lJWde6gzb4ZZoZAs0GEpvV7dH+jkSXVl2UTVqqkzgjp1UleVrVz5DvXqmahXz4EyZdRjcuVK2q8nORmQK4QQSb777jv69u1LREQEefLkYdWqVbRp00brsKxKkhsr++dh0p20eQXfpxwprl1TN+O7fh0WLlTLVq9W9yDy8VFnIyVuSJc3r9r6cviwOhXa0RF27TLi7CzNLUIIkR5xcXGMHDmShY/+4NarV4+vv/6aoo9vtZ4NyJ3BysJjk55XK5rGJjE5mMmkTrVOtG0bjB8PW7Yk7dOTOGU6KgqSbyF27x7cvw+vvZZp4QohRLYyatQoc2IzZswYdu3alS0TG5DkxuruxastN27ODhSUPaTMjhxRW1tWr1ZX2zWZ1EQmTx51ltJ//yUdazKpM5tefVW7eIUQIruZMGECL774Ir/88gsfffQRzs7Zd5kS6ZayIkVRuB+vPi+SxwNdDh3sERsLbdrAjh1qC4ynp7qpobe3ur3Anj3q10BtjXlcDq02IYSwqtjYWL7//nu6desGgJ+fHydOnMAhB8yeyP6fMBNFxRvRm9Q7s693zmm1efNNda+kI0fU12fOJG1FMHmyOgXbyUkdLxMXl5TYCCGEsI0zZ85Qp04dunfvzoYNG8zlOSGxAUlurOpBbIL5eW53Fw0jsZ3ISJgyJWllX0VRV/f980/45x+1rFo1dWXeunVh5Eh1U0iAcuWSdq8WQghhG6tXr6ZGjRqcPHmSggULmlcdzkmkW8qK7kQmJTcFsuF4m+hotWsJ1JaZ4sXVtWPefFPdlTr5uk89e6oPIYQQmSM6OpqhQ4eyYsUKAJo0acJXX32Fv7+/xpFlPmm5saL/YpKSm3ye9t9y88cf8MorSZtGurio2xmUKKEOCnZzU8s7dlQX2AuQ/UGFEEITf//9N7Vr12bFihU4ODgwdepUtm3bliMTG5CWG6tK3CwTwMvNPqtWUdcfRKeD5cvVcTSHD6v7MBUuDE2bqtO2k6/sK4QQQlsXLlzg1KlT+Pv7s3btWl555RWtQ9KUfd6Bs6j/YvTm53ntsOXm/fchPFzddNLDAxYvhvbt1fVmEhfTA0lshBAiK1AUxTwrt23btnzxxRe0adOGggULahyZ9qRbyoruRsWbnxfwsq8xNw8eqK0zixapqwMn7s0UGKhuKCmEECLrOHHiBA0aNODatWvmsn79+kli84gkN1YUpzeZn7u5ZP0NM9evV5MXRYHcudWxNdWqqWvTJG+pEUIIkTUoisJnn31GnTp12L9/P++++67WIWVJ0i1lRVHxSWNuPF2ydtX+9BN06aI+P3QIXnpJne107Ji2cQkhhEhbREQEb775JuvXrwegdevWLFq0SOOosiZpubGi5LOl8nhmvWWt33oLvvhCfd62rfoa1G0RhBBCZF3Hjh2jRo0arF+/HicnJ2bNmsVPP/1E/vz5tQ4tS5Lkxooexia13OTxyFoDik+ehM8/hxUrkvZxWrxY7ZKqVUvb2IQQQjzZzp07qVu3LufPn6do0aLs2bOHkSNH5pjVhjMia/ed2JmYR91Srk4OODtmrR+6SpWgZk11fRofH7VM9nASQois76WXXqJcuXKULFmS5cuX58gVhy2Vte7Adi720YBid+es0c+zdKm6/UGixYvVvZ0k2RdCiKzt77//xmg0AuDu7s7OnTv5/vvvJbFJJ7nNWVGcQf1BdHXSvlpjY+GTT2D1ajh1Si2rWVP2dhJCiKxMURTmzp1LtWrVmD59urk8X7585jVtxLNpfxfORgxGdXlfZ0ftfwCdnKBXL7hzJ2mbBCGEEFnX/fv3adeuHcHBwej1ev766y+UxGXjhUUkubGiBKPaLeWiYcvN+vUQEQHOztCokbpVQsmSmoUjhBAiHfbv30/VqlXZtGkTLi4uLFy4kK+//lpaazJIkhsrikl41D+qwQJ+MTHqAOEuXeDTT9Vuqdq11d26hRBCZE0mk4mZM2fSqFEjrl27RunSpTl48CCDBg2SxOY5SHJjJXqjCaNJbT7UYkCxh4e6fUKlSjB1aqZfXgghRAZcuHCByZMnYzQa6dq1K8eOHaNatWpah2X3ZCq4lcQbkrZecMnEaeB79sC//6obXPr7Q2go+Ppm2uWFEEI8hzJlyrBgwQIURaF///7SWmMl0nJjJQZjUnKTWWvcHDyojqv57jswmdRuKUlshBAi6zKZTEybNo3Dhw+by/r378+AAQMksbEiSW6sRG9MGtHu6JA5P6AlSkDTprBjByQkPPt4IYQQ2rl9+zYtWrRgwoQJdO7cmejoaK1DyrYkubGSyDi9+Xkut8zp7fP1hQED4I8/ZLq3EEJkZb/99htVq1YlNDQUd3d3QkJC8PT01DqsbEuSGytJSNYtZetF/JYuhchI9XnnzlC0qE0vJ4QQIoOMRiNTpkwhMDCQ8PBwKlasyJEjR+jdu7fWoWVrktxYiSGTuqVWrIA33wRvb7h+3WaXEUII8ZwiIiIIDAxk6tSpKIpC3759OXz4MBUqVNA6tGxPZktZicGUlNw42XBAcfHi0L+/Ou27SBGbXUYIIcRz8vLywtPTE09PT5YsWcL//vc/rUPKMSS5sRKjKalbyskGLTeJK3A3bgzly0OuXFa/hBBCiOdkMBjQ6/W4u7vj4ODAqlWruHv3LuXKldM6tBxFuqWsxJbdUoqiJjOHDqmv/fxAxqEJIUTWcv36dZo0acLbb79tLsuXL58kNhqQ5MZKkvVKYe2Gm2++gehoGD0a7t617rmFEEI8vy1btlC1alX27NnD999/z+XLl7UOKUeT5MYGdFg3u3n1VXjjDQgLg/z5rXpqIYQQz0Gv1zN69Ghat27NvXv3qF69OseOHaN48eJah5ajSXJjJQq225bewwOGDYMrV2x2CSGEEBa6evUqL7/8MrNmzQJg6NCh7N+/n9KlS2scmZABxVaiJO+WslLKaDCAo6O6rUKjRtY5pxBCiOdnMplo0aIFp0+fxsfHh+XLl/PGG29oHZZ4RFpurCT5VHBHK+0PMmkSFC4MH31kldMJIYSwEgcHB+bPn89LL73E8ePHJbHJYiS5sRJTsuTGwUojiqtUgagoyJPHKqcTQgjxHC5evEhoaKj5dbNmzdi3bx8lSpTQMCqRFklurCTFIn5WSm66dIGAAOjY0SqnE0IIkUHfffcd1apVo0OHDly4cMFc7mCtcQjCquS7YiXG5C03z9ktlXz8zokTkDfvc51OCCFEBsXFxTFkyBA6dOhAREQEFStWxNnZWeuwxDNIcmMlipK8W+r5ztWzJ6xdC7Gx4CRDvoUQQhP//PMP9erVY+HChQCMHj2a3bt3U1R2K87yJLmxkpSL+GW85ebPP+Grr6B7d9ixwwqBCSGEsNi6deuoUaMGx48fJ1++fGzevJkZM2ZIq42dkOTGSkzJWm6ep1PK1xfWr4egIGjd+vnjEkIIYblDhw4RGRlJw4YNCQsLo1WrVlqHJCwgnR5Wkjy5eZ69pXx9oVMn9SGEECLzKIqC7lHL+4wZMyhdujRvvfUWTjI+wO5Iy42VJB9QnNHkJiwMrl61UkBCCCHS7auvvqJ169YYDAYAXFxcGDx4sCQ2dkqSGytJPsMpo+02rVrBBx/Av/9aJSQhhBDPEB0dTd++fenRowe//PILK1as0DokYQWS3FhJip2lMjCgOC4OwsNh6VI4fdpqYQkhhHiCv//+m9q1a7NixQp0Oh1Tpkyhb9++WoclrEDa22wgIy03bm5w4wYcPSr7SAkhhC0pisKKFSsYPHgwsbGx+Pn5sXbtWho3bqx1aMJKNG+5WbhwIcWLF8fNzY06depw+PDhpx4/b948ypUrh7u7OwEBAYwYMYK4uLhMivbJkq9zk1H+/vDqqxlq+BFCCJFO77//Pn379iU2NpZmzZpx4sQJSWyyGU2Tm/Xr1xMcHExISAjHjh2jSpUqBAUF8e8TBp2sXbuWsWPHEhISwunTp1m2bBnr169n/PjxmRz501manNy6pQ4mFkIIYXsdO3bE29ubDz/8kK1bt1KwYEGtQxJWpmm31Jw5cxgwYAB9+vQBYMmSJWzevJnly5czduzYVMfv37+f+vXr061bNwCKFy9O165dOXTo0BOvER8fT3x8vPl1REQEAHq9Hr1eb7XPYjAazc+NRqNF5373XUdCQ3X06mVi+nST1WLKrhLr1prfP5Ga1HPmkHq2PUVROHHiBBUrVgSgdOnSnDt3jrx582I0GjEm+/stnp+tfqYtOZ9myU1CQgJHjx5l3Lhx5jIHBwcCAwM5cOBAmu+pV68eX331FYcPH6Z27dpcvHiRLVu20KNHjydeZ/r06UydOjVV+bZt2/Dw8Hj+D/LIids6wBGAM6dPs+X+qXS/9/Dhl7l7NzdXr55ny5YzVospu0u+O6+wHannzCH1bBsxMTEsXryYffv28f7771OxYkWp60xi7XqOiYlJ97GaJTd3797FaDTi6+ubotzX15czZ9K+wXfr1o27d+/SoEEDFEXBYDDw9ttvP7Vbaty4cQQHB5tfR0REEBAQQPPmzfH29rbOhwGijlxn3UU1oalYoQKt6hRL93tbtoTwcD158pTEza2k1WLKrvR6PaGhoTRr1kyWQrchqefMIfVsO8ePH6d79+6cP38eR0dH8998qWvbstXPdGLPS3rY1WypXbt2MW3aNBYtWkSdOnU4f/48w4YN4/3332fSpElpvsfV1RVXV9dU5c7OzlatdAdHR/NzJydHi88t+7BZztrfQ5E2qefMIfVsPYqisGjRIoKDg0lISKBo0aKsW7eOmjVrsmXLFqnrTGLterbkXJolN/nz58fR0ZHbt2+nKL99+zZ+fn5pvmfSpEn06NGD/v37A1CpUiWio6N58803mTBhAg7Pux33c0i+/YIlG2eePQtly8oMKSGEsIYHDx7Qv39/vvvuOwDatm3LihUryJs3r4xrykE0ywZcXFyoUaMGO5JtfW0ymdixYwd169ZN8z0xMTGpEhjHRy0m1piK/TyS7wqe3kTl33+halWoVw+SjXkWQgiRQT/88APfffcdzs7OzJ07lx9++IG8efNqHZbIZJp2SwUHB9OrVy9q1qxJ7dq1mTdvHtHR0ebZUz179qRw4cJMnz4dgDZt2jBnzhyqVatm7paaNGkSbdq0MSc5WlEy0HKzd6+6MnFYGKTRcyaEEMJCvXr14s8//6Rr167UqlVL63CERjRNbjp37sydO3eYPHky4eHhVK1ala1bt5oHGV+9ejVFS83EiRPR6XRMnDiRGzduUKBAAdq0acOHH36o1UcwMxgt3zizZUv4+2+QllIhhMiY+/fvM3HiRKZPn46Pjw86nY45c+ZoHZbQmOYDiocMGcKQIUPS/NquXbtSvHZyciIkJISQkJBMiCzj0jt8xt0dKlSwaShCCJFtHThwgC5dunD16lUePnzImjVrtA5JZBGab78ghBBCWMJkMjFr1iwaNWrE1atXKVWqFO+++67WYYksRJIbK7F0OPPu3dClC6xbZ5NwhBAiW7p79y5t2rRh9OjRGAwGOnfuzLFjx6hevbrWoYksRJIbG9ClY0DxkiWwfj38+msmBCSEENlAWFgYVatWZcuWLbi6uvLZZ5/x9ddfW3VBVpE9aD7mJruwdCr6kCHQoIEs3ieEEOlVpEgRAMqVK8eGDRuoXLmyxhGJrEqSGxtIz4Di+vXVhxBCiCeLiIgwt8zkz5+fX3/9lWLFiuHl5aVxZCIrk24pIYQQWdLOnTspV64cq1atMpdVrFhREhvxTJLc2MCzhtz89ResWqUu3ieEECIlo9HI1KlTCQwMJDw8nIULF2IymbQOS9gRSW6sxJIhN/37Q+/e8MknNgtHCCHs0q1bt2jevDlTpkzBZDLRp08fdu7cqenegcL+WDTm5vTp06xbt449e/Zw5coVYmJiKFCgANWqVSMoKIj27dunuQO3SKlXL/D0hJIltY5ECCGyjtDQUP73v//x77//4unpyeLFi+nRo4fWYQk7lK7k5tixY4wePZq9e/dSv3596tSpw+uvv467uzv379/nr7/+YsKECQwdOpTRo0czfPhwSXKeYuBACAyUmVJCCJHo4sWLtGzZEqPRSKVKldiwYQMvvPCC1mEJO5Wu5KZ9+/aMGjWKb7/9lty5cz/xuAMHDjB//nw+/vhjxo8fb60Ys6UyZbSOQAghso6SJUsyZswY7t27x9y5c3F3d9c6JGHH0pXcnDt3Dmdn52ceV7duXerWrYtedoJ8ovXroUgRqFIFZMC/ECIn++WXXyhXrhwlH/XRf/DBB+laBFWIZ0nXCK30JDbPc3x2oKRzA4YxY9TF+375xcYBCSFEFqXX6xk9ejStWrWiS5cuJCQkAOlb3V2I9EhXy80nFkzreeeddzIcjD1LPlvqSb+gigIvvghXrkClSpkUmBBCZCFXr16lS5cuHDhwAIDatWtbvMK7EM+SruRm7ty56TqZTqfLsclNck/6t4dOBz//nKmhCCFElvHTTz/Ru3dv/vvvP3x8fFi2bBnt27fXOiyRDaUrubl06ZKt47B78u8OIYRIW0JCAmPHjjX/Q7lWrVqsW7fOPNZGCGuTVZFs4Endxvfvw3//ZW4sQgihNUVR+P333wEYPnw4e/fulcRG2FS6Wm6Cg4PTfcI5c+ZkOBh7lp4u49Klwc0Npk6FAQNsH5MQQmhJURR0Oh2urq5s2LCBkydP0q5dO63DEjlAupKb48ePp+tkMtJdpXvCqJvEVptHEwOEECJbio+PZ+TIkeTOnZv3338fUNexkdYakVnSldzs3LnT1nHkCPHxsGULNG2qdSRCCGEb58+fp3Pnzhw7dgwHBwd69epF6dKltQ5L5DAy5iYTubjAa69BrlxaRyKEENa3YcMGqlevzrFjx8iXLx8//fSTJDZCExZtnJnoyJEjbNiwgatXr5oXX0q0ceNGqwSW3SjKkwcaCyGEPYuNjWXEiBF89tlnADRo0ICvv/6aIkWKaByZyKksbrlZt24d9erV4/Tp03z//ffo9Xr+/vtvfvvtN3x8fGwRo1141grFW7bAO+/At99mUkBCCJEJFEUhMDCQzz77DJ1Ox/jx49m5c6ckNkJTFic306ZNY+7cuWzatAkXFxfmz5/PmTNn6NSpE0Vz8DbXKVcoTv31Tz6BTz+FgwczLyYhhLA1nU7HgAEDKFCgAFu3buXDDz/EySlDnQJCWI3Fyc2FCxdo3bo1AC4uLkRHR6PT6RgxYgSff/651QO0R2klN8OGwdtvw0svZX48QghhTTExMZw+fdr8unfv3pw9e5bmzZtrGJUQSSxOr/PkyUNkZCQAhQsX5q+//qJSpUo8ePCAmJgYqwdoL0ympKYbhzSym1atoFkzyIF7igohspFTp07RqVMnHj58SFhYGPny5QPUe4MQWYXFLTeNGjUiNDQUgI4dOzJs2DAGDBhA165daZqD5zibntEtBZLYCCHs28qVK6lZsyZ///03BoOBy5cvax2SEGmyuOVmwYIFxMXFATBhwgScnZ3Zv38/7du3Z+LEiVYP0F4kH1D8eG6TkAB//gk+PlCmTObGJYQQzysqKorBgwezevVqAAIDA/nqq6/w9fXVODIh0mZxcpM3b17zcwcHB8aOHWvVgOxV8pabx7ulrlyBWrXUrRdiYzM5MCGEeA4nT56kU6dOnDlzBgcHB9577z3GjRuHg4MskyayLouTmy1btuDo6EhQUFCK8m3btmE0GmnZsqXVgrMrKaZLpfySwQBFi8LVq5kbkhBCPK8ZM2Zw5swZChUqxNdff02jRo20DkmIZ7I49R47dixGozFVuclkklacRx7fW6p8ebX1Jj2bawohRFaycOFC+vfvT1hYmCQ2wm5YnNz8888/VKhQIVX5Cy+8wPnz560SlBBCCG0cP36cUaNGoTz615iPjw9Lly6lQIECGkcmRPpZnNz4+Phw8eLFVOXnz5/H09PTKkFlN+Hh8PCh1lEIIcSTKYrCokWLeOmll5g9ezYrV67UOiQhMszi5KZdu3YMHz6cCxcumMvOnz/Pu+++S9u2ba0aXHbRqRN06ADr12sdiRBCpPbw4UM6derE4MGDSUhIoE2bNrRr107rsITIMIuTm5kzZ+Lp6ckLL7xAiRIlKFGiBOXLlydfvnzMnj3bFjHahacNp9HrYft2OHcu08IRQoh0+eOPP6hWrRrffvstzs7OzJkzhx9//DHFzFgh7I3Fs6V8fHzYv38/oaGhnDhxAnd3dypXriwDzZJ5fBG/efPg0CF47TUtohFCiLQtX76ct99+G71eT/HixVm/fj21a9fWOiwhnluGdjfT6XQ0b96cRo0a4erqiu5JS/LmIE+bCVWnjvoQQoispHTp0hiNRt544w2WLVtG7ty5tQ5JCKuwuFvKZDLx/vvvU7hwYby8vLh06RIAkyZNYtmyZVYP0B5JqieEyKoePHhgft6oUSMOHTrEt99+K4mNyFYsTm4++OADVq5cycyZM3FxcTGXv/jii3zxxRdWDS47uHULQkLg4EGtIxFC5GQmk4nZs2dTokQJzpw5Yy6vWbOmtL6LbMfi5Gb16tV8/vnndO/eHUdHR3N5lSpVUvzCCNXJk/Dee/Dyy1pHIoTIqe7evUvbtm0ZNWoUDx484Msvv9Q6JCFsyuIxNzdu3KB06dKpyk0mE3q93ipBZSclS0Lr1nDzptaRCCFyor1799K1a1euX7+Oq6sr8+fP580339Q6LCFsyuKWmwoVKrBnz55U5d9++y3VqlWzSlDZSenS8PPPsH+/1pEIIXISk8nE9OnTeeWVV7h+/Tply5bl0KFDvPXWW9INJbI9i1tuJk+eTK9evbhx4wYmk4mNGzdy9uxZVq9ezc8//2yLGLMFNzetIxBC5CQrV65k/PjxAPzvf/9j8eLFeHl5aRyVEJkjQysUb9q0ie3bt+Pp6cnkyZM5ffo0mzZtolmzZraI0a7JZplCCC307NmTZs2asWzZMlavXi2JjchRMrTOTcOGDQkNDU1VfuTIEWrWrPncQWUnHTvCkSPw0UfQpYvW0Qghsiuj0ciyZcvo3bs3Li4uODk58euvv0oXlMiRLG65iYqKIjY2NkVZWFgYbdq0oY6sVJfKn3/ClSsQH691JEKI7Co8PJzmzZvz1ltvMXbsWHO5JDYip0p3cnPt2jXq1q2Lj48PPj4+BAcHExMTQ8+ePalTpw6enp7sl1GzqWzdCitXylRwIYRtbN++napVq/Lbb7/h4eEhEzuEwIJuqVGjRhEXF8f8+fPZuHEj8+fPZ8+ePdSpU4cLFy5QpEgRW8Zpt0qWVB9CCGFNBoOBqVOn8uGHH6IoCpUqVWLDhg288MILWocmhObSndz8/vvvbNy4kZdeeolOnTrh5+dH9+7dGT58uA3DE0II8bgbN27QrVs3fv/9dwAGDBjA/PnzcXd31zgyIbKGdHdL3b59mxIlSgBQsGBBPDw8aNmypc0Cyw7i4qBPH9i4EUwmraMRQmQXsbGxHD9+HC8vL9auXcvnn38uiY0QyVg0W8rBwSHF8+R7S4nUrl9Xx9usXCnJjRDi+SiKYh4gXLp0aTZs2ECpUqUoU6aMxpEJkfWkO7lRFIWyZcuaf7mioqKoVq1aioQH4P79+9aN0I4ZDNCyJeh06kMIITLi2rVrdO/encmTJxMYGAhAixYtNI5KiKwr3cnNihUrbBmH3VNIvVrfCy/Apk1w+rQGAQkhsoVNmzbRu3dv7t+/z+DBgzl16lSKTYuFEKmlO7np1auXLePIVpK30jg6wosvaheLEMI+JSQkMG7cOObMmQNAzZo1Wb9+vSQ2QqRDupKb5H29QgghbOvy5ct07tyZw4cPAzBs2DBmzJiBq6urxpEJYR/SNVuqYsWKrFu3joSEhKce988//zBw4EA++ugjqwRnT9LaQ+rDDyFvXni0d50QQjzTtWvXqFatGocPHyZ37tx8//33zJs3TxIbISyQruTm008/Zfbs2fj5+dG5c2dmzZrFmjVr+O677/jiiy8IDg6mdu3aVK1aFW9vbwYOHJjuABYuXEjx4sVxc3OjTp065n+pPMmDBw8YPHgw/v7+uLq6UrZsWbZs2ZLu62UGHWor14MH8N9/8IycUAghzIoUKUKbNm146aWXCAsL47XXXtM6JCHsTrq6pZo2bcqRI0fYu3cv69evZ82aNVy5coXY2Fjy589PtWrV6NmzJ927dydPnjzpvvj69esJDg5myZIl1KlTh3nz5hEUFMTZs2cpWLBgquMTEhJo1qwZBQsW5Ntvv6Vw4cJcuXKF3Llzp/uamWncOOjXD3x8tI5ECJGV3bp1i3v37uHn54dOp2PJkiU4Ozvj7OysdWhC2CWL1rlp0KABDRo0sNrF58yZw4ABA+jTpw8AS5YsYfPmzSxfvjzF5m+Jli9fzv3799m/f7/5l7548eJPvUZ8fDzxyXatjIiIAECv16PX6630ScCUbCEbg8GAXq8nVy7IlYtH17PapXK8xO+bNb9/IjWp58yxbt06goOD2bRpE99//z06nc78903q3rrkZzpz2KqeLTmfTlHSGi2SPnFxcbi5uWXovQkJCXh4ePDtt9+maHbt1asXDx484Mcff0z1nlatWpE3b148PDz48ccfKVCgAN26dWPMmDFPnEEwZcoUpk6dmqp87dq1eHh4ZCj2tGy66sD2G2ov35AKRsr4ZLhahRA5QEJCAsuXL2fr1q0AlC9fnokTJ+Lp6alxZEJkTTExMXTr1o2HDx/i7e391GMtarkBMBqNTJs2jSVLlnD79m3OnTtHyZIlmTRpEsWLF6dfv37pOs/du3cxGo34+vqmKPf19eXMmTNpvufixYv89ttvdO/enS1btnD+/HkGDRqEXq8nJCQkzfeMGzeO4OBg8+uIiAgCAgJo3rz5MyvHEqdD/2H7jUsA1KhRgwZlC7J6tY6HD3W0a2eiaFGrXSrH0+v1hIaG0qxZM2m2tyGpZ9s5d+4c3bp1488//wSgffv2LF++XLZQsDH5mc4ctqrnxJ6X9LA4ufnwww9ZtWoVM2fOZMCAAebyF198kXnz5qU7uckIk8lEwYIF+fzzz3F0dKRGjRrcuHGDWbNmPTG5cXV1TXOWgbX7s5Ov1Ozk5ISzszMffQTnz0ONGo6UKmW1S4lHZExC5pB6tq41a9bw1ltvER0dTYECBVixYgUGgwF3d3ep50wiP9OZw9r1bMm50r1xZqLVq1fz+eef07179xRdQVWqVHlii0ta8ufPj6OjI7dv305Rfvv2bfz8/NJ8j7+/P2XLlk1x3fLlyxMeHv7MaepaaNcOunSBgACtIxFCZAUxMTFMnDiR6OhoXnnlFcLCwmjevLnWYQmR7Vic3Ny4cYPSpUunKjeZTBYN9nFxcaFGjRrs2LEjxTl27NhB3bp103xP/fr1OX/+fIrBu+fOncPf31/zTTzTGmEzezZ8/TWULJnp4QghsiAPDw/Wr19PSEgI27dvp1ChQlqHJES2ZHFyU6FCBfbs2ZOq/Ntvv6VatWoWnSs4OJilS5eyatUqTp8+zcCBA4mOjjbPnurZsyfjxo0zHz9w4EDu37/PsGHDOHfuHJs3b2batGkMHjzY0o9hdcmHZctizkKIRKtWrWL58uXm17Vr12bKlCmyjYIQNmTxmJvJkyfTq1cvbty4gclkYuPGjZw9e5bVq1fz888/W3Suzp07c+fOHSZPnkx4eDhVq1Zl69at5kHGV69eTTGWJSAggF9//ZURI0ZQuXJlChcuzLBhwxgzZoylH8PmFEV9OFicPgohsoOoqCgGDx7M6tWrcXV1pUGDBpQtW1brsITIESxObtq1a8emTZt477338PT0ZPLkyVSvXp1NmzbRrFkziwMYMmQIQ4YMSfNru3btSlVWt25dDh48aPF1bO3xXcHPnIEKFdTxNlevahSUEEITJ0+epFOnTpw5cwYHBwcmTpxIKZlVIESmsTi5AWjYsCGhoaHWjsW+PdYtde+e+vzaNW3CEUJkPkVRWLZsGUOHDiUuLo5ChQqxdu1aXn75Za1DEyJHsbjTpGTJktxLvHMn8+DBA0rKyFlA3VvqpZcgPBweLWMhhMjmFEWhV69eDBgwgLi4OFq0aEFYWJgkNkJowOLk5vLlyxiNxlTl8fHx3LhxwypBZQdOTuDrC5UqaR2JECIz6HQ6ypQpg6OjIx999BGbN2+mQIECWoclRI6U7m6pn376yfz8119/xSfZbpBGo5EdO3Y8c58nIYTIThRF4cGDB+YNg8ePH0/btm2pUqWKxpEJkbOlO7lJ3P9Jp9PRq1evFF9zdnamePHifPzxx1YNzp5t2ADbtkGbNupifkKI7OXhw4cMGDCAs2fPcvDgQdzd3XF0dJTERogsIN3JTeLCeSVKlOCPP/4gf/78NgsqO/jnH1i2DE6elORGiOzmyJEjdO7cmYsXL+Lk5MS+ffsIDAzUOiwhxCMWz5a6dOmSLeLIdmrVgkGDoEEDrSMRQliLoih8+umnjBw5Er1eT7FixVi/fj116tTROjQhRDIZmgoeHR3N7t27uXr1aqo9nd555x2rBGbvmjdXH0KI7OG///6jb9++/PDDD4DaVb98+XLzeBshRNZhcXJz/PhxWrVqRUxMDNHR0eTNm5e7d+/i4eFBwYIFJbkRQmRLgwYN4ocffsDFxYXZs2czZMgQdLLXihBZksVTwUeMGEGbNm3477//cHd35+DBg1y5coUaNWowe/ZsW8Rol6KiIC4u5Z5TQgj7NWPGDGrVqsX+/fsZOnSoJDZCZGEWJzdhYWG8++67ODg44OjoSHx8PAEBAcycOZPx48fbIka7FBgI7u7wqAVbCGFn7t27x8qVK82vixYtyqFDh6hRo4Z2QQkh0sXi5MbZ2dm8mWXBggW5+mjjJB8fH67JXgNmDx6o/0+2HJAQwk7s27ePqlWr0qdPHzZt2mQul9YaIeyDxWNuqlWrxh9//EGZMmV4+eWXmTx5Mnfv3uXLL7/kxRdftEWMdun0aYiJARcXrSMRQqSXyWRi5syZTJw4EaPRSJkyZQgICNA6LCGEhSxuuZk2bRr+/v4AfPjhh+TJk4eBAwdy584dPvvsM6sHaK90OvD0BGdnrSMRQqTHv//+S6tWrRg3bhxGo5Fu3bpx9OhRqlatqnVoQggLWdxyU7NmTfPzggULsnXrVqsGJIQQmW337t107dqVW7du4ebmxoIFC+jbt690QwlhpyxuuXmSY8eO8eqrr1rrdHYtNhaGDIHRo2W2lBD24NatW9y6dYvy5cvzxx9/0K9fP0lshLBjFrXc/Prrr4SGhuLi4kL//v0pWbIkZ86cYezYsWzatImgoCBbxWlXoqJg4UJ1vM3MmVpHI4RIi6Io5gSmS5cuJCQk0L59ezw9PTWOTAjxvNLdcrNs2TJatmzJypUrmTFjBi+99BJfffUVdevWxc/Pj7/++ostW7bYMla74eoKkybByJFaRyKESMuOHTuoXr064eHh5rKePXtKYiNENpHu5Gb+/PnMmDGDu3fvsmHDBu7evcuiRYs4efIkS5YsoXz58raM0654e8N778GHH2odiRAiOaPRyOTJk2nWrBlhYWFMnTpV65CEEDaQ7m6pCxcu0LFjRwDeeOMNnJycmDVrFkWKFLFZcEIIYS03b96kW7du7N69G4D+/fvz8ccfaxyVEMIW0p3cxMbG4uHhAagLWbm6upqnhIuUDAaIiFBXKJap4EJo79dff+V///sfd+/excvLi88++4xu3bppHZYQwkYsGlD8xRdf4OXlBYDBYGDlypXkz58/xTGycSbs2aPj5begalU4flzraITI2b755hs6deoEQJUqVdiwYQNly5bVOCohhC2lO7kpWrQoS5cuNb/28/Pjyy+/THGMTqeT5AZwdFT/f/q0tnEIIaBFixaULVuWwMBAPv74Y9zc3LQOSQhhY+lObi5fvmzDMLKXRo0UEhJAr9c6EiFypoMHD1KnTh10Oh25cuXijz/+wNvbW+uwhBCZxGqL+OV0j6/V5+wMj4YoCSEySUJCAiNHjqRu3brMmzfPXC6JjRA5i8XbL4hnk4VNhch8ly9fpkuXLhw6dAiAGzduaByREEIr0nJjJUqyfRa2btXRsyf8+KOGAQmRg/zwww9Uq1aNQ4cOkTt3br7//ntmz56tdVhCCI1IcmMlyfeQ+vOEA19+CT/8oFk4QuQI8fHxDBs2jNdff50HDx5Qp04djh8/zmuvvaZ1aEIIDUm3lA00bmzCzwnq19c6EiGyt1OnTrFo0SIA3n33XaZNm4aLi4vGUQkhtJah5ObChQusWLGCCxcuMH/+fAoWLMgvv/xC0aJFqVixorVjtAvJBxTXqQPvyPpgQthctWrV+PTTTylSpAivvvqq1uEIIbIIi7uldu/eTaVKlTh06BAbN24kKioKgBMnThASEmL1AO1F8m4pGU8shG3ExcUxbNgw/vzzT3PZ22+/LYmNECIFi5ObsWPH8sEHHxAaGpqi+bdJkyYcPHjQqsHZq4gIePAAjEatIxEi+zh37hwvvfQSn3zyCZ07d8ZgMGgdkhAii7I4uTl58iSvv/56qvKCBQty9+5dqwRlj5RkHVPvvutAnjzwyy8aBiRENrJ27Vpq1KjBiRMnKFCgAPPmzcPJSYYMCiHSZnFykzt3bm7dupWq/Pjx4xQuXNgqQdm7uDi1WnPn1jYOIexdTEwMAwYMoHv37kRFRfHyyy8TFhZGUFCQ1qEJIbIwi5ObLl26MGbMGMLDw9HpdJhMJvbt28fIkSPp2bOnLWK0O199aSA2Fl56SetIhLBf4eHh1KlThy+++AKdTsfkyZPZvn07hQoV0jo0IUQWZ3G77rRp0xg8eDABAQEYjUYqVKiA0WikW7duTJw40RYx2oXkA4rRgezNJ8TzKVCgAAULFsTX15c1a9bQtGlTrUMSQtgJi5MbFxcXli5dyqRJk/jrr7+IioqiWrVqlClTxhbx2SWd7L8gRIZER0fj6OiIm5sbjo6OrFmzBgA/Pz+NIxNC2BOLu6X27t0LQNGiRWnVqhWdOnWSxOYx/fs7MGoUxMRoHYkQ9uOvv/6iVq1ajBgxwlzm5+cniY0QwmIWJzdNmjShRIkSjB8/nlOnTtkiJrt3+rQDs2c/1lUlhEiToigsW7aMWrVqcfr0aX766Sfu3bundVhCCDtmcXJz8+ZN3n33XXbv3s2LL75I1apVmTVrFtevX7dFfHapX18j77wD7u5aRyJE1hYZGUmPHj3o378/cXFxBAUFERYWRr58+bQOTQhhxyxObvLnz8+QIUPYt28fFy5coGPHjqxatYrixYvTpEkTW8Rod/r2U5g/HxxkW1IhnujEiRPUrFmTNWvW4OjoyPTp09myZQsFChTQOjQhhJ17rlWwSpQowdixY6lSpQqTJk1i9+7d1opLCJGNxcfH06pVK27evEmRIkVYt24d9WWnWSGElWS4bWHfvn0MGjQIf39/unXrxosvvsjmzZutGZvdevgQYmO1jkKIrMvV1ZXFixfz6quvEhYWJomNEMKqLG65GTduHOvWrePmzZs0a9aM+fPn065dOzw8PGwRn11q1cqJop7wzz9aRyJE1nH06FH+++8/AgMDAWjbti1t2rSRpROEEFZncXLz+++/M2rUKDp16kT+/PltEVO2cP681hEIkTUoisKCBQsYOXIkXl5ehIWFERAQAMiaUEII27A4udm3b58t4shWdu00UEm22RKC//77j379+vH9998D0KhRI7y8vDSOSgiR3aUrufnpp59o2bIlzs7O/PTTT089tm3btlYJzJ45u4D8/RY53aFDh+jSpQuXL1/GxcWF2bNnM2TIEGmtEULYXLqSm9dee43w8HAKFizIa6+99sTjdDodRqPRWrEJIeyQoijMnTuXMWPGYDAYKFmyJBs2bKBGjRpahyaEyCHSNVvKZDJRsGBB8/MnPSSxUc2cqePzz7WOQght6HQ6zpw5g8FgoGPHjhw7dkwSGyFEprJ4Kvjq1auJj49PVZ6QkMDq1autEpS9+/FHR775RusohMhcJpPJ/Hz+/Pl89dVXrF+/Hh8fHw2jEkLkRBYnN3369OHhw4epyiMjI+nTp49VgrJ3nTsZqVhR6yiEyBwmk4kZM2bw6quvmhMcd3d3unfvLuNrhBCasHi2lKIoaf7Bun79uvwL7ZF3hinULKF1FELY3p07d+jZsydbt24F4Mcff+T111/XOCohRE6X7uSmWrVq6HQ6dDodTZs2xckp6a1Go5FLly7RokULmwQphMh6fv/9d7p27crNmzdxc3NjwYIFT51wIIQQmSXdyU3iH62wsDCCgoJSrFXh4uJC8eLFad++vdUDtEexsaDXg7Oz1pEIYX1Go5Hp06cTEhKCyWSifPnybNiwgRdffFHr0IQQArAguQkJCQGgePHidO7cGTc3N5sFZe8CA53o3gKWL9c6EiGsb9CgQXz+aDpg7969WbBgAZ6enhpHJYQQSSweUNyrVy9JbNLhv/+0jkAI2xg4cCB58+Zl1apVrFixQhIbIUSWk67kJm/evNy9exeAPHnykDdv3ic+MmLhwoUUL14cNzc36tSpw+HDh9P1vnXr1qHT6bJcP/+mTQY+/VTrKISwDqPRyIEDB8yvq1atypUrV+jZs6eGUQkhxJOlq1tq7ty55MqVy/zcmtM7169fT3BwMEuWLKFOnTrMmzePoKAgzp49a144MC2XL19m5MiRNGzY0GqxWEvevFCkiNZRCPH87t+/T1BQEAcOHGDfvn3UqlULQPaHEkJkaelKbnr16mV+3rt3b6sGMGfOHAYMGGBeI2fJkiVs3ryZ5cuXM3bs2DTfYzQa6d69O1OnTmXPnj08ePDAqjEJIWDbtm2MGDGChw8f4uXlxc2bN7UOSQgh0sXidW6OHTuGs7MzlSpVAtR1LVasWEGFChWYMmUKLi4u6T5XQkICR48eZdy4ceYyBwcHAgMDUzSDP+69996jYMGC9OvXjz179jz1GvHx8SlWVI6IiABAr9ej1+vTHeuzJF+ddcYMGPCGgcaNFaudXyRJ/L5Z8/snkhgMBkJCQpg1axYAlSpV4uuvv6Zs2bJS5zYgP8+ZR+o6c9iqni05n8XJzVtvvcXYsWOpVKkSFy9epHPnzrzxxht88803xMTEMG/evHSf6+7duxiNRnx9fVOU+/r6cubMmTTfs3fvXpYtW0ZYWFi6rjF9+nSmTp2aqnzbtm14eHikO9ZnuXzZgcQhTD/95ETMtavExp602vlFaqGhoVqHkO3cuXOHOXPmcPr0aQBatmxJnz59OH/+POfPn9c4uuxNfp4zj9R15rB2PcfExKT7WIuTm3PnzlG1alUAvvnmG15++WXWrl3Lvn376NKli0XJjaUiIyPp0aMHS5cuJX/+/Ol6z7hx4wgODja/joiIICAggObNm+Pt7W212I5uPsPv4VcBaN7cQL/XitKqVYDVzi+S6PV6QkNDadasGc6ymJBVffrpp5w+fRpvb28WLlxIrly5pJ5tTH6eM4/UdeawVT0n9rykR4a2X0jsgtm+fTuvvvoqAAEBAeYZVemVP39+HB0duX37dory27dv4+fnl+r4CxcucPnyZdq0aWMuS4zFycmJs2fPUqpUqRTvcXV1xdXVNdW5nJ2drVrpDg5JE89CQqBmCYurVljI2t9DAcOHD+f27du8+eabFC1alC1btkg9ZxKp58wjdZ05rF3PlpzL4nVuatasyQcffMCXX37J7t27ad26NQCXLl1K1b30LC4uLtSoUYMdO3aYy0wmEzt27KBu3bqpjn/hhRc4efIkYWFh5kfbtm1p3LgxYWFhBARIS4kQlkic0h0VFQWoSfqMGTNS/SNBCCHsicXNC/PmzaN79+788MMPTJgwgdKlSwPw7bffUq9ePYsDCA4OplevXtSsWZPatWszb948oqOjzbOnevbsSeHChZk+fTpubm6plnjPnTs3QJZa+j06GhQFZENkkZX9+OOP9O7dmwcPHuDl5cWiRYu0DkkIIazC4uSmcuXKnDyZeqDsrFmzcHR0tDiAzp07c+fOHSZPnkx4eDhVq1Zl69at5lagq1evpujysQfNmzsxcwwMG6Z1JEKklpCQwOjRo5k/fz4AtWvXZvTo0RpHJYQQ1pPhgSFHjx41z6ioUKEC1atXz3AQQ4YMYciQIWl+bdeuXU9978qVKzN8XVsqUEDrCIRILXGG45EjRwB49913mTZtmkVLOAghRFZncXLz77//0rlzZ3bv3m3uEnrw4AGNGzdm3bp1FJC7Otu2GahbVusohEhp165dtGvXjoiICPPeUIkTAoQQIjuxuL9n6NChREVF8ffff3P//n3u37/PX3/9RUREBO+8844tYrQ7np4g/xAWWU25cuVwc3Ojfv36hIWFSWIjhMi2LG652bp1K9u3b6d8+fLmsgoVKrBw4UKaN29u1eCEEM/n7t275jWh/P392b17N6VKlZJpsEKIbM3ilhuTyZTmH0ZnZ+cUWxDkZJ9/puPKFa2jEDnd119/TcmSJfn222/NZS+88IIkNkKIbM/i5KZJkyYMGzYsxSZ6N27cYMSIETRt2tSqwdmrVasdkT0GhVZiY2N588036datG5GRkaxevVrrkIQQIlNZnNwsWLCAiIgIihcvTqlSpShVqhQlSpQgIiKCTz/91BYx2p0OHYwUKqR1FCInOnPmDHXq1GHp0qXodDomTZrExo0btQ5LCCEylcVjbgICAjh27Bg7duwwTwUvX748gYGBVg/OXo0YoVCsmNZRiJxm9erVDBw4kJiYGHx9ffnqq6/k91IIkSNZlNysX7+en376iYSEBJo2bcrQoUNtFZcQwgLHjh2jV69egNp1vGbNmjT3ZxNCiJwg3cnN4sWLGTx4MGXKlMHd3Z2NGzdy4cIFZs2aZcv47JKiaB2ByGmqV6/Ou+++i4+PD+PHj8/QauFCCJFdpHvMzYIFCwgJCeHs2bOEhYWxatUq2YvmCRo0cOLqVa2jENmZoiisWrWK69evm8tmz57NpEmTJLERQuR46U5uLl68aG72BujWrRsGg4Fbt27ZJDB7J4v4CVuJjIykR48e9O7dm65du2IwGLQOSQghspR0d0vFx8fj6elpfu3g4ICLiwuxsbE2Ccyebf7ZIHtLCZs4ceIEnTp14ty5czg6OtK6dWu721hWCCFszaIBxZMmTcLDw8P8OiEhgQ8//BAfHx9z2Zw5c6wXnZ3KnQekZ0BYk6IofP755wwbNoz4+HiKFCnCunXrqF+/vtahCSFElpPu5KZRo0acPXs2RVm9evW4ePGi+bVOp7NeZEIIQO2G6t+/Pxs2bADg1VdfZeXKleTLl0/jyIQQImtKd3Kza9cuG4aRvSxfrqPGeyC5nrAGR0dHTp06hZOTEx999BHBwcHyDwkhhHgKixfxE8+2YrkDi97XOgphzxRFQVEUHBwc8PDwYMOGDTx8+JCXXnpJ69CEECLLk5GINvDa67KBqMi4Bw8e0KFDB2bMmGEuK1++vCQ2QgiRTpLc2MC778oqfiJjDh8+TLVq1di4cSPvv/8+t2/f1jokIYSwO5LcCJEFKIrC3LlzadCgAZcvX6ZkyZL8/vvv+Pr6ah2aEELYHRlzI4TG7t+/T+/evdm0aRMAHTp04IsvvkixxIIQQoj0y1DLzZ49e/jf//5H3bp1uXHjBgBffvkle/futWpw9qpVK1nkRqRPQkICL730Eps2bcLV1ZVFixaxYcMGSWyEEOI5WJzcfPfddwQFBeHu7s7x48eJj48H4OHDh0ybNs3qAdqjhw9lmq5IHxcXF4YPH06ZMmU4ePAgAwcOlGneQgjxnCxObj744AOWLFnC0qVLcXZ2NpfXr1+fY8eOWTU4e/Xjj7LXj3iyu3fvcurUKfPrgQMHEhYWRtWqVbULSgghshGLk5uzZ8/SqFGjVOU+Pj48ePDAGjHZvfz5tY5AZFV79uyhSpUqtGnThocPHwLqyt7JtzURQgjxfCxObvz8/Dh//nyq8r1791KyZEmrBCVEdmMymfjwww955ZVXuHnzJi4uLty5c0frsIQQIluyOLkZMGAAw4YN49ChQ+h0Om7evMmaNWsYOXIkAwcOtEWMdufnn7WOQGQlt2/fpkWLFkycOBGTyUSvXr04cuQIpUuX1jo0IYTIliyeCj527FhMJhNNmzYlJiaGRo0a4erqysiRIxk6dKgtYrQ7G9Y7MkWqQgC//fYb3bt3Jzw8HA8PDxYtWkSvXr20DksIIbI1i5MbnU7HhAkTGDVqFOfPnycqKooKFSrg5eVli/jsUsNGsv2CUM2dO5fw8HAqVqzIhg0bqFChgtYhCSFEtpfhRfxcXFzkD/UTDBgg2y8I1YoVK5gxYwZTp06VQcNCCJFJLE5uGjdu/NR1OH777bfnCkgIe7Zt2za2bdvG7NmzAcifPz+zZs3SOCohhMhZLE5uHl+LQ6/XExYWxl9//SVjCUSOZTAYCAkJYfr06SiKQr169XjjjTe0DksIIXIki5ObuXPnplk+ZcoUoqKinjug7GDUKAd2fqt1FCKzXL9+nW7durFnzx4A3n77bVq2bKlxVEIIkXNZbVfw//3vfyxfvtxap7Nrer3WEYjMsmXLFqpWrcqePXvIlSsX69evZ/Hixbi7u2sdmhBC5FhWS24OHDiAm5ubtU5n1yZOkNlSOcG0adNo3bo19+7do0aNGhw/fpxOnTppHZYQQuR4FndLPT6OQFEUbt26xZEjR5g0aZLVArNn+QtoHYHIDDVq1ECn0zFkyBBmzZqFq6ur1iEJIYQgA8mNj49PitcODg6UK1eO9957j+bNm1stMCGyon///ZeCBQsCEBQUxN9//0358uU1jkoIIURyFiU3RqORPn36UKlSJfLkyWOrmOzegQNQs4TWUQhrSkhIYMyYMaxcuZKjR4+a91GTxEYIIbIei8bcODo60rx5c9n9+xl+3Wq1oUwiC7h06RINGjRg3rx5PHjwgF9++UXrkIQQQjyFxXfhF198kYsXL9oilmyjUmVZoTi7+O6776hWrRp//PEHefPm5aeffmLw4MFahyWEEOIpLE5uPvjgA0aOHMnPP//MrVu3iIiISPEQ0L69JDf2Li4ujiFDhtChQwcePnxIvXr1OH78OG3atNE6NCGEEM+Q7uTmvffeIzo6mlatWnHixAnatm1LkSJFyJMnD3ny5CF37twyDkdkG5988gkLFy4EYMyYMezatYuiRYtqHJUQQoj0SPeA4qlTp/L222+zc+dOW8YjRJYwbNgwdu7cyTvvvCOrDQshhJ1Jd3KjKGpXy8svv2yzYLKLzz/TUfMjraMQloiNjWXhwoUMHz4cJycnXF1dZeCwEELYKYumgj9tN3CR5J9/pJ7syZkzZ+jUqRMnT57kwYMHfPDBB1qHJIQQ4jlYlNyULVv2mQnO/fv3nyug7GDceNl+wV58+eWXDBw4kOjoaHx9fXnllVe0DkkIIcRzsii5mTp1aqoVikVqefNqHYF4lujoaIYOHcqKFSsAaNKkCWvWrMHPz0/jyIQQQjwvi5KbLl26mJeeF8JenT59mg4dOnDq1CkcHBwICQlhwoQJODo6ah2aEEIIK0h3ciPjbdLvzBnZfiErM5lMXLp0CX9/f9auXStdUUIIkc2ke52bxNlS4tn275dEMKsxGo3m5xUrVuT7778nLCxMEhshhMiG0p3cmEwm6ZJKp+LFJRHMSk6cOEHlypXZu3evuSwoKEh+noUQIpuSHR5toEkTrSMQoLY2fvbZZ9SpU4dTp04xatQoaYEUQogcQJIbkS1FRETQtWtX3n77beLj42nVqhWbNm2SsWNCCJEDSHIjsp1jx45Ro0YN1q9fj5OTE7NmzWLTpk3kz59f69CEEEJkAoumgov02bIFag7WOoqc6a+//qJu3bokJCRQtGhR1q1bR926dbUOSwghRCaS5MYGEuKl60MrFStW5NVXX8VgMLBixQryyoqKQgiR40hyYwOvvCKDVjPTkSNHKFOmDD4+Puh0Or766ivc3NxkfI0QQuRQWWLMzcKFCylevDhubm7UqVOHw4cPP/HYpUuX0rBhQ/LkyUOePHkIDAx86vFayJ1H6whyBkVRmDt3LvXq1ePNN980z4Ryd3eXxEYIIXIwzZOb9evXExwcTEhICMeOHaNKlSoEBQXx77//pnn8rl276Nq1Kzt37uTAgQMEBATQvHlzbty4kcmRCy1FRkbSvn17goOD0ev1mEwmEhIStA5LCCFEFqB5cjNnzhwGDBhAnz59qFChAkuWLMHDw4Ply5enefyaNWsYNGgQVatW5YUXXuCLL77AZDKxY8eOTI78ycJvaR1B9nbw4EFGjBjBzz//jIuLCwsXLmTDhg24urpqHZoQQogsQNMxNwkJCRw9epRx48aZyxwcHAgMDOTAgQPpOkdMTAx6vf6JA0fj4+OJj483v46IiABAr9ej1+ufI/qUTCaT+fmhQyaCalnv3EJlMpmYM2cOkyZNwmg0UqpUKdauXUu1atUwGAxah5ftJP5+WPP3RKQm9Zx5pK4zh63q2ZLzaZrc3L17F6PRiK+vb4pyX19fzpw5k65zjBkzhkKFChEYGJjm16dPn87UqVNTlW/btg0PDw/Lg36Cy5cdSGwIi439my1bIqx2bqGKjIxk9uzZGI1GGjZsyKBBg7h16xa3bklTmS2FhoZqHUKOIPWceaSuM4e16zkmJibdx9r1bKmPPvqIdevWsWvXLtzc3NI8Zty4cQQHB5tfR0REmMfpeHt7Wy2Wo5vP8Hv4VQA6dCxPjeL5rHZukaRQoUKcPn2awoUL07x5c5ydnbUOKdvS6/WEhobSrFkzqWcbknrOPFLXmcNW9ZzY85IemiY3+fPnx9HRkdu3b6cov337Nn5+fk997+zZs/noo4/Yvn07lStXfuJxrq6uaY7FcHZ2tmqlOzgkDV9ycnKSXxwrMJlMTJ8+nWLFivG///0PgCZNmtCwYUO2bNli9e+hSJvUc+aQes48UteZw9r1bMm5NB1Q7OLiQo0aNVIMBk4cHPy0VWVnzpzJ+++/z9atW6lZs2ZmhCoy2e3bt2nRogUTJ07krbfektlwQggh0k3z2VLBwcEsXbqUVatWcfr0aQYOHEh0dDR9+vQBoGfPnikGHM+YMYNJkyaxfPlyihcvTnh4OOHh4URFRWn1EVI5m77hQuIJdu7cSdWqVQkNDcXd3Z0FCxZQqFAhrcMSQghhJzQfc9O5c2fu3LnD5MmTCQ8Pp2rVqmzdutU8yPjq1aspunwWL15MQkICHTp0SHGekJAQpkyZkpmhP5FM3MkYo9HIBx98wHvvvYfJZKJixYps2LCBChUqaB2aEEIIO6J5cgMwZMgQhgwZkubXdu3aleL15cuXbR/QcypcROsI7I/BYKBFixbmLsp+/frxySefWHVGmxBCiJxB826p7Ch3bq0jsD9OTk7UqlULT09PvvrqK7744gtJbIQQQmSIJDdCMwaDgTt37phfv/fee5w4cYLu3btrGJUQQgh7J8mNDURFah1B1nf9+nUaN25M69atzXtCOTs7U6pUKY0jE0IIYe8kubGBf+88+5icbMuWLVStWpW9e/dy5swZ/vrrL61DEkIIkY1IcmMDHu5aR5A16fV6Ro8eTevWrbl37x7Vq1fn2LFjVK9eXevQhBBCZCNZYrZUduPnr3UEWc+VK1fo0qULBw8eBGDo0KHMmjVLdvIWQghhdZLciEzRv39/Dh48iI+PD8uXL+eNN97QOiQhhBDZlHRLiUyxePFiAgMDOX78uCQ2QgghbEqSGxt4bB/QHOnSpUt88cUX5telS5cmNDSUEiVKaBiVEEKInEC6pWzAZNI6Am1999139OvXj4iICIoXL05gYKDWIQkhhMhBpOXGBvLl1ToCbcTFxTFkyBA6dOjAw4cPeemllyhTpozWYQkhhMhhJLmxAZccOAHo/Pnz1KtXj4ULFwIwevRodu/eTbFixTSOTAghRE4j3VLiuX3zzTf069ePyMhI8uXLx+rVq2nVqpXWYQkhhMihJLmxAX2C1hFkrqioKCIjI2nYsCFr166lSBHZFl0IIYR2JLmxgehorSOwPYPBgJOT+uPTu3dvvLy8eP31181lQgghhFZkzI0NZPf7+5dffknlypW5d+8eADqdjo4dO0piI4QQIkuQ5MYGvHJpHYFtREdH07dvX3r27Mnp06f55JNPtA5JCCGESEX+qS3S5e+//6ZTp06cOnUKnU5HSEgIEydO1DosIYQQIhVJbsRTKYrCypUrGTx4MLGxsfj5+bF27VoaN26sdWhCCCFEmqRbykqSr0ocE6NdHNa2aNEi+vbtS2xsLM2aNSMsLEwSGyGEEFmaJDdWoihJz7PT9gvdu3endOnSfPjhh2zduhVfX1+tQxJCCCGeSrqlrMQhWZro5qZdHM9LURS2b99OYGAgOp2O3Llzc/LkSdzs+UMJIYTIUaTlxkp0uqTn9jojOiIigm7dutG8eXOWLl1qLpfERgghhD2x09uwsLbjx4/TqVMnzp8/j5OTE7GxsVqHJITIgkwmEwkJ9rkMu16vx8nJibi4OIxGo9bhZFvPU88uLi44ODx/u4skN1aSYsyNHf3OKIrCokWLCA4OJiEhgaJFi7Ju3Trq1q2rdWhCiCwmISGBS5cuYbLTgYWKouDn58e1a9fQJW9uF1b1PPXs4OBAiRIlcHFxea4YJLmxEr0+2XODdnFY4sGDB/Tv35/vvvsOgLZt27JixQry5s2rcWRCiKxGURRu3bqFo6MjAQEBVvnXdWYzmUxERUXh5eVll/Hbi4zWs8lk4ubNm9y6dYuiRYs+VwIqyY2VJP8eONjJPwhOnjzJ999/j7OzMzNnzmTYsGHyrxkhRJoMBgMxMTEUKlQIDw8PrcPJkMQuNTc3N0lubOh56rlAgQLcvHkTg8GAs7NzhmOQ5MZKkn8PnJ+vNS3TNGzYkAULFlCzZk1q1aqldThCiCwscezE83YXCPE0iT9fRqPxuZIbSV1zkPv379OtWzfOnj1rLhs4cKAkNkKIdJPWXWFL1vr5kpabHOLAgQN06dKFq1evcv78eQ4dOiR/pIQQQmRL0nJjJclnRipZaCKByWRi1qxZNGrUiKtXr1KqVCmWLFkiiY0QQohsS5IbK8mKMyPv3r1LmzZtGD16NAaDgc6dO3Ps2DGqV6+udWhCCJFp7ty5w8CBAylevDi+vr4UKlSIoKAg9u3bZz7m+PHjdO7cGX9/f1xdXSlWrBivvvoqmzZtQnm01sfly5fR6XTmR65cuahYsSKDBw/mn3/+0erjiTRIt5SVpBj3lAUaRc6fP88rr7zCjRs3cHNzY/78+QwYMEBabIQQOU779u1JSEhgxYoVFChQgJiYGHbu3Mm9e/cA+PHHH+nUqROBgYGsWrWK0qVLEx8fz/79+5k4cSINGzYkd+7c5vNt376dihUrEhMTw8mTJ5k/fz5VqlRh06ZNNG3aVKNPKZKT5MZKHB2TnmeF/KFYsWIUK1YMLy8vNmzYQOXKlbUOSQiRDUVHq//38Ej625eQoK795eQErq6pj3V3T9qPT69Xj3d0TLkv35OOtXQCzYMHD9izZw+7du2iYcOGRERE4O3tzUsvvfToOtH069eP1q1bs3HjxhTvLV++PP369TO33CTKly8ffn5+AJQsWZI2bdrQtGlT+vXrx4ULF3BMfkMQmpBuqWzkzp075mXRnZ2d+fbbbzly5IgkNkIIm/HyUh937yaVzZqllg0ZkvLYggXV8qtXk8oWLlTL+vVLeWzx4mr56dNJZStXZiQ+L7y8vPjhhx+Ij49P9fVt27Zx7949Ro8e/cRzPKvF28HBgWHDhnHlyhWOHj1qeZDC6iS5sRKtx9zs3LmTypUrM378eHOZv78/Xl5eGkYlhBDacnJyYuXKlaxatYq8efMSFBTEhAkT+PPPPwE4d+4cAOXKlTO/548//jAnRV5eXvz888/PvM4LL7wAqONyhPYkubESrfaZNBqNTJ06lcDAQMLDw9m6dSsxMTHaBCOEyHGiotRH/vxJZaNGqWULFqQ89t9/1fKiRZPKBg9Wy5YtS3ns5ctqefnySWW9e2csxvbt23Pz5k1++OEHAgMD2b17N9WrV2flE5qCKleuTFhYGGFhYURHR2MwPHtPncSuKxnXmDVIcmPHbt26RfPmzZkyZQomk4m+ffty+PBhu10aXQhhfzw91Ufye7qLi1qWfLxN8mOTr8jv7KyWJR9v87RjM8rNzY1mzZoxatQo9u7dS+/evQkJCaFMmTIAKRY3dXV1pXTp0pQuXTrd5z/9qP+sRIkSGQ9SWI0kN1bi6Zm51wsNDaVq1ar89ttveHp6snr1apYtWyaJjRBCpEOFChWIjo6mefPm5M2blxkzZmT4XCaTiU8++YQSJUpQrVo1K0YpMkpmS9mhBw8e0LFjRx4+fEilSpXYsGGDub9XCCFEknv37tGxY0f69u3Liy++iE6n48yZM8ycOZN27drh5eXFF198QefOnWndujXvvPMOZcqUISoqiq1btwKkmv107949wsPDiYmJ4a+//mLevHkcPnyYzZs3y0ypLEKSGzuUO3dulixZws6dO5k3bx7u7u5ahySEEFmSl5cXderUYe7cuVy4cAG9Xk9AQAADBgwwT8B4/fXX2b9/PzNmzKBnz57cv38fHx8fatasybp163j11VdTnDMwMBAADw8PihUrRuPGjfn8888t6sYStiXJjZXYegzvL7/8gpubG40bNwagS5cudOnSxbYXFUIIO+fq6sr06dOZPn06JpPJvM6Ng0PKURk1a9bkm2++eeq5ihcvnmrNG5E1yZgbK9HrbXVePWPGjKFVq1Z07dqV27dv2+ZCQgghRDYhLTdWYotxvFevXqVLly4cOHAAgA4dOuDj42P9CwkhhBDZiCQ3VvI8UxTT8tNPP9G7d2/+++8/fHx8WLZsGe3bt7fuRYQQQohsSLqlshij0UhwcDDt2rXjv//+o1atWhw7dkwSGyGEECKdJLmxEuOzF7BMFwcHB/79918Ahg8fzt69eylZsqR1Ti6EEELkANItZSUPI57v/QaDAScnJ3Q6HYsXL6Z79+60bNnSOsEJIYQQOYi03FiJQwZrMj4+nqFDh9K+fXvzFMNcuXJJYiOEEEJkkLTcWEnu3Ja/5/z583Tu3Jljx44BsHfvXho2bGjdwIQQQogcRlpuNLJ+/XqqV6/OsWPHyJcvHz///LMkNkIIIYQVSHKTyWJjY3n77bfp0qULkZGRNGjQgLCwMFq3bq11aEIIIaxMp9Pxww8/aB1GjiPJjZVEpHNAcZcuXfjss8/Q6XSMHz+enTt3UqRIEdsGJ4QQOVTv3r3R6XTodDpcXV2pUqUKY8aMIS4uTuvQbCr5507+OH/+vKYxvfbaa5lyLRlzYyXx8ek7bvz48Rw9epTly5fTvHlz2wYlhBCCFi1asGLFCuLj49m7dy+DBg3CwcGBGTNmaB2aTSV+7uQKFCiQoXMlJCTg4uJijbAyhbTcWEmuXGmXx8TEsHv3bvPrOnXqcOHCBUlshBAik7i6uuLn50dAQACtW7emadOmhIaGmr9+7949unbtSuHChfHw8KBSpUp8/fXXKc7xyiuv8M477zB69Gjy5s2Ln58fU6ZMSXHMP//8Q6NGjXBzc6NChQoprpHo5MmTNGnSBHd3d/Lly8ebb75JVFSU+euJrRvTpk3D19eX3Llz895772EwGBg1ahR58+alSJEiqZKWp33u5A9HR0cAdu/eTe3atXF1dcXf35+xY8diMCQt2PbKK68wZMgQhg8fTv78+QkKCgLgr7/+omXLlnh5eeHr60uPHj24e/eu+X3ffvstVapUwd/fnwIFChAYGEh0dDRTpkxh1apV/Pjjj+ZWpF27dj3zM2SUtNxYiZtb6rJTp07RqVMnLly4wKFDh6hcuTKg/sAJIYS9a/PpXu5EprPZ2ooK5HJl09AGGXrvqVOnOHDgAMWKFTOXxcXFUaNGDcaMGYO3tzebN2+mR48elCpVitq1a5uPW7VqFcHBwRw6dIgDBw7Qu3dv6tevT7NmzTCZTLzxxhv4+vpy6NAhHj58yPDhw1NcOzo6mqCgIOrWrcsff/zBv//+S//+/RkyZAgrV640H/fbb79RpEgRfv/9d/bt20e/fv3Yv38/jRo14tChQ6xfv5633nqLZs2aZWhYw40bN2jVqhW9e/dm9erVnDlzhgEDBuDm5pYiYVu1ahUDBw5k3759ADx48IAmTZrQv39/5s6dS2xsLGPGjKFTp0789ttv3Lp1i65duzJjxgwCAwNRFIV9+/ahKAojR47k9OnTREREmBOzvHnzWhx7emWJ5GbhwoXMmjWL8PBwqlSpwqeffpriB+px33zzDZMmTeLy5cuUKVOGGTNm0KpVq0yM+OkURWHFihUMHjyY2NhY/Pz8iEjvoBwhhLATdyLjCY/I+mNXfv75Z7y8vDAYDMTHx+Pg4MCCBQvMXy9cuDAjR440vx46dCi//vorGzZsSHEvqly5MiEhIQCUKVOGBQsWsGPHDpo1a8b27ds5c+YMv/76K4UKFQJg2rRpKdYsW7t2LXFxcaxevRpPT08AFixYQJs2bZgxYwa+vr6AetP/5JNPcHBwoFy5csycOZOYmBjGjx8PwLhx4/joo4/Yu3cvXbp0eebnTtSyZUu++eYbFi1aREBAAAsWLECn0/HCCy9w8+ZNxowZw+TJk3F4tHBbmTJlmDlzpvn9H3zwAdWqVWPatGnmsuXLlxMQEMC5c+eIiorCYDDw+uuvkydPHry9valSpYr5WHd3d+Lj4/Hz83vm9+x5aZ7crF+/nuDgYJYsWUKdOnWYN28eQUFBnD17loIFC6Y6fv/+/XTt2pXp06fz6quvsnbtWl577TWOHTvGiy++qMEnUCVuv2BKiGXKyCFs+X4DAM2aNePLL780/9AKIUR2USCXNq3Qll63cePGLF68mMjISGbNmoW7u3uK/fqMRiPTpk1jw4YN3Lhxg4SEBOLj4/Hw8EhxnsTW90T+/v7m7XJOnz5NQECAObEBqFu3borjT58+TZUqVcyJDUD9+vUxmUycPXvWfJ+oWLGiOcEA8PX1TXF/c3R0JF++fOZrP+tzJ0q87unTp6lbty46nS5FHFFRUVy/fp2iRYsCUKNGjRTnO3HiBDt37kyRMCVKHG7RtGlTqlSpQpMmTWjZsiWdOnUiT548T43TFjRPbubMmcOAAQPo06cPAEuWLGHz5s0sX76csWPHpjp+/vz5tGjRglGjRgHw/vvvExoayoIFC1iyZEmmxp7c3XuQ8O8l7vw4g2v3r+Pg4MB7773HuHHjUvyQCiFEdpHRrqHM5unpSenSpTGZTCxYsICXX36ZZcuW0a9fPwBmzZrF/PnzmTdvHpUqVcLT05Phw4eTkJCQ4jzOzs4pXut0Okwmk9XjTes6Gbl24ufOqORJGEBUVJS5lelx/v7+ODo6Ehoayt69e/n5559ZuHAhkyZN4tChQ5QoUSLDcWSEpslNQkICR48eZdy4ceYyBwcHAgMDOXDgQJrvOXDgAMHBwSnKgoKCnriOQHx8PPHJpjIldg/p9Xr0ev1zfoLkTMT8cxDD/evkL+jH+q/X0LBhQ4xGI0aj0YrXEYnfN+t+/8TjpJ4zh73Us16vR1EUTCaTTW7otqIoijluRVFwcHBg7NixjBw5ki5duuDu7s7evXtp27Yt3bp1A8BkMnHu3DnKly+f4rMmnietc5crV45r165x48YN/P39AbWnIfF8icesXLmSyMhIc+KwZ88eHBwcKFOmjDnGx6+T1rWfVJZWbI974YUX2LhxI0aj0dx6s3fvXnLlykWhQoXM73n8/dWqVWPjxo0ULVoUJ6fU6UPisfXq1aNSpUq8//77lCxZko0bNzJixAicnZ0xGAxP/flJrAO9Xm8e/JzIkt8RTZObu3fvYjQaU3XZ+Pr6cubMmTTfEx4enubx4eHhaR4/ffp0pk6dmqp827ZtqZocn0d0tAM+dTuB0cCI/7UiMjKSLVu2WO38IrW0ZiII65N6zhxZvZ6dnJzw8/MjKioqVYtGVqbX6zEYDCnGPQYFBTF69GjmzJnD0KFDKVasGD/++COhoaHkzp2bRYsWER4eTpkyZczvMxgMJCQkpDiPwWBAr9cTERFB7dq1KV26ND169GDq1KlERkYyYcIEQF28NSIigjZt2jBlyhT+97//MWbMGO7du8c777xD586dcXd3JyIiIs1407q2yWQiLi7uieM50zpPov/973/Mnz+ft99+mwEDBnD+/HlCQkIYNGiQeeZWWtfs0aMHS5cupVOnTrzzzjvkyZOHixcvsnHjRj755BOOHz/O7t27adKkCfnz5+fo0aPcuXOHokWLEhERgZ+fH1u3buXo0aPkzZsXb2/vVC1SCQkJxMbG8vvvv6eYvQXq7OP00rxbytbGjRuXoqUnIiKCgIAAmjdvjre3t9WuUzI8kg53IjlRHvq0fZn83tZLnERKer2e0NBQmjVrluoXQ1iP1HPmsJd6jouL49q1a3h5eeGW1vTQLMrZ2RknJye8vb1RFIXIyEjy5MnDkCFDmDt3LsOHD2fq1Klcv36dDh064OHhwYABA3jttdd4+PCh+T7h5OSEi4tLivuGk5MTzs7O5rLvv/+eAQMGEBgYSPHixZk3bx6tWrXC3d0db29vvL292bp1KyNGjKBp06Z4eHjwxhtv8PHHH5vHsSSPN/l1Hr+2g4MDbm5uT7yPpXWeRN7e3vz888+MGTOGhg0bkjdvXvr168d7771nbpFJ65re3t7s3buXsWPH0r59e+Lj4ylWrBhBQUHkzp0bf39/Dh8+zGeffUZERATFihVj9uzZ5vFNQ4YM4eDBgzRp0oSoqCh27NjBK6+8kiK2uLg43N3dzVPqk7NkYo5OSdyKWgMJCQl4eHjw7bffpli1sFevXjx48IAff/wx1XuKFi1KcHBwiil2ISEh/PDDD5w4ceKZ14yIiMDHxyfFD6216PV6tmzZQqtWrbL0Hyl7J/WcOaSeM4e91HNcXByXLl2iRIkSdpXcJGcymYiIiMDb21vGQtrQ89Tz037OLLl/a/rddXFxoUaNGuzYscNcZjKZ2LFjR6pR5onq1q2b4nhQm3OfdLwQQgghchbNu6WCg4Pp1asXNWvWpHbt2sybN4/o6Gjz7KmePXtSuHBhpk+fDsCwYcN4+eWX+fjjj2ndujXr1q3jyJEjfP7551p+DCGEEEJkEZonN507d+bOnTtMnjyZ8PBwqlatytatW82Dhq9evZqiWatevXqsXbuWiRMnMn78eMqUKcMPP/yg6Ro3QgghhMg6NE9uQB1kNGTIkDS/ltbeEx07dqRjx442jkoIIYQQ9khGVAkhhEg3DeegiBzAWj9fktwIIYR4psQF1expjRthfxJ/vh5fwM9SWaJbSgghRNbm5OSEh4cHd+7cwdnZ2S6nUptMJhISEoiLi7PL+O1FRuvZZDJx584dPDw80lwB2RKS3AghhHgmnU6Hv78/ly5d4sqVK1qHkyGKohAbG4u7u3uKTSOFdT1PPTs4OFC0aNHn/v5IciOEECJdXFxcKFOmjN12Ten1en7//XcaNWqUpRdMtHfPU88uLi5WaVWT5EYIIUS6JS77b48cHR0xGAy4ublJcmNDWaGepdNRCCGEENmKJDdCCCGEyFYkuRFCCCFEtpLjxtwkLhBkydbp6aXX64mJiSEiIkL6c21I6jlzSD1nDqnnzCN1nTlsVc+J9+30LPSX45KbyMhIAAICAjSORAghhBCWioyMxMfH56nH6JQctpa2yWTi5s2b5MqVy+rrHERERBAQEMC1a9fw9va26rlFEqnnzCH1nDmknjOP1HXmsFU9K4pCZGQkhQoVeuZ08RzXcuPg4ECRIkVseg1vb2/5xckEUs+ZQ+o5c0g9Zx6p68xhi3p+VotNIhlQLIQQQohsRZIbIYQQQmQrktxYkaurKyEhIbi6umodSrYm9Zw5pJ4zh9Rz5pG6zhxZoZ5z3IBiIYQQQmRv0nIjhBBCiGxFkhshhBBCZCuS3AghhBAiW5HkRgghhBDZiiQ3Flq4cCHFixfHzc2NOnXqcPjw4ace/8033/DCCy/g5uZGpUqV2LJlSyZFat8sqeelS5fSsGFD8uTJQ548eQgMDHzm90WoLP15TrRu3Tp0Oh2vvfaabQPMJiyt5wcPHjB48GD8/f1xdXWlbNmy8rcjHSyt53nz5lGuXDnc3d0JCAhgxIgRxMXFZVK09un333+nTZs2FCpUCJ1Oxw8//PDM9+zatYvq1avj6upK6dKlWblypc3jRBHptm7dOsXFxUVZvny58vfffysDBgxQcufOrdy+fTvN4/ft26c4OjoqM2fOVE6dOqVMnDhRcXZ2Vk6ePJnJkdsXS+u5W7duysKFC5Xjx48rp0+fVnr37q34+Pgo169fz+TI7Yul9Zzo0qVLSuHChZWGDRsq7dq1y5xg7Zil9RwfH6/UrFlTadWqlbJ3717l0qVLyq5du5SwsLBMjty+WFrPa9asUVxdXZU1a9Yoly5dUn799VfF399fGTFiRCZHbl+2bNmiTJgwQdm4caMCKN9///1Tj7948aLi4eGhBAcHK6dOnVI+/fRTxdHRUdm6datN45TkxgK1a9dWBg8ebH5tNBqVQoUKKdOnT0/z+E6dOimtW7dOUVanTh3lrbfesmmc9s7Sen6cwWBQcuXKpaxatcpWIWYLGalng8Gg1KtXT/niiy+UXr16SXKTDpbW8+LFi5WSJUsqCQkJmRVitmBpPQ8ePFhp0qRJirLg4GClfv36No0zO0lPcjN69GilYsWKKco6d+6sBAUF2TAyRZFuqXRKSEjg6NGjBAYGmsscHBwIDAzkwIEDab7nwIEDKY4HCAoKeuLxImP1/LiYmBj0ej158+a1VZh2L6P1/N5771GwYEH69euXGWHavYzU808//UTdunUZPHgwvr6+vPjii0ybNg2j0ZhZYdudjNRzvXr1OHr0qLnr6uLFi2zZsoVWrVplSsw5hVb3wRy3cWZG3b17F6PRiK+vb4pyX19fzpw5k+Z7wsPD0zw+PDzcZnHau4zU8+PGjBlDoUKFUv1CiSQZqee9e/eybNkywsLCMiHC7CEj9Xzx4kV+++03unfvzpYtWzh//jyDBg1Cr9cTEhKSGWHbnYzUc7du3bh79y4NGjRAURQMBgNvv/0248ePz4yQc4wn3QcjIiKIjY3F3d3dJteVlhuRrXz00UesW7eO77//Hjc3N63DyTYiIyPp0aMHS5cuJX/+/FqHk62ZTCYKFizI559/To0aNejcuTMTJkxgyZIlWoeWrezatYtp06axaNEijh07xsaNG9m8eTPvv/++1qEJK5CWm3TKnz8/jo6O3L59O0X57du38fPzS/M9fn5+Fh0vMlbPiWbPns1HH33E9u3bqVy5si3DtHuW1vOFCxe4fPkybdq0MZeZTCYAnJycOHv2LKVKlbJt0HYoIz/P/v7+ODs74+joaC4rX7484eHhJCQk4OLiYtOY7VFG6nnSpEn06NGD/v37A1CpUiWio6N58803mTBhAg4O8m9/a3jSfdDb29tmrTYgLTfp5uLiQo0aNdixY4e5zGQysWPHDurWrZvme+rWrZvieIDQ0NAnHi8yVs8AM2fO5P3332fr1q3UrFkzM0K1a5bW8wsvvMDJkycJCwszP9q2bUvjxo0JCwsjICAgM8O3Gxn5ea5fvz7nz583J48A586dw9/fXxKbJ8hIPcfExKRKYBITSkW2XLQaze6DNh2unM2sW7dOcXV1VVauXKmcOnVKefPNN5XcuXMr4eHhiqIoSo8ePZSxY8eaj9+3b5/i5OSkzJ49Wzl9+rQSEhIiU8HTwdJ6/uijjxQXFxfl22+/VW7dumV+REZGavUR7IKl9fw4mS2VPpbW89WrV5VcuXIpQ4YMUc6ePav8/PPPSsGCBZUPPvhAq49gFyyt55CQECVXrlzK119/rVy8eFHZtm2bUqpUKaVTp05afQS7EBkZqRw/flw5fvy4Aihz5sxRjh8/rly5ckVRFEUZO3as0qNHD/PxiVPBR40apZw+fVpZuHChTAXPij799FOlaNGiiouLi1K7dm3l4MGD5q+9/PLLSq9evVIcv2HDBqVs2bKKi4uLUrFiRWXz5s2ZHLF9sqSeixUrpgCpHiEhIZkfuJ2x9Oc5OUlu0s/Set6/f79Sp04dxdXVVSlZsqTy4YcfKgaDIZOjtj+W1LNer1emTJmilCpVSnFzc1MCAgKUQYMGKf/991/mB25Hdu7cmebf28S67dWrl/Lyyy+nek/VqlUVFxcXpWTJksqKFStsHqdOUaT9TQghhBDZh4y5EUIIIUS2IsmNEEIIIbIVSW6EEEIIka1IciOEEEKIbEWSGyGEEEJkK5LcCCGEECJbkeRGCCGEENmKJDdCCCGEyFYkuRFCpLBy5Upy586tdRgZptPp+OGHH556TO/evXnttdcyJR4hROaT5EaIbKh3797odLpUj/Pnz2sdGitXrjTH4+DgQJEiRejTpw///vuvVc5/69YtWrZsCcDly5fR6XSEhYWlOGb+/PmsXLnSKtd7kilTppg/p6OjIwEBAbz55pvcv3/fovNIIiaE5Zy0DkAIYRstWrRgxYoVKcoKFCigUTQpeXt7c/bsWUwmEydOnKBPnz7cvHmTX3/99bnP7efn98xjfHx8nvs66VGxYkW2b9+O0Wjk9OnT9O3bl4cPH7J+/fpMub4QOZW03AiRTbm6uuLn55fi4ejoyJw5c6hUqRKenp4EBAQwaNAgoqKinnieEydO0LhxY3LlyoW3tzc1atTgyJEj5q/v3buXhg0b4u7uTkBAAO+88w7R0dFPjU2n0+Hn50ehQoVo2bIl77zzDtu3byc2NhaTycR7771HkSJFcHV1pWrVqmzdutX83oSEBIYMGYK/vz9ubm4UK1aM6dOnpzh3YrdUiRIlAKhWrRo6nY5XXnkFSNka8vnnn1OoUCFMJlOKGNu1a0ffvn3Nr3/88UeqV6+Om5sbJUuWZOrUqRgMhqd+TicnJ/z8/ChcuDCBgYF07NiR0NBQ89eNRiP9+vWjRIkSuLu7U65cOebPn2/++pQpU1i1ahU//vijuRVo165dAFy7do1OnTqRO3du8ubNS7t27bh8+fJT4xEip5DkRogcxsHBgU8++YS///6bVatW8dtvvzF69OgnHt+9e3eKFCnCH3/8wdGjRxk7dizOzs4AXLhwgRYtWtC+fXv+/PNP1q9fz969exkyZIhFMbm7u2MymTAYDMyfP5+PP/6Y2bNn8+effxIUFETbtm35559/APjkk0/46aef2LBhA2fPnmXNmjUUL148zfMePnwYgO3bt3Pr1i02btyY6piOHTty7949du7caS67f/8+W7dupXv37gDs2bOHnj17MmzYME6dOsVnn33GypUr+fDDD9P9GS9fvsyvv/6Ki4uLucxkMlGkSBG++eYbTp06xeTJkxk/fjwbNmwAYOTIkXTq1IkWLVpw69Ytbt26Rb169dDr9QQFBZErVy727NnDvn378PLyokWLFiQkJKQ7JiGyLZvvOy6EyHS9evVSHB0dFU9PT/OjQ4cOaR77zTffKPny5TO/XrFiheLj42N+nStXLmXlypVpvrdfv37Km2++maJsz549ioODgxIbG5vmex4//7lz55SyZcsqNWvWVBRFUQoVKqR8+OGHKd5Tq1YtZdCgQYqiKMrQoUOVJk2aKCaTKc3zA8r333+vKIqiXLp0SQGU48ePpzimV69eSrt27cyv27Vrp/Tt29f8+rPPPlMKFSqkGI1GRVEUpWnTpsq0adNSnOPLL79U/P3904xBURQlJCREcXBwUDw9PRU3NzcFUABlzpw5T3yPoijK4MGDlfbt2z8x1sRrlytXLkUdxMfHK+7u7sqvv/761PMLkRPImBshsqnGjRuzePFi82tPT09AbcWYPn06Z86cISIiAoPBQFxcHDExMXh4eKQ6T3BwMP379+fLL780d62UKlUKULus/vzzT9asWWM+XlEUTCYTly5donz58mnG9vDhQ7y8vDCZTMTFxdGgQQO++OILIiIiuHnzJvXr109xfP369Tlx4gSgdik1a9aMcuXK0aJFC1599VWaN2/+XHXVvXt3BgwYwKJFi3B1dWXNmjV06dIFBwcH8+fct29fipYao9H41HoDKFeuHD/99BNxcXF89dVXhIWFMXTo0BTHLFy4kOXLl3P16lViY2NJSEigatWqT433xIkTnD9/nly5cqUoj4uL48KFCxmoASGyF0luhMimPD09KV26dIqyy5cv8+qrrzJw4EA+/PBD8ubNy969e+nXrx8JCQlp3qSnTJlCt27d2Lx5M7/88gshISGsW7eO119/naioKN566y3eeeedVO8rWrToE2PLlSsXx44dw8HBAX9/f9zd3QGIiIh45ueqXr06ly5d4pdffmH79u106tSJwMBAvv3222e+90natGmDoihs3ryZWrVqsWfPHubOnWv+elRUFFOnTuWNN95I9V43N7cnntfFxcX8Pfjoo49o3bo1U6dO5f333wdg3bp1jBw5ko8//pi6deuSK1cuZs2axaFDh54ab1RUFDVq1EiRVCbKKoPGhdCSJDdC5CBHjx7FZDLx8ccfm1slEsd3PE3ZsmUpW7YsI0aMoGvXrqxYsYLXX3+d6tWrc+rUqVRJ1LM4ODik+R5vb28KFSrEvn37ePnll83l+/bto3bt2imO69y5M507d6ZDhw60aNGC+/fvkzdv3hTnSxzfYjQanxqPm5sbb7zxBmvWrOH8+fOUK1eO6tWrm79evXp1zp49a/HnfNzEiRNp0qQJAwcONH/OevXqMWjQIPMxj7e8uLi4pIq/evXqrF+/noIFC+Lt7f1cMQmRHcmAYiFykNKlS6PX6/n000+5ePEiX375JUuWLHni8bGxsQwZMoRdu3Zx5coV9u3bxx9//GHubhozZgz79+9nyJAhhIWF8c8///Djjz9aPKA4uVGjRjFjxgzWr1/P2bNnGTt2LGFhYQwbNgyAOXPm8PXXX3PmzBnOnTvHN998g5+fX5oLDxYsWBB3d3e2bt3K7du3efjw4ROv2717dzZv3szy5cvNA4kTTZ48mdWrVzN16lT+/vtvTp8+zbp165g4caJFn61u3bpUrlyZadOmAVCmTBmOHDnCr7/+yrlz55g0aRJ//PFHivcUL16cP//8k7Nnz3L37l30ej3du3cnf/78tGvXjj179nDp0iV27drFO++8w/Xr1y2KSYhsSetBP0II60trEGqiOXPmKP7+/oq7u7sSFBSkrF69WgGU//77T1GUlAN+4+PjlS5duigBAQGKi4uLUqhQIWXIkCEpBgsfPnxYadasmeLl5aV4enoqlStXTjUgOLnHBxQ/zmg0KlOmTFEKFy6sODs7K1WqVFF++eUX89c///xzpWrVqoqnp6fi7e2tNG3aVDl27Jj56yQbUKwoirJ06VIlICBAcXBwUF5++eUn1o/RaFT8/f0VQLlw4UKquLZu3arUq1dPcXd3V7y9vZXatWsrn3/++RM/R0hIiFKlSpVU5V9//bXi6uqqXL16VYmLi1N69+6t+Pj4KLlz51YGDhyojB07NsX7/v33X3P9AsrOnTsVRVGUW7duKT179lTy58+vuLq6KiVLllQGDBigPHz48IkxCZFT6BRFUbRNr4QQQgghrEe6pYQQQgiRrUhyI4QQQohsRZIbIYQQQmQrktwIIYQQIluR5EYIIYQQ2YokN0IIIYTIViS5EUIIIUS2IsmNEEIIIbIVSW6EEEIIka1IciOEEEKIbEWSGyGEEEJkK/8HaO5Q1WUJbSMAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotar a curva ROC, ajuda se representar a primeira curva ROC\n",
    "plt.plot(fpr, tpr, \"b:\", label='SGD')\n",
    "plot_roc_curve(fpr_forest,tpr_forest,'Random Forest')\n",
    "plt.legend(loc ='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "20f7fda4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9983436731328145"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# verificar o score\n",
    "roc_auc_score(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "5202fb93",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SVC was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# CLASSIFICAÇÃO MULTICLASSE \n",
    "\n",
    "# Estratégia One-versus-the-rest (OvR):no caso dos numeros de 0 a 9, seria treinar 10 classificadores, um pra cada alg\n",
    "# Quando quiser classificar a imagem, basta obter o score de decisão pra cada imagem e seleciona a classe que tem a saída\n",
    "# de maior score\n",
    "\n",
    "#Estratégia one-versus-one (OvO): treinar o classificador binário pra cada par de algs. \n",
    "from sklearn.svm import SVC\n",
    "\n",
    "svm_clf = SVC()\n",
    "svm_clf.fit(X_train, y_train) # y_train e não o y_train_5( para as multiclasses)\n",
    "svm_clf.predict([some_digit])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "a14d7158",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SVC was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 1.72501977,  2.72809088,  7.2510018 ,  8.3076379 , -0.31087254,\n",
       "         9.3132482 ,  1.70975103,  2.76765202,  6.23049537,  4.84771048]])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_digit_scores = svm_clf.decision_function([some_digit])\n",
    "some_digit_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "dbb6f398",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argmax(some_digit_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "754cde40",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_clf.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "2916fb20",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_clf.classes_[5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "eba86fc1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SGDClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([3], dtype=uint8)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# treinar um SGD\n",
    "sgd_clf.fit(X_train, y_train)\n",
    "sgd_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c22c0ce7",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but SGDClassifier was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[-31893.03095419, -34419.69069632,  -9530.63950739,\n",
       "          1823.73154031, -22320.14822878,  -1385.80478895,\n",
       "        -26188.91070951, -16147.51323997,  -4604.35491274,\n",
       "        -12050.767298  ]])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.decision_function([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "55dd847e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.87365, 0.85835, 0.8689 ])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Não precisou rodar o OvR ou Ovo, pq o classificador SGD pode classificar direto as instancias em varias classes.\n",
    "# Os scores estão bastante confiante, o menor valor em módulo é do alg 5, há uma dúvida no alg 3 que possui um valor positivo\n",
    "# Usar a validação cruzada\n",
    "\n",
    "cross_val_score(sgd_clf, X_train, y_train, cv=3, scoring='accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "eb024642",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Para melhorar essa acurácia vamos aplicar o escalonamento das entradas\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler =  StandardScaler()\n",
    "X_train_scaled = scaler.fit_transform(X_train.astype(np.float64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "4287e6d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.8983, 0.891 , 0.9018])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(sgd_clf, X_train_scaled, y_train, cv=3,scoring='accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "a16d6c87",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Análise de Erro\n",
    "# Encontrei o modelo ideal e quero melhora-lo -> Tipos de erros que o modelo comete\n",
    "# ANALISE DA MATRIZ DE CONFUSÃO \n",
    "y_train_pred = cross_val_predict(sgd_clf, X_train_scaled, y_train, cv=3)\n",
    "conf_mx = confusion_matrix(y_train, y_train_pred)\n",
    "conf_mx # Análise ruim, usa o matplotlib\n",
    "plt.matshow(conf_mx, cmap = plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "0cf699fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Esta matriz de confusao é boa, pq parte as imagens estão na diagonal principal\n",
    "# significa que foram classificadas corretamente\n",
    "# 5s são um pouco mais escuro, pode indicar que possui menos imagens 5s no conjunto de dados ou que n funciona tão bem com os 5s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "bf202804",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dito isso, agr precisamos \"normalizar\" esse dados da matriz de confusão\n",
    "# vamos dividir pelo numero de imagens da classe correspondente de modo a ter uma taxa e não um valor absoluto(pois faria com \n",
    "# que classes numerosas parecessem ruins)\n",
    "row_sums = conf_mx.sum(axis=1, keepdims=True)\n",
    "norm_conf_mx = conf_mx /row_sums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "d598eeb8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAGkCAYAAAAIduO+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAWiUlEQVR4nO3de2zVhd3H8W+ptiCWBnCghKLMLEEuXpBLlMRdJBqjbmZGZ4YJwWRZtiIgiRlsUWccVpbNkIhDMZsjGXhJFuYl08WwKHPKQBAjcYqLiRYdIAm2iq5iz3n+WB4e+xzQHuTL7xx8vZITw+Ecfp/8in3n15ZzGsrlcjkAIMmAogcAcGwTGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUtVtaO6+++447bTTYuDAgTF9+vTYuHFj0ZNqSkdHR0ydOjVaWlpixIgRccUVV8Rrr71W9Kyad8cdd0RDQ0MsWLCg6Ck16e23345rr702hg8fHoMGDYpJkybFCy+8UPSsmtLb2xs33XRTjB07NgYNGhSnn3563HbbbfFlfrWvugzNQw89FAsXLoxbbrkltmzZEmeddVZcfPHFsXv37qKn1Yxnnnkm2tvbY8OGDfHUU0/F/v3746KLLop9+/YVPa1mbdq0Ke69994488wzi55Sk/bu3RszZsyI448/Pp544ol45ZVX4te//nUMHTq06Gk1ZenSpbFixYpYvnx5/POf/4ylS5fGL3/5y7jrrruKnlaYhnp8Uc3p06fH1KlTY/ny5RERUSqVoq2tLa6//vpYtGhRwetq07vvvhsjRoyIZ555Ji644IKi59ScDz74ICZPnhy/+c1v4he/+EWcffbZsWzZsqJn1ZRFixbF3//+9/jb3/5W9JSadtlll8XIkSPjt7/97YH7rrzyyhg0aFD84Q9/KHBZceruiubjjz+OzZs3x8yZMw/cN2DAgJg5c2Y8//zzBS6rbV1dXRERMWzYsIKX1Kb29va49NJL+/y9oq9HH300pkyZEldddVWMGDEizjnnnLjvvvuKnlVzzj///Fi3bl1s3749IiJeeumlePbZZ+OSSy4peFlxjit6QLX27NkTvb29MXLkyD73jxw5Ml599dWCVtW2UqkUCxYsiBkzZsTEiROLnlNzHnzwwdiyZUts2rSp6Ck17Y033ogVK1bEwoUL46c//Wls2rQp5s2bF01NTTF79uyi59WMRYsWRXd3d4wbNy4aGxujt7c3lixZErNmzSp6WmHqLjRUr729PbZt2xbPPvts0VNqTmdnZ8yfPz+eeuqpGDhwYNFzalqpVIopU6bE7bffHhER55xzTmzbti3uueceofmUhx9+OFavXh1r1qyJCRMmxNatW2PBggUxatSoL+15qrvQnHTSSdHY2Bi7du3qc/+uXbvi5JNPLmhV7Zo7d248/vjjsX79+hg9enTRc2rO5s2bY/fu3TF58uQD9/X29sb69etj+fLl0dPTE42NjQUurB2nnHJKjB8/vs99Z5xxRvzxj38saFFtuvHGG2PRokVxzTXXRETEpEmT4s0334yOjo4vbWjq7ns0TU1Nce6558a6desO3FcqlWLdunVx3nnnFbistpTL5Zg7d26sXbs2/vrXv8bYsWOLnlSTLrzwwnj55Zdj69atB25TpkyJWbNmxdatW0XmU2bMmFHxI/Lbt2+PU089taBFtenDDz+MAQP6fmptbGyMUqlU0KLi1d0VTUTEwoULY/bs2TFlypSYNm1aLFu2LPbt2xdz5swpelrNaG9vjzVr1sQjjzwSLS0tsXPnzoiIaG1tjUGDBhW8rna0tLRUfN9q8ODBMXz4cN/P+n9uuOGGOP/88+P222+Pq6++OjZu3BgrV66MlStXFj2tplx++eWxZMmSGDNmTEyYMCFefPHFuPPOO+O6664relpxynXqrrvuKo8ZM6bc1NRUnjZtWnnDhg1FT6opEXHQ2/3331/0tJr39a9/vTx//vyiZ9Skxx57rDxx4sRyc3Nzedy4ceWVK1cWPanmdHd3l+fPn18eM2ZMeeDAgeWvfvWr5Z/97Gflnp6eoqcVpi7/HQ0A9aPuvkcDQH0RGgBSCQ0AqYQGgFRCA0AqoQEgVd2GpqenJ37+859HT09P0VNqnnPVP85T/zhP/edc/Vfd/jua7u7uaG1tja6urhgyZEjRc2qac9U/zlP/OE/951z9V91e0QBQH4QGgFRH/UU1S6VSvPPOO9HS0hINDQ2H/ed0d3f3+S+H5lz1j/PUP85T/x3r56pcLsf7778fo0aNqnjF6k876t+j2bFjR7S1tR3NQwKQqLOz8zPf7+qoX9G0tLQc7UPWrVp9x8da/AmaL3J1nKW1tbXoCRV6e3uLnlDh/fffL3rCQX3/+98vekKF1atXFz3hoD7v8/pRD00tfkKoVbV6rmpxl039Y1P/NTU1FT2hbnzex9APAwCQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUh1WaO6+++447bTTYuDAgTF9+vTYuHHjkd4FwDGi6tA89NBDsXDhwrjllltiy5YtcdZZZ8XFF18cu3fvztgHQJ2rOjR33nln/OAHP4g5c+bE+PHj45577okTTjghfve732XsA6DOVRWajz/+ODZv3hwzZ878vz9gwICYOXNmPP/88wd9Tk9PT3R3d/e5AfDlUVVo9uzZE729vTFy5Mg+948cOTJ27tx50Od0dHREa2vrgZu3cQb4ckn/qbPFixdHV1fXgVtnZ2f2IQGoIVW9lfNJJ50UjY2NsWvXrj7379q1K04++eSDPqe5uTmam5sPfyEAda2qK5qmpqY499xzY926dQfuK5VKsW7dujjvvPOO+DgA6l9VVzQREQsXLozZs2fHlClTYtq0abFs2bLYt29fzJkzJ2MfAHWu6tB873vfi3fffTduvvnm2LlzZ5x99tnx5JNPVvyAAABEHEZoIiLmzp0bc+fOPdJbADgGea0zAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFSH9VpnHB3HH3980RPqxn/+85+iJ1QolUpFT6hQi2+l3tDQUPSEg3rvvfeKnlChsbGx6Al9lMvlfv09d0UDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEh1XFEHPuGEE6KhoaGow9eF7u7uoicc1Lhx44qeUOGjjz4qekKFPXv2FD2hwujRo4ueUKFW/55/97vfLXpChccee6zoCYfFFQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIVVVoOjo6YurUqdHS0hIjRoyIK664Il577bWsbQAcA6oKzTPPPBPt7e2xYcOGeOqpp2L//v1x0UUXxb59+7L2AVDnqnrjsyeffLLPr3//+9/HiBEjYvPmzXHBBRcc0WEAHBu+0DtsdnV1RUTEsGHDDvmYnp6e6OnpOfDrWn03PQByHPYPA5RKpViwYEHMmDEjJk6ceMjHdXR0RGtr64FbW1vb4R4SgDp02KFpb2+Pbdu2xYMPPviZj1u8eHF0dXUduHV2dh7uIQGoQ4f1pbO5c+fG448/HuvXr4/Ro0d/5mObm5ujubn5sMYBUP+qCk25XI7rr78+1q5dG08//XSMHTs2axcAx4iqQtPe3h5r1qyJRx55JFpaWmLnzp0REdHa2hqDBg1KGQhAfavqezQrVqyIrq6u+MY3vhGnnHLKgdtDDz2UtQ+AOlf1l84AoBpe6wyAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkOqw3vjsSCiVStHQ0FDU4SuUSqWiJ1Q4++yzi55wUFu3bi16Ql246qqrip5Q4c9//nPREyocf/zxRU84qFp8w8ZRo0YVPaGPUqkUO3bs+NzHuaIBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQ6rqgDf/LJJ9HQ0FDU4SvU0pb/9dZbbxU94aAaGxuLnlCht7e36AkV/vSnPxU9ocIJJ5xQ9IQK+/fvL3rCQb333ntFT6gwfvz4oif08cknn8SOHTs+93GuaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0CqLxSaO+64IxoaGmLBggVHaA4Ax5rDDs2mTZvi3nvvjTPPPPNI7gHgGHNYofnggw9i1qxZcd9998XQoUOP9CYAjiGHFZr29va49NJLY+bMmZ/72J6enuju7u5zA+DLo+q3cn7wwQdjy5YtsWnTpn49vqOjI2699daqhwFwbKjqiqazszPmz58fq1evjoEDB/brOYsXL46urq4Dt87OzsMaCkB9quqKZvPmzbF79+6YPHnygft6e3tj/fr1sXz58ujp6YnGxsY+z2lubo7m5uYjsxaAulNVaC688MJ4+eWX+9w3Z86cGDduXPzkJz+piAwAVBWalpaWmDhxYp/7Bg8eHMOHD6+4HwAivDIAAMmq/qmz/+/pp58+AjMAOFa5ogEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABI1VAul8tH84Dd3d3R2toagwcPjoaGhqN56M+0f//+oidUqNX38fnWt75V9IQK//jHP4qeUOHf//530RMqnH766UVPqFCrb4bY09NT9IQKbW1tRU/oo1QqxTvvvBNdXV0xZMiQQz7OFQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBINVxRR24oaEhGhoaijp8hXK5XPSECscdV9iH5zM98sgjRU+oMGzYsKInVBg/fnzREyq88sorRU+oUEufBz7tvPPOK3pChR07dhQ94bC4ogEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCpqg7N22+/Hddee20MHz48Bg0aFJMmTYoXXnghYxsAx4Cq3vBk7969MWPGjPjmN78ZTzzxRHzlK1+J119/PYYOHZq1D4A6V1Voli5dGm1tbXH//fcfuG/s2LFHfBQAx46qvnT26KOPxpQpU+Kqq66KESNGxDnnnBP33XffZz6np6cnuru7+9wA+PKoKjRvvPFGrFixIr72ta/FX/7yl/jRj34U8+bNi1WrVh3yOR0dHdHa2nrg1tbW9oVHA1A/Gsrlcrm/D25qaoopU6bEc889d+C+efPmxaZNm+L5558/6HN6enqip6fnwK+7u7ujra0tTjzxxJp6r/BPb6wVJ554YtETDmrv3r1FT6gwbNiwoidUGDlyZNETKrzyyitFT6hQS58HPm369OlFT6iwYcOGoiccVFdXVwwZMuSQv1/VFc0pp5wS48eP73PfGWecEW+99dYhn9Pc3BxDhgzpcwPgy6Oq0MyYMSNee+21Pvdt3749Tj311CM6CoBjR1WhueGGG2LDhg1x++23x7/+9a9Ys2ZNrFy5Mtrb27P2AVDnqgrN1KlTY+3atfHAAw/ExIkT47bbbotly5bFrFmzsvYBUOeq+nc0ERGXXXZZXHbZZRlbADgGea0zAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRVv9bZkbJ///6aesOjUqlU9IQKjY2NRU84qMGDBxc9oUIt/V36X9u3by96QoVaPE9VvPfiUXXyyScXPaHC1VdfXfSEPvbv3x9r16793Me5ogEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFI1lMvl8tE8YHd3d7S2tkZTU1M0NDQczUN/ppNOOqnoCRU+/vjjoicc1NChQ4ueUOH1118vekKFadOmFT2hwrZt24qeUKGnp6foCQf1ySefFD2hwvDhw4ue0EepVIq9e/dGV1dXDBky5JCPc0UDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkqio0vb29cdNNN8XYsWNj0KBBcfrpp8dtt90WR/kFoAGoI8dV8+ClS5fGihUrYtWqVTFhwoR44YUXYs6cOdHa2hrz5s3L2ghAHasqNM8991x85zvfiUsvvTQiIk477bR44IEHYuPGjSnjAKh/VX3p7Pzzz49169bF9u3bIyLipZdeimeffTYuueSSQz6np6cnuru7+9wA+PKo6opm0aJF0d3dHePGjYvGxsbo7e2NJUuWxKxZsw75nI6Ojrj11lu/8FAA6lNVVzQPP/xwrF69OtasWRNbtmyJVatWxa9+9atYtWrVIZ+zePHi6OrqOnDr7Oz8wqMBqB9VXdHceOONsWjRorjmmmsiImLSpEnx5ptvRkdHR8yePfugz2lubo7m5uYvvhSAulTVFc2HH34YAwb0fUpjY2OUSqUjOgqAY0dVVzSXX355LFmyJMaMGRMTJkyIF198Me6888647rrrsvYBUOeqCs1dd90VN910U/z4xz+O3bt3x6hRo+KHP/xh3HzzzVn7AKhzVYWmpaUlli1bFsuWLUuaA8CxxmudAZBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSVfWimkdSY2NjNDQ0FHX4Cnv27Cl6QoWmpqaiJxzUvn37ip5QYfLkyUVPqPDqq68WPaFCLX7svv3tbxc94aCuvPLKoidUONQbTNY6VzQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEglNACkEhoAUgkNAKmEBoBUQgNAKqEBIJXQAJBKaABIJTQApBIaAFIJDQCphAaAVEIDQCqhASCV0ACQSmgASCU0AKQSGgBSCQ0AqY472gcsl8t9/lsram1PRG1uiogolUpFT6jQ29tb9IQKtfrxqzX79+8vesJBffTRR0VPqBuf93e9oXyU/2/YsWNHtLW1Hc1DApCos7MzRo8efcjfP+qhKZVK8c4770RLS0s0NDQc9p/T3d0dbW1t0dnZGUOGDDmCC489zlX/OE/94zz137F+rsrlcrz//vsxatSoGDDg0N+JOepfOhswYMBnlq9aQ4YMOSY/gBmcq/5xnvrHeeq/Y/lctba2fu5j/DAAAKmEBoBUdRua5ubmuOWWW6K5ubnoKTXPueof56l/nKf+c67+66j/MAAAXy51e0UDQH0QGgBSCQ0AqYQGgFRCA0AqoQEgldAAkEpoAEj1P1wmZ0z0V3v5AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# preencher a diagonal com zeros para manter somente os erros e plotar o resultado\n",
    "np.fill_diagonal(norm_conf_mx, 0)\n",
    "plt.matshow(norm_conf_mx, cmap= plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "aa6f1b47",
   "metadata": {},
   "outputs": [],
   "source": [
    "# As linhas representam classes reais, colunas representam classes previstas\n",
    "# Mais clareza indica que tem uma classificação errada."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "15616b40",
   "metadata": {},
   "outputs": [],
   "source": [
    "# melhorando o modelo\n",
    "\n",
    "cl_a, cl_b = 3,5\n",
    "X_aa = X_train[(y_train == cl_a) & (y_train_pred ==cl_a)]\n",
    "X_ab = X_train[(y_train == cl_a) & (y_train_pred ==cl_b)]\n",
    "X_ba = X_train[(y_train == cl_b) & (y_train_pred ==cl_a)]\n",
    "X_bb = X_train[(y_train == cl_b) & (y_train_pred ==cl_b)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "631eec2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_digits(instances, images_per_row=10, **options):\n",
    "    size = 28\n",
    "    images_per_row = min(len(instances), images_per_row)\n",
    "    images = [np.array(instances.iloc[i]).reshape(size, size) for i in range(instances.shape[0])] #change done here\n",
    "    \n",
    "    if images_per_row == 0:\n",
    "       images_per_row = 0.1\n",
    "    \n",
    "    n_rows = (len(instances) - 1) // images_per_row + 1\n",
    "    row_images = []\n",
    "    n_empty = n_rows * images_per_row - len(instances)\n",
    "    images.append(np.zeros((size, size * n_empty)))\n",
    "    for row in range(n_rows):\n",
    "        rimages = images[row * images_per_row : (row + 1) * images_per_row]\n",
    "        row_images.append(np.concatenate(rimages, axis=1))\n",
    "    image = np.concatenate(row_images, axis=0)\n",
    "    plt.imshow(image, cmap = plt.cm.binary, **options)\n",
    "    plt.axis(\"off\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "dae5b314",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_digit(data):\n",
    "    image = data.reshape(28, 28)\n",
    "    plt.imshow(image, cmap = mpl.cm.binary,\n",
    "               interpolation=\"nearest\")\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "ece6ca61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "plt.subplot(221);plot_digits(X_aa[:25], images_per_row=5)\n",
    "plt.subplot(222);plot_digits(X_ab[:25], images_per_row=5)\n",
    "plt.subplot(223);plot_digits(X_ba[:25], images_per_row=5)\n",
    "plt.subplot(224);plot_digits(X_bb[:25], images_per_row=5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "4cc301dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>KNeighborsClassifier()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">KNeighborsClassifier</label><div class=\"sk-toggleable__content\"><pre>KNeighborsClassifier()</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Multirótulo\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "y_train_large = (y_train>=7)\n",
    "y_train_odd = (y_train % 2 == 1)\n",
    "y_multilabel = np.c_[y_train_large, y_train_odd]\n",
    "\n",
    "knn_clf = KNeighborsClassifier()\n",
    "knn_clf.fit(X_train, y_multilabel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "3e7e8050",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\base.py:464: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n",
      "  warnings.warn(\n",
      "c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\joblib\\externals\\loky\\backend\\context.py:136: UserWarning: Could not find the number of physical cores for the following reason:\n",
      "found 0 physical cores < 1\n",
      "Returning the number of logical cores instead. You can silence this warning by setting LOKY_MAX_CPU_COUNT to the number of cores you want to use.\n",
      "  warnings.warn(\n",
      "  File \"c:\\Users\\hudso\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\joblib\\externals\\loky\\backend\\context.py\", line 282, in _count_physical_cores\n",
      "    raise ValueError(f\"found {cpu_count_physical} physical cores < 1\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[False,  True]])"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "f6759dc8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uma abordagem para o classificador multirótulo, calculo do F1 score para rotulo individual \n",
    "# Depois calcular a média \n",
    "# y_train_knn_pred = cross_val_predict(knn_clf, X_train, y_multilabel, cv=3)\n",
    "# f1_score(y_multilabel, y_train_knn_pred, average=\"macro\")\n",
    "# error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "c534495c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# np.zeros(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "6fd5cc69",
   "metadata": {},
   "outputs": [],
   "source": [
    "noise = np.random.randint(0,100,(len(X_train),784))\n",
    "X_train_mod = X_train + noise\n",
    "noise = np.random.randint(0,100,(len(X_test),784))\n",
    "X_test_mod = X_test + noise\n",
    "y_train_mod = X_train\n",
    "y_test_mod = X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "2fc732bd",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# some_index = 0\n",
    "# plt.subplot(121); plot_digit(X_test_mod[some_index])\n",
    "# plt.subplot(122); plot_digit(y_test_mod[some_index])\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a81b6a5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "4f01c197",
   "metadata": {},
   "outputs": [],
   "source": [
    "# treinar o classificador que limpa a imagem\n",
    "# knn_clf.fit(X_train_mod, y_train_mod)\n",
    "# clean_digit = knn_clf.predict([X_test_mod[some_index]])\n",
    "# plot_digit(clean_digit)\n",
    "# save_fig(\"cleaned_digit_example_plot\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "770316b5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.3.0\n"
     ]
    }
   ],
   "source": [
    "import sklearn\n",
    "print(sklearn.__version__)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0cc90cda",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5b4ba80",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "483e818c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5f65b7a7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "60693434",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9d83e059",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "917b17c9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "036bac4e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "afea9086",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "242e30dd",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b74e9f1c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "de94b146",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b589173",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3043ebaa",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "987d1640",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4ef03236",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}